WEBVTT 00:00:00.000 --> 00:00:02.560 align:middle line:84% The following content is provided under a Creative 00:00:02.560 --> 00:00:03.920 align:middle line:90% Commons license. 00:00:03.920 --> 00:00:06.360 align:middle line:84% Your support will help MIT OpenCourseWare 00:00:06.360 --> 00:00:10.000 align:middle line:84% continue to offer high-quality educational resources for free. 00:00:10.000 --> 00:00:13.280 align:middle line:84% To make a donation or view additional materials 00:00:13.280 --> 00:00:17.240 align:middle line:84% from hundreds of MIT courses, visit MIT OpenCourseWare 00:00:17.240 --> 00:00:20.760 align:middle line:90% at ocw.mit.edu. 00:00:20.760 --> 00:00:23.120 align:middle line:84% IRVING WLADAWSKY-BERGER: So, we've 00:00:23.120 --> 00:00:25.680 align:middle line:84% been talking for the last three weeks, 00:00:25.680 --> 00:00:29.120 align:middle line:84% about formulating a market strategy 00:00:29.120 --> 00:00:33.400 align:middle line:84% around a complex, disruptive technology formulating. 00:00:33.400 --> 00:00:36.620 align:middle line:84% And now we'll start on the next major subject, 00:00:36.620 --> 00:00:38.440 align:middle line:90% which is executing. 00:00:38.440 --> 00:00:41.940 align:middle line:84% But before that, let me just briefly go over it. 00:00:41.940 --> 00:00:46.040 align:middle line:84% And remember, there are different kinds 00:00:46.040 --> 00:00:50.880 align:middle line:84% of reasons you do a strategy, some more incremental. 00:00:50.880 --> 00:00:54.780 align:middle line:84% We are after disruptive innovation in this class. 00:00:54.780 --> 00:00:57.520 align:middle line:90% That's what we are after here. 00:00:57.520 --> 00:01:01.060 align:middle line:84% Not because it's the best, but it's a big world out there, 00:01:01.060 --> 00:01:04.300 align:middle line:84% and we're focusing on disruptive innovation. 00:01:04.300 --> 00:01:07.020 align:middle line:84% And we are using, as our overriding case 00:01:07.020 --> 00:01:08.400 align:middle line:90% study, the internet. 00:01:08.400 --> 00:01:09.740 align:middle line:90% We all know that. 00:01:09.740 --> 00:01:14.340 align:middle line:84% And then we looked at IBM in particular, 00:01:14.340 --> 00:01:17.620 align:middle line:84% and the decisions we went through 00:01:17.620 --> 00:01:21.020 align:middle line:90% in embracing the internet. 00:01:21.020 --> 00:01:28.460 align:middle line:84% And in the seminar too, we talked about the key market 00:01:28.460 --> 00:01:30.580 align:middle line:90% factors, which were this. 00:01:30.580 --> 00:01:33.020 align:middle line:90% Was the technology ready? 00:01:33.020 --> 00:01:34.200 align:middle line:90% Yes, it was ready. 00:01:34.200 --> 00:01:36.020 align:middle line:90% How did you know it was ready? 00:01:36.020 --> 00:01:39.240 align:middle line:84% Because the marketplace was embracing it, dummy. 00:01:39.240 --> 00:01:45.340 align:middle line:84% That's often the most important reason why something is ready. 00:01:45.340 --> 00:01:47.460 align:middle line:84% And even if you say, I have no idea 00:01:47.460 --> 00:01:49.760 align:middle line:84% what those people are doing out there, well, 00:01:49.760 --> 00:01:52.180 align:middle line:84% then you better get them find out 00:01:52.180 --> 00:01:54.700 align:middle line:84% because they are doing something. 00:01:54.700 --> 00:01:57.590 align:middle line:84% And now, you may decide it's not for you, 00:01:57.590 --> 00:02:00.490 align:middle line:84% but at least you better analyze it. 00:02:00.490 --> 00:02:03.170 align:middle line:84% And that's essentially what we did, 00:02:03.170 --> 00:02:08.090 align:middle line:84% is understand what's going on out there. 00:02:08.090 --> 00:02:09.949 align:middle line:90% What are the marketplaces doing? 00:02:09.949 --> 00:02:11.750 align:middle line:90% What are competitors doing? 00:02:11.750 --> 00:02:14.050 align:middle line:90% What are our clients doing? 00:02:14.050 --> 00:02:17.330 align:middle line:84% And once we looked at that, it was obvious 00:02:17.330 --> 00:02:22.350 align:middle line:84% that we had no choice but to go into it. 00:02:22.350 --> 00:02:25.090 align:middle line:84% And of course, once you have no choice, 00:02:25.090 --> 00:02:30.250 align:middle line:84% then you get to the head of the parade and embrace it. 00:02:30.250 --> 00:02:35.690 align:middle line:84% And embrace it, not cynically, but you really embrace it. 00:02:35.690 --> 00:02:39.050 align:middle line:84% And then the second major part are 00:02:39.050 --> 00:02:42.010 align:middle line:84% what I would call the organizational factors, which 00:02:42.010 --> 00:02:48.810 align:middle line:84% is now you have to pick a way of framing your strategy that 00:02:48.810 --> 00:02:50.850 align:middle line:90% fits you. 00:02:50.850 --> 00:02:54.130 align:middle line:84% And with disruptive innovations, there 00:02:54.130 --> 00:02:57.130 align:middle line:90% isn't just one way to do it. 00:02:57.130 --> 00:03:05.750 align:middle line:84% And each company has to fit what best fits their competencies, 00:03:05.750 --> 00:03:10.790 align:middle line:84% their products and install base, their organization and culture, 00:03:10.790 --> 00:03:15.310 align:middle line:84% and what they have brand permission to go after. 00:03:15.310 --> 00:03:19.550 align:middle line:84% And sometimes you cannot find a total fit, 00:03:19.550 --> 00:03:24.110 align:middle line:84% but the more of the things you can fit, that is, 00:03:24.110 --> 00:03:27.270 align:middle line:84% where you want to go with where you are, 00:03:27.270 --> 00:03:31.730 align:middle line:84% the easier it's going to be to move in those directions. 00:03:31.730 --> 00:03:38.990 align:middle line:84% And as we said, with IBM, I think the breakthrough-- 00:03:38.990 --> 00:03:42.310 align:middle line:84% and it was a breakthrough we learned 00:03:42.310 --> 00:03:45.470 align:middle line:84% by watching what our customers were doing, 00:03:45.470 --> 00:03:48.310 align:middle line:90% like UPS, FedEx, and others. 00:03:48.310 --> 00:03:51.310 align:middle line:90% The breakthrough was to say-- 00:03:51.310 --> 00:03:54.700 align:middle line:84% OK, there are a lot of people out there 00:03:54.700 --> 00:04:00.180 align:middle line:84% that are saying this web stuff is brand new. 00:04:00.180 --> 00:04:02.980 align:middle line:90% It's a whole new infrastructure. 00:04:02.980 --> 00:04:06.900 align:middle line:84% It has nothing to do with existing IT. 00:04:06.900 --> 00:04:11.300 align:middle line:84% In, fact, all the existing businesses are going to die 00:04:11.300 --> 00:04:14.740 align:middle line:84% because the new economy will take over, 00:04:14.740 --> 00:04:18.180 align:middle line:84% and all the born-to-the-web businesses will replace all 00:04:18.180 --> 00:04:19.720 align:middle line:90% the existing businesses. 00:04:19.720 --> 00:04:23.500 align:middle line:84% Trust me, people were saying that because that 00:04:23.500 --> 00:04:25.060 align:middle line:84% was a way of getting a lot of money 00:04:25.060 --> 00:04:28.860 align:middle line:84% from VCs, which most of them squandered. 00:04:28.860 --> 00:04:32.820 align:middle line:84% I don't know how, but they squandered lots of money. 00:04:32.820 --> 00:04:37.020 align:middle line:84% And what we did, and I thought this 00:04:37.020 --> 00:04:41.500 align:middle line:84% was very smart is, as you can imagine, 00:04:41.500 --> 00:04:45.060 align:middle line:84% our clients didn't like to hear that message. 00:04:45.060 --> 00:04:47.840 align:middle line:84% Do you mean everything I've built? 00:04:47.840 --> 00:04:52.120 align:middle line:84% I'm a retailer, and I have physical stores. 00:04:52.120 --> 00:04:57.120 align:middle line:84% I was in Paris at a meeting in '97, and I was talking 00:04:57.120 --> 00:04:59.840 align:middle line:90% to the CIO-- 00:04:59.840 --> 00:05:06.080 align:middle line:84% it was a conference of a major Spanish retailer, El Corte 00:05:06.080 --> 00:05:06.700 align:middle line:90% Inglés. 00:05:06.700 --> 00:05:12.420 align:middle line:90% It's a huge retailer in Spain. 00:05:12.420 --> 00:05:15.360 align:middle line:84% And they said, we just remodeled all these stores. 00:05:15.360 --> 00:05:18.260 align:middle line:84% And all these people are saying, it's over. 00:05:18.260 --> 00:05:21.000 align:middle line:90% Everybody will buy from home. 00:05:21.000 --> 00:05:24.600 align:middle line:90% I said, well, I don't think so. 00:05:24.600 --> 00:05:30.920 align:middle line:84% And we said, look, everything you have is a wonderful base 00:05:30.920 --> 00:05:32.600 align:middle line:90% to build on. 00:05:32.600 --> 00:05:38.320 align:middle line:84% What is critical is that you embrace the new technologies 00:05:38.320 --> 00:05:43.400 align:middle line:84% because the web now gives everything 00:05:43.400 --> 00:05:47.200 align:middle line:84% you have in your IT world, universal connectivity 00:05:47.200 --> 00:05:48.440 align:middle line:90% and reach. 00:05:48.440 --> 00:05:54.110 align:middle line:84% And so all this stuff you have is far more valuable if you 00:05:54.110 --> 00:06:00.230 align:middle line:84% integrate it with the web, the internet, browsers, and so on. 00:06:00.230 --> 00:06:02.590 align:middle line:84% And that's what we call the business. 00:06:02.590 --> 00:06:06.390 align:middle line:84% And that was a strategy that resonated 00:06:06.390 --> 00:06:08.290 align:middle line:90% at least with our clients. 00:06:08.290 --> 00:06:11.810 align:middle line:84% It wasn't a good strategy for everybody. 00:06:11.810 --> 00:06:14.870 align:middle line:84% If you were a brand new business, 00:06:14.870 --> 00:06:16.450 align:middle line:90% it made no sense for you. 00:06:16.450 --> 00:06:19.830 align:middle line:84% But for a company like IBM, that made a lot of sense. 00:06:19.830 --> 00:06:25.670 align:middle line:84% We did a business, which turned out to be pretty successful. 00:06:25.670 --> 00:06:29.190 align:middle line:84% And then let me spend a minute on this slide, which I went 00:06:29.190 --> 00:06:32.450 align:middle line:90% through very quickly last time. 00:06:32.450 --> 00:06:36.790 align:middle line:90% 00:06:36.790 --> 00:06:41.510 align:middle line:84% I think that one of the hardest things 00:06:41.510 --> 00:06:45.630 align:middle line:84% to do with formulating disruptive, 00:06:45.630 --> 00:06:51.490 align:middle line:84% innovation strategies is this notion of striking a balance. 00:06:51.490 --> 00:07:00.570 align:middle line:84% Because it is as bad to ignore the innovation as it 00:07:00.570 --> 00:07:06.570 align:middle line:84% is to fall head over heels in love with the innovation 00:07:06.570 --> 00:07:08.590 align:middle line:90% and ignore your base. 00:07:08.590 --> 00:07:12.070 align:middle line:90% 00:07:12.070 --> 00:07:13.290 align:middle line:90% It's equally bad. 00:07:13.290 --> 00:07:16.490 align:middle line:84% If you ignore your innovation, you'll 00:07:16.490 --> 00:07:20.210 align:middle line:84% die because over time, you won't be able to get any growth, 00:07:20.210 --> 00:07:23.170 align:middle line:84% and everybody else will move into the future. 00:07:23.170 --> 00:07:26.330 align:middle line:84% If you ignore your base, you'll die 00:07:26.330 --> 00:07:29.810 align:middle line:90% because you'll run out of cash. 00:07:29.810 --> 00:07:34.650 align:middle line:84% You may be having a ball in all this innovation for a while, 00:07:34.650 --> 00:07:40.490 align:middle line:84% but eventually, you'll notice that new businesses don't 00:07:40.490 --> 00:07:42.770 align:middle line:90% make money for a few years. 00:07:42.770 --> 00:07:46.520 align:middle line:84% They take money for a few-- you have to invest in them, 00:07:46.520 --> 00:07:50.200 align:middle line:84% whether it's an entrepreneurial business or a new business 00:07:50.200 --> 00:07:52.240 align:middle line:90% within a company. 00:07:52.240 --> 00:07:57.820 align:middle line:84% And where does the money come from to feed the new business? 00:07:57.820 --> 00:08:01.240 align:middle line:84% Well, it comes from your legacy businesses, which 00:08:01.240 --> 00:08:04.360 align:middle line:90% is where you generate cash. 00:08:04.360 --> 00:08:07.340 align:middle line:84% They have high profit margins because they are more mature. 00:08:07.340 --> 00:08:09.400 align:middle line:84% You don't need to invest as much. 00:08:09.400 --> 00:08:11.880 align:middle line:84% They are not growth businesses, but they 00:08:11.880 --> 00:08:14.480 align:middle line:90% are cash-generating businesses. 00:08:14.480 --> 00:08:20.440 align:middle line:84% And then you can take that cash and invest it in your new stuff, 00:08:20.440 --> 00:08:25.500 align:middle line:84% let alone, leverage your skills, your talents, 00:08:25.500 --> 00:08:27.160 align:middle line:90% start moving them over. 00:08:27.160 --> 00:08:30.360 align:middle line:84% So this notion of striking a balance 00:08:30.360 --> 00:08:36.860 align:middle line:84% is incredibly important and not easy to do. 00:08:36.860 --> 00:08:41.240 align:middle line:90% It's easy to fall in either way. 00:08:41.240 --> 00:08:44.140 align:middle line:84% And there is no recipe for balance. 00:08:44.140 --> 00:08:46.600 align:middle line:84% Each case will be a little bit different, 00:08:46.600 --> 00:08:49.580 align:middle line:90% but you really need a balance. 00:08:49.580 --> 00:08:51.960 align:middle line:84% Before we go on to the next section, 00:08:51.960 --> 00:08:53.940 align:middle line:90% any comments or questions? 00:08:53.940 --> 00:08:55.480 align:middle line:90% Please. 00:08:55.480 --> 00:09:00.700 align:middle line:90% AUDIENCE: [INAUDIBLE] with. 00:09:00.700 --> 00:09:03.180 align:middle line:84% You have to handle your customer relationships very, very 00:09:03.180 --> 00:09:03.800 align:middle line:90% delicately. 00:09:03.800 --> 00:09:05.300 align:middle line:84% IRVING WLADAWSKY-BERGER: Absolutely, 00:09:05.300 --> 00:09:08.820 align:middle line:84% you have to handle your customer relations very delicately 00:09:08.820 --> 00:09:11.100 align:middle line:90% because-- 00:09:11.100 --> 00:09:16.660 align:middle line:84% in fact, the ideal is to help your customers strike 00:09:16.660 --> 00:09:20.460 align:middle line:84% a similar balance because you are trying 00:09:20.460 --> 00:09:22.900 align:middle line:84% to convince your customers to stay 00:09:22.900 --> 00:09:28.820 align:middle line:84% with you for the install base and stuff like that. 00:09:28.820 --> 00:09:31.540 align:middle line:84% But you also have to make the promise that you'll 00:09:31.540 --> 00:09:34.260 align:middle line:90% take them into the future. 00:09:34.260 --> 00:09:38.580 align:middle line:90% And that's the ideal. 00:09:38.580 --> 00:09:40.580 align:middle line:84% And in fact, if you have a really good relation 00:09:40.580 --> 00:09:43.810 align:middle line:84% with your customer, one of the things 00:09:43.810 --> 00:09:45.890 align:middle line:84% that we love about the relationship 00:09:45.890 --> 00:09:49.490 align:middle line:84% is that you are making a promise to keep bringing them 00:09:49.490 --> 00:09:52.810 align:middle line:84% new technologies for the future, while helping them 00:09:52.810 --> 00:09:57.410 align:middle line:84% continue with their present businesses. 00:09:57.410 --> 00:10:03.090 align:middle line:84% But all that takes tremendous nurturing, which is why-- 00:10:03.090 --> 00:10:06.170 align:middle line:84% don't forget the premise, I've said all along, 00:10:06.170 --> 00:10:13.490 align:middle line:84% is that you cannot be a leader of the business, 00:10:13.490 --> 00:10:17.530 align:middle line:84% no matter how great an engineer, and not have marketplace 00:10:17.530 --> 00:10:18.550 align:middle line:90% contact. 00:10:18.550 --> 00:10:22.690 align:middle line:84% You need to do great jobs in engineering things. 00:10:22.690 --> 00:10:27.170 align:middle line:84% But how do you learn this balance? 00:10:27.170 --> 00:10:30.130 align:middle line:84% The biggest learning is in the marketplace 00:10:30.130 --> 00:10:32.290 align:middle line:84% because by watching your customers 00:10:32.290 --> 00:10:35.210 align:middle line:84% and what works for them, chances are 00:10:35.210 --> 00:10:37.970 align:middle line:84% that's what you should be doing inside the business 00:10:37.970 --> 00:10:42.113 align:middle line:84% because in the end, you want to serve them. 00:10:42.113 --> 00:10:43.530 align:middle line:84% AUDIENCE: In this particular case, 00:10:43.530 --> 00:10:45.230 align:middle line:84% we're talking about a type of technology 00:10:45.230 --> 00:10:50.355 align:middle line:84% that was very complementary to VIT [INAUDIBLE] underlying. 00:10:50.355 --> 00:10:52.730 align:middle line:84% IRVING WLADAWSKY-BERGER: Remember, we framed it that way. 00:10:52.730 --> 00:10:54.370 align:middle line:90% It didn't have to be. 00:10:54.370 --> 00:10:57.470 align:middle line:84% AUDIENCE: So what if it's not complementary 00:10:57.470 --> 00:11:02.072 align:middle line:84% when you're replacing in a way you-- then 00:11:02.072 --> 00:11:03.530 align:middle line:84% I don't know how you would do that. 00:11:03.530 --> 00:11:06.945 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, can we take an example? 00:11:06.945 --> 00:11:08.570 align:middle line:84% AUDIENCE: I was trying to think of one, 00:11:08.570 --> 00:11:11.930 align:middle line:84% and I was thinking, OK, if he had just taken a DVD player. 00:11:11.930 --> 00:11:13.227 align:middle line:90% That's not that good of one. 00:11:13.227 --> 00:11:15.310 align:middle line:84% IRVING WLADAWSKY-BERGER: No, no, no, that is good. 00:11:15.310 --> 00:11:19.630 align:middle line:84% AUDIENCE: [INAUDIBLE] like that one, I mean. 00:11:19.630 --> 00:11:23.270 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, token ring, remember, 00:11:23.270 --> 00:11:27.550 align:middle line:84% one strategy is we embraced ethernet. 00:11:27.550 --> 00:11:29.890 align:middle line:90% So sometimes we saw the light. 00:11:29.890 --> 00:11:31.290 align:middle line:90% We embraced the internet. 00:11:31.290 --> 00:11:37.690 align:middle line:84% We said that was good or screwed up or whatever. 00:11:37.690 --> 00:11:40.820 align:middle line:84% Now, if we were only in the token-ring business, 00:11:40.820 --> 00:11:42.360 align:middle line:90% that would have been a problem. 00:11:42.360 --> 00:11:44.960 align:middle line:84% But no, no, but that's a very important thing. 00:11:44.960 --> 00:11:51.620 align:middle line:84% But for example, right now, let's stay with DVDs. 00:11:51.620 --> 00:11:53.740 align:middle line:84% Right now, as you know, there is this huge battle 00:11:53.740 --> 00:11:59.340 align:middle line:90% between Blu-ray and HD DVD. 00:11:59.340 --> 00:12:02.320 align:middle line:84% Now, my expectation is at the end of the day, 00:12:02.320 --> 00:12:04.740 align:middle line:90% the customers could care less. 00:12:04.740 --> 00:12:07.900 align:middle line:90% They just want one to win. 00:12:07.900 --> 00:12:13.260 align:middle line:84% And whoever loses should quickly embrace the other and move on. 00:12:13.260 --> 00:12:15.820 align:middle line:84% They will lose some revenues because I think you get 00:12:15.820 --> 00:12:17.940 align:middle line:90% a license something or another. 00:12:17.940 --> 00:12:23.640 align:middle line:90% But you just have to move on. 00:12:23.640 --> 00:12:26.140 align:middle line:84% AUDIENCE: So they continue to bail out until the market says 00:12:26.140 --> 00:12:26.960 align:middle line:90% who's the winner? 00:12:26.960 --> 00:12:31.400 align:middle line:84% Or should one of them just begin to embrace technology? 00:12:31.400 --> 00:12:33.900 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, and that's a very interesting 00:12:33.900 --> 00:12:37.520 align:middle line:90% mating dance, which is-- 00:12:37.520 --> 00:12:42.280 align:middle line:84% I hope Sony and Toshiba, which are the champions of each, 00:12:42.280 --> 00:12:44.920 align:middle line:90% have plan B's. 00:12:44.920 --> 00:12:47.540 align:middle line:84% In other words, at the same time they say, 00:12:47.540 --> 00:12:50.160 align:middle line:84% I hate it, in the back room, they 00:12:50.160 --> 00:12:54.720 align:middle line:84% have the people that have the Blu-ray strategy for Toshiba. 00:12:54.720 --> 00:12:59.640 align:middle line:84% I know that while Microsoft hates Linux, 00:12:59.640 --> 00:13:02.560 align:middle line:84% there are people in Microsoft that could care less, 00:13:02.560 --> 00:13:05.320 align:middle line:90% that Linux is Linux. 00:13:05.320 --> 00:13:09.160 align:middle line:84% And so they would have strategies so that should Linux 00:13:09.160 --> 00:13:12.840 align:middle line:84% gain more and more acceptance, how about Office on Linux. 00:13:12.840 --> 00:13:16.120 align:middle line:90% So you do that. 00:13:16.120 --> 00:13:22.520 align:middle line:84% And Lou Gerstner, on these cases, 00:13:22.520 --> 00:13:26.400 align:middle line:84% he always talked about the notion 00:13:26.400 --> 00:13:31.800 align:middle line:84% of sophisticated competition because 00:13:31.800 --> 00:13:35.830 align:middle line:84% at the extreme, sophisticated competition, 00:13:35.830 --> 00:13:39.630 align:middle line:84% an example is, so you're playing a tennis match, 00:13:39.630 --> 00:13:44.350 align:middle line:84% and you want to kill the person at the other side of the net, 00:13:44.350 --> 00:13:47.590 align:middle line:84% but after the match over, you go have a beer. 00:13:47.590 --> 00:13:55.710 align:middle line:84% And in most businesses, let's say, IBM and Oracle, we 00:13:55.710 --> 00:14:00.990 align:middle line:84% compete like crazy in certain areas, relational databases. 00:14:00.990 --> 00:14:07.430 align:middle line:84% But Oracle and DB2 are the top leaders in relational databases. 00:14:07.430 --> 00:14:11.090 align:middle line:84% But we also cooperate a lot in other areas. 00:14:11.090 --> 00:14:15.150 align:middle line:84% And Oracle, especially as they've made 00:14:15.150 --> 00:14:19.950 align:middle line:84% more and more acquisitions PeopleSoft and JD Edwards, boy, 00:14:19.950 --> 00:14:22.310 align:middle line:90% it's critical to cooperate. 00:14:22.310 --> 00:14:25.430 align:middle line:84% And we had lots of discussions, that's 00:14:25.430 --> 00:14:27.730 align:middle line:84% why Lou said that in senior management meetings, 00:14:27.730 --> 00:14:29.890 align:middle line:84% well, Lou, are they competitors or are they not? 00:14:29.890 --> 00:14:32.310 align:middle line:90% The answer is, well, it depends. 00:14:32.310 --> 00:14:34.010 align:middle line:84% Are they on the other side of the net, 00:14:34.010 --> 00:14:35.950 align:middle line:84% and you're playing a game against them, 00:14:35.950 --> 00:14:38.710 align:middle line:84% that is, you're competing for business with a client? 00:14:38.710 --> 00:14:41.630 align:middle line:84% Then you want to take the ball and shove it down their throat 00:14:41.630 --> 00:14:42.730 align:middle line:90% if you can. 00:14:42.730 --> 00:14:47.130 align:middle line:84% But are you working with standard bodies 00:14:47.130 --> 00:14:48.870 align:middle line:90% on areas of mutual things? 00:14:48.870 --> 00:14:50.650 align:middle line:84% Well, then you hang out together. 00:14:50.650 --> 00:14:57.390 align:middle line:84% And it's really important to be able to do this. 00:14:57.390 --> 00:15:00.510 align:middle line:90% And it takes emotional maturity. 00:15:00.510 --> 00:15:04.930 align:middle line:84% I say emotional maturity because it's really 00:15:04.930 --> 00:15:09.190 align:middle line:84% professional maturity because you have to on the one hand, 00:15:09.190 --> 00:15:10.510 align:middle line:90% really want to kill them. 00:15:10.510 --> 00:15:13.410 align:middle line:84% I mean, if you're competing for $100 million 00:15:13.410 --> 00:15:18.570 align:middle line:84% relational database deal and they win, it's really painful. 00:15:18.570 --> 00:15:20.930 align:middle line:84% But if you say, the bastards, I'm not 00:15:20.930 --> 00:15:22.910 align:middle line:84% going to do business with them again, 00:15:22.910 --> 00:15:25.650 align:middle line:90% now you're cutting your nose. 00:15:25.650 --> 00:15:30.970 align:middle line:84% So, that's why there is really no good answer. 00:15:30.970 --> 00:15:32.890 align:middle line:84% That was not exactly your question, 00:15:32.890 --> 00:15:41.210 align:middle line:84% but framing a strategy to your advantage, in my mind, 00:15:41.210 --> 00:15:44.630 align:middle line:90% is a huge part of innovation. 00:15:44.630 --> 00:15:50.430 align:middle line:84% And again, most of you are either 00:15:50.430 --> 00:15:57.190 align:middle line:84% engineers or honorary engineers for the purpose of this course. 00:15:57.190 --> 00:16:04.230 align:middle line:84% As we have said, it takes as much hard thinking 00:16:04.230 --> 00:16:06.670 align:middle line:84% to know how to do that as it does 00:16:06.670 --> 00:16:11.350 align:middle line:84% to solve a big signal-processing problem with fast Fourier 00:16:11.350 --> 00:16:12.130 align:middle line:90% transforms. 00:16:12.130 --> 00:16:15.390 align:middle line:84% You really need to put your brain 00:16:15.390 --> 00:16:19.830 align:middle line:84% and think through as if you were thinking about the toughest 00:16:19.830 --> 00:16:21.130 align:middle line:90% technical problem. 00:16:21.130 --> 00:16:23.830 align:middle line:84% And if you do that, and engineers 00:16:23.830 --> 00:16:28.230 align:middle line:84% are really good at solving complex problems, 00:16:28.230 --> 00:16:30.460 align:middle line:90% you usually will do very well. 00:16:30.460 --> 00:16:34.180 align:middle line:90% So yes, and then we go here. 00:16:34.180 --> 00:16:37.193 align:middle line:84% AUDIENCE: So it sounds like if I'm a disruptive guy, 00:16:37.193 --> 00:16:37.860 align:middle line:90% it sounds like-- 00:16:37.860 --> 00:16:39.443 align:middle line:84% IRVING WLADAWSKY-BERGER: If you are a? 00:16:39.443 --> 00:16:41.100 align:middle line:84% AUDIENCE: Disruptive technology, it 00:16:41.100 --> 00:16:43.860 align:middle line:84% sounds like my best philosophy should 00:16:43.860 --> 00:16:48.100 align:middle line:84% be to either fly under the radar until I get big enough 00:16:48.100 --> 00:16:52.420 align:middle line:84% or partner with or complement the existing thing, 00:16:52.420 --> 00:16:54.058 align:middle line:90% until I can take over. 00:16:54.058 --> 00:16:56.100 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, that's a good one. 00:16:56.100 --> 00:16:58.860 align:middle line:84% AUDIENCE: Example would be the batteries. 00:16:58.860 --> 00:17:02.700 align:middle line:84% They had, these new HydraCell batteries 00:17:02.700 --> 00:17:04.400 align:middle line:90% get charged by whatever. 00:17:04.400 --> 00:17:06.800 align:middle line:84% But instead of saying, OK, you can charge your battery, 00:17:06.800 --> 00:17:08.359 align:middle line:90% you can run your laptop on it. 00:17:08.359 --> 00:17:11.057 align:middle line:84% They say, you can charge the battery that you have. 00:17:11.057 --> 00:17:13.640 align:middle line:84% So if you're on the road, and you need to charge your battery, 00:17:13.640 --> 00:17:14.050 align:middle line:90% you can just plug this in. 00:17:14.050 --> 00:17:14.719 align:middle line:90% IRVING WLADAWSKY-BERGER: Yeah. 00:17:14.719 --> 00:17:17.136 align:middle line:84% AUDIENCE: But eventually, they want to get rid of the door 00:17:17.136 --> 00:17:17.920 align:middle line:90% cells or whatever. 00:17:17.920 --> 00:17:20.940 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, and that's reasonable because-- 00:17:20.940 --> 00:17:23.579 align:middle line:84% and by the way, both sides can know it 00:17:23.579 --> 00:17:28.880 align:middle line:84% that this is a dating of convenience 00:17:28.880 --> 00:17:32.020 align:middle line:84% because you're learning from them, 00:17:32.020 --> 00:17:34.380 align:middle line:84% and I'm assuming they are getting something from you. 00:17:34.380 --> 00:17:37.480 align:middle line:84% But they know you can part companies, 00:17:37.480 --> 00:17:40.120 align:middle line:84% and that's totally fine, exactly. 00:17:40.120 --> 00:17:45.160 align:middle line:84% And the maturity is that you need 00:17:45.160 --> 00:17:48.040 align:middle line:90% to be comfortable with that. 00:17:48.040 --> 00:17:50.680 align:middle line:84% It's just a very comfortable thing. 00:17:50.680 --> 00:17:54.880 align:middle line:84% And I'm saying that because often when people especially, 00:17:54.880 --> 00:17:56.700 align:middle line:84% are more junior, it's like, well, 00:17:56.700 --> 00:17:59.340 align:middle line:84% but how can I deal with the bastards? 00:17:59.340 --> 00:18:02.240 align:middle line:84% We have to-- and then, especially 00:18:02.240 --> 00:18:06.320 align:middle line:84% within IBM, where different groups hate 00:18:06.320 --> 00:18:11.560 align:middle line:84% versus cooperate, they send email messages at midnight 00:18:11.560 --> 00:18:13.720 align:middle line:90% attacking. 00:18:13.720 --> 00:18:15.280 align:middle line:90% Trust me. 00:18:15.280 --> 00:18:17.160 align:middle line:84% I don't know if you've all gotten messages 00:18:17.160 --> 00:18:21.340 align:middle line:84% from some of your colleagues saying, how dare you do this. 00:18:21.340 --> 00:18:22.380 align:middle line:90% You're an idiot. 00:18:22.380 --> 00:18:24.140 align:middle line:84% These people are trying to kill me. 00:18:24.140 --> 00:18:26.910 align:middle line:84% And the answer is, well, that's it, 00:18:26.910 --> 00:18:29.230 align:middle line:84% I'm not playing tennis against them. 00:18:29.230 --> 00:18:32.090 align:middle line:90% We're actually doubles partners. 00:18:32.090 --> 00:18:32.590 align:middle line:90% [LAUGHS] 00:18:32.590 --> 00:18:33.510 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:18:33.510 --> 00:18:34.843 align:middle line:90% IRVING WLADAWSKY-BERGER: Please. 00:18:34.843 --> 00:18:37.350 align:middle line:84% AUDIENCE: So I'm taking up on the need 00:18:37.350 --> 00:18:42.110 align:middle line:84% to strike a balance between not falling in love with [INAUDIBLE] 00:18:42.110 --> 00:18:43.470 align:middle line:90% and doing that. 00:18:43.470 --> 00:18:46.070 align:middle line:84% So what kind of framework can we do, What. 00:18:46.070 --> 00:18:48.990 align:middle line:84% Kind of checks and balances can we do inside an organization, 00:18:48.990 --> 00:18:52.363 align:middle line:90% to make sure that happens? 00:18:52.363 --> 00:18:54.030 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, it really 00:18:54.030 --> 00:19:01.710 align:middle line:84% gets down to having a very sober analysis of what cards 00:19:01.710 --> 00:19:03.110 align:middle line:90% you are holding. 00:19:03.110 --> 00:19:05.270 align:middle line:90% There's no substitute for that. 00:19:05.270 --> 00:19:08.870 align:middle line:84% And the cards will come different kinds of cards. 00:19:08.870 --> 00:19:14.150 align:middle line:84% To be honest, the most important one, I think, 00:19:14.150 --> 00:19:16.310 align:middle line:90% is revenue and profit. 00:19:16.310 --> 00:19:18.230 align:middle line:84% The reason it's the most important one 00:19:18.230 --> 00:19:22.030 align:middle line:84% is if you run out of money, you can 00:19:22.030 --> 00:19:25.410 align:middle line:84% have the most beautiful strategy in the world, 00:19:25.410 --> 00:19:28.530 align:middle line:90% but you're going to die. 00:19:28.530 --> 00:19:30.450 align:middle line:90% I mean, you just will. 00:19:30.450 --> 00:19:32.970 align:middle line:84% And as I said, I think that happened to the old 00:19:32.970 --> 00:19:39.250 align:middle line:84% AT&T. They called the right shots, 00:19:39.250 --> 00:19:46.290 align:middle line:84% but they took their eye off the ball of their legacy businesses. 00:19:46.290 --> 00:19:48.850 align:middle line:84% And as a result, remember, it's my opinion, 00:19:48.850 --> 00:19:52.090 align:middle line:84% somebody else may see it differently, 00:19:52.090 --> 00:19:59.210 align:middle line:84% they didn't have enough time to see the new strategies through. 00:19:59.210 --> 00:20:02.970 align:middle line:84% And remember, it's much easier to predict 00:20:02.970 --> 00:20:08.530 align:middle line:84% what will happen than how quickly it will happen. 00:20:08.530 --> 00:20:16.290 align:middle line:84% For example, let me take some concrete things 00:20:16.290 --> 00:20:18.510 align:middle line:90% and some different ones. 00:20:18.510 --> 00:20:20.810 align:middle line:90% Let's take virtual worlds. 00:20:20.810 --> 00:20:22.563 align:middle line:90% You know what I'm talking about? 00:20:22.563 --> 00:20:23.480 align:middle line:90% AUDIENCE: Second Life. 00:20:23.480 --> 00:20:26.120 align:middle line:84% IRVING WLADAWSKY-BERGER: Second Life and lots of things. 00:20:26.120 --> 00:20:31.820 align:middle line:84% I am convinced, and so are a lot of people, 00:20:31.820 --> 00:20:34.960 align:middle line:84% that this is going to be a very big deal. 00:20:34.960 --> 00:20:37.040 align:middle line:90% But if you ask me how quickly? 00:20:37.040 --> 00:20:40.760 align:middle line:90% When will-- I have no idea. 00:20:40.760 --> 00:20:41.900 align:middle line:90% Is it two years? 00:20:41.900 --> 00:20:42.900 align:middle line:90% Is it one year? 00:20:42.900 --> 00:20:43.780 align:middle line:90% Is it five years? 00:20:43.780 --> 00:20:45.820 align:middle line:90% I honestly have no idea. 00:20:45.820 --> 00:20:48.780 align:middle line:90% 00:20:48.780 --> 00:20:50.100 align:middle line:90% IBM is very involved. 00:20:50.100 --> 00:20:52.480 align:middle line:84% And I worked hard to get us there. 00:20:52.480 --> 00:20:58.480 align:middle line:84% And I tell lots of companies to get on the learning curve, 00:20:58.480 --> 00:21:01.300 align:middle line:84% but I don't know when it will take off. 00:21:01.300 --> 00:21:03.000 align:middle line:90% That's an example. 00:21:03.000 --> 00:21:06.760 align:middle line:84% Most of what people said would happen 00:21:06.760 --> 00:21:10.320 align:middle line:84% during the height of the dotcom bubble, 00:21:10.320 --> 00:21:14.880 align:middle line:84% that we're building a digital economy and things like that, 00:21:14.880 --> 00:21:16.880 align:middle line:90% they're happening. 00:21:16.880 --> 00:21:23.660 align:middle line:84% It's just that those people said it would happen in two years. 00:21:23.660 --> 00:21:27.620 align:middle line:84% The reality is you don't build a digital economy 00:21:27.620 --> 00:21:29.480 align:middle line:90% for the world in two years. 00:21:29.480 --> 00:21:31.700 align:middle line:90% It takes a while to do that. 00:21:31.700 --> 00:21:36.260 align:middle line:84% And because the new things are unpredictable, how quickly-- 00:21:36.260 --> 00:21:37.980 align:middle line:90% and remember, part of-- 00:21:37.980 --> 00:21:42.260 align:middle line:84% how quickly will they start generating serious profits 00:21:42.260 --> 00:21:46.940 align:middle line:84% and cash, you need to rely on what you have now. 00:21:46.940 --> 00:21:52.140 align:middle line:84% And you need to plan so that you keep living off this while this 00:21:52.140 --> 00:21:54.540 align:middle line:90% is nurtured and growing. 00:21:54.540 --> 00:21:56.160 align:middle line:90% So cash is one. 00:21:56.160 --> 00:21:58.340 align:middle line:90% The other is talent. 00:21:58.340 --> 00:22:01.300 align:middle line:84% Do you have talented people that you 00:22:01.300 --> 00:22:04.820 align:middle line:84% can start moving into these areas, both technical people, 00:22:04.820 --> 00:22:08.700 align:middle line:90% management talent, and so on? 00:22:08.700 --> 00:22:10.440 align:middle line:90% And I've seen people do that. 00:22:10.440 --> 00:22:13.580 align:middle line:84% If you said no, our existing people will just 00:22:13.580 --> 00:22:17.240 align:middle line:84% put on the legacy crap and let them do the money, 00:22:17.240 --> 00:22:20.970 align:middle line:84% we'll hire all new people to do the exciting stuff, that 00:22:20.970 --> 00:22:25.110 align:middle line:90% is not a good idea. 00:22:25.110 --> 00:22:27.430 align:middle line:84% You do want to acquire new talents, 00:22:27.430 --> 00:22:30.990 align:middle line:84% but the new talent usually has zero loyalty to the company. 00:22:30.990 --> 00:22:33.170 align:middle line:90% They don't even know-- 00:22:33.170 --> 00:22:38.390 align:middle line:84% you cannot show up on day one and say, I love this company. 00:22:38.390 --> 00:22:40.770 align:middle line:90% That would be suspect. 00:22:40.770 --> 00:22:43.810 align:middle line:84% It's like you go on the first date and say, let's get married. 00:22:43.810 --> 00:22:49.770 align:middle line:84% That takes a little while to fall in love. 00:22:49.770 --> 00:22:57.070 align:middle line:84% And so you want some people that can start bringing the culture. 00:22:57.070 --> 00:22:58.570 align:middle line:84% And if you don't have anybody who 00:22:58.570 --> 00:23:04.130 align:middle line:84% knows anything about this area, then you're out of balance. 00:23:04.130 --> 00:23:07.510 align:middle line:84% Brand and history is extremely important. 00:23:07.510 --> 00:23:10.090 align:middle line:84% In fact, I just finished a conference call 00:23:10.090 --> 00:23:14.290 align:middle line:84% with Chris Wall, who is the person from O&M, Ogilvy 00:23:14.290 --> 00:23:18.130 align:middle line:84% & Mather, that will be the guest lecture on November 1, 00:23:18.130 --> 00:23:19.990 align:middle line:84% and I'm really looking forward to what 00:23:19.990 --> 00:23:25.550 align:middle line:84% he's going to say because he's an expert in brand and things 00:23:25.550 --> 00:23:26.210 align:middle line:90% like that. 00:23:26.210 --> 00:23:29.590 align:middle line:90% That is so important. 00:23:29.590 --> 00:23:35.510 align:middle line:84% We talk quite a bit about major screw ups we've seen. 00:23:35.510 --> 00:23:38.470 align:middle line:90% Ford fixed Jaguar. 00:23:38.470 --> 00:23:41.090 align:middle line:84% Jaguars used to have horrible engineering. 00:23:41.090 --> 00:23:43.590 align:middle line:90% Now they are very reliable. 00:23:43.590 --> 00:23:46.590 align:middle line:84% You all know what I'm talking about? 00:23:46.590 --> 00:23:52.710 align:middle line:84% Ford screwed up the Jaguar brand because Jaguar, 00:23:52.710 --> 00:23:57.190 align:middle line:90% it's a high-end brand. 00:23:57.190 --> 00:24:01.150 align:middle line:84% And they started developing low-end Jaguars 00:24:01.150 --> 00:24:05.670 align:middle line:90% on top of Ford chassis. 00:24:05.670 --> 00:24:09.190 align:middle line:84% Well, this is as if I tell you I'm 00:24:09.190 --> 00:24:13.170 align:middle line:84% going to give you Porsches on VW chassis. 00:24:13.170 --> 00:24:16.840 align:middle line:84% And VW is a very good car, but if you buy a Porsche, 00:24:16.840 --> 00:24:19.220 align:middle line:90% you want a goddamned Porsche. 00:24:19.220 --> 00:24:22.280 align:middle line:84% And if you start bringing Porsche to downscale, 00:24:22.280 --> 00:24:24.220 align:middle line:90% eventually you break the brand. 00:24:24.220 --> 00:24:25.880 align:middle line:84% People say, what the hell is a Porsche? 00:24:25.880 --> 00:24:27.140 align:middle line:90% I don't know anymore. 00:24:27.140 --> 00:24:29.380 align:middle line:90% And they did that to Jaguar. 00:24:29.380 --> 00:24:33.020 align:middle line:90% So you have to be very careful. 00:24:33.020 --> 00:24:36.720 align:middle line:84% Is what you're trying to do, does it fit the brand? 00:24:36.720 --> 00:24:38.980 align:middle line:84% And if it doesn't fit the brand-- 00:24:38.980 --> 00:24:42.500 align:middle line:84% I don't know --then you're probably better off having 00:24:42.500 --> 00:24:44.860 align:middle line:84% a totally separate company and managing 00:24:44.860 --> 00:24:47.980 align:middle line:90% as a totally separate company. 00:24:47.980 --> 00:24:53.300 align:middle line:84% So these are all the myriad of strategic decisions 00:24:53.300 --> 00:24:55.020 align:middle line:90% you need to make. 00:24:55.020 --> 00:24:58.220 align:middle line:90% And they take a lot of thinking. 00:24:58.220 --> 00:25:00.960 align:middle line:84% And let me keep repeating, sometimes people say, well, 00:25:00.960 --> 00:25:02.460 align:middle line:90% you must be an MBA. 00:25:02.460 --> 00:25:06.540 align:middle line:84% Well, you can be an MBA and have a lot of respect for MBAs. 00:25:06.540 --> 00:25:10.900 align:middle line:84% But people who are really good with complex systems, when they 00:25:10.900 --> 00:25:13.840 align:middle line:84% put their mind to it are superb at these 00:25:13.840 --> 00:25:17.760 align:middle line:84% because these are God-awful, complex problems. 00:25:17.760 --> 00:25:24.320 align:middle line:84% That's why I keep saying so much how we need more and more people 00:25:24.320 --> 00:25:29.440 align:middle line:84% with talented system skills, engineering skills 00:25:29.440 --> 00:25:33.680 align:middle line:84% to be in major management positions leading 00:25:33.680 --> 00:25:37.480 align:middle line:90% these kinds of areas. 00:25:37.480 --> 00:25:42.580 align:middle line:84% So let's move and start talking about execution. 00:25:42.580 --> 00:25:48.540 align:middle line:84% And as a colleague of mine, Steve Mills, 00:25:48.540 --> 00:25:51.500 align:middle line:84% who runs IBM software business, says, 00:25:51.500 --> 00:25:59.520 align:middle line:84% often a strategy that you cannot execute is a fantasy. 00:25:59.520 --> 00:26:04.800 align:middle line:84% And fantasies are OK but not in the business world. 00:26:04.800 --> 00:26:08.760 align:middle line:84% Then you really have a bad strategy. 00:26:08.760 --> 00:26:16.550 align:middle line:84% So let me start by talking about some of the questions 00:26:16.550 --> 00:26:21.470 align:middle line:84% that Clay Christensen and Michael Raynor have posed. 00:26:21.470 --> 00:26:29.070 align:middle line:84% And one of the major questions about executing a new strategy 00:26:29.070 --> 00:26:35.110 align:middle line:84% is to decide what are you going to do yourself, 00:26:35.110 --> 00:26:37.710 align:middle line:84% and what do you want to partner for? 00:26:37.710 --> 00:26:41.470 align:middle line:84% Remember, this is really important for something new 00:26:41.470 --> 00:26:44.630 align:middle line:84% because you've built incredible skills here, 00:26:44.630 --> 00:26:47.070 align:middle line:84% and you want to bring some of these skills 00:26:47.070 --> 00:26:49.830 align:middle line:90% to execute the new strategy. 00:26:49.830 --> 00:26:53.270 align:middle line:84% But by definition of new and disruptive, 00:26:53.270 --> 00:26:55.750 align:middle line:84% you probably don't have all these skills. 00:26:55.750 --> 00:26:58.890 align:middle line:84% And if you say, well, I'm going to take my time, build it up, 00:26:58.890 --> 00:27:00.590 align:middle line:90% you don't have time. 00:27:00.590 --> 00:27:07.710 align:middle line:84% So you need to decide partnering and acquisition and so on. 00:27:07.710 --> 00:27:12.050 align:middle line:84% A key question, especially now more and more 00:27:12.050 --> 00:27:20.850 align:middle line:84% is, how about proprietary versus standard? 00:27:20.850 --> 00:27:24.190 align:middle line:84% And as you know, we're making those decisions more and more. 00:27:24.190 --> 00:27:29.890 align:middle line:84% This is fairly new in the IT world. 00:27:29.890 --> 00:27:35.890 align:middle line:84% Actually in every engineering discipline 00:27:35.890 --> 00:27:38.070 align:middle line:90% goes through similar questions. 00:27:38.070 --> 00:27:43.130 align:middle line:84% So what do you build using standard components versus what 00:27:43.130 --> 00:27:45.530 align:middle line:84% should you build making it more proprietary 00:27:45.530 --> 00:27:48.690 align:middle line:84% with automobiles, whatever you want to? 00:27:48.690 --> 00:27:52.370 align:middle line:84% And often it's very related to the first question. 00:27:52.370 --> 00:27:55.210 align:middle line:90% We'll talk about that. 00:27:55.210 --> 00:27:56.910 align:middle line:90% What's the dynamics? 00:27:56.910 --> 00:28:00.170 align:middle line:84% How do you gravitate between them? 00:28:00.170 --> 00:28:05.690 align:middle line:84% And once you decide one way, are you stuck, or do you often 00:28:05.690 --> 00:28:06.310 align:middle line:90% come back? 00:28:06.310 --> 00:28:11.030 align:middle line:84% So let's explore some of these questions 00:28:11.030 --> 00:28:13.990 align:middle line:90% that Christensen has posed. 00:28:13.990 --> 00:28:23.950 align:middle line:84% And as I said, the decisions about what to insource 00:28:23.950 --> 00:28:27.350 align:middle line:84% and what to procure from others has 00:28:27.350 --> 00:28:34.190 align:middle line:84% a huge impact on the chance of success for any new venture. 00:28:34.190 --> 00:28:39.310 align:middle line:84% I'd love to hear your thoughts on why you think that is. 00:28:39.310 --> 00:28:43.930 align:middle line:84% I agree totally, but why would you think that is? 00:28:43.930 --> 00:28:49.776 align:middle line:90% 00:28:49.776 --> 00:28:52.203 align:middle line:84% AUDIENCE: [INAUDIBLE] a very good idea. 00:28:52.203 --> 00:28:53.870 align:middle line:84% IRVING WLADAWSKY-BERGER: Basically what? 00:28:53.870 --> 00:28:56.270 align:middle line:84% AUDIENCE: I'm saying [INAUDIBLE]. 00:28:56.270 --> 00:28:59.990 align:middle line:90% It's a ripe area. 00:28:59.990 --> 00:29:01.550 align:middle line:84% IRVING WLADAWSKY-BERGER: So where 00:29:01.550 --> 00:29:03.210 align:middle line:90% are you going to put effort? 00:29:03.210 --> 00:29:08.420 align:middle line:90% 00:29:08.420 --> 00:29:09.100 align:middle line:90% Yes? 00:29:09.100 --> 00:29:12.380 align:middle line:84% AUDIENCE: It's really about keeping the knowledge and the IP 00:29:12.380 --> 00:29:15.980 align:middle line:84% within your own company and outsourcing 00:29:15.980 --> 00:29:17.780 align:middle line:90% all the complementary tasks. 00:29:17.780 --> 00:29:19.200 align:middle line:90% If you want to be-- 00:29:19.200 --> 00:29:21.940 align:middle line:90% 00:29:21.940 --> 00:29:23.820 align:middle line:84% as the next generation evolves, you 00:29:23.820 --> 00:29:26.860 align:middle line:84% need to have the knowledge in-house, 00:29:26.860 --> 00:29:28.960 align:middle line:84% rather than giving it out to your partners. 00:29:28.960 --> 00:29:33.740 align:middle line:84% IRVING WLADAWSKY-BERGER: Mm-hmm, but now, 00:29:33.740 --> 00:29:38.980 align:middle line:84% what if your partners have the knowledge you want? 00:29:38.980 --> 00:29:40.700 align:middle line:84% AUDIENCE: Then you are in trouble. 00:29:40.700 --> 00:29:46.100 align:middle line:84% IRVING WLADAWSKY-BERGER: Or you make them your partner. 00:29:46.100 --> 00:29:47.940 align:middle line:84% AUDIENCE: Yeah, but there's no guarantee 00:29:47.940 --> 00:29:50.033 align:middle line:84% that they won't work with your competitors. 00:29:50.033 --> 00:29:51.700 align:middle line:84% IRVING WLADAWSKY-BERGER: Oh, no, no, you 00:29:51.700 --> 00:29:53.860 align:middle line:90% have to assume they will. 00:29:53.860 --> 00:29:59.020 align:middle line:90% But yeah, I'll get back with-- 00:29:59.020 --> 00:30:00.340 align:middle line:90% yes, go ahead. 00:30:00.340 --> 00:30:02.940 align:middle line:84% AUDIENCE: I was thinking about Boeing. 00:30:02.940 --> 00:30:07.360 align:middle line:84% For Boeing 787, the person who was in charge of the program 00:30:07.360 --> 00:30:08.907 align:middle line:90% had come here and given a talk. 00:30:08.907 --> 00:30:10.240 align:middle line:90% IRVING WLADAWSKY-BERGER: At MIT. 00:30:10.240 --> 00:30:13.040 align:middle line:84% AUDIENCE: At MIT, and one of the things that they had mentioned 00:30:13.040 --> 00:30:16.040 align:middle line:84% is they made this massive decision for 787 00:30:16.040 --> 00:30:18.680 align:middle line:84% that they are going to outsource a lot of things 00:30:18.680 --> 00:30:20.680 align:middle line:90% across the world. 00:30:20.680 --> 00:30:24.120 align:middle line:84% And one of the crown jewels, which is the wing development, 00:30:24.120 --> 00:30:26.400 align:middle line:90% they have given out. 00:30:26.400 --> 00:30:27.640 align:middle line:90% And one of the questions-- 00:30:27.640 --> 00:30:29.500 align:middle line:84% IRVING WLADAWSKY-BERGER: Who does that, Japan, am I correct? 00:30:29.500 --> 00:30:30.840 align:middle line:90% AUDIENCE: I think so. 00:30:30.840 --> 00:30:31.932 align:middle line:90% It's Japanese, I think. 00:30:31.932 --> 00:30:33.140 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 00:30:33.140 --> 00:30:34.300 align:middle line:90% AUDIENCE: Matsushita, I think. 00:30:34.300 --> 00:30:35.550 align:middle line:90% IRVING WLADAWSKY-BERGER: Yeah. 00:30:35.550 --> 00:30:39.220 align:middle line:84% AUDIENCE: And one of the questions was, China, Brazil, 00:30:39.220 --> 00:30:42.960 align:middle line:84% and a bunch of others are developing their own jets 00:30:42.960 --> 00:30:44.060 align:middle line:90% and would they learn? 00:30:44.060 --> 00:30:47.020 align:middle line:84% And he actually said the same thing, 00:30:47.020 --> 00:30:50.280 align:middle line:84% which is they would never have been able to make the capital 00:30:50.280 --> 00:30:53.040 align:middle line:84% investments and ramp as fast to beat Airbus 00:30:53.040 --> 00:30:54.640 align:middle line:90% as they did with this. 00:30:54.640 --> 00:30:57.000 align:middle line:84% And they are already making an assumption, 00:30:57.000 --> 00:30:59.720 align:middle line:84% over a certain time frame, that people are 00:30:59.720 --> 00:31:03.670 align:middle line:90% going to absorb that knowledge. 00:31:03.670 --> 00:31:08.430 align:middle line:84% Is their skill really the heavy engineering aspect, or just 00:31:08.430 --> 00:31:10.270 align:middle line:90% the system integration aspect? 00:31:10.270 --> 00:31:12.865 align:middle line:90% It's a difficult balance. 00:31:12.865 --> 00:31:14.990 align:middle line:84% IRVING WLADAWSKY-BERGER: Let me rephrase it and see 00:31:14.990 --> 00:31:18.950 align:middle line:84% if I got the gist, which is, you would 00:31:18.950 --> 00:31:21.410 align:middle line:84% say, everything else being equal, 00:31:21.410 --> 00:31:26.830 align:middle line:84% I want to do it all by myself, but what if that's impossible? 00:31:26.830 --> 00:31:27.970 align:middle line:90% Now, you have two choices. 00:31:27.970 --> 00:31:29.710 align:middle line:84% Either you say, then I'm not going 00:31:29.710 --> 00:31:33.510 align:middle line:84% to do it because I only do things I do by myself. 00:31:33.510 --> 00:31:38.550 align:middle line:84% And the second one is, I'm going to partner with other people. 00:31:38.550 --> 00:31:42.470 align:middle line:84% Then you say, well, you idiot, but the Chinese just 00:31:42.470 --> 00:31:45.830 align:middle line:84% want to partner with you to learn from you. 00:31:45.830 --> 00:31:49.150 align:middle line:84% The answer is, well, let me repeat, 00:31:49.150 --> 00:31:53.450 align:middle line:84% if I don't partner with them-- and by the way, the Japanese, 00:31:53.450 --> 00:31:57.390 align:middle line:84% the Koreans, the Brazilians it's not like the Chinese are 00:31:57.390 --> 00:31:58.850 align:middle line:90% particularly nefarious. 00:31:58.850 --> 00:32:04.010 align:middle line:84% And everybody wants to learn and then go their own way. 00:32:04.010 --> 00:32:08.270 align:middle line:84% If you don't do that, then you don't play. 00:32:08.270 --> 00:32:14.210 align:middle line:90% 00:32:14.210 --> 00:32:21.770 align:middle line:84% My feeling is that when you are in a leadership position, 00:32:21.770 --> 00:32:27.090 align:middle line:84% in a management position, if you keep thinking about the problem 00:32:27.090 --> 00:32:30.290 align:middle line:84% and thinking, and I don't just mean thinking, and you analyze 00:32:30.290 --> 00:32:36.490 align:middle line:84% and you ponder, once you get convinced this is the only way 00:32:36.490 --> 00:32:40.050 align:middle line:84% to get it done, then the decision is obvious 00:32:40.050 --> 00:32:41.750 align:middle line:84% because the alternative doesn't work. 00:32:41.750 --> 00:32:44.390 align:middle line:84% Now, you don't know that at the beginning. 00:32:44.390 --> 00:32:48.850 align:middle line:84% So often what happened through this process of thinking 00:32:48.850 --> 00:32:52.270 align:middle line:84% is you start deciding, do I have alternatives? 00:32:52.270 --> 00:32:54.130 align:middle line:90% What would be the alternative? 00:32:54.130 --> 00:32:58.330 align:middle line:84% But once you've thought it through, 00:32:58.330 --> 00:33:00.880 align:middle line:84% and this is the only way you're able to do 00:33:00.880 --> 00:33:04.360 align:middle line:90% that, it's like it's over. 00:33:04.360 --> 00:33:07.280 align:middle line:84% Somebody says yeah, but real men-- 00:33:07.280 --> 00:33:08.720 align:middle line:90% screw it. 00:33:08.720 --> 00:33:10.860 align:middle line:90% That was 50 years ago. 00:33:10.860 --> 00:33:16.160 align:middle line:90% Or do a product that is simpler. 00:33:16.160 --> 00:33:17.980 align:middle line:90% Don't do a 787. 00:33:17.980 --> 00:33:21.800 align:middle line:90% Do a Cessna. 00:33:21.800 --> 00:33:23.140 align:middle line:90% No, I want to do 78-- 00:33:23.140 --> 00:33:24.740 align:middle line:90% well, you cannot do that. 00:33:24.740 --> 00:33:30.480 align:middle line:84% And those are very interesting debates and discussions 00:33:30.480 --> 00:33:32.280 align:middle line:90% that you have to do. 00:33:32.280 --> 00:33:37.380 align:middle line:84% But then your point is extremely well taken because once you say, 00:33:37.380 --> 00:33:40.400 align:middle line:84% well, I'm going to partner, then often 00:33:40.400 --> 00:33:45.120 align:middle line:84% what then happens is you start thinking of your business 00:33:45.120 --> 00:33:48.760 align:middle line:84% at the next level, which is you are 00:33:48.760 --> 00:33:52.520 align:middle line:84% the best in serving the market, serving the airlines. 00:33:52.520 --> 00:33:56.080 align:middle line:84% And, we'll talk about it throughout. 00:33:56.080 --> 00:34:00.640 align:middle line:84% And these people may learn to make wings, maybe airplanes, 00:34:00.640 --> 00:34:02.980 align:middle line:84% but I'm not sure they'll learn how 00:34:02.980 --> 00:34:05.620 align:middle line:84% to have loyal customer relationships, 00:34:05.620 --> 00:34:07.120 align:middle line:90% or it will take longer. 00:34:07.120 --> 00:34:09.460 align:middle line:84% So it's a very interesting thing. 00:34:09.460 --> 00:34:12.500 align:middle line:84% AUDIENCE: I think this question is, at the very core 00:34:12.500 --> 00:34:14.820 align:middle line:84% of defining your company, because it 00:34:14.820 --> 00:34:19.564 align:middle line:84% defines how you're different from your competitors. 00:34:19.564 --> 00:34:20.939 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly. 00:34:20.939 --> 00:34:25.020 align:middle line:84% AUDIENCE: So basically, if you are designing-- 00:34:25.020 --> 00:34:27.699 align:middle line:84% if you're making compilers and you're 00:34:27.699 --> 00:34:29.699 align:middle line:84% buying lots of other things, you have 00:34:29.699 --> 00:34:32.412 align:middle line:84% to know what is your core business? 00:34:32.412 --> 00:34:33.620 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 00:34:33.620 --> 00:34:36.120 align:middle line:84% AUDIENCE: That it is very different from everybody else. 00:34:36.120 --> 00:34:37.328 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 00:34:37.328 --> 00:34:39.340 align:middle line:84% AUDIENCE: So it's very important to know. 00:34:39.340 --> 00:34:40.798 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, it's 00:34:40.798 --> 00:34:43.900 align:middle line:84% very important to know and remember, any changes over time. 00:34:43.900 --> 00:34:45.360 align:middle line:90% This is very important. 00:34:45.360 --> 00:34:46.733 align:middle line:90% Yes, I agree totally. 00:34:46.733 --> 00:34:49.400 align:middle line:84% AUDIENCE: My question is, are we talking in terms of disruptive? 00:34:49.400 --> 00:34:52.380 align:middle line:84% So if you have something that's brand new in the market, 00:34:52.380 --> 00:34:53.500 align:middle line:90% it's hard. 00:34:53.500 --> 00:34:57.590 align:middle line:84% My guess is you may not be able to partner with the best. 00:34:57.590 --> 00:34:59.530 align:middle line:90% And so the only option you may-- 00:34:59.530 --> 00:35:00.050 align:middle line:84% IRVING WLADAWSKY-BERGER: Say that again. 00:35:00.050 --> 00:35:01.430 align:middle line:90% You may not be able to partner. 00:35:01.430 --> 00:35:03.930 align:middle line:84% AUDIENCE: You get partners, you may be able to get supplies, 00:35:03.930 --> 00:35:05.250 align:middle line:90% you say, make something new. 00:35:05.250 --> 00:35:07.345 align:middle line:90% So what do you have-- 00:35:07.345 --> 00:35:08.970 align:middle line:84% I mean, would there be any other choice 00:35:08.970 --> 00:35:13.810 align:middle line:90% but to develop it yourself? 00:35:13.810 --> 00:35:15.230 align:middle line:84% Only when it comes to disruptive. 00:35:15.230 --> 00:35:17.930 align:middle line:84% I mean, there are other cases where it's too messy. 00:35:17.930 --> 00:35:26.210 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, but usually not the whole 00:35:26.210 --> 00:35:30.730 align:middle line:90% thing is disruptive, one part. 00:35:30.730 --> 00:35:33.550 align:middle line:84% And so part of the thinking is, OK, 00:35:33.550 --> 00:35:38.850 align:middle line:90% so keep to yourself the jewel. 00:35:38.850 --> 00:35:43.850 align:middle line:84% And then there is usually, again, in engineering projects 00:35:43.850 --> 00:35:46.340 align:middle line:84% there are a lot of, I was going to say crap, 00:35:46.340 --> 00:35:47.590 align:middle line:90% but that's not the right term. 00:35:47.590 --> 00:35:51.570 align:middle line:84% A lot of other important things that 00:35:51.570 --> 00:35:57.350 align:middle line:84% surround a system, a business, a product. 00:35:57.350 --> 00:36:01.210 align:middle line:84% And the more you know where is the jewel in that crown, 00:36:01.210 --> 00:36:04.110 align:middle line:84% the more you keep that to yourself, that's 00:36:04.110 --> 00:36:07.590 align:middle line:84% where you protect the IP like crazy. 00:36:07.590 --> 00:36:11.130 align:middle line:84% You may not even share that with your partners. 00:36:11.130 --> 00:36:14.410 align:middle line:84% You may say, the partnership stops here. 00:36:14.410 --> 00:36:16.090 align:middle line:90% You don't go into my bedroom. 00:36:16.090 --> 00:36:20.070 align:middle line:84% You can come into the living room, 00:36:20.070 --> 00:36:24.390 align:middle line:84% but that's where you stay because there 00:36:24.390 --> 00:36:27.750 align:middle line:84% are certain things you're not willing to share, 00:36:27.750 --> 00:36:33.050 align:middle line:84% or there are partners, and there are partners. 00:36:33.050 --> 00:36:39.790 align:middle line:84% And sometimes you say, but if you invest $1 billion here, 00:36:39.790 --> 00:36:42.070 align:middle line:90% now it's different. 00:36:42.070 --> 00:36:44.050 align:middle line:84% And by the way, for example, IBM, 00:36:44.050 --> 00:36:51.670 align:middle line:84% we decided the fabs in Fishkill got too expensive 00:36:51.670 --> 00:36:56.060 align:middle line:84% and we needed to share the cost, just got too expensive. 00:36:56.060 --> 00:36:57.820 align:middle line:90% Not unlike Boeing's decision. 00:36:57.820 --> 00:37:00.460 align:middle line:84% In principle Boeing could have done that, but now 00:37:00.460 --> 00:37:02.940 align:middle line:90% the risk is this high. 00:37:02.940 --> 00:37:05.860 align:middle line:84% You all know what I mean by a fab where you build something? 00:37:05.860 --> 00:37:08.300 align:middle line:90% They are frightfully expensive. 00:37:08.300 --> 00:37:14.940 align:middle line:84% And so a number of companies, Samsung and Sony and others 00:37:14.940 --> 00:37:16.560 align:middle line:90% pay-- not just pay. 00:37:16.560 --> 00:37:18.100 align:middle line:90% They are partners. 00:37:18.100 --> 00:37:22.268 align:middle line:90% They co-own the facility. 00:37:22.268 --> 00:37:24.060 align:middle line:84% I don't know if they co-own it, but I don't 00:37:24.060 --> 00:37:25.400 align:middle line:90% know what the arrangement is. 00:37:25.400 --> 00:37:28.380 align:middle line:90% 00:37:28.380 --> 00:37:30.040 align:middle line:84% AUDIENCE: When you share like that, 00:37:30.040 --> 00:37:32.780 align:middle line:84% you also share a lot of the other information 00:37:32.780 --> 00:37:33.760 align:middle line:90% with your partner. 00:37:33.760 --> 00:37:34.960 align:middle line:90% So aren't you giving up-- 00:37:34.960 --> 00:37:35.780 align:middle line:84% IRVING WLADAWSKY-BERGER: You are. 00:37:35.780 --> 00:37:36.880 align:middle line:84% AUDIENCE: So I mean, for example-- 00:37:36.880 --> 00:37:38.755 align:middle line:84% IRVING WLADAWSKY-BERGER: Gets back to Boeing, 00:37:38.755 --> 00:37:41.060 align:middle line:84% so do you want the fab, or don't you want the fab? 00:37:41.060 --> 00:37:45.100 align:middle line:84% And the only way to have the fab is to partner. 00:37:45.100 --> 00:37:48.100 align:middle line:84% AUDIENCE: And so cost is the only driver to go out, meaning-- 00:37:48.100 --> 00:37:50.800 align:middle line:84% IRVING WLADAWSKY-BERGER: No. no, because you now 00:37:50.800 --> 00:37:54.440 align:middle line:84% have to think of a business model that 00:37:54.440 --> 00:37:56.860 align:middle line:84% lets you stay in business with partners, 00:37:56.860 --> 00:37:59.340 align:middle line:84% versus a vertically integrated business model. 00:37:59.340 --> 00:38:02.200 align:middle line:84% You have to rethink your business model. 00:38:02.200 --> 00:38:08.240 align:middle line:84% And if you cannot make the business model close, 00:38:08.240 --> 00:38:11.560 align:middle line:84% then you'll get eaten up by your partner. 00:38:11.560 --> 00:38:14.520 align:middle line:90% So you have to decide is there-- 00:38:14.520 --> 00:38:17.480 align:middle line:90% 00:38:17.480 --> 00:38:18.260 align:middle line:90% just one second. 00:38:18.260 --> 00:38:24.080 align:middle line:84% Let's assume that your partner starts undercutting you. 00:38:24.080 --> 00:38:26.620 align:middle line:84% They learn from you, and now they start undercutting you. 00:38:26.620 --> 00:38:30.032 align:middle line:84% Is there something better you can do from them? 00:38:30.032 --> 00:38:31.740 align:middle line:84% By the way, sometimes you make a mistake. 00:38:31.740 --> 00:38:33.540 align:middle line:84% You think I can, and before you know it, 00:38:33.540 --> 00:38:37.480 align:middle line:84% you're dead because your partner is much better than you. 00:38:37.480 --> 00:38:41.120 align:middle line:84% But at least you shouldn't die quietly. 00:38:41.120 --> 00:38:45.440 align:middle line:84% You should say, well, I think I can do better than them. 00:38:45.440 --> 00:38:49.050 align:middle line:84% This gets played out over and over. 00:38:49.050 --> 00:38:52.030 align:middle line:90% 00:38:52.030 --> 00:39:02.750 align:middle line:84% The Chinese cars, people said, well, sorry, nothing cars. 00:39:02.750 --> 00:39:05.590 align:middle line:90% And the Western companies-- 00:39:05.590 --> 00:39:07.350 align:middle line:84% sure the Japanese companies also, I 00:39:07.350 --> 00:39:10.090 align:middle line:84% don't know if Korean companies have business in China, 00:39:10.090 --> 00:39:11.790 align:middle line:90% but I'm assuming they do. 00:39:11.790 --> 00:39:15.190 align:middle line:84% To do that, you probably have to take Chinese partners. 00:39:15.190 --> 00:39:18.270 align:middle line:84% To do that you say, why did they do it, 00:39:18.270 --> 00:39:21.050 align:middle line:84% even though they knew the partners are learning from them? 00:39:21.050 --> 00:39:24.430 align:middle line:84% Well, because China is a huge market. 00:39:24.430 --> 00:39:28.790 align:middle line:84% And if you don't partner, you don't get to play in the market. 00:39:28.790 --> 00:39:32.690 align:middle line:84% But just like the Japanese car companies, 00:39:32.690 --> 00:39:35.710 align:middle line:84% then the Korean car companies, is there any question 00:39:35.710 --> 00:39:39.270 align:middle line:84% that the Chinese car companies and the Indian car companies 00:39:39.270 --> 00:39:43.410 align:middle line:84% will start producing very good, high quality cars over time? 00:39:43.410 --> 00:39:49.050 align:middle line:90% 00:39:49.050 --> 00:39:49.690 align:middle line:90% How can I say? 00:39:49.690 --> 00:39:51.990 align:middle line:84% That's the mating dance of business. 00:39:51.990 --> 00:39:54.030 align:middle line:84% At some level, you're circling each other. 00:39:54.030 --> 00:39:56.050 align:middle line:90% Yes, please. 00:39:56.050 --> 00:39:58.370 align:middle line:84% AUDIENCE: [INAUDIBLE] the two important factor 00:39:58.370 --> 00:40:01.282 align:middle line:84% is the speed to market and the cost advantage. 00:40:01.282 --> 00:40:02.490 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 00:40:02.490 --> 00:40:04.615 align:middle line:84% AUDIENCE: I think, personally, from my experience-- 00:40:04.615 --> 00:40:06.570 align:middle line:90% I working in Siemens. 00:40:06.570 --> 00:40:11.850 align:middle line:84% But in 2000, 2002, 2003, we decided that for architecture 00:40:11.850 --> 00:40:14.170 align:middle line:84% with Siemens and we would outsource whatever 00:40:14.170 --> 00:40:17.490 align:middle line:84% is the best come first to get in the market 00:40:17.490 --> 00:40:19.450 align:middle line:90% and have the cost advantage. 00:40:19.450 --> 00:40:24.210 align:middle line:84% And I've seen that if you develop everything in house, 00:40:24.210 --> 00:40:26.330 align:middle line:84% it might be you're lagging the market. 00:40:26.330 --> 00:40:28.030 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, that's a very good point. 00:40:28.030 --> 00:40:29.447 align:middle line:84% AUDIENCE: In a way, the fast mover 00:40:29.447 --> 00:40:32.150 align:middle line:84% will always have some advantages, second mover. 00:40:32.150 --> 00:40:33.530 align:middle line:90% But if you have third mover-- 00:40:33.530 --> 00:40:34.610 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, you're dead. 00:40:34.610 --> 00:40:37.628 align:middle line:84% AUDIENCE: So it's not necessary that if you develop everything, 00:40:37.628 --> 00:40:38.670 align:middle line:90% you can be the first one. 00:40:38.670 --> 00:40:39.350 align:middle line:84% IRVING WLADAWSKY-BERGER: No, that's right. 00:40:39.350 --> 00:40:41.090 align:middle line:84% AUDIENCE: From that point if you have a big advantage. 00:40:41.090 --> 00:40:43.610 align:middle line:84% Second point, I thought that from the cost advantage, 00:40:43.610 --> 00:40:47.280 align:middle line:84% I have seen that the same system has [INAUDIBLE] business 00:40:47.280 --> 00:40:49.320 align:middle line:90% their own [INAUDIBLE] server. 00:40:49.320 --> 00:40:53.580 align:middle line:84% But if we find it, we are in an enormous cost ratio from IBM, 00:40:53.580 --> 00:40:55.360 align:middle line:90% from Oracle, from Sun. 00:40:55.360 --> 00:41:01.520 align:middle line:84% So what we decided that the way customer wants, we'll build it. 00:41:01.520 --> 00:41:03.700 align:middle line:84% If you choose this one, this is the price. 00:41:03.700 --> 00:41:06.860 align:middle line:84% If you [INAUDIBLE] this will be price if you sign the price. 00:41:06.860 --> 00:41:09.000 align:middle line:84% So the customer has the option and you 00:41:09.000 --> 00:41:12.000 align:middle line:84% can play with your competitor, [INAUDIBLE]. 00:41:12.000 --> 00:41:13.268 align:middle line:90% So to speak to market then-- 00:41:13.268 --> 00:41:15.060 align:middle line:84% IRVING WLADAWSKY-BERGER: They're very good. 00:41:15.060 --> 00:41:16.435 align:middle line:84% I think that's a very good point. 00:41:16.435 --> 00:41:19.000 align:middle line:84% And we'll get to speak to market when we now 00:41:19.000 --> 00:41:21.280 align:middle line:90% talk about modular components. 00:41:21.280 --> 00:41:22.840 align:middle line:84% Because as you can imagine, that's 00:41:22.840 --> 00:41:25.580 align:middle line:90% another huge thing that comes. 00:41:25.580 --> 00:41:26.470 align:middle line:90% Yes, please. 00:41:26.470 --> 00:41:28.720 align:middle line:84% AUDIENCE: I wanted to make some comments about the fab 00:41:28.720 --> 00:41:31.460 align:middle line:84% because I worked at the official fab before I came here. 00:41:31.460 --> 00:41:32.960 align:middle line:90% IRVING WLADAWSKY-BERGER: OK. 00:41:32.960 --> 00:41:36.080 align:middle line:90% AUDIENCE: It's interesting. 00:41:36.080 --> 00:41:38.420 align:middle line:84% Like you mentioned, everything is shared. 00:41:38.420 --> 00:41:40.820 align:middle line:84% It's a partnership in the true sense. 00:41:40.820 --> 00:41:44.660 align:middle line:84% We share defect density, everything. 00:41:44.660 --> 00:41:47.980 align:middle line:84% But you also get information from your partner. 00:41:47.980 --> 00:41:52.300 align:middle line:84% So you know, like, half the worth of fabs, 00:41:52.300 --> 00:41:53.560 align:middle line:90% where they're [INAUDIBLE] 00:41:53.560 --> 00:41:54.600 align:middle line:84% IRVING WLADAWSKY-BERGER: No, that's very good. 00:41:54.600 --> 00:41:56.183 align:middle line:84% AUDIENCE: So there is shared learning. 00:41:56.183 --> 00:41:56.900 align:middle line:90% That's one. 00:41:56.900 --> 00:41:58.900 align:middle line:84% The cost of developing a technology right 00:41:58.900 --> 00:41:59.800 align:middle line:90% now is $1 billion. 00:41:59.800 --> 00:42:01.740 align:middle line:84% It's going to $2 billion for every node. 00:42:01.740 --> 00:42:03.413 align:middle line:84% And the cost of a fab is doubling. 00:42:03.413 --> 00:42:04.080 align:middle line:90% It's $2 billion. 00:42:04.080 --> 00:42:05.440 align:middle line:90% Now, it's going to $4 billion. 00:42:05.440 --> 00:42:10.220 align:middle line:84% So there is no way we can even be in this market sustainably 00:42:10.220 --> 00:42:11.860 align:middle line:90% without that. 00:42:11.860 --> 00:42:17.060 align:middle line:84% And thirdly, we can build on businesses which normally 00:42:17.060 --> 00:42:19.780 align:middle line:84% would not have chosen us, would have chosen TSMC, 00:42:19.780 --> 00:42:21.840 align:middle line:84% because we just don't have the capacity. 00:42:21.840 --> 00:42:23.940 align:middle line:84% So a lot of businesses, for example, 00:42:23.940 --> 00:42:26.620 align:middle line:84% Xbox with Christmas, you have nothing 00:42:26.620 --> 00:42:29.137 align:middle line:84% until November, then November, December, January, 00:42:29.137 --> 00:42:31.220 align:middle line:84% there is this big peak, and then there is nothing. 00:42:31.220 --> 00:42:34.060 align:middle line:84% And so if you are delayed by a month or so, 00:42:34.060 --> 00:42:36.380 align:middle line:84% so you have to have multiple segments. 00:42:36.380 --> 00:42:38.130 align:middle line:84% IRVING WLADAWSKY-BERGER: That's very good. 00:42:38.130 --> 00:42:39.780 align:middle line:84% AUDIENCE: And the fact that we have, 00:42:39.780 --> 00:42:45.100 align:middle line:84% all these virtual fabs helps us compete with TSMC. 00:42:45.100 --> 00:42:47.080 align:middle line:84% So we got our foot in the door, and a lot 00:42:47.080 --> 00:42:50.600 align:middle line:84% of businesses which would not normally have considered us. 00:42:50.600 --> 00:42:54.180 align:middle line:84% And we have some similar cost structures. 00:42:54.180 --> 00:42:56.920 align:middle line:84% But when we go to market, a lot of times 00:42:56.920 --> 00:43:00.520 align:middle line:84% we have an order desk, which is one of these fabs. 00:43:00.520 --> 00:43:06.320 align:middle line:84% And you go and typically the firm that wins the business 00:43:06.320 --> 00:43:08.580 align:middle line:90% has the right of first refusal. 00:43:08.580 --> 00:43:10.920 align:middle line:84% So if you cannot deal with it, then it can-- 00:43:10.920 --> 00:43:13.920 align:middle line:84% so a lot of times our partners will also win business, 00:43:13.920 --> 00:43:16.128 align:middle line:84% and they can't handle it because of some other order. 00:43:16.128 --> 00:43:17.420 align:middle line:90% IRVING WLADAWSKY-BERGER: I see. 00:43:17.420 --> 00:43:19.380 align:middle line:84% AUDIENCE: So there's a lot of synergy there. 00:43:19.380 --> 00:43:20.963 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, but you 00:43:20.963 --> 00:43:22.740 align:middle line:84% said a number of very good things, 00:43:22.740 --> 00:43:26.000 align:middle line:84% but the first one I should have said, which is you 00:43:26.000 --> 00:43:29.180 align:middle line:84% pick your partners very carefully. 00:43:29.180 --> 00:43:32.280 align:middle line:84% And let's say Samsung is a partner. 00:43:32.280 --> 00:43:36.080 align:middle line:84% Samsung is a wonderful company, as you know. 00:43:36.080 --> 00:43:38.660 align:middle line:84% So you can say, well, Samsung will learn from you. 00:43:38.660 --> 00:43:40.110 align:middle line:90% Well, guess what? 00:43:40.110 --> 00:43:41.790 align:middle line:90% We learn from Samsung. 00:43:41.790 --> 00:43:43.470 align:middle line:90% Sony is a partner. 00:43:43.470 --> 00:43:44.850 align:middle line:90% Philips is a partner. 00:43:44.850 --> 00:43:49.070 align:middle line:84% AUDIENCE: So we have [INAUDIBLE] and USTA, so IBM chartered, 00:43:49.070 --> 00:43:50.910 align:middle line:90% Samsung Infineon is one. 00:43:50.910 --> 00:43:55.573 align:middle line:84% And then AMD, Toshiba is [INAUDIBLE]. 00:43:55.573 --> 00:43:57.490 align:middle line:84% IRVING WLADAWSKY-BERGER: How about the Boeing, 00:43:57.490 --> 00:43:59.010 align:middle line:90% did they make that point also? 00:43:59.010 --> 00:44:00.610 align:middle line:90% So let's take the wing. 00:44:00.610 --> 00:44:06.790 align:middle line:84% The Japanese are superb engineers, the most incredibly 00:44:06.790 --> 00:44:08.870 align:middle line:90% disciplined engineers. 00:44:08.870 --> 00:44:13.000 align:middle line:84% So Boeing must be learning how the Japanese do these things. 00:44:13.000 --> 00:44:14.750 align:middle line:84% AUDIENCE: I think their big thing was they 00:44:14.750 --> 00:44:17.470 align:middle line:90% were trying to hit this date. 00:44:17.470 --> 00:44:20.650 align:middle line:84% So it was primarily time to market and capital investment. 00:44:20.650 --> 00:44:22.150 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, but I 00:44:22.150 --> 00:44:26.310 align:middle line:84% should have mentioned that learning from your partners 00:44:26.310 --> 00:44:28.110 align:middle line:90% is critical. 00:44:28.110 --> 00:44:32.190 align:middle line:84% And in fact, the reason I'm sticking to that's so important 00:44:32.190 --> 00:44:39.010 align:middle line:84% is because the worst thing you can do is to have, let me use, 00:44:39.010 --> 00:44:42.090 align:middle line:84% a colonial attitude toward your partner. 00:44:42.090 --> 00:44:48.290 align:middle line:84% Say, well, we need their money, and they don't know anything. 00:44:48.290 --> 00:44:50.650 align:middle line:90% What a mistake. 00:44:50.650 --> 00:44:52.530 align:middle line:90% Look at Detroit. 00:44:52.530 --> 00:44:55.390 align:middle line:84% So it's much better to say, oh my God, 00:44:55.390 --> 00:44:57.950 align:middle line:84% look what Samsung has done, which I mean, 00:44:57.950 --> 00:45:01.330 align:middle line:84% it's incredible what they have done. 00:45:01.330 --> 00:45:02.710 align:middle line:90% How did they do it? 00:45:02.710 --> 00:45:05.710 align:middle line:84% And now you learn from them and so on 00:45:05.710 --> 00:45:08.910 align:middle line:84% or now, more and more the Indian companies. 00:45:08.910 --> 00:45:12.610 align:middle line:90% Look at what they've been doing. 00:45:12.610 --> 00:45:14.070 align:middle line:90% So how do they do that? 00:45:14.070 --> 00:45:19.130 align:middle line:84% So the learning is [INAUDIBLE], but I cannot say enough. 00:45:19.130 --> 00:45:24.090 align:middle line:84% And this is especially important in global markets 00:45:24.090 --> 00:45:31.570 align:middle line:84% because often what happens, like, everybody in the US will 00:45:31.570 --> 00:45:34.210 align:middle line:84% more or less, they hire the same MBAs 00:45:34.210 --> 00:45:40.040 align:middle line:84% from the same Sloan and Harvard and Stanford, Wharton. 00:45:40.040 --> 00:45:42.360 align:middle line:84% So pretty much they look the same. 00:45:42.360 --> 00:45:47.920 align:middle line:84% And that's why often the real totally different things 00:45:47.920 --> 00:45:51.520 align:middle line:84% are being done by people someplace else, 00:45:51.520 --> 00:45:54.000 align:middle line:90% China and India and Japan. 00:45:54.000 --> 00:45:57.040 align:middle line:84% And they do the business differently. 00:45:57.040 --> 00:46:00.480 align:middle line:84% And you can say, well, but they're different. 00:46:00.480 --> 00:46:06.000 align:middle line:84% It's a lot they do that's much better than you forgot. 00:46:06.000 --> 00:46:10.440 align:middle line:84% And so in a global world, the learning 00:46:10.440 --> 00:46:15.360 align:middle line:84% by partnering with companies from very different regions 00:46:15.360 --> 00:46:19.120 align:middle line:84% is particularly important because they often 00:46:19.120 --> 00:46:24.040 align:middle line:84% bring tremendous management techniques that you never 00:46:24.040 --> 00:46:29.160 align:middle line:84% had before because that wasn't the Sloan way or the Harvard 00:46:29.160 --> 00:46:29.680 align:middle line:90% way. 00:46:29.680 --> 00:46:31.780 align:middle line:90% And now it should. 00:46:31.780 --> 00:46:37.060 align:middle line:90% 00:46:37.060 --> 00:46:42.240 align:middle line:84% So as Clay Christensen says, so if it fits your core, do it. 00:46:42.240 --> 00:46:44.620 align:middle line:90% Otherwise, do it outside. 00:46:44.620 --> 00:46:49.500 align:middle line:84% But if you read the chapter, he and Raynor 00:46:49.500 --> 00:46:54.060 align:middle line:84% make the point that don't assume you're 00:46:54.060 --> 00:46:56.620 align:middle line:90% making a decision forever. 00:46:56.620 --> 00:47:01.580 align:middle line:84% Because often you're making the decision now. 00:47:01.580 --> 00:47:04.340 align:middle line:84% But often markets will change again, 00:47:04.340 --> 00:47:06.440 align:middle line:84% and you may need to take it back. 00:47:06.440 --> 00:47:09.740 align:middle line:84% So make a decision for sure, but keep your mind 00:47:09.740 --> 00:47:15.940 align:middle line:84% open because you need to decide strategically. 00:47:15.940 --> 00:47:20.780 align:middle line:84% What that really says is if you-- 00:47:20.780 --> 00:47:21.740 align:middle line:90% what's the term? 00:47:21.740 --> 00:47:22.740 align:middle line:90% I forgot. 00:47:22.740 --> 00:47:27.860 align:middle line:84% When you lose all your skills in an area, what's the term? 00:47:27.860 --> 00:47:31.460 align:middle line:84% Deskill yourself, but there is another term. 00:47:31.460 --> 00:47:35.890 align:middle line:84% If you lose all your skills in an area 00:47:35.890 --> 00:47:40.690 align:middle line:84% because you have decided to outsource, 00:47:40.690 --> 00:47:43.370 align:middle line:84% be very careful because if in the future 00:47:43.370 --> 00:47:49.730 align:middle line:84% you may decide to re-enter it, you lost all your skills. 00:47:49.730 --> 00:47:51.730 align:middle line:84% And in fact, one of the things that 00:47:51.730 --> 00:47:55.210 align:middle line:84% often happens in competition is your competitor 00:47:55.210 --> 00:47:59.950 align:middle line:84% will want you to get rid of all your skills in an area, 00:47:59.950 --> 00:48:01.770 align:middle line:84% and they can even do that by selling 00:48:01.770 --> 00:48:05.630 align:middle line:84% you stuff very cheaply until you have no more skills. 00:48:05.630 --> 00:48:07.990 align:middle line:90% And then you're in trouble. 00:48:07.990 --> 00:48:12.830 align:middle line:84% So it's just something to watch out for. 00:48:12.830 --> 00:48:13.330 align:middle line:90% Please. 00:48:13.330 --> 00:48:16.650 align:middle line:84% AUDIENCE: [INAUDIBLE] show me that 00:48:16.650 --> 00:48:20.490 align:middle line:84% by how well your competency has gone 00:48:20.490 --> 00:48:26.090 align:middle line:84% for several hours every single day for probably for whoever 00:48:26.090 --> 00:48:28.250 align:middle line:84% doesn't need to have [INAUDIBLE]. 00:48:28.250 --> 00:48:34.590 align:middle line:90% 00:48:34.590 --> 00:48:42.790 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, but again, sometimes you totally 00:48:42.790 --> 00:48:44.670 align:middle line:90% outsource. 00:48:44.670 --> 00:48:46.730 align:middle line:84% Other times, like, let's say, BMW, 00:48:46.730 --> 00:48:52.530 align:middle line:84% I believe the X3 they don't manufacture, but the X5 they do. 00:48:52.530 --> 00:48:55.310 align:middle line:90% Those are SUVs. 00:48:55.310 --> 00:49:02.870 align:middle line:84% And so maybe there is a balance point. 00:49:02.870 --> 00:49:05.350 align:middle line:84% Now, for some things, you said, well, 00:49:05.350 --> 00:49:08.390 align:middle line:90% we outsource the cafeteria. 00:49:08.390 --> 00:49:10.390 align:middle line:90% Should we retain some chefs? 00:49:10.390 --> 00:49:15.390 align:middle line:84% Sometimes you would say, that is not a core competency. 00:49:15.390 --> 00:49:16.810 align:middle line:90% No, I'm being very serious. 00:49:16.810 --> 00:49:20.990 align:middle line:84% Once upon a time, companies were so well 00:49:20.990 --> 00:49:23.750 align:middle line:84% integrated that they had their own cafeterias 00:49:23.750 --> 00:49:25.530 align:middle line:90% and their own people. 00:49:25.530 --> 00:49:28.090 align:middle line:84% Most companies now have outsourced that, 00:49:28.090 --> 00:49:29.760 align:middle line:90% but they don't need to retain. 00:49:29.760 --> 00:49:32.420 align:middle line:84% They need people in procurement who 00:49:32.420 --> 00:49:37.120 align:middle line:84% understand how to negotiate with the best and so on, 00:49:37.120 --> 00:49:39.100 align:middle line:84% but otherwise, they don't need that. 00:49:39.100 --> 00:49:42.380 align:middle line:84% Manufacturing, marketing, you just 00:49:42.380 --> 00:49:48.100 align:middle line:84% have to be a little more careful because it's so core 00:49:48.100 --> 00:49:51.780 align:middle line:84% to so much of what you do that you have to be careful. 00:49:51.780 --> 00:49:57.380 align:middle line:84% And again, this is the warning that Raynor and Christensen 00:49:57.380 --> 00:50:01.020 align:middle line:84% give, that if you want to go back, 00:50:01.020 --> 00:50:04.740 align:middle line:84% and you just get rid of it all together, 00:50:04.740 --> 00:50:07.380 align:middle line:84% it's going to be very difficult to go back. 00:50:07.380 --> 00:50:09.160 align:middle line:84% And sometimes you do want to go back. 00:50:09.160 --> 00:50:13.800 align:middle line:84% Whereas, if you've retained it, at a certain level, 00:50:13.800 --> 00:50:17.300 align:middle line:84% it becomes easier to now expand that. 00:50:17.300 --> 00:50:19.200 align:middle line:90% Now, let's talk about-- 00:50:19.200 --> 00:50:22.740 align:middle line:90% 00:50:22.740 --> 00:50:25.900 align:middle line:90% OK, so we talked about this. 00:50:25.900 --> 00:50:33.120 align:middle line:84% The next major decisions, which it's really 00:50:33.120 --> 00:50:37.240 align:middle line:84% vertical integrated designs, proprietary 00:50:37.240 --> 00:50:42.560 align:middle line:84% vertically integrated designs versus modular designs. 00:50:42.560 --> 00:50:47.920 align:middle line:84% It's a huge, huge, huge decision. 00:50:47.920 --> 00:50:54.200 align:middle line:84% And then I thought chapter 5 had a very good discussion 00:50:54.200 --> 00:50:58.280 align:middle line:84% of vertical integration versus modular. 00:50:58.280 --> 00:51:04.580 align:middle line:84% And when do you use vertical integration, 00:51:04.580 --> 00:51:07.440 align:middle line:84% which interdependent architectures, 00:51:07.440 --> 00:51:11.040 align:middle line:90% and when do you use modular? 00:51:11.040 --> 00:51:18.040 align:middle line:84% And as you know, he gives a very good definition 00:51:18.040 --> 00:51:25.830 align:middle line:84% of vertically integrated ones and what you're optimizing for. 00:51:25.830 --> 00:51:28.930 align:middle line:84% But by definition of proprietary, 00:51:28.930 --> 00:51:31.590 align:middle line:84% you're committing yourself to pretty much do it 00:51:31.590 --> 00:51:41.630 align:middle line:84% all versus when it's modular, which is much more you can 00:51:41.630 --> 00:51:47.870 align:middle line:84% distribute it more than you are relying on other people more, 00:51:47.870 --> 00:51:51.550 align:middle line:84% becomes easier to decompose it in time to market, please. 00:51:51.550 --> 00:51:54.830 align:middle line:84% AUDIENCE: There's two things I would like to bring up. 00:51:54.830 --> 00:51:58.102 align:middle line:84% As I was reading this, the word performance jumped out at me. 00:51:58.102 --> 00:51:59.310 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 00:51:59.310 --> 00:52:01.590 align:middle line:84% AUDIENCE: And the reason why, if you ask anyone 00:52:01.590 --> 00:52:04.070 align:middle line:84% in this room their definition of performance 00:52:04.070 --> 00:52:05.390 align:middle line:90% is going to be different. 00:52:05.390 --> 00:52:10.070 align:middle line:84% So as far as performance, how does Christensen 00:52:10.070 --> 00:52:11.490 align:middle line:90% define performance. 00:52:11.490 --> 00:52:15.010 align:middle line:84% Because performance to me, I mean, I do a lot of graphics. 00:52:15.010 --> 00:52:18.457 align:middle line:84% So I need a graphics card that has much more memory. 00:52:18.457 --> 00:52:19.790 align:middle line:90% IRVING WLADAWSKY-BERGER: Mm-hmm. 00:52:19.790 --> 00:52:23.950 align:middle line:84% AUDIENCE: But the new types of performances, for instance, 00:52:23.950 --> 00:52:25.290 align:middle line:90% let's look at the iPhone. 00:52:25.290 --> 00:52:28.170 align:middle line:84% Someone can look at the iPhone and say, OK, this performs well 00:52:28.170 --> 00:52:30.490 align:middle line:84% because it has an interface, and it 00:52:30.490 --> 00:52:33.147 align:middle line:84% has something that drives the aesthetic value. 00:52:33.147 --> 00:52:34.730 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, I don't 00:52:34.730 --> 00:52:36.150 align:middle line:90% think that's what he meant. 00:52:36.150 --> 00:52:37.490 align:middle line:90% AUDIENCE: But I mean but it's-- 00:52:37.490 --> 00:52:39.970 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree violently with-- 00:52:39.970 --> 00:52:41.867 align:middle line:84% AUDIENCE: Performance is being redefined. 00:52:41.867 --> 00:52:43.450 align:middle line:84% IRVING WLADAWSKY-BERGER: That's right. 00:52:43.450 --> 00:52:44.410 align:middle line:90% That's right. 00:52:44.410 --> 00:52:47.150 align:middle line:84% AUDIENCE: So what are we targeting our market? 00:52:47.150 --> 00:52:48.790 align:middle line:84% I mean, what markets are we targeting? 00:52:48.790 --> 00:52:50.623 align:middle line:84% IRVING WLADAWSKY-BERGER: You're bringing up, 00:52:50.623 --> 00:52:52.070 align:middle line:90% and we'll discuss it much more. 00:52:52.070 --> 00:52:55.130 align:middle line:90% 00:52:55.130 --> 00:52:58.950 align:middle line:84% I think when Christensen is talking about performance, 00:52:58.950 --> 00:53:02.690 align:middle line:84% he's talking about classic product performance 00:53:02.690 --> 00:53:08.590 align:middle line:84% speed, reliability, and things like that. 00:53:08.590 --> 00:53:13.450 align:middle line:84% The kinds of performance where you need to carefully control 00:53:13.450 --> 00:53:14.870 align:middle line:90% different components. 00:53:14.870 --> 00:53:22.180 align:middle line:84% For example, if the iPod in order to run, 00:53:22.180 --> 00:53:31.120 align:middle line:84% needed a custom micro and custom memory and custom storage, 00:53:31.120 --> 00:53:36.920 align:middle line:84% it could have been, then whoever is designing the iPod 00:53:36.920 --> 00:53:40.740 align:middle line:84% probably had to design all these components at the same time. 00:53:40.740 --> 00:53:44.960 align:middle line:84% By the way, in the old days, that's how we did computers. 00:53:44.960 --> 00:53:50.040 align:middle line:84% The PC was probably the first time 00:53:50.040 --> 00:53:52.660 align:middle line:84% that those things got disaggregated. 00:53:52.660 --> 00:53:56.210 align:middle line:90% 00:53:56.210 --> 00:54:00.900 align:middle line:84% But before, you did minicomputers, you did it all. 00:54:00.900 --> 00:54:04.120 align:middle line:84% Everybody did it all, pretty much. 00:54:04.120 --> 00:54:09.360 align:middle line:84% Because the components were of a quality 00:54:09.360 --> 00:54:15.040 align:middle line:84% that you gave away too much performance and quality 00:54:15.040 --> 00:54:16.900 align:middle line:90% if you didn't do it yourself. 00:54:16.900 --> 00:54:21.100 align:middle line:84% And this stuff, because of the immaturity, 00:54:21.100 --> 00:54:23.660 align:middle line:90% you couldn't do that. 00:54:23.660 --> 00:54:28.640 align:middle line:84% But you're bringing out an incredibly important point, 00:54:28.640 --> 00:54:30.320 align:middle line:84% and we'll talk about it in a second, 00:54:30.320 --> 00:54:36.580 align:middle line:84% that once you get to the level of good enough components, 00:54:36.580 --> 00:54:39.160 align:middle line:84% other criteria need to take over. 00:54:39.160 --> 00:54:45.540 align:middle line:84% Otherwise, you're spending money on something 00:54:45.540 --> 00:54:50.020 align:middle line:84% that your clients could care less about. 00:54:50.020 --> 00:54:54.360 align:middle line:84% You can say, well, maps on the iPhone, take a second. 00:54:54.360 --> 00:54:59.260 align:middle line:84% And in my machine, take 0.2 seconds. 00:54:59.260 --> 00:55:03.780 align:middle line:84% But of course, you need to learn assembly language 00:55:03.780 --> 00:55:08.420 align:middle line:84% to deal with site maps, instead of whatever 00:55:08.420 --> 00:55:10.220 align:middle line:90% you do with the iPhone. 00:55:10.220 --> 00:55:12.260 align:middle line:90% That's a simple trade-off. 00:55:12.260 --> 00:55:15.540 align:middle line:84% AUDIENCE: And in addition to that, 00:55:15.540 --> 00:55:19.560 align:middle line:84% I guess he never explains or talks 00:55:19.560 --> 00:55:20.870 align:middle line:90% about the design component. 00:55:20.870 --> 00:55:22.620 align:middle line:84% IRVING WLADAWSKY-BERGER: No, I understand. 00:55:22.620 --> 00:55:24.520 align:middle line:84% AUDIENCE: Because if you talk about 00:55:24.520 --> 00:55:27.940 align:middle line:84% interdependent architecture and modular architecture, 00:55:27.940 --> 00:55:30.940 align:middle line:84% compromising a lot when you talk about modular architecture. 00:55:30.940 --> 00:55:34.300 align:middle line:90% You're compromising design. 00:55:34.300 --> 00:55:38.720 align:middle line:84% Because if you take modular out of the context of technology, 00:55:38.720 --> 00:55:41.340 align:middle line:84% let us look at modular in the context of architecture, 00:55:41.340 --> 00:55:42.120 align:middle line:90% for instance. 00:55:42.120 --> 00:55:46.060 align:middle line:84% When you talk about modular design in architecture, I mean, 00:55:46.060 --> 00:55:48.680 align:middle line:84% building architecture, you're compromising a lot as far 00:55:48.680 --> 00:55:51.160 align:middle line:90% as the aesthetics of it. 00:55:51.160 --> 00:55:52.650 align:middle line:90% So modular compromise-- 00:55:52.650 --> 00:55:54.900 align:middle line:84% IRVING WLADAWSKY-BERGER: Wait, wait, wait, wait, wait, 00:55:54.900 --> 00:56:00.420 align:middle line:84% and you're an architect, so you know a lot about this. 00:56:00.420 --> 00:56:01.600 align:middle line:90% AUDIENCE: Mm-hmm. 00:56:01.600 --> 00:56:02.975 align:middle line:84% IRVING WLADAWSKY-BERGER: I wonder 00:56:02.975 --> 00:56:05.880 align:middle line:84% if it depends what we mean by modular? 00:56:05.880 --> 00:56:08.320 align:middle line:90% Because-- 00:56:08.320 --> 00:56:10.240 align:middle line:84% AUDIENCE: It's component architecture 00:56:10.240 --> 00:56:12.740 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, yeah, yeah, but no, no, but I 00:56:12.740 --> 00:56:18.890 align:middle line:90% mean is let me go to an extreme. 00:56:18.890 --> 00:56:21.390 align:middle line:90% So let's take the Pru Tower. 00:56:21.390 --> 00:56:25.790 align:middle line:84% And if I ask you how much of what goes into the Pru Tower 00:56:25.790 --> 00:56:33.390 align:middle line:84% are standard things in the building ecosystem. 00:56:33.390 --> 00:56:38.970 align:middle line:84% Steel and bricks, all that is standard, am I correct? 00:56:38.970 --> 00:56:39.803 align:middle line:90% AUDIENCE: Yes it is. 00:56:39.803 --> 00:56:41.678 align:middle line:84% IRVING WLADAWSKY-BERGER: So no, no, but those 00:56:41.678 --> 00:56:42.690 align:middle line:90% are modular components. 00:56:42.690 --> 00:56:44.350 align:middle line:84% Don't underestimate-- so that's why 00:56:44.350 --> 00:56:45.830 align:middle line:90% I say it depends what we mean. 00:56:45.830 --> 00:56:55.510 align:middle line:84% Now, I think modular components is 00:56:55.510 --> 00:56:58.310 align:middle line:84% what we call a fractal concept, I 00:56:58.310 --> 00:57:03.730 align:middle line:84% don't know if you all know about fractal, which is at any level, 00:57:03.730 --> 00:57:06.730 align:middle line:84% you look at it, you will find a similar problem. 00:57:06.730 --> 00:57:08.790 align:middle line:84% Remember, once upon a time, there 00:57:08.790 --> 00:57:11.310 align:middle line:84% may have been building contractors 00:57:11.310 --> 00:57:13.810 align:middle line:84% that had their own steel factories 00:57:13.810 --> 00:57:17.790 align:middle line:84% and did custom steal for their building. 00:57:17.790 --> 00:57:20.490 align:middle line:84% I bet you that happened once upon a time. 00:57:20.490 --> 00:57:24.770 align:middle line:84% Or they ask you a steel to do them a certain kind of steal. 00:57:24.770 --> 00:57:27.570 align:middle line:90% You no longer need to do that. 00:57:27.570 --> 00:57:33.450 align:middle line:84% But what would you say in a building 00:57:33.450 --> 00:57:39.232 align:middle line:84% like the Pru Center is custom versus standard components? 00:57:39.232 --> 00:57:41.190 align:middle line:84% AUDIENCE: As far as the design component of it? 00:57:41.190 --> 00:57:45.330 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, yeah, by design components, 00:57:45.330 --> 00:57:51.250 align:middle line:84% AUDIENCE: I mean you talk about glass and steel. 00:57:51.250 --> 00:57:53.690 align:middle line:84% Back to what you talk about component issues, 00:57:53.690 --> 00:57:55.947 align:middle line:84% if you boil it down to components, steel is steel. 00:57:55.947 --> 00:57:58.030 align:middle line:84% IRVING WLADAWSKY-BERGER: No, but that's important, 00:57:58.030 --> 00:57:59.493 align:middle line:90% and glass is glass, right? 00:57:59.493 --> 00:58:00.076 align:middle line:90% AUDIENCE: Yes. 00:58:00.076 --> 00:58:01.350 align:middle line:84% IRVING WLADAWSKY-BERGER: No, but that's very important. 00:58:01.350 --> 00:58:02.530 align:middle line:84% AUDIENCE: But it's how you cut the deal-- 00:58:02.530 --> 00:58:03.530 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree. 00:58:03.530 --> 00:58:04.470 align:middle line:90% No, no, I agree. 00:58:04.470 --> 00:58:08.330 align:middle line:90% 00:58:08.330 --> 00:58:17.680 align:middle line:84% My point, and this is incredibly important for businesses, 00:58:17.680 --> 00:58:23.760 align:middle line:84% I bet you, and I wish I had quantification of this, 00:58:23.760 --> 00:58:28.600 align:middle line:90% that 90% of what goes-- 00:58:28.600 --> 00:58:30.700 align:middle line:84% I don't know how much the Pru Center costs. 00:58:30.700 --> 00:58:33.080 align:middle line:84% Let's assume it costs $1 billion. 00:58:33.080 --> 00:58:37.880 align:middle line:84% I bet you between 80% and 90% of that cost 00:58:37.880 --> 00:58:41.000 align:middle line:84% was based on standard components, 00:58:41.000 --> 00:58:47.040 align:middle line:84% and the amount that is custom is 10% to 20%. 00:58:47.040 --> 00:58:48.760 align:middle line:90% That's my feeling. 00:58:48.760 --> 00:58:52.600 align:middle line:84% That in mature engineering disciplines, 00:58:52.600 --> 00:58:58.120 align:middle line:84% and what I'm saying applies to automobiles and so on, 00:58:58.120 --> 00:59:04.720 align:middle line:84% the majority of the pieces are components. 00:59:04.720 --> 00:59:09.680 align:middle line:84% And the design, how you assemble them 00:59:09.680 --> 00:59:13.140 align:middle line:84% and unique things that you do and unique things 00:59:13.140 --> 00:59:18.260 align:middle line:84% that you hire expert people to do for the first time for you 00:59:18.260 --> 00:59:21.160 align:middle line:84% is a relatively smaller percentage. 00:59:21.160 --> 00:59:22.480 align:middle line:90% That's my expectation. 00:59:22.480 --> 00:59:25.060 align:middle line:90% Do you agree with that? 00:59:25.060 --> 00:59:28.780 align:middle line:84% AUDIENCE: I guess my argument it still impacts the design result. 00:59:28.780 --> 00:59:30.280 align:middle line:84% IRVING WLADAWSKY-BERGER: It totally. 00:59:30.280 --> 00:59:33.480 align:middle line:84% AUDIENCE: And you compromise the design based on your components. 00:59:33.480 --> 00:59:37.700 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, yes, I agree, but remember, 00:59:37.700 --> 00:59:41.220 align:middle line:84% you also have a budget and a time to market. 00:59:41.220 --> 00:59:43.900 align:middle line:84% Just a second, just a second, we'll take everybody. 00:59:43.900 --> 00:59:48.680 align:middle line:84% Because no, this is an incredibly important discussion. 00:59:48.680 --> 00:59:52.520 align:middle line:90% Incredibly important because. 00:59:52.520 --> 00:59:56.300 align:middle line:90% 00:59:56.300 --> 01:00:02.700 align:middle line:84% I believe-- let's say, I believe that every discipline 01:00:02.700 --> 01:00:08.810 align:middle line:84% starts doing their own steel and glass. 01:00:08.810 --> 01:00:11.610 align:middle line:84% And then it takes forever to do that. 01:00:11.610 --> 01:00:14.290 align:middle line:90% 01:00:14.290 --> 01:00:19.090 align:middle line:90% Cathedrals were pretty custom. 01:00:19.090 --> 01:00:24.170 align:middle line:84% But eventually it becomes an engineering discipline, 01:00:24.170 --> 01:00:27.330 align:middle line:84% and part of becoming an engineering discipline, one 01:00:27.330 --> 01:00:32.190 align:middle line:84% of the major pieces is the notion of modular components. 01:00:32.190 --> 01:00:34.730 align:middle line:84% I think the Industrial Revolution, 01:00:34.730 --> 01:00:37.890 align:middle line:84% a major step was the introduction 01:00:37.890 --> 01:00:40.090 align:middle line:90% of standard components. 01:00:40.090 --> 01:00:42.730 align:middle line:84% You all know what I'm talking about with the, I think, 01:00:42.730 --> 01:00:48.250 align:middle line:84% it may have been the Winchester rifles that could be assembled? 01:00:48.250 --> 01:00:55.570 align:middle line:84% And if you look at mechanical engineering, civil engineering, 01:00:55.570 --> 01:00:59.970 align:middle line:84% they all start developing a discipline 01:00:59.970 --> 01:01:03.150 align:middle line:84% based on standards and standard components. 01:01:03.150 --> 01:01:07.710 align:middle line:84% And then the tremendous uniqueness 01:01:07.710 --> 01:01:11.130 align:middle line:90% is design at multiple levels. 01:01:11.130 --> 01:01:14.190 align:middle line:84% So I'm agreeing violently with you 01:01:14.190 --> 01:01:17.250 align:middle line:84% that it's no longer in the components. 01:01:17.250 --> 01:01:19.470 align:middle line:90% It's in the design. 01:01:19.470 --> 01:01:22.310 align:middle line:90% That's what I'm saying. 01:01:22.310 --> 01:01:28.390 align:middle line:84% Immature disciplines spend too much time in the components 01:01:28.390 --> 01:01:30.230 align:middle line:90% and not enough in the design. 01:01:30.230 --> 01:01:32.390 align:middle line:90% That's my feeling. 01:01:32.390 --> 01:01:34.850 align:middle line:84% And the reason I find this so fascinating, 01:01:34.850 --> 01:01:37.390 align:middle line:84% and then we'll get back to all the discussions, 01:01:37.390 --> 01:01:43.570 align:middle line:84% is that if you look at business as an engineering discipline, 01:01:43.570 --> 01:01:49.030 align:middle line:84% which I know, people, we haven't, it is perhaps 01:01:49.030 --> 01:01:51.510 align:middle line:90% the most immature form. 01:01:51.510 --> 01:01:55.350 align:middle line:84% But you notice that every goddamn company 01:01:55.350 --> 01:01:59.630 align:middle line:84% reinvents the processes for running the business. 01:01:59.630 --> 01:02:01.330 align:middle line:84% And you say, well, why do you do that? 01:02:01.330 --> 01:02:03.450 align:middle line:90% Well, because that's a business. 01:02:03.450 --> 01:02:08.880 align:middle line:84% You cannot build this business using somebody else's steel. 01:02:08.880 --> 01:02:10.260 align:middle line:90% Now that's changing. 01:02:10.260 --> 01:02:12.080 align:middle line:84% That's where business process out. 01:02:12.080 --> 01:02:13.720 align:middle line:90% It's beginning to change. 01:02:13.720 --> 01:02:16.540 align:middle line:90% But boy, is it-- 01:02:16.540 --> 01:02:21.340 align:middle line:84% I view that to be, this is my opinion, 01:02:21.340 --> 01:02:25.380 align:middle line:84% the mother of all disruptive innovations in the 21st century 01:02:25.380 --> 01:02:31.240 align:middle line:84% is how business is transforming, becoming more distributed, 01:02:31.240 --> 01:02:39.420 align:middle line:84% more deconstructed, which then brings into business design, 01:02:39.420 --> 01:02:41.620 align:middle line:84% which is what you really want to have. 01:02:41.620 --> 01:02:44.600 align:middle line:84% And business design and business strategy, in my mind, 01:02:44.600 --> 01:02:46.140 align:middle line:90% are totally intertwined. 01:02:46.140 --> 01:02:51.540 align:middle line:84% So that's why I was probing so much, because I 01:02:51.540 --> 01:02:56.820 align:middle line:84% think design is really, really, really critical. 01:02:56.820 --> 01:02:58.240 align:middle line:90% It's really, really critical. 01:02:58.240 --> 01:03:01.180 align:middle line:84% In fact, I would say it's everything. 01:03:01.180 --> 01:03:04.820 align:middle line:84% Yes, let me ask Soon-Yu to comment first. 01:03:04.820 --> 01:03:07.960 align:middle line:84% AUDIENCE: I guess the first thing that you were trying 01:03:07.960 --> 01:03:09.760 align:middle line:84% to mention was that the trade-offs 01:03:09.760 --> 01:03:13.840 align:middle line:84% between modular design and interdependent design, which 01:03:13.840 --> 01:03:17.520 align:middle line:84% I think you guys are mostly aware of, you guys, 01:03:17.520 --> 01:03:19.160 align:middle line:90% from a system dynamics. 01:03:19.160 --> 01:03:20.960 align:middle line:84% And what [INAUDIBLE] brought about 01:03:20.960 --> 01:03:25.600 align:middle line:84% was the interaction or trade-off between the design, which 01:03:25.600 --> 01:03:27.600 align:middle line:84% is more like an aesthetic design, 01:03:27.600 --> 01:03:30.020 align:middle line:90% and then the architecture of it. 01:03:30.020 --> 01:03:34.080 align:middle line:84% So I think, from my experience, I talked to a lot of people 01:03:34.080 --> 01:03:38.640 align:middle line:84% from Samsung and LG and these Korean electronics companies, 01:03:38.640 --> 01:03:42.340 align:middle line:84% which is a similar structure with Apple. 01:03:42.340 --> 01:03:44.900 align:middle line:84% So they're making this electronic stuff, 01:03:44.900 --> 01:03:48.440 align:middle line:84% which I think the core of the competition 01:03:48.440 --> 01:03:51.360 align:middle line:84% has shifted from engineering problem solving 01:03:51.360 --> 01:03:52.860 align:middle line:90% into more like a design-- 01:03:52.860 --> 01:03:54.000 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree, violently. 01:03:54.000 --> 01:03:54.940 align:middle line:90% AUDIENCE: --aspect. 01:03:54.940 --> 01:04:00.790 align:middle line:84% What they changed was that these industrial designers, who 01:04:00.790 --> 01:04:04.150 align:middle line:84% design the form factor, not the function factor, 01:04:04.150 --> 01:04:07.550 align:middle line:84% they are getting more and more and more 01:04:07.550 --> 01:04:09.210 align:middle line:84% importance in their organization. 01:04:09.210 --> 01:04:10.418 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 01:04:10.418 --> 01:04:13.030 align:middle line:84% AUDIENCE: So what they do is they usually-- 01:04:13.030 --> 01:04:16.310 align:middle line:84% 10 years before, LG and Samsung didn't really 01:04:16.310 --> 01:04:19.350 align:middle line:84% concern about this aesthetic aspect. 01:04:19.350 --> 01:04:21.430 align:middle line:84% So what they did was their research 01:04:21.430 --> 01:04:22.472 align:middle line:90% centers and the product-- 01:04:22.472 --> 01:04:24.180 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree totally. 01:04:24.180 --> 01:04:25.725 align:middle line:84% AUDIENCE: They started the initiation 01:04:25.725 --> 01:04:26.850 align:middle line:90% of the product development. 01:04:26.850 --> 01:04:27.470 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree. 01:04:27.470 --> 01:04:29.710 align:middle line:84% AUDIENCE: So they came up with nice functionalities 01:04:29.710 --> 01:04:31.650 align:middle line:84% and features and things like that. 01:04:31.650 --> 01:04:33.930 align:middle line:84% And then they thought about how to wrap it up. 01:04:33.930 --> 01:04:37.250 align:middle line:84% But now they are getting more of these industrial designers. 01:04:37.250 --> 01:04:38.630 align:middle line:84% They are getting more importance. 01:04:38.630 --> 01:04:40.410 align:middle line:84% IRVING WLADAWSKY-BERGER: I couldn't agree with you more. 01:04:40.410 --> 01:04:40.940 align:middle line:90% In fact-- 01:04:40.940 --> 01:04:42.690 align:middle line:84% AUDIENCE: It's the same for Apple as well. 01:04:42.690 --> 01:04:45.150 align:middle line:84% IRVING WLADAWSKY-BERGER: In fact, let me tell you, 01:04:45.150 --> 01:04:48.630 align:middle line:84% one of the projects that I've seen recently 01:04:48.630 --> 01:04:53.910 align:middle line:84% that I'm most excited about is a project in the UK 01:04:53.910 --> 01:05:02.170 align:middle line:84% called Design London that Gordon Brown has initiated. 01:05:02.170 --> 01:05:04.690 align:middle line:84% He initiated when he was Chancellor of the Exchequer. 01:05:04.690 --> 01:05:10.090 align:middle line:84% And I'm involved with Imperial College, and I was there in July 01:05:10.090 --> 01:05:11.690 align:middle line:90% and learned about it. 01:05:11.690 --> 01:05:14.330 align:middle line:84% Now, Design London is a partnership 01:05:14.330 --> 01:05:16.610 align:middle line:90% of three institutions. 01:05:16.610 --> 01:05:20.610 align:middle line:84% Two are what you would expect Imperial College 01:05:20.610 --> 01:05:24.250 align:middle line:84% School of Engineering, [INAUDIBLE] College School 01:05:24.250 --> 01:05:25.730 align:middle line:90% of Management. 01:05:25.730 --> 01:05:27.370 align:middle line:90% That's what you would expect. 01:05:27.370 --> 01:05:32.030 align:middle line:84% The third is the RCA, the Royal College of Arts, 01:05:32.030 --> 01:05:34.290 align:middle line:84% which I'm sure you would say, it's 01:05:34.290 --> 01:05:37.190 align:middle line:84% one of the world's best design schools. 01:05:37.190 --> 01:05:39.210 align:middle line:90% Am I correct? 01:05:39.210 --> 01:05:42.810 align:middle line:84% So the fact that they are an equal partner 01:05:42.810 --> 01:05:46.210 align:middle line:84% is what makes this so innovative that they 01:05:46.210 --> 01:05:50.530 align:middle line:84% have a school of design equal partner with engineering 01:05:50.530 --> 01:05:55.590 align:middle line:84% and management precisely for all the reasons we're discussing. 01:05:55.590 --> 01:05:57.110 align:middle line:84% Yeah, you had a comment in the back? 01:05:57.110 --> 01:05:57.900 align:middle line:90% Yes, please. 01:05:57.900 --> 01:06:00.440 align:middle line:84% AUDIENCE: Yeah, I was going to find 01:06:00.440 --> 01:06:04.440 align:middle line:84% an analogy in civil engineering and software, excuse me. 01:06:04.440 --> 01:06:07.080 align:middle line:84% Civil engineering, about 20, 30 years back, bridges 01:06:07.080 --> 01:06:09.820 align:middle line:90% were created on site, in situ. 01:06:09.820 --> 01:06:14.080 align:middle line:84% And recently, you've seen that prefabricated and prestressed 01:06:14.080 --> 01:06:19.440 align:middle line:84% concrete arrangements and panels have increased the construction 01:06:19.440 --> 01:06:21.840 align:middle line:84% of bridge time to maybe less than a year, which 01:06:21.840 --> 01:06:23.160 align:middle line:90% is really fabulous. 01:06:23.160 --> 01:06:26.160 align:middle line:84% Now, even the aesthetics of prefabricated bridges 01:06:26.160 --> 01:06:27.760 align:middle line:90% are kind of better. 01:06:27.760 --> 01:06:28.360 align:middle line:90% That's one. 01:06:28.360 --> 01:06:33.960 align:middle line:84% Second advantage is since the prefabricated panels are created 01:06:33.960 --> 01:06:36.120 align:middle line:84% in industries and not on site, you 01:06:36.120 --> 01:06:38.080 align:middle line:84% have better advanced composite materials 01:06:38.080 --> 01:06:39.620 align:middle line:90% that are used to create them. 01:06:39.620 --> 01:06:40.000 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 01:06:40.000 --> 01:06:42.380 align:middle line:84% AUDIENCE: So from a reliability perspective, from-- 01:06:42.380 --> 01:06:43.140 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, perfect. 01:06:43.140 --> 01:06:43.945 align:middle line:90% I understand. 01:06:43.945 --> 01:06:45.320 align:middle line:84% AUDIENCE: It's a perfect scenario 01:06:45.320 --> 01:06:47.660 align:middle line:84% where things have changed drastically over decades. 01:06:47.660 --> 01:06:48.940 align:middle line:84% IRVING WLADAWSKY-BERGER: Yes that's perfect. 01:06:48.940 --> 01:06:51.520 align:middle line:84% AUDIENCE: If you compare this to software architecture, 01:06:51.520 --> 01:06:53.100 align:middle line:90% open source has come up a lot. 01:06:53.100 --> 01:06:56.820 align:middle line:84% And it has allowed people to work in smaller groups 01:06:56.820 --> 01:07:00.120 align:middle line:84% and create smaller products and then model products. 01:07:00.120 --> 01:07:03.020 align:middle line:84% And if you interface them and create a larger product, 01:07:03.020 --> 01:07:05.700 align:middle line:84% you have a quicker time to market. 01:07:05.700 --> 01:07:09.140 align:middle line:84% You have better reliable products because each of them 01:07:09.140 --> 01:07:11.500 align:middle line:84% has been tested and you have costs. 01:07:11.500 --> 01:07:13.500 align:middle line:84% IRVING WLADAWSKY-BERGER: But you also brought up 01:07:13.500 --> 01:07:18.620 align:middle line:84% something very important, which is comparative advantage. 01:07:18.620 --> 01:07:22.460 align:middle line:84% Who did that, Schumpeter, who talked about comparative? 01:07:22.460 --> 01:07:23.780 align:middle line:90% AUDIENCE: Porter. 01:07:23.780 --> 01:07:25.420 align:middle line:84% IRVING WLADAWSKY-BERGER: Porter also, 01:07:25.420 --> 01:07:29.540 align:middle line:84% which is once you can deconstruct it-- 01:07:29.540 --> 01:07:31.580 align:middle line:90% oh, I'm sorry. 01:07:31.580 --> 01:07:33.500 align:middle line:84% The one who first probably brought it up 01:07:33.500 --> 01:07:38.300 align:middle line:84% was Adam Smith in The Wealth of Nations, which 01:07:38.300 --> 01:07:41.880 align:middle line:90% is that you start specializing. 01:07:41.880 --> 01:07:46.660 align:middle line:84% So to your point, if you're a bridge builder 01:07:46.660 --> 01:07:51.140 align:middle line:84% and you are also building steel, or cables, whatever, 01:07:51.140 --> 01:07:57.770 align:middle line:84% you may be a superb bridge builder, but a medium-rate steel 01:07:57.770 --> 01:07:59.250 align:middle line:90% builder. 01:07:59.250 --> 01:08:02.530 align:middle line:84% But if you are working with another company that 01:08:02.530 --> 01:08:07.730 align:middle line:84% only does steel, they may be world class in steel. 01:08:07.730 --> 01:08:12.330 align:middle line:84% So now if you bring your design skills, your project management 01:08:12.330 --> 01:08:18.970 align:middle line:84% skills with their composite whatever, 01:08:18.970 --> 01:08:21.830 align:middle line:90% the world's quality goes up. 01:08:21.830 --> 01:08:25.729 align:middle line:84% Adam Smith actually first formulated this principle 01:08:25.729 --> 01:08:27.790 align:middle line:90% in The Wealth of Nations. 01:08:27.790 --> 01:08:32.970 align:middle line:84% And then through the ages, Rowland, Coles, and Schumpeter 01:08:32.970 --> 01:08:36.910 align:middle line:84% and others, which is very, very important. 01:08:36.910 --> 01:08:39.550 align:middle line:84% And that's why if somebody said, well, 01:08:39.550 --> 01:08:44.649 align:middle line:84% I prefer to compete this way, the penalties eventually 01:08:44.649 --> 01:08:46.870 align:middle line:90% become extraordinary. 01:08:46.870 --> 01:08:50.069 align:middle line:84% So you have a good advantage initially, but eventually, 01:08:50.069 --> 01:08:55.569 align:middle line:84% as you pointed out, it's very difficult. You had a-- 01:08:55.569 --> 01:09:00.069 align:middle line:84% AUDIENCE: I guess it's a point about the architectural argument 01:09:00.069 --> 01:09:01.290 align:middle line:90% just then. 01:09:01.290 --> 01:09:04.430 align:middle line:90% We talked in our last class. 01:09:04.430 --> 01:09:06.310 align:middle line:90% This is our intention. 01:09:06.310 --> 01:09:09.430 align:middle line:90% We have an architect coming in. 01:09:09.430 --> 01:09:12.149 align:middle line:90% And he stresses a lot like how-- 01:09:12.149 --> 01:09:15.510 align:middle line:84% IRVING WLADAWSKY-BERGER: In the last class? 01:09:15.510 --> 01:09:21.350 align:middle line:84% AUDIENCE: System architecture, and what he talks about is how-- 01:09:21.350 --> 01:09:24.790 align:middle line:84% IRVING WLADAWSKY-BERGER: You mean like a civil engineer? 01:09:24.790 --> 01:09:27.250 align:middle line:84% AUDIENCE: Yeah it's a real architect who came in. 01:09:27.250 --> 01:09:29.649 align:middle line:90% IRVING WLADAWSKY-BERGER: OK. 01:09:29.649 --> 01:09:30.736 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:09:30.736 --> 01:09:32.069 align:middle line:90% IRVING WLADAWSKY-BERGER: OK, OK. 01:09:32.069 --> 01:09:33.652 align:middle line:84% AUDIENCE: So he talked a lot about how 01:09:33.652 --> 01:09:38.670 align:middle line:84% traditional architecture gives a form follows function. 01:09:38.670 --> 01:09:41.109 align:middle line:84% And you as an architect need to understand 01:09:41.109 --> 01:09:42.370 align:middle line:90% what kind of functions it is. 01:09:42.370 --> 01:09:43.029 align:middle line:90% [INAUDIBLE] 01:09:43.029 --> 01:09:44.410 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 01:09:44.410 --> 01:09:47.029 align:middle line:84% AUDIENCE: And that's the same for a lot of engineering stuff. 01:09:47.029 --> 01:09:48.737 align:middle line:84% IRVING WLADAWSKY-BERGER: Yes, absolutely. 01:09:48.737 --> 01:09:51.609 align:middle line:84% AUDIENCE: [INAUDIBLE] for in the [INAUDIBLE]. 01:09:51.609 --> 01:09:53.809 align:middle line:84% But right now, what he talks more about 01:09:53.809 --> 01:09:57.110 align:middle line:84% is there's an intertwined between form and function. 01:09:57.110 --> 01:09:58.662 align:middle line:90% You can't just do one. 01:09:58.662 --> 01:10:00.370 align:middle line:84% IRVING WLADAWSKY-BERGER: I agree totally. 01:10:00.370 --> 01:10:05.370 align:middle line:84% AUDIENCE: And in fact, when you look at the Christensen 01:10:05.370 --> 01:10:11.210 align:middle line:84% traditional Christensen, there's disruptive innovation and all 01:10:11.210 --> 01:10:13.530 align:middle line:84% this how performance keep going up. 01:10:13.530 --> 01:10:15.850 align:middle line:84% And then a lot of companies actually 01:10:15.850 --> 01:10:18.850 align:middle line:84% now start to have a surplus because now they 01:10:18.850 --> 01:10:20.070 align:middle line:90% reached a certain point. 01:10:20.070 --> 01:10:21.445 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly. 01:10:21.445 --> 01:10:24.030 align:middle line:84% AUDIENCE: And where do they compete and compete and fall. 01:10:24.030 --> 01:10:24.850 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly. 01:10:24.850 --> 01:10:26.950 align:middle line:84% AUDIENCE: Beauty rather than the functionality. 01:10:26.950 --> 01:10:30.530 align:middle line:84% IRVING WLADAWSKY-BERGER: No, I agree violently. 01:10:30.530 --> 01:10:31.590 align:middle line:90% Perfect, I agree. 01:10:31.590 --> 01:10:33.970 align:middle line:90% Let me take a-- please. 01:10:33.970 --> 01:10:37.330 align:middle line:84% AUDIENCE: I wanted to give another example of the trade-off 01:10:37.330 --> 01:10:40.690 align:middle line:84% that you talked about between performance and modular. 01:10:40.690 --> 01:10:44.130 align:middle line:84% So I used to work in chip design and specifically 01:10:44.130 --> 01:10:45.590 align:middle line:90% in application specific design. 01:10:45.590 --> 01:10:48.680 align:middle line:84% So different components that IBM designed, and we also designed 01:10:48.680 --> 01:10:52.160 align:middle line:90% our own memory designs. 01:10:52.160 --> 01:10:55.580 align:middle line:84% And it was all customized, so it was really high performance. 01:10:55.580 --> 01:10:58.800 align:middle line:84% One of the things we did is we changed the architecture 01:10:58.800 --> 01:11:01.840 align:middle line:84% of our system to put all these things together 01:11:01.840 --> 01:11:04.360 align:middle line:90% to get third-party IP. 01:11:04.360 --> 01:11:07.040 align:middle line:84% And when we made that change, we had the option 01:11:07.040 --> 01:11:10.320 align:middle line:84% of making it modular, and accepting 01:11:10.320 --> 01:11:14.040 align:middle line:84% third-party drastically reduced performance. 01:11:14.040 --> 01:11:19.025 align:middle line:84% But the fact was you could now have tens of companies making-- 01:11:19.025 --> 01:11:20.400 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly. 01:11:20.400 --> 01:11:22.880 align:middle line:84% AUDIENCE: And a lot of times, each of these designs 01:11:22.880 --> 01:11:25.740 align:middle line:84% are optimized for various applications. 01:11:25.740 --> 01:11:28.120 align:middle line:84% So while the putting together of the thing 01:11:28.120 --> 01:11:29.800 align:middle line:84% reduced the performance, the fact 01:11:29.800 --> 01:11:32.540 align:middle line:84% that these memories were optimized increased it. 01:11:32.540 --> 01:11:36.880 align:middle line:84% And plus we were able to go after a much broader section 01:11:36.880 --> 01:11:38.320 align:middle line:84% of the market, which we otherwise 01:11:38.320 --> 01:11:39.520 align:middle line:90% wouldn't have been able to. 01:11:39.520 --> 01:11:42.478 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, yeah, please. 01:11:42.478 --> 01:11:44.520 align:middle line:84% AUDIENCE: The last time, we talked about decision 01:11:44.520 --> 01:11:47.052 align:middle line:90% to enforce and outsource. 01:11:47.052 --> 01:11:48.260 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 01:11:48.260 --> 01:11:50.620 align:middle line:84% AUDIENCE: In this slide, we talk about factors. 01:11:50.620 --> 01:11:52.660 align:middle line:84% [INAUDIBLE] we choose to outsource. 01:11:52.660 --> 01:11:54.960 align:middle line:84% So then we have to use modern architecture. 01:11:54.960 --> 01:11:57.040 align:middle line:84% IRVING WLADAWSKY-BERGER: Yes, I would say that-- 01:11:57.040 --> 01:12:00.100 align:middle line:90% 01:12:00.100 --> 01:12:02.860 align:middle line:90% otherwise, it's really risky. 01:12:02.860 --> 01:12:07.840 align:middle line:84% Because if it's proprietary, but you're just contract-- 01:12:07.840 --> 01:12:11.540 align:middle line:84% otherwise, it's more contract hiring somebody. 01:12:11.540 --> 01:12:15.700 align:middle line:84% Whereas in fact, the modular architectures 01:12:15.700 --> 01:12:19.900 align:middle line:84% facilitate building an industry ecosystem 01:12:19.900 --> 01:12:25.500 align:middle line:84% because also most companies that you would outsource to, 01:12:25.500 --> 01:12:29.840 align:middle line:84% most companies don't want to be trapped by you. 01:12:29.840 --> 01:12:32.360 align:middle line:90% 01:12:32.360 --> 01:12:34.900 align:middle line:84% And so if it's standard, then they 01:12:34.900 --> 01:12:38.700 align:middle line:84% have multiple clients themselves. 01:12:38.700 --> 01:12:42.180 align:middle line:84% And especially if they are the best companies 01:12:42.180 --> 01:12:51.730 align:middle line:84% in the world in that, then those are some of the trade-offs. 01:12:51.730 --> 01:12:56.510 align:middle line:84% Let's talk a little bit about, again, just very quickly, 01:12:56.510 --> 01:13:00.450 align:middle line:84% I'm just summarizing some of what the chapter says 01:13:00.450 --> 01:13:10.130 align:middle line:84% that the proprietary vertical integrated 01:13:10.130 --> 01:13:15.910 align:middle line:84% systems or products, are clearly what you do in the early stages. 01:13:15.910 --> 01:13:17.630 align:middle line:84% You just don't know what you're doing. 01:13:17.630 --> 01:13:22.170 align:middle line:84% You haven't had the chance to deconstruct it yet. 01:13:22.170 --> 01:13:24.890 align:middle line:90% So that's what you need to do. 01:13:24.890 --> 01:13:31.930 align:middle line:84% And usually that's what lets you do the product. 01:13:31.930 --> 01:13:34.310 align:middle line:84% But you have to invest a lot yourself. 01:13:34.310 --> 01:13:35.930 align:middle line:90% You're taking a lot of risk. 01:13:35.930 --> 01:13:37.730 align:middle line:84% And it takes awhile because you need 01:13:37.730 --> 01:13:40.490 align:middle line:90% to do all these other things. 01:13:40.490 --> 01:13:44.270 align:middle line:84% And at that point, you are vertically integrated, 01:13:44.270 --> 01:13:47.950 align:middle line:90% and you control all the pieces. 01:13:47.950 --> 01:13:52.470 align:middle line:84% When you start moving to modular, open, flexible 01:13:52.470 --> 01:13:57.150 align:middle line:84% architecture, usually that requires a certain level 01:13:57.150 --> 01:14:00.990 align:middle line:84% of maturity because what you're saying 01:14:00.990 --> 01:14:08.090 align:middle line:84% is most of the components, not all, most are good enough. 01:14:08.090 --> 01:14:14.750 align:middle line:84% Which means, as you pointed out, if you keep making them-- 01:14:14.750 --> 01:14:17.470 align:middle line:84% if you say, well, no, no, screw the iPod. 01:14:17.470 --> 01:14:19.970 align:middle line:84% I'm going to keep doing a better micro. 01:14:19.970 --> 01:14:24.450 align:middle line:84% Well, people who buy iPods, the micro is good enough. 01:14:24.450 --> 01:14:28.970 align:middle line:84% Or for most laptops, the micro is good enough. 01:14:28.970 --> 01:14:33.110 align:middle line:84% So just continuing the emphasis on the micro, 01:14:33.110 --> 01:14:35.250 align:middle line:90% you are overshooting the market. 01:14:35.250 --> 01:14:36.950 align:middle line:84% So that's the point when you say, 01:14:36.950 --> 01:14:40.270 align:middle line:84% I can go with standard components 01:14:40.270 --> 01:14:44.940 align:middle line:84% and then things begin to disaggregate, 01:14:44.940 --> 01:14:47.120 align:middle line:84% and you start building ecosystems. 01:14:47.120 --> 01:14:51.680 align:middle line:84% And this is what most disciplines do. 01:14:51.680 --> 01:14:57.180 align:middle line:90% 01:14:57.180 --> 01:15:02.160 align:middle line:84% And now the question comes, which should you do? 01:15:02.160 --> 01:15:03.760 align:middle line:90% And it all depends. 01:15:03.760 --> 01:15:10.500 align:middle line:84% It's really the only answer, which is for the new things 01:15:10.500 --> 01:15:15.580 align:middle line:84% that you are innovating by yourself, the really new pieces, 01:15:15.580 --> 01:15:22.140 align:middle line:84% in architecture maybe an architect 01:15:22.140 --> 01:15:26.380 align:middle line:84% has an idea for an incredible new window, 01:15:26.380 --> 01:15:29.840 align:middle line:84% and they do the specs and get a company to do it for them. 01:15:29.840 --> 01:15:32.660 align:middle line:84% And for a while they are the only ones who have it. 01:15:32.660 --> 01:15:34.480 align:middle line:90% And that's a custom window. 01:15:34.480 --> 01:15:39.060 align:middle line:84% And that makes the whole building. 01:15:39.060 --> 01:15:41.060 align:middle line:90% A few of those are totally fine. 01:15:41.060 --> 01:15:42.080 align:middle line:90% In fact, a few-- 01:15:42.080 --> 01:15:42.980 align:middle line:90% people do that. 01:15:42.980 --> 01:15:49.200 align:middle line:84% Am I correct that you may have some absolutely things? 01:15:49.200 --> 01:15:55.200 align:middle line:84% But that's a market excess, and my toilets are also-- eventually 01:15:55.200 --> 01:15:58.520 align:middle line:90% your client runs out of money. 01:15:58.520 --> 01:16:01.040 align:middle line:84% You say, no, I need special toilets. 01:16:01.040 --> 01:16:04.520 align:middle line:90% So you have to pick. 01:16:04.520 --> 01:16:05.620 align:middle line:90% Am I not correct? 01:16:05.620 --> 01:16:10.000 align:middle line:84% There are a few things that may be the unique differentiators, 01:16:10.000 --> 01:16:13.680 align:middle line:84% and the rest, you get from your ecosystem. 01:16:13.680 --> 01:16:18.120 align:middle line:84% And I'm assuming if you're a really good contractor, 01:16:18.120 --> 01:16:22.560 align:middle line:84% or a really good, let's say, Toyota is famous for this, 01:16:22.560 --> 01:16:26.720 align:middle line:84% your partners are so good that they bring you 01:16:26.720 --> 01:16:29.680 align:middle line:90% their innovations. 01:16:29.680 --> 01:16:31.220 align:middle line:84% So you don't need to do new toilets. 01:16:31.220 --> 01:16:35.080 align:middle line:84% Whoever those toilets for you brings you 01:16:35.080 --> 01:16:36.740 align:middle line:90% the best toilets in the world. 01:16:36.740 --> 01:16:38.605 align:middle line:84% So that's what you're using now, please. 01:16:38.605 --> 01:16:40.230 align:middle line:84% AUDIENCE: For example, I think actually 01:16:40.230 --> 01:16:43.310 align:middle line:90% Steve Jobs working on the new-- 01:16:43.310 --> 01:16:47.470 align:middle line:84% I mean, the store in Manhattan on Fifth Avenue 01:16:47.470 --> 01:16:51.523 align:middle line:84% redefined the windows system and patented it. 01:16:51.523 --> 01:16:53.190 align:middle line:84% IRVING WLADAWSKY-BERGER: Redefined the-- 01:16:53.190 --> 01:16:55.050 align:middle line:84% AUDIENCE: The windows system that he had, 01:16:55.050 --> 01:16:57.155 align:middle line:84% the caulking was actually outside. 01:16:57.155 --> 01:16:59.530 align:middle line:84% IRVING WLADAWSKY-BERGER: You don't mean redefine Windows? 01:16:59.530 --> 01:17:01.050 align:middle line:84% AUDIENCE: No, the windows system. 01:17:01.050 --> 01:17:02.575 align:middle line:90% The window-- 01:17:02.575 --> 01:17:04.450 align:middle line:84% IRVING WLADAWSKY-BERGER: Wow, that would be-- 01:17:04.450 --> 01:17:07.127 align:middle line:84% AUDIENCE: No, the actual physical windows system. 01:17:07.127 --> 01:17:08.710 align:middle line:84% IRVING WLADAWSKY-BERGER: Of the store? 01:17:08.710 --> 01:17:09.510 align:middle line:90% AUDIENCE: Of the store. 01:17:09.510 --> 01:17:11.343 align:middle line:84% IRVING WLADAWSKY-BERGER: I never heard that. 01:17:11.343 --> 01:17:12.350 align:middle line:90% AUDIENCE: Yeah. 01:17:12.350 --> 01:17:13.830 align:middle line:90% IRVING WLADAWSKY-BERGER: How so? 01:17:13.830 --> 01:17:15.288 align:middle line:84% AUDIENCE: The caulking was actually 01:17:15.288 --> 01:17:16.770 align:middle line:84% outside and not concealed enough, 01:17:16.770 --> 01:17:19.430 align:middle line:84% and he did not want it to be seen from the outside. 01:17:19.430 --> 01:17:22.570 align:middle line:84% So he said, hey, you have to get rid of this caulking. 01:17:22.570 --> 01:17:24.430 align:middle line:84% You have to do something about it. 01:17:24.430 --> 01:17:26.190 align:middle line:84% Why don't you just put it inside? 01:17:26.190 --> 01:17:29.630 align:middle line:84% And he's like, OK, you cannot do that because it's a rain 01:17:29.630 --> 01:17:30.510 align:middle line:90% sealant. 01:17:30.510 --> 01:17:32.530 align:middle line:84% But he said, you have to come up with something. 01:17:32.530 --> 01:17:36.070 align:middle line:84% So he got people to do it, and it redefined how it looked. 01:17:36.070 --> 01:17:39.170 align:middle line:84% And now you cannot see the sealant from outside. 01:17:39.170 --> 01:17:43.590 align:middle line:84% So it's pushing the envelope based on the design factor. 01:17:43.590 --> 01:17:45.610 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, yeah. 01:17:45.610 --> 01:17:52.770 align:middle line:84% and then other people concentrate 01:17:52.770 --> 01:17:55.810 align:middle line:90% on the architectural dimension. 01:17:55.810 --> 01:18:01.210 align:middle line:90% And this gets back to-- 01:18:01.210 --> 01:18:04.570 align:middle line:84% in my mind, it all gets back to design in my mind. 01:18:04.570 --> 01:18:10.570 align:middle line:84% That the reason this is so important 01:18:10.570 --> 01:18:16.130 align:middle line:84% is because I'm convinced that in the knowledge age 01:18:16.130 --> 01:18:20.490 align:middle line:84% we're in now, because the components we have to work with 01:18:20.490 --> 01:18:25.090 align:middle line:84% are so good, the incredible competition 01:18:25.090 --> 01:18:28.610 align:middle line:90% is design and its form. 01:18:28.610 --> 01:18:32.370 align:middle line:84% But now this is what gets so interesting. 01:18:32.370 --> 01:18:39.760 align:middle line:84% Classically, we apply design to physical things, 01:18:39.760 --> 01:18:44.120 align:middle line:84% whether it's an iPod or a BMW 3 Series. 01:18:44.120 --> 01:18:49.420 align:middle line:84% But how about, let's say, a customer service in a business? 01:18:49.420 --> 01:18:52.760 align:middle line:84% Could you think of customer service 01:18:52.760 --> 01:18:54.800 align:middle line:84% from the point of view of design? 01:18:54.800 --> 01:19:00.300 align:middle line:90% 01:19:00.300 --> 01:19:04.560 align:middle line:90% And apply similar criteria of-- 01:19:04.560 --> 01:19:07.100 align:middle line:90% 01:19:07.100 --> 01:19:09.880 align:middle line:84% I'm assuming one of the things in design 01:19:09.880 --> 01:19:14.320 align:middle line:84% have to be you put yourself in the shoes of the people looking 01:19:14.320 --> 01:19:15.960 align:middle line:90% at the design, correct? 01:19:15.960 --> 01:19:16.980 align:middle line:90% AUDIENCE: Yes. 01:19:16.980 --> 01:19:18.980 align:middle line:84% IRVING WLADAWSKY-BERGER: So in customer service, 01:19:18.980 --> 01:19:21.800 align:middle line:84% you put yourself in the shoes of the person receiving 01:19:21.800 --> 01:19:24.600 align:middle line:90% the customer service. 01:19:24.600 --> 01:19:28.520 align:middle line:84% And as we know, a lot of times when you do that, you say, 01:19:28.520 --> 01:19:32.180 align:middle line:84% it's amazing anybody does business with us 01:19:32.180 --> 01:19:37.140 align:middle line:90% because we do such a poor job. 01:19:37.140 --> 01:19:41.380 align:middle line:84% Now, the reason this is so important 01:19:41.380 --> 01:19:47.420 align:middle line:90% is because the industry-- 01:19:47.420 --> 01:19:52.420 align:middle line:84% I mean, again, if you think of business as a system, 01:19:52.420 --> 01:19:54.000 align:middle line:90% we're building cathedrals. 01:19:54.000 --> 01:19:55.880 align:middle line:90% I don't know how else to say it. 01:19:55.880 --> 01:20:00.820 align:middle line:84% We are in the cathedral stage of businesses. 01:20:00.820 --> 01:20:06.460 align:middle line:84% And the biggest opportunity that I see 01:20:06.460 --> 01:20:12.060 align:middle line:84% is the ability to start taking this into the future 01:20:12.060 --> 01:20:18.140 align:middle line:84% by systematically doing certain processes far 01:20:18.140 --> 01:20:21.900 align:middle line:90% better than anybody else. 01:20:21.900 --> 01:20:25.980 align:middle line:84% And now the reason we might have a chance 01:20:25.980 --> 01:20:33.770 align:middle line:84% to do that is because this notion of modularity 01:20:33.770 --> 01:20:38.450 align:middle line:84% are now beginning to show up in business processes. 01:20:38.450 --> 01:20:40.050 align:middle line:90% Before, you didn't have to-- 01:20:40.050 --> 01:20:42.530 align:middle line:84% remember, you needed the internet. 01:20:42.530 --> 01:20:45.430 align:middle line:90% You needed a SOA. 01:20:45.430 --> 01:20:48.730 align:middle line:84% There's a whole set of things that you 01:20:48.730 --> 01:20:51.970 align:middle line:84% needed to start getting a certain level 01:20:51.970 --> 01:20:55.210 align:middle line:84% of modular business components, but they 01:20:55.210 --> 01:20:58.730 align:middle line:90% are beginning to show up. 01:20:58.730 --> 01:21:03.210 align:middle line:84% And part of them showing up, just like with the iPod 01:21:03.210 --> 01:21:07.210 align:middle line:90% and with the Boeing 787. 01:21:07.210 --> 01:21:14.410 align:middle line:84% if you can free up the people, the master architects, business 01:21:14.410 --> 01:21:18.690 align:middle line:84% architects from having to do every goddamn thing, 01:21:18.690 --> 01:21:22.970 align:middle line:84% then they can start focusing on a few unique advantages 01:21:22.970 --> 01:21:24.190 align:middle line:90% and partner for the other. 01:21:24.190 --> 01:21:26.190 align:middle line:90% Remember how new this all is. 01:21:26.190 --> 01:21:29.570 align:middle line:90% This is barely happening now. 01:21:29.570 --> 01:21:32.990 align:middle line:84% So the opportunities for business 01:21:32.990 --> 01:21:37.210 align:middle line:84% to move from here to here, I think, are enormous. 01:21:37.210 --> 01:21:39.770 align:middle line:84% I don't know how you all feel about that, 01:21:39.770 --> 01:21:46.590 align:middle line:84% but I think that's-- in my mind, this is my feeling in the 21st 01:21:46.590 --> 01:21:49.750 align:middle line:90% century, that's the game. 01:21:49.750 --> 01:21:53.250 align:middle line:84% And before we take the break, let me give you one example. 01:21:53.250 --> 01:21:58.790 align:middle line:84% I think I already mentioned it once, but let me give it again, 01:21:58.790 --> 01:22:07.390 align:middle line:84% which is Sunil Mittal, who is the chief of Bharti. 01:22:07.390 --> 01:22:11.130 align:middle line:84% And Bharti now is a conglomerate. 01:22:11.130 --> 01:22:13.630 align:middle line:84% They are a big conglomerate, but they are also 01:22:13.630 --> 01:22:18.190 align:middle line:84% the biggest private sector cellular company in India 01:22:18.190 --> 01:22:21.470 align:middle line:90% growing by leaps and bounds. 01:22:21.470 --> 01:22:25.570 align:middle line:84% And Sunil Mittal said, I needed to grow fast. 01:22:25.570 --> 01:22:29.430 align:middle line:84% I didn't have enough energy to worry about all the other stuff, 01:22:29.430 --> 01:22:34.010 align:middle line:84% so he outsourced everything, except customer acquisition 01:22:34.010 --> 01:22:34.878 align:middle line:90% and retention. 01:22:34.878 --> 01:22:35.670 align:middle line:90% That's what he did. 01:22:35.670 --> 01:22:37.190 align:middle line:90% He said, that's it. 01:22:37.190 --> 01:22:39.870 align:middle line:90% That's my unique window. 01:22:39.870 --> 01:22:43.130 align:middle line:84% It's the customer acquisition and retention. 01:22:43.130 --> 01:22:47.650 align:middle line:84% And to do that, he got, I believe, Eriksson and somebody 01:22:47.650 --> 01:22:49.270 align:middle line:90% else to do the network. 01:22:49.270 --> 01:22:54.530 align:middle line:84% And IBM is doing the computers, and other people 01:22:54.530 --> 01:22:58.990 align:middle line:84% are doing other stuff, which he said, 01:22:58.990 --> 01:23:03.450 align:middle line:84% I heard him personally give a talk at an IBM CEO conference, 01:23:03.450 --> 01:23:06.730 align:middle line:84% he said, when he did this, everybody else in the telecom 01:23:06.730 --> 01:23:10.330 align:middle line:84% industry thought he was crazy, that how can you 01:23:10.330 --> 01:23:13.430 align:middle line:84% be a telco without owning your network? 01:23:13.430 --> 01:23:16.770 align:middle line:84% They thought, you don't know what you're doing. 01:23:16.770 --> 01:23:21.690 align:middle line:84% He said, well, the network is a commodity. 01:23:21.690 --> 01:23:24.720 align:middle line:90% Network is a commodity. 01:23:24.720 --> 01:23:27.760 align:middle line:84% Nobody had ever thought that you can 01:23:27.760 --> 01:23:31.100 align:middle line:84% be a telecommunications company and not own the network. 01:23:31.100 --> 01:23:33.720 align:middle line:90% Nobody ever thought of that. 01:23:33.720 --> 01:23:39.160 align:middle line:84% He reshaped the definition of a telecommunications company 01:23:39.160 --> 01:23:41.300 align:middle line:90% in terms of customers. 01:23:41.300 --> 01:23:46.460 align:middle line:84% He just went out and did it in terms of the customers. 01:23:46.460 --> 01:23:52.800 align:middle line:84% And would you all agree Sunil Mittal is very, very successful? 01:23:52.800 --> 01:23:58.800 align:middle line:84% And I think that's a very good example of what I expect to see 01:23:58.800 --> 01:24:05.400 align:middle line:84% more and more in business innovation in the 21st century, 01:24:05.400 --> 01:24:09.180 align:middle line:84% precisely because we're moving from here to there. 01:24:09.180 --> 01:24:16.240 align:middle line:84% Now, here is the problem, and here is where you all come 01:24:16.240 --> 01:24:30.780 align:middle line:84% in, which is don't forget that in order to design 787s, chips, 01:24:30.780 --> 01:24:36.940 align:middle line:84% buildings, bridges you use really sophisticated tools. 01:24:36.940 --> 01:24:41.940 align:middle line:84% You all know that EDA tools, CAD/CAM 01:24:41.940 --> 01:24:46.420 align:middle line:90% tools, all kinds of tools. 01:24:46.420 --> 01:24:52.700 align:middle line:84% And you do a lot of simulation, massive simulation. 01:24:52.700 --> 01:24:57.940 align:middle line:84% Where are the tools for business design and simulation 01:24:57.940 --> 01:24:59.360 align:middle line:90% of different strategies? 01:24:59.360 --> 01:25:01.060 align:middle line:90% Where are they? 01:25:01.060 --> 01:25:05.420 align:middle line:84% And the answer is somebody has to invent them. 01:25:05.420 --> 01:25:08.700 align:middle line:84% It's a little bit of work in that. 01:25:08.700 --> 01:25:11.860 align:middle line:84% So none of this can happen because if what 01:25:11.860 --> 01:25:15.260 align:middle line:84% you say, oh, now that it's modular, 01:25:15.260 --> 01:25:23.530 align:middle line:84% you'll have people with drafting paper, architecting fidelity. 01:25:23.530 --> 01:25:25.730 align:middle line:90% It doesn't work that way. 01:25:25.730 --> 01:25:30.850 align:middle line:84% We really need sophisticated tools. 01:25:30.850 --> 01:25:32.910 align:middle line:90% And that's all ahead of us. 01:25:32.910 --> 01:25:37.570 align:middle line:84% And, we'll talk more about it, but let's take a break now, 01:25:37.570 --> 01:25:40.830 align:middle line:84% and then we'll continue in 10 minutes. 01:25:40.830 --> 01:25:43.890 align:middle line:90% 01:25:43.890 --> 01:25:50.410 align:middle line:84% Before we get to, so what did we do in IBM in operationalizing 01:25:50.410 --> 01:25:53.930 align:middle line:84% a business and go through some of the items, 01:25:53.930 --> 01:26:02.770 align:middle line:84% let me talk about how I've been-- 01:26:02.770 --> 01:26:09.530 align:middle line:84% when I'm talking about some of the biggest changes going on 01:26:09.530 --> 01:26:17.650 align:middle line:84% in innovation, what I tell people, which very much is 01:26:17.650 --> 01:26:21.570 align:middle line:84% the evolution of systems, it's the way I think about it, 01:26:21.570 --> 01:26:23.350 align:middle line:84% but remember, I have a very expanded view 01:26:23.350 --> 01:26:28.430 align:middle line:84% of systems that are pushing us to have 01:26:28.430 --> 01:26:31.050 align:middle line:84% to keep moving into this dimension, 01:26:31.050 --> 01:26:36.030 align:middle line:84% the architectural-design dimension more and more. 01:26:36.030 --> 01:26:40.970 align:middle line:84% And let me talk about this in two critical dimensions. 01:26:40.970 --> 01:26:43.630 align:middle line:84% And then I'll tell you the third dimension that I usually 01:26:43.630 --> 01:26:47.590 align:middle line:90% add that I didn't include here. 01:26:47.590 --> 01:26:55.990 align:middle line:84% So one of the dimensions is breadth and scope. 01:26:55.990 --> 01:26:59.550 align:middle line:84% Now, let me tell you what I mean by that 01:26:59.550 --> 01:27:04.950 align:middle line:84% and illustrate it with IT systems. 01:27:04.950 --> 01:27:10.090 align:middle line:84% When I first started working with computers in the 1960s, 01:27:10.090 --> 01:27:15.150 align:middle line:84% I was first in college and then a graduate student in Chicago, 01:27:15.150 --> 01:27:20.180 align:middle line:84% a system was just the computer and the way it was connected 01:27:20.180 --> 01:27:21.400 align:middle line:90% to its I/O gear. 01:27:21.400 --> 01:27:24.060 align:middle line:90% That was the system. 01:27:24.060 --> 01:27:31.080 align:middle line:84% So it was very much just a piece of technology that was isolated. 01:27:31.080 --> 01:27:32.840 align:middle line:84% We never thought of it as isolated. 01:27:32.840 --> 01:27:34.860 align:middle line:84% We thought of it as wonderful new stuff 01:27:34.860 --> 01:27:37.380 align:middle line:84% because you can do a lot of calculations, 01:27:37.380 --> 01:27:40.620 align:middle line:84% but you brought your deck of cards. 01:27:40.620 --> 01:27:45.460 align:middle line:84% Then I usually went off to have a pizza downtown at Uno's, came 01:27:45.460 --> 01:27:48.260 align:middle line:84% back, got the results, remember, oh, God, I 01:27:48.260 --> 01:27:52.160 align:middle line:84% forgot a comma in Fortran, so had to resubmit it. 01:27:52.160 --> 01:27:57.240 align:middle line:90% 01:27:57.240 --> 01:27:58.760 align:middle line:90% You're too young to remember. 01:27:58.760 --> 01:28:01.060 align:middle line:90% AUDIENCE: I remember that day. 01:28:01.060 --> 01:28:03.540 align:middle line:84% IRVING WLADAWSKY-BERGER: You used punch cards? 01:28:03.540 --> 01:28:04.560 align:middle line:90% AUDIENCE: Yeah, 43. 01:28:04.560 --> 01:28:06.775 align:middle line:90% 01:28:06.775 --> 01:28:08.400 align:middle line:84% IRVING WLADAWSKY-BERGER: Wait a second. 01:28:08.400 --> 01:28:10.060 align:middle line:90% When did you use? 01:28:10.060 --> 01:28:11.500 align:middle line:90% [LAUGHTER] 01:28:11.500 --> 01:28:15.020 align:middle line:84% AUDIENCE: At the Academy, we still had a [INAUDIBLE]. 01:28:15.020 --> 01:28:17.260 align:middle line:84% IRVING WLADAWSKY-BERGER: So it was a history course. 01:28:17.260 --> 01:28:18.840 align:middle line:90% AUDIENCE: Yeah. 01:28:18.840 --> 01:28:21.280 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, in 1962-- 01:28:21.280 --> 01:28:22.522 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:28:22.522 --> 01:28:23.980 align:middle line:84% IRVING WLADAWSKY-BERGER: I'm sorry? 01:28:23.980 --> 01:28:25.952 align:middle line:90% [LAUGHTER] 01:28:25.952 --> 01:28:26.660 align:middle line:90% What did you say? 01:28:26.660 --> 01:28:27.340 align:middle line:90% I'm sorry. 01:28:27.340 --> 01:28:29.220 align:middle line:84% AUDIENCE: I have seen five-inch disks. 01:28:29.220 --> 01:28:31.640 align:middle line:90% [LAUGHTER] 01:28:31.640 --> 01:28:33.500 align:middle line:84% IRVING WLADAWSKY-BERGER: But in 1962, 01:28:33.500 --> 01:28:35.547 align:middle line:90% when I was entering college-- 01:28:35.547 --> 01:28:36.880 align:middle line:90% AUDIENCE: [INAUDIBLE] every day. 01:28:36.880 --> 01:28:38.672 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, that was it. 01:28:38.672 --> 01:28:44.920 align:middle line:84% The interactive terminals didn't show up until late '60s, early 01:28:44.920 --> 01:28:47.760 align:middle line:90% '70s, text terminals. 01:28:47.760 --> 01:28:52.940 align:middle line:84% And then cathode ray tubes showed up in the '70s. 01:28:52.940 --> 01:28:56.400 align:middle line:90% But before that, it was cards. 01:28:56.400 --> 01:29:02.280 align:middle line:84% But then the notion of a system changed radically 01:29:02.280 --> 01:29:07.800 align:middle line:84% over the next 20 years with the rise of networks, local area 01:29:07.800 --> 01:29:11.120 align:middle line:90% networks, wide area networks. 01:29:11.120 --> 01:29:16.670 align:middle line:84% It was a much more bigger view of systems. 01:29:16.670 --> 01:29:22.350 align:middle line:84% And people could talk about, for example, banking systems 01:29:22.350 --> 01:29:24.210 align:middle line:90% with ATM terminals. 01:29:24.210 --> 01:29:27.430 align:middle line:84% Remember those showed up in the '70s were the first-- 01:29:27.430 --> 01:29:28.230 align:middle line:90% that I remember. 01:29:28.230 --> 01:29:32.190 align:middle line:84% You don't remember, but you must have read about it. 01:29:32.190 --> 01:29:38.510 align:middle line:84% And the first ATMs appeared in the '70s, connected over 01:29:38.510 --> 01:29:42.990 align:middle line:84% proprietary networks like IBM's systems network architecture, 01:29:42.990 --> 01:29:45.630 align:middle line:90% manufacturing systems. 01:29:45.630 --> 01:29:51.350 align:middle line:84% So people started looking more specialized systems. 01:29:51.350 --> 01:29:56.590 align:middle line:84% Now, and this is very important, no question this 01:29:56.590 --> 01:29:58.290 align:middle line:90% is totally proprietary. 01:29:58.290 --> 01:30:00.910 align:middle line:84% I mean, this was almost the essence 01:30:00.910 --> 01:30:06.370 align:middle line:84% of functional performance guiding everything. 01:30:06.370 --> 01:30:09.390 align:middle line:84% And here, it was all pretty proprietary 01:30:09.390 --> 01:30:11.690 align:middle line:84% because networks were totally new. 01:30:11.690 --> 01:30:14.770 align:middle line:90% So they were proprietary. 01:30:14.770 --> 01:30:19.850 align:middle line:84% Systems received an incredible boost 01:30:19.850 --> 01:30:27.410 align:middle line:84% when the internet and the web got adopted in the mid-90s 01:30:27.410 --> 01:30:32.250 align:middle line:84% because for the first time, for the very first time, 01:30:32.250 --> 01:30:40.170 align:middle line:84% you can connect your clients, to your systems. 01:30:40.170 --> 01:30:44.250 align:middle line:84% Remember, before the browser, you 01:30:44.250 --> 01:30:48.310 align:middle line:84% can only interact with your systems 01:30:48.310 --> 01:30:53.370 align:middle line:84% if you were inside the company, through very specialized 01:30:53.370 --> 01:31:00.450 align:middle line:84% applications like ATMs, or through some proprietary things 01:31:00.450 --> 01:31:03.650 align:middle line:90% that led you-- 01:31:03.650 --> 01:31:06.810 align:middle line:84% if you were a partner, you had maybe a special PC 01:31:06.810 --> 01:31:08.930 align:middle line:90% that let you in. 01:31:08.930 --> 01:31:16.000 align:middle line:84% If in 1994, you wanted to find out 01:31:16.000 --> 01:31:22.420 align:middle line:84% where your FedEx package was, and if you were a normal person, 01:31:22.420 --> 01:31:24.560 align:middle line:84% as opposed to, let's say, a business that 01:31:24.560 --> 01:31:30.400 align:middle line:84% did a lot of business with them, you had to call a rep. 01:31:30.400 --> 01:31:36.600 align:middle line:84% And the rep typed into the proprietary network 01:31:36.600 --> 01:31:38.260 align:middle line:90% the number of the package. 01:31:38.260 --> 01:31:41.760 align:middle line:84% It went into its transaction system on the mainframe 01:31:41.760 --> 01:31:45.600 align:middle line:84% and got back the answer, and the rep told you. 01:31:45.600 --> 01:31:50.720 align:middle line:84% For large customers, they would install a special PC 01:31:50.720 --> 01:31:53.920 align:middle line:84% with special software, proprietary, 01:31:53.920 --> 01:31:55.540 align:middle line:90% that link you directly. 01:31:55.540 --> 01:31:59.360 align:middle line:84% So that you essentially can be viewed by their system 01:31:59.360 --> 01:32:02.600 align:middle line:90% as if you were the rep. 01:32:02.600 --> 01:32:07.320 align:middle line:84% The browser, the web revolutionized that. 01:32:07.320 --> 01:32:11.380 align:middle line:84% Because now everybody was using a browser, 01:32:11.380 --> 01:32:15.220 align:middle line:84% and everybody could connect to a web server. 01:32:15.220 --> 01:32:19.260 align:middle line:84% And now the web server, all it was doing initially 01:32:19.260 --> 01:32:22.860 align:middle line:84% was almost played the role of the operator that 01:32:22.860 --> 01:32:25.140 align:middle line:90% took your call. 01:32:25.140 --> 01:32:27.240 align:middle line:90% So you typed it in the browser. 01:32:27.240 --> 01:32:28.800 align:middle line:90% It went to the web server. 01:32:28.800 --> 01:32:32.180 align:middle line:84% The web server said, oh, this is a request to the mainframe, 01:32:32.180 --> 01:32:35.740 align:middle line:84% a little bit of software went there got you back the answer. 01:32:35.740 --> 01:32:38.420 align:middle line:84% And you said, God, this is magic. 01:32:38.420 --> 01:32:44.200 align:middle line:90% But it was a massive expansion. 01:32:44.200 --> 01:32:47.220 align:middle line:84% And in fact, the reason the bubble 01:32:47.220 --> 01:32:49.140 align:middle line:84% became a bubble is because it felt 01:32:49.140 --> 01:32:55.420 align:middle line:84% like magic, that these standards could let you do so much. 01:32:55.420 --> 01:33:06.080 align:middle line:84% I remember, in the mid '90s, maybe '96, '97, I was in Japan. 01:33:06.080 --> 01:33:09.090 align:middle line:90% I had flown in. 01:33:09.090 --> 01:33:15.870 align:middle line:84% I flew in, it must have been a Saturday out of here. 01:33:15.870 --> 01:33:22.250 align:middle line:84% Landed in Japan Sunday evening, and then woke up like 3:00 AM, 01:33:22.250 --> 01:33:27.370 align:middle line:90% 4:00 AM Monday morning in Japan. 01:33:27.370 --> 01:33:30.910 align:middle line:84% And I remember there was a baseball game back in New York. 01:33:30.910 --> 01:33:32.190 align:middle line:90% The Yankees or Mets? 01:33:32.190 --> 01:33:33.410 align:middle line:90% I don't remember. 01:33:33.410 --> 01:33:35.450 align:middle line:84% That I was very interested, but obviously, 01:33:35.450 --> 01:33:37.450 align:middle line:90% you wouldn't get it in Japan. 01:33:37.450 --> 01:33:45.490 align:middle line:84% And then I remembered that oh, but my PC 01:33:45.490 --> 01:33:47.690 align:middle line:90% is connected to the internet. 01:33:47.690 --> 01:33:56.170 align:middle line:84% They were beginning to link the games over PCs so I could be 01:33:56.170 --> 01:33:59.630 align:middle line:84% drinking green tea at 3:00 AM in my room in Japan, 01:33:59.630 --> 01:34:02.190 align:middle line:84% listening to the baseball game back here. 01:34:02.190 --> 01:34:07.970 align:middle line:84% Now today, that is, like, why are you telling me this? 01:34:07.970 --> 01:34:08.750 align:middle line:90% [LAUGHTER] 01:34:08.750 --> 01:34:13.070 align:middle line:84% But when it first happened, it was magic. 01:34:13.070 --> 01:34:18.290 align:middle line:84% And it was this deconstruction of the system. 01:34:18.290 --> 01:34:21.070 align:middle line:84% That's the only way I can say that. 01:34:21.070 --> 01:34:29.470 align:middle line:84% Now, where we are heading is that we really 01:34:29.470 --> 01:34:35.550 align:middle line:84% want to take every single aspect of a business 01:34:35.550 --> 01:34:40.390 align:middle line:84% and integrate it together, every single aspect of a business. 01:34:40.390 --> 01:34:42.090 align:middle line:84% We say, well, what do you mean by that? 01:34:42.090 --> 01:34:48.470 align:middle line:84% Well, a business has processes, information, people, and now 01:34:48.470 --> 01:34:52.550 align:middle line:84% we have all the work on standards, 01:34:52.550 --> 01:34:54.990 align:middle line:84% SOA Of course, the internet, we can 01:34:54.990 --> 01:34:58.150 align:middle line:90% begin to couple them together. 01:34:58.150 --> 01:35:03.740 align:middle line:84% So you can begin to look at the business as a system. 01:35:03.740 --> 01:35:08.360 align:middle line:84% But now, once you start thinking that way, 01:35:08.360 --> 01:35:11.780 align:middle line:84% you realize, wait a second, there 01:35:11.780 --> 01:35:17.300 align:middle line:84% is no business that doesn't exist within an industry 01:35:17.300 --> 01:35:19.500 align:middle line:90% ecosystem. 01:35:19.500 --> 01:35:22.500 align:middle line:84% So if this is going to be useful, 01:35:22.500 --> 01:35:27.380 align:middle line:84% your definition of a system has to include the industry 01:35:27.380 --> 01:35:31.100 align:middle line:84% ecosystem because otherwise, if you 01:35:31.100 --> 01:35:34.500 align:middle line:84% say everybody inside the company can work with you, 01:35:34.500 --> 01:35:37.980 align:middle line:84% but the people making the wings of the Boeing 01:35:37.980 --> 01:35:42.560 align:middle line:84% have to make phone calls, it's all over. 01:35:42.560 --> 01:35:45.580 align:middle line:84% So you have to stretch what you're 01:35:45.580 --> 01:35:51.020 align:middle line:84% doing to all your suppliers and all your partners. 01:35:51.020 --> 01:35:57.340 align:middle line:84% And because more and more industries like healthcare 01:35:57.340 --> 01:36:00.560 align:middle line:84% are a mix of different industries. 01:36:00.560 --> 01:36:04.080 align:middle line:84% There's the hospitals, the doctors, the insurance 01:36:04.080 --> 01:36:06.480 align:middle line:90% companies, and others. 01:36:06.480 --> 01:36:09.320 align:middle line:84% That's what's driving the digital economy 01:36:09.320 --> 01:36:15.960 align:middle line:84% that we're trying to look at all the things, it's a business. 01:36:15.960 --> 01:36:20.680 align:middle line:84% Now, the technology to help us start 01:36:20.680 --> 01:36:24.780 align:middle line:84% doing that it's beginning to happen. 01:36:24.780 --> 01:36:27.700 align:middle line:84% I mentioned several times, more and more 01:36:27.700 --> 01:36:31.840 align:middle line:84% the internet, broadband, wireless, even software 01:36:31.840 --> 01:36:38.520 align:middle line:84% standards, SOA, the really hard question 01:36:38.520 --> 01:36:43.120 align:middle line:84% is, who the hell is designing this? 01:36:43.120 --> 01:36:47.240 align:middle line:84% Now, if you said to me, well, Irving, biology. 01:36:47.240 --> 01:36:49.640 align:middle line:90% This is not biology. 01:36:49.640 --> 01:36:54.360 align:middle line:84% Remember in biology, yes, you designed this, 01:36:54.360 --> 01:36:58.900 align:middle line:84% but if you're the CEO of a company, 01:36:58.900 --> 01:37:01.700 align:middle line:84% you cannot go to the shareholders, and they say, 01:37:01.700 --> 01:37:05.440 align:middle line:84% how come you lost the records of all your clients? 01:37:05.440 --> 01:37:07.380 align:middle line:84% And you say, well, you know, shit happens. 01:37:07.380 --> 01:37:08.980 align:middle line:90% [LAUGHTER] 01:37:08.980 --> 01:37:10.120 align:middle line:90% We got a virus. 01:37:10.120 --> 01:37:13.300 align:middle line:84% Everybody knows in biology-- you cannot do that. 01:37:13.300 --> 01:37:18.900 align:middle line:84% Somebody has to design, an architect. 01:37:18.900 --> 01:37:25.260 align:middle line:84% And that gets back to the question, how do we do that? 01:37:25.260 --> 01:37:28.220 align:middle line:84% And by the way, there is no answer. 01:37:28.220 --> 01:37:33.920 align:middle line:84% Remember, for a long time, planes and cars, 01:37:33.920 --> 01:37:38.420 align:middle line:84% there were drafts, people drafting by hand. 01:37:38.420 --> 01:37:42.020 align:middle line:84% And I'm assuming for a long time, 01:37:42.020 --> 01:37:45.800 align:middle line:84% civil engineering architects were drafting by hand. 01:37:45.800 --> 01:37:51.260 align:middle line:84% It's only later that the tools came in and the simulations. 01:37:51.260 --> 01:37:55.060 align:middle line:90% In business, who designs it? 01:37:55.060 --> 01:37:56.280 align:middle line:90% I don't know who designed. 01:37:56.280 --> 01:37:57.990 align:middle line:90% It's a fascinating question. 01:37:57.990 --> 01:37:59.810 align:middle line:90% I don't know. 01:37:59.810 --> 01:38:02.330 align:middle line:84% Soon-Yu, is there a secret course 01:38:02.330 --> 01:38:05.450 align:middle line:84% in Sloan on designing business processes? 01:38:05.450 --> 01:38:08.130 align:middle line:90% 01:38:08.130 --> 01:38:10.930 align:middle line:84% And the reason, by the way, is that the whole notion 01:38:10.930 --> 01:38:17.190 align:middle line:84% of designing business systems, it's so new. 01:38:17.190 --> 01:38:20.070 align:middle line:90% I think, it's really pushing us. 01:38:20.070 --> 01:38:24.370 align:middle line:90% But why do we not have a choice? 01:38:24.370 --> 01:38:28.090 align:middle line:84% Because of the scope, the complexity here, 01:38:28.090 --> 01:38:34.250 align:middle line:84% is such that if you don't have some kind of design 01:38:34.250 --> 01:38:39.410 align:middle line:84% on top of that, and in fact, we'll get to this in a second, 01:38:39.410 --> 01:38:45.130 align:middle line:84% the notion of architecture that you apply to these incredibly 01:38:45.130 --> 01:38:50.930 align:middle line:84% complex systems has to be more flexible and adaptable 01:38:50.930 --> 01:38:54.170 align:middle line:84% than the notion of architecture that you 01:38:54.170 --> 01:38:57.530 align:middle line:90% use for a much simpler product. 01:38:57.530 --> 01:38:59.670 align:middle line:84% But I guess you would say, the notion 01:38:59.670 --> 01:39:03.470 align:middle line:84% of architecture in an incredibly large microprocessor 01:39:03.470 --> 01:39:05.950 align:middle line:84% with a huge number of things has to be 01:39:05.950 --> 01:39:10.610 align:middle line:84% more flexible and adaptable than just a small one. 01:39:10.610 --> 01:39:11.870 align:middle line:90% It gets back to modularity. 01:39:11.870 --> 01:39:12.370 align:middle line:90% Soon-Yu? 01:39:12.370 --> 01:39:15.190 align:middle line:84% AUDIENCE: Yes, just remember there's two professors who 01:39:15.190 --> 01:39:16.670 align:middle line:90% are related to this. 01:39:16.670 --> 01:39:19.098 align:middle line:84% So one is Steve Eppinger, who I'm 01:39:19.098 --> 01:39:21.390 align:middle line:84% pretty sure you guys are familiar with Steve Eppinger's 01:39:21.390 --> 01:39:22.928 align:middle line:90% work. 01:39:22.928 --> 01:39:24.470 align:middle line:84% What he's trying to do is he's trying 01:39:24.470 --> 01:39:26.990 align:middle line:84% to reduce the complexity of the architecture 01:39:26.990 --> 01:39:32.297 align:middle line:84% and use it as an organization framework. 01:39:32.297 --> 01:39:33.630 align:middle line:90% And the other one is Tom Malone. 01:39:33.630 --> 01:39:35.990 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly, exactly. 01:39:35.990 --> 01:39:38.310 align:middle line:84% AUDIENCE: Make the primitives of these processes. 01:39:38.310 --> 01:39:40.670 align:middle line:84% It's a very small unit of processes. 01:39:40.670 --> 01:39:45.732 align:middle line:84% And trying to use it as how to exactly build more complex. 01:39:45.732 --> 01:39:47.690 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, I agree totally. 01:39:47.690 --> 01:39:52.610 align:middle line:84% Steve Eppinger and Tom Malone are leading the way. 01:39:52.610 --> 01:39:55.640 align:middle line:84% But how many professors are there in Sloan? 01:39:55.640 --> 01:39:58.340 align:middle line:90% These are two out of-- 01:39:58.340 --> 01:40:03.380 align:middle line:84% By the way, at some point, engineering 01:40:03.380 --> 01:40:05.400 align:middle line:84% started coming into the cathedral. 01:40:05.400 --> 01:40:07.600 align:middle line:90% I think we're seeing that. 01:40:07.600 --> 01:40:08.460 align:middle line:90% Yes. 01:40:08.460 --> 01:40:11.020 align:middle line:84% AUDIENCE: I think it was Murphy who 01:40:11.020 --> 01:40:13.620 align:middle line:84% said that if you look out 20 years, 20 to 30 years 01:40:13.620 --> 01:40:18.220 align:middle line:84% on the macro economy, things will shift down 01:40:18.220 --> 01:40:19.960 align:middle line:90% into buckets, if you will. 01:40:19.960 --> 01:40:22.980 align:middle line:84% The first bucket will be that Americans will 01:40:22.980 --> 01:40:25.500 align:middle line:90% continue to be the innovators. 01:40:25.500 --> 01:40:28.820 align:middle line:84% The Chinese will be the blue-collar workers of America, 01:40:28.820 --> 01:40:31.360 align:middle line:84% and the Indians will be the white-collar workers here. 01:40:31.360 --> 01:40:32.180 align:middle line:84% IRVING WLADAWSKY-BERGER: [LAUGHS] 01:40:32.180 --> 01:40:33.880 align:middle line:84% AUDIENCE: And I think at the heart of all of that-- 01:40:33.880 --> 01:40:34.680 align:middle line:84% IRVING WLADAWSKY-BERGER: Who said that? 01:40:34.680 --> 01:40:36.460 align:middle line:84% AUDIENCE: I think it was Murphy, I think who said it. 01:40:36.460 --> 01:40:38.377 align:middle line:84% I'm trying to remember, but someone said that. 01:40:38.377 --> 01:40:39.752 align:middle line:84% IRVING WLADAWSKY-BERGER: I wonder 01:40:39.752 --> 01:40:41.660 align:middle line:84% if he really believes that or was trying 01:40:41.660 --> 01:40:43.840 align:middle line:90% to lull the Americans to sleep. 01:40:43.840 --> 01:40:45.348 align:middle line:90% [LAUGHTER] 01:40:45.348 --> 01:40:47.140 align:middle line:84% AUDIENCE: I think, though, do believe that. 01:40:47.140 --> 01:40:49.155 align:middle line:84% If you look at the innovation side of America, 01:40:49.155 --> 01:40:50.780 align:middle line:84% at the heart of all that, the principle 01:40:50.780 --> 01:40:52.280 align:middle line:84% is going to be mass customization. 01:40:52.280 --> 01:40:53.300 align:middle line:84% IRVING WLADAWSKY-BERGER: No, no, I understand. 01:40:53.300 --> 01:40:54.540 align:middle line:90% I understand totally. 01:40:54.540 --> 01:40:56.560 align:middle line:84% I just want to say that, let's say, 01:40:56.560 --> 01:40:59.120 align:middle line:84% if you go to Tsinghua University, 01:40:59.120 --> 01:41:02.880 align:middle line:84% and you say you guys are the blue-collar workers, 01:41:02.880 --> 01:41:07.040 align:middle line:84% Tsinghua is a really superb university in Beijing or Peking 01:41:07.040 --> 01:41:11.400 align:middle line:84% or Xidian Actually, India needs to do 01:41:11.400 --> 01:41:16.320 align:middle line:84% much better at the high end, but I think you all know that. 01:41:16.320 --> 01:41:18.740 align:middle line:84% But if you go to China and then India, 01:41:18.740 --> 01:41:21.880 align:middle line:90% hopefully, as universities-- 01:41:21.880 --> 01:41:24.740 align:middle line:84% I've been asking my Indian friends, why is it like that? 01:41:24.740 --> 01:41:27.880 align:middle line:84% And they say for whatever reason, 01:41:27.880 --> 01:41:34.760 align:middle line:84% the Indian universities still cannot pay faculty, top faculty, 01:41:34.760 --> 01:41:36.960 align:middle line:90% enough to attract them. 01:41:36.960 --> 01:41:38.360 align:middle line:90% AUDIENCE: They were public. 01:41:38.360 --> 01:41:39.920 align:middle line:84% Most of the [INAUDIBLE] universities 01:41:39.920 --> 01:41:42.720 align:middle line:84% are public universities, funded by the government. 01:41:42.720 --> 01:41:46.400 align:middle line:84% But even then, they do not have a lot of control 01:41:46.400 --> 01:41:49.000 align:middle line:90% in the technical domain. 01:41:49.000 --> 01:41:52.007 align:middle line:84% Salary wise, they do have nothing. 01:41:52.007 --> 01:41:53.590 align:middle line:84% IRVING WLADAWSKY-BERGER: But everybody 01:41:53.590 --> 01:41:58.990 align:middle line:84% knows India-- and that's why most Indian students go 01:41:58.990 --> 01:42:01.830 align:middle line:90% to the US to get PhDs. 01:42:01.830 --> 01:42:03.990 align:middle line:90% Am I correct? 01:42:03.990 --> 01:42:05.510 align:middle line:90% AUDIENCE: The high end-- 01:42:05.510 --> 01:42:07.843 align:middle line:84% IRVING WLADAWSKY-BERGER: I'm talking about the high end. 01:42:07.843 --> 01:42:13.210 align:middle line:84% In China, Xidian, wouldn't you agree with me? 01:42:13.210 --> 01:42:16.970 align:middle line:84% They are becoming world class is my opinion. 01:42:16.970 --> 01:42:21.630 align:middle line:84% So I'm just saying, any US executive that really 01:42:21.630 --> 01:42:23.290 align:middle line:90% believes that is in trouble. 01:42:23.290 --> 01:42:24.430 align:middle line:90% That's all I'm saying. 01:42:24.430 --> 01:42:27.065 align:middle line:90% 01:42:27.065 --> 01:42:28.690 align:middle line:84% AUDIENCE: Let's take it out of context. 01:42:28.690 --> 01:42:31.290 align:middle line:84% IRVING WLADAWSKY-BERGER: By the way, I agree totally with you. 01:42:31.290 --> 01:42:34.330 align:middle line:84% And in fact, it's an interesting discussion to have. 01:42:34.330 --> 01:42:37.830 align:middle line:84% Let me go to the next section and then we'll talk about that. 01:42:37.830 --> 01:42:40.210 align:middle line:84% Because it's a very interesting discussion. 01:42:40.210 --> 01:42:42.870 align:middle line:90% 01:42:42.870 --> 01:42:44.530 align:middle line:90% I've talked about breadth. 01:42:44.530 --> 01:42:49.930 align:middle line:84% Let me now go in the vertical dimension my view of systems. 01:42:49.930 --> 01:42:52.510 align:middle line:84% And I'll explain the shapes later. 01:42:52.510 --> 01:42:55.410 align:middle line:90% But if you look at a stack-- 01:42:55.410 --> 01:42:59.570 align:middle line:84% so we start with the technology component microprocessors, 01:42:59.570 --> 01:43:00.550 align:middle line:90% memories. 01:43:00.550 --> 01:43:04.750 align:middle line:84% And we know how to do those fairly well. 01:43:04.750 --> 01:43:08.410 align:middle line:84% We have tools, deterministic stuff, and so on. 01:43:08.410 --> 01:43:12.850 align:middle line:84% And then we use this to build products, hardware 01:43:12.850 --> 01:43:14.230 align:middle line:90% and software products. 01:43:14.230 --> 01:43:21.690 align:middle line:84% And again, we pretty much know how to do that. 01:43:21.690 --> 01:43:29.170 align:middle line:84% And as you know, once upon a time, the technology 01:43:29.170 --> 01:43:34.170 align:middle line:84% and the products, I'm talking in IT now, 01:43:34.170 --> 01:43:40.350 align:middle line:84% were very proprietary, very unique. 01:43:40.350 --> 01:43:42.530 align:middle line:90% That was the competition. 01:43:42.530 --> 01:43:44.850 align:middle line:84% But as you know, more and mow, a lot of these 01:43:44.850 --> 01:43:48.680 align:middle line:84% are totally commoditized, like, DRAMs, HDD, 01:43:48.680 --> 01:43:50.720 align:middle line:90% a lot of microprocessors. 01:43:50.720 --> 01:43:54.680 align:middle line:84% And even here, you can see standardization 01:43:54.680 --> 01:43:55.900 align:middle line:90% beginning to show up. 01:43:55.900 --> 01:43:59.940 align:middle line:84% And I think as SOA takes hold, that will happen more and more. 01:43:59.940 --> 01:44:05.800 align:middle line:84% But you can see the rise of Linux and other things. 01:44:05.800 --> 01:44:10.800 align:middle line:84% I think that what's happening is that the action is going up 01:44:10.800 --> 01:44:13.160 align:middle line:90% the stack. 01:44:13.160 --> 01:44:19.040 align:middle line:84% And one of the major things is the design 01:44:19.040 --> 01:44:22.840 align:middle line:84% of applications using technologies and products, which 01:44:22.840 --> 01:44:27.120 align:middle line:84% keeps expanding, more and more and more and more, 01:44:27.120 --> 01:44:30.720 align:middle line:84% because as these things keep getting cheaper and cheaper 01:44:30.720 --> 01:44:34.360 align:middle line:84% and cheaper and more and more powerful and more reliable, 01:44:34.360 --> 01:44:40.280 align:middle line:84% there is more and more you can do in brand new areas. 01:44:40.280 --> 01:44:47.460 align:middle line:84% And then even more, is to now use all this stuff 01:44:47.460 --> 01:44:51.060 align:middle line:84% to help companies build business systems. 01:44:51.060 --> 01:44:56.060 align:middle line:84% Do it for companies, aggregate it. 01:44:56.060 --> 01:45:03.460 align:middle line:84% What's interesting, when I look at these systems of the stack, 01:45:03.460 --> 01:45:06.780 align:middle line:84% what goes on down here, and what goes on up there, 01:45:06.780 --> 01:45:10.540 align:middle line:90% is very different. 01:45:10.540 --> 01:45:14.600 align:middle line:84% Systems, as you go down, they are deterministic. 01:45:14.600 --> 01:45:16.480 align:middle line:90% They are physical things. 01:45:16.480 --> 01:45:19.660 align:middle line:84% You know exactly what you're dealing with. 01:45:19.660 --> 01:45:21.960 align:middle line:84% It's like you're building a microprocessor. 01:45:21.960 --> 01:45:23.820 align:middle line:90% You're building a bridge. 01:45:23.820 --> 01:45:28.940 align:middle line:84% You're engineering, designing physical things. 01:45:28.940 --> 01:45:38.900 align:middle line:84% As you go up the stack, you are designing market-facing systems 01:45:38.900 --> 01:45:43.930 align:middle line:84% that involve people, and that involve 01:45:43.930 --> 01:45:47.450 align:middle line:84% people dealing with each other, and that 01:45:47.450 --> 01:45:50.970 align:middle line:90% renders them unpredictable. 01:45:50.970 --> 01:45:54.650 align:middle line:84% And when I say unpredictable, I mean emergent 01:45:54.650 --> 01:45:57.510 align:middle line:84% in a biological sense of unpredictable. 01:45:57.510 --> 01:46:02.690 align:middle line:84% Things are very complicated and very fast changing. 01:46:02.690 --> 01:46:05.690 align:middle line:84% You don't have control over them. 01:46:05.690 --> 01:46:07.230 align:middle line:90% They are emergent. 01:46:07.230 --> 01:46:10.550 align:middle line:84% Now, does that mean you cannot apply technology? 01:46:10.550 --> 01:46:13.970 align:middle line:84% No, of course, you can apply a lot of technology. 01:46:13.970 --> 01:46:20.110 align:middle line:84% But it's dynamic instead of static. 01:46:20.110 --> 01:46:23.170 align:middle line:84% When you're dealing with dynamic systems, 01:46:23.170 --> 01:46:25.670 align:middle line:84% you have to constantly be watching them. 01:46:25.670 --> 01:46:27.730 align:middle line:84% Otherwise, they'll get away from you. 01:46:27.730 --> 01:46:31.970 align:middle line:84% You have to constantly be monitoring them. 01:46:31.970 --> 01:46:34.850 align:middle line:90% Let's take customer service. 01:46:34.850 --> 01:46:38.970 align:middle line:84% Somebody said, well, customer service is deterministic. 01:46:38.970 --> 01:46:42.670 align:middle line:84% When a client calls, I'll say, I'm 01:46:42.670 --> 01:46:47.550 align:middle line:84% programmed to answer in one of these five questions. 01:46:47.550 --> 01:46:49.610 align:middle line:84% What if the client has a sixth question? 01:46:49.610 --> 01:46:52.950 align:middle line:84% Well, I'm not programmed to answer that. 01:46:52.950 --> 01:46:57.590 align:middle line:84% Guess what, you have a very unhappy client. 01:46:57.590 --> 01:47:02.550 align:middle line:84% That's an example of a market-facing system that is 01:47:02.550 --> 01:47:06.990 align:middle line:84% unpredictable because you cannot control what your clients are 01:47:06.990 --> 01:47:09.230 align:middle line:90% going to want from you. 01:47:09.230 --> 01:47:11.890 align:middle line:84% And think of a healthcare system. 01:47:11.890 --> 01:47:15.590 align:middle line:84% Think of systems to help nurses, doctors. 01:47:15.590 --> 01:47:22.630 align:middle line:84% So the design, and this is why I like the thinking of design, 01:47:22.630 --> 01:47:31.350 align:middle line:84% the architecture and design of systems at these higher levels 01:47:31.350 --> 01:47:37.490 align:middle line:84% with some predictability, have to be far more frameworks. 01:47:37.490 --> 01:47:39.370 align:middle line:90% I think it has much more. 01:47:39.370 --> 01:47:44.380 align:middle line:84% It's almost like urban planning rather than physical building. 01:47:44.380 --> 01:47:46.980 align:middle line:84% Now, I don't know if they teach urban planning 01:47:46.980 --> 01:47:48.480 align:middle line:90% in schools of architecture. 01:47:48.480 --> 01:47:50.060 align:middle line:90% Do they do that? 01:47:50.060 --> 01:47:51.948 align:middle line:90% AUDIENCE: Mm-hmm. 01:47:51.948 --> 01:47:53.962 align:middle line:84% IRVING WLADAWSKY-BERGER: Did they teach-- 01:47:53.962 --> 01:47:55.420 align:middle line:84% AUDIENCE: At least in some schools. 01:47:55.420 --> 01:47:58.260 align:middle line:90% I remember in my undergrad days. 01:47:58.260 --> 01:48:02.240 align:middle line:84% IRVING WLADAWSKY-BERGER: But you all have to agree, 01:48:02.240 --> 01:48:06.100 align:middle line:84% there is architecture of a different sort when 01:48:06.100 --> 01:48:09.100 align:middle line:90% you think of urban planning. 01:48:09.100 --> 01:48:14.180 align:middle line:84% And so when you start doing business systems-- 01:48:14.180 --> 01:48:20.540 align:middle line:84% now, what's so exciting is that before there was nothing 01:48:20.540 --> 01:48:21.320 align:middle line:90% we could do. 01:48:21.320 --> 01:48:26.780 align:middle line:84% So for the most part, as you go up, 01:48:26.780 --> 01:48:28.700 align:middle line:90% you have to do it with labor. 01:48:28.700 --> 01:48:32.980 align:middle line:84% Because you say, well, if it's unpredictable, it's all people. 01:48:32.980 --> 01:48:35.520 align:middle line:90% It's labor, and guess what? 01:48:35.520 --> 01:48:38.320 align:middle line:84% As we get better and better and better, 01:48:38.320 --> 01:48:42.680 align:middle line:84% the industrial sector of the economy 01:48:42.680 --> 01:48:46.040 align:middle line:84% is smaller and smaller and smaller, 01:48:46.040 --> 01:48:49.540 align:middle line:84% and the services sector of the economy keeps growing. 01:48:49.540 --> 01:48:55.320 align:middle line:84% In most countries like the US, the UK, it's over 70% or 80% 01:48:55.320 --> 01:48:56.920 align:middle line:90% of the economy. 01:48:56.920 --> 01:49:02.680 align:middle line:84% Also, more and more businesses, if you look at IBM, over 55% 01:49:02.680 --> 01:49:05.420 align:middle line:84% of IBM's revenue is now services. 01:49:05.420 --> 01:49:10.560 align:middle line:84% But that's true of almost every business, because in these areas 01:49:10.560 --> 01:49:13.500 align:middle line:84% they keep applying automation and technology, 01:49:13.500 --> 01:49:16.160 align:middle line:90% and those areas is labor. 01:49:16.160 --> 01:49:22.240 align:middle line:84% What is changing is with tools and collaborative systems, 01:49:22.240 --> 01:49:29.420 align:middle line:84% you can now begin to attack the labor part of those systems, 01:49:29.420 --> 01:49:30.820 align:middle line:90% the people-oriented part. 01:49:30.820 --> 01:49:34.320 align:middle line:84% When I say attack, I mean make people more productive. 01:49:34.320 --> 01:49:42.060 align:middle line:84% Not by automation, but by giving them better and better tools. 01:49:42.060 --> 01:49:47.580 align:middle line:84% And in my opinion, the hardest part in all this 01:49:47.580 --> 01:49:49.180 align:middle line:90% gets back to design. 01:49:49.180 --> 01:49:52.380 align:middle line:90% Who is architecting, designing? 01:49:52.380 --> 01:49:55.100 align:middle line:84% How much knowledge do you need to do 01:49:55.100 --> 01:50:02.220 align:middle line:84% by industry vertical healthcare, investment banking, and so on? 01:50:02.220 --> 01:50:04.100 align:middle line:84% And the answer, I believe, is going 01:50:04.100 --> 01:50:08.100 align:middle line:84% to be very vertical for quite a while. 01:50:08.100 --> 01:50:10.540 align:middle line:84% Let me explain this shape because the first time I 01:50:10.540 --> 01:50:17.140 align:middle line:84% did this was to go to Washington to talk at some government thing 01:50:17.140 --> 01:50:21.180 align:middle line:84% where people were saying, there were no jobs in technology, 01:50:21.180 --> 01:50:23.980 align:middle line:84% or there were, but they questioned all technology. 01:50:23.980 --> 01:50:29.260 align:middle line:84% It's bad to be a technologist because all jobs are going on. 01:50:29.260 --> 01:50:36.970 align:middle line:84% And my hypothesis is everyone at these level, 01:50:36.970 --> 01:50:41.290 align:middle line:84% the requirement goes up by an order of magnitude. 01:50:41.290 --> 01:50:44.030 align:middle line:84% So for every one person doing this, 01:50:44.030 --> 01:50:46.410 align:middle line:90% you have 10 people doing that. 01:50:46.410 --> 01:50:50.290 align:middle line:84% And then you have 100 people doing that or 1,000 people doing 01:50:50.290 --> 01:50:50.830 align:middle line:90% that. 01:50:50.830 --> 01:50:52.970 align:middle line:84% So if your definition of technology 01:50:52.970 --> 01:50:58.970 align:middle line:84% is down here, designing microprocessors, 01:50:58.970 --> 01:51:03.410 align:middle line:84% you better put up, what I tell people, David Ortiz 01:51:03.410 --> 01:51:05.410 align:middle line:90% kind of numbers. 01:51:05.410 --> 01:51:08.890 align:middle line:84% You have to be a really, really good microprocessor designed 01:51:08.890 --> 01:51:11.430 align:middle line:90% because we don't need that many. 01:51:11.430 --> 01:51:14.690 align:middle line:84% The competition is very, very strong. 01:51:14.690 --> 01:51:20.010 align:middle line:84% And more and more, even at the product level, the more mature 01:51:20.010 --> 01:51:22.690 align:middle line:84% industries, the less people you need. 01:51:22.690 --> 01:51:27.090 align:middle line:84% Also, the more you learn, the more it gets automated. 01:51:27.090 --> 01:51:32.630 align:middle line:84% The bulk of the jobs are in application systems 01:51:32.630 --> 01:51:35.430 align:middle line:90% and then in business systems. 01:51:35.430 --> 01:51:40.790 align:middle line:84% Just to give you a view, in IBM, I just came back from IBM 01:51:40.790 --> 01:51:47.510 align:middle line:84% Academy of Technology meeting, there are maybe 30,000 people 01:51:47.510 --> 01:51:50.070 align:middle line:90% doing this stuff in labs. 01:51:50.070 --> 01:52:00.190 align:middle line:84% And there are maybe 120,000, 140,000 people up there. 01:52:00.190 --> 01:52:03.610 align:middle line:84% And by the way, IBM has really good labs. 01:52:03.610 --> 01:52:09.170 align:middle line:84% So it's just the action is in the marketplace, 01:52:09.170 --> 01:52:11.310 align:middle line:90% and that's where the jobs are. 01:52:11.310 --> 01:52:15.090 align:middle line:84% And this stuff is infinitely more complicated. 01:52:15.090 --> 01:52:19.710 align:middle line:84% I mean, you put together the previous slides I show, 01:52:19.710 --> 01:52:27.710 align:middle line:84% which is the scope of the systems that we try to design, 01:52:27.710 --> 01:52:34.060 align:middle line:84% with the fact that we can now attack systems at this much more 01:52:34.060 --> 01:52:39.180 align:middle line:84% market-facing, people-oriented level, and this is-- 01:52:39.180 --> 01:52:40.180 align:middle line:90% it's pretty interesting. 01:52:40.180 --> 01:52:45.600 align:middle line:84% In ESD, MIT's engineering systems division, 01:52:45.600 --> 01:52:48.500 align:middle line:84% we're trying to put together some pretty nifty healthcare 01:52:48.500 --> 01:52:52.340 align:middle line:84% system initiatives because we can begin 01:52:52.340 --> 01:52:58.140 align:middle line:84% to try to attack healthcare as a systems problem 01:52:58.140 --> 01:53:00.160 align:middle line:84% because there is so much we can bring, 01:53:00.160 --> 01:53:03.420 align:middle line:84% although it's a very, very complicated problem. 01:53:03.420 --> 01:53:12.020 align:middle line:84% So now, once you talk about things of this scope I'm talking 01:53:12.020 --> 01:53:18.140 align:middle line:84% about, the question of modularization versus vertically 01:53:18.140 --> 01:53:24.780 align:middle line:84% integrated disappear because you just cannot possibly do things 01:53:24.780 --> 01:53:28.440 align:middle line:84% of this scope if they are not standard based and modular. 01:53:28.440 --> 01:53:30.820 align:middle line:90% You just can't anymore. 01:53:30.820 --> 01:53:33.740 align:middle line:84% You say, well, why is the internet standard? 01:53:33.740 --> 01:53:37.680 align:middle line:84% Well, it's like almost asking why is biology built out 01:53:37.680 --> 01:53:39.100 align:middle line:90% of standard components? 01:53:39.100 --> 01:53:45.040 align:middle line:84% Which it is cells, DNA, pretty standard. 01:53:45.040 --> 01:53:47.520 align:middle line:84% And it's because that's the only way 01:53:47.520 --> 01:53:52.020 align:middle line:84% you can scale to the complexity we're trying to scale. 01:53:52.020 --> 01:53:55.640 align:middle line:90% 01:53:55.640 --> 01:54:01.080 align:middle line:84% So we'll see where all this goes. 01:54:01.080 --> 01:54:08.320 align:middle line:84% Now I'll move on to the specific implementations 01:54:08.320 --> 01:54:10.220 align:middle line:90% of a business in IBM. 01:54:10.220 --> 01:54:12.180 align:middle line:90% But any questions before that? 01:54:12.180 --> 01:54:17.000 align:middle line:90% 01:54:17.000 --> 01:54:19.000 align:middle line:84% AUDIENCE: So you were saying that there are not 01:54:19.000 --> 01:54:24.600 align:middle line:84% many systems which actually do the business process management. 01:54:24.600 --> 01:54:27.270 align:middle line:84% But there have been right from [INAUDIBLE] 01:54:27.270 --> 01:54:31.590 align:middle line:84% we have CRM systems and business process engineering. 01:54:31.590 --> 01:54:32.578 align:middle line:90% BPM seems to be. 01:54:32.578 --> 01:54:34.370 align:middle line:84% IRVING WLADAWSKY-BERGER: They're primitive. 01:54:34.370 --> 01:54:35.830 align:middle line:90% That's the way I would say. 01:54:35.830 --> 01:54:38.570 align:middle line:90% People are doing them, ERPs. 01:54:38.570 --> 01:54:41.630 align:middle line:90% We have ERP systems. 01:54:41.630 --> 01:54:49.550 align:middle line:90% But I think they work. 01:54:49.550 --> 01:54:52.890 align:middle line:84% If you give them a grade by form, 01:54:52.890 --> 01:54:58.130 align:middle line:90% you would give them a minus 10. 01:54:58.130 --> 01:55:03.510 align:middle line:84% I don't know how many of you are familiar with CRP systems. 01:55:03.510 --> 01:55:07.590 align:middle line:84% Would you agree that most people are not very happy with them? 01:55:07.590 --> 01:55:08.990 align:middle line:90% AUDIENCE: They work, yeah. 01:55:08.990 --> 01:55:10.790 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, they work. 01:55:10.790 --> 01:55:13.690 align:middle line:84% That would be as if we were talking about, 01:55:13.690 --> 01:55:18.430 align:middle line:84% well, look, I know this building looks horrendous, 01:55:18.430 --> 01:55:20.790 align:middle line:90% but it won't fall down. 01:55:20.790 --> 01:55:23.710 align:middle line:90% But it really looks horrendous. 01:55:23.710 --> 01:55:27.230 align:middle line:84% ERP systems, would you agree they have that quality? 01:55:27.230 --> 01:55:28.430 align:middle line:90% AUDIENCE: They all have-- 01:55:28.430 --> 01:55:30.370 align:middle line:84% you have to know what they're capable of, 01:55:30.370 --> 01:55:32.930 align:middle line:84% and then you can ask them whatever you need to, 01:55:32.930 --> 01:55:34.770 align:middle line:84% and they'll give you a reasonable answer. 01:55:34.770 --> 01:55:36.270 align:middle line:84% AUDIENCE: This is foundation, right? 01:55:36.270 --> 01:55:36.930 align:middle line:90% AUDIENCE: Yeah. 01:55:36.930 --> 01:55:38.388 align:middle line:84% IRVING WLADAWSKY-BERGER: That's it. 01:55:38.388 --> 01:55:40.013 align:middle line:84% AUDIENCE: You have to be very critical. 01:55:40.013 --> 01:55:42.596 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, but, no, no, no, the foundation 01:55:42.596 --> 01:55:43.850 align:middle line:90% doesn't have to be pretty. 01:55:43.850 --> 01:55:50.230 align:middle line:84% But remember, you need to do interior decorating, 01:55:50.230 --> 01:55:52.250 align:middle line:90% so where is that? 01:55:52.250 --> 01:55:55.250 align:middle line:84% And so I think what's happening is 01:55:55.250 --> 01:55:57.850 align:middle line:84% the end users are dealing with the foundation. 01:55:57.850 --> 01:56:00.410 align:middle line:90% As a foundation, it's wonderful. 01:56:00.410 --> 01:56:05.250 align:middle line:84% But there is nothing above the foundation. 01:56:05.250 --> 01:56:09.330 align:middle line:84% Nothing good about above the foundation. 01:56:09.330 --> 01:56:11.570 align:middle line:90% That's OK. 01:56:11.570 --> 01:56:15.650 align:middle line:84% When I say that's OK, I mean that's an opportunity for all 01:56:15.650 --> 01:56:20.930 align:middle line:84% of you to go do some innovative stuff and invent new things 01:56:20.930 --> 01:56:25.400 align:middle line:84% and make systems that are much better, taking advantage of all 01:56:25.400 --> 01:56:27.800 align:middle line:90% the advances in technologies. 01:56:27.800 --> 01:56:34.880 align:middle line:84% But what we have today is very, very primitive because 01:56:34.880 --> 01:56:38.200 align:middle line:84% of the lack of form, the lack of-- again, 01:56:38.200 --> 01:56:43.420 align:middle line:90% they are not people oriented. 01:56:43.420 --> 01:56:46.640 align:middle line:84% They are foundational oriented, not people oriented. 01:56:46.640 --> 01:56:50.200 align:middle line:84% Would you agree with that for the most part? 01:56:50.200 --> 01:56:51.880 align:middle line:90% Look at Windows. 01:56:51.880 --> 01:56:54.160 align:middle line:90% Windows works. 01:56:54.160 --> 01:56:57.580 align:middle line:84% We use it, but there aren't too many people that say, 01:56:57.580 --> 01:57:01.340 align:middle line:84% oh, my God, do I love my Windows system. 01:57:01.340 --> 01:57:04.580 align:middle line:84% I'm sure there are people who say that, Bill Gates, 01:57:04.580 --> 01:57:07.820 align:middle line:84% but it's just-- now, why is that? 01:57:07.820 --> 01:57:10.800 align:middle line:90% Well, when was Windows designed? 01:57:10.800 --> 01:57:17.280 align:middle line:84% Windows was designed in the late '80s. 01:57:17.280 --> 01:57:23.360 align:middle line:84% Look at the advances in microprocessor technology, 01:57:23.360 --> 01:57:28.580 align:middle line:84% storage technology, everything else since the late '80s. 01:57:28.580 --> 01:57:30.940 align:middle line:90% Let's take user interfaces. 01:57:30.940 --> 01:57:34.660 align:middle line:84% The reason I'm so excited about virtual worlds, which 01:57:34.660 --> 01:57:36.580 align:middle line:84% comes from game playing and so on, 01:57:36.580 --> 01:57:40.880 align:middle line:84% is I think that's the biggest advance since the GUI. 01:57:40.880 --> 01:57:44.540 align:middle line:84% You know when the GUI was designed? 01:57:44.540 --> 01:57:46.580 align:middle line:84% AUDIENCE: Xerox did it in the '70s. 01:57:46.580 --> 01:57:51.220 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, so here we are with the GUI, 01:57:51.220 --> 01:57:56.500 align:middle line:84% which was designed in a world that was down here, 01:57:56.500 --> 01:57:59.500 align:middle line:84% and we're trying to use it to this-- 01:57:59.500 --> 01:58:02.760 align:middle line:84% so it's not surprising that it just doesn't work. 01:58:02.760 --> 01:58:05.020 align:middle line:84% And it's because of the lack of-- 01:58:05.020 --> 01:58:09.437 align:middle line:90% people haven't focused on form. 01:58:09.437 --> 01:58:11.020 align:middle line:84% AUDIENCE: But also, with their models, 01:58:11.020 --> 01:58:15.800 align:middle line:84% maybe the other objection that people have as you go up, 01:58:15.800 --> 01:58:19.060 align:middle line:84% it's the service economy, and the kind of jobs you get 01:58:19.060 --> 01:58:23.150 align:middle line:84% are probably not high-paying jobs. 01:58:23.150 --> 01:58:26.810 align:middle line:84% And probably you are talking about customer-facing jobs, 01:58:26.810 --> 01:58:27.310 align:middle line:90% right? 01:58:27.310 --> 01:58:30.090 align:middle line:90% 01:58:30.090 --> 01:58:33.867 align:middle line:84% They probably don't require as much intellect as-- 01:58:33.867 --> 01:58:35.450 align:middle line:84% IRVING WLADAWSKY-BERGER: Remember, I'm 01:58:35.450 --> 01:58:40.210 align:middle line:84% talking about somebody designing hospital management systems 01:58:40.210 --> 01:58:42.730 align:middle line:90% for hospitals. 01:58:42.730 --> 01:58:44.867 align:middle line:84% Would you agree with me that is an intellectual-- 01:58:44.867 --> 01:58:46.950 align:middle line:84% AUDIENCE: People have been doing that for a while. 01:58:46.950 --> 01:58:51.330 align:middle line:84% I mean, you have CRM systems for healthcare. 01:58:51.330 --> 01:58:54.410 align:middle line:84% You have expert systems, rule-based systems 01:58:54.410 --> 01:58:55.550 align:middle line:90% for all of those things. 01:58:55.550 --> 01:58:57.530 align:middle line:90% And they do work. 01:58:57.530 --> 01:59:02.250 align:middle line:84% Now, I don't know how elegant the healthcare system is. 01:59:02.250 --> 01:59:07.118 align:middle line:84% IRVING WLADAWSKY-BERGER: I think they work, but they are-- 01:59:07.118 --> 01:59:08.410 align:middle line:90% AUDIENCE: They are inefficient. 01:59:08.410 --> 01:59:12.410 align:middle line:84% IRVING WLADAWSKY-BERGER: Not inefficient, no, no, 01:59:12.410 --> 01:59:15.370 align:middle line:84% this gets back to the conversation we had before. 01:59:15.370 --> 01:59:17.290 align:middle line:90% It's not the performance. 01:59:17.290 --> 01:59:20.150 align:middle line:90% It's that they are user hostile. 01:59:20.150 --> 01:59:22.710 align:middle line:90% So no, no, I'm serious. 01:59:22.710 --> 01:59:25.270 align:middle line:84% And by user, I don't mean programmers, 01:59:25.270 --> 01:59:29.390 align:middle line:84% or I don't mean graduates of SDM. 01:59:29.390 --> 01:59:32.830 align:middle line:84% I mean nurses and doctors who are just 01:59:32.830 --> 01:59:36.050 align:middle line:84% trying to do their job or insurance people. 01:59:36.050 --> 01:59:42.910 align:middle line:84% I don't think they have properly taken them into account. 01:59:42.910 --> 01:59:43.930 align:middle line:90% That's my feeling. 01:59:43.930 --> 01:59:46.110 align:middle line:90% They are foundational. 01:59:46.110 --> 01:59:49.270 align:middle line:90% They were built bottoms up. 01:59:49.270 --> 01:59:52.510 align:middle line:84% Well, let me ask you, do you think the iPod was designed 01:59:52.510 --> 01:59:55.550 align:middle line:90% bottoms-up or tops-down? 01:59:55.550 --> 01:59:57.670 align:middle line:90% What would you think? 01:59:57.670 --> 01:59:58.532 align:middle line:90% AUDIENCE: Tops-down. 01:59:58.532 --> 01:59:59.990 align:middle line:84% IRVING WLADAWSKY-BERGER: I'm sorry? 01:59:59.990 --> 02:00:00.823 align:middle line:90% AUDIENCE: Tops-down. 02:00:00.823 --> 02:00:02.390 align:middle line:84% IRVING WLADAWSKY-BERGER: Tops-down, I 02:00:02.390 --> 02:00:09.070 align:middle line:84% think that elegant design, this is my feeling, is tops-down. 02:00:09.070 --> 02:00:12.972 align:middle line:84% Now, obviously, you need to have a pretty good idea. 02:00:12.972 --> 02:00:14.430 align:middle line:84% AUDIENCE: But most of these systems 02:00:14.430 --> 02:00:17.460 align:middle line:84% are designed for [INAUDIBLE] to use the computer. 02:00:17.460 --> 02:00:20.960 align:middle line:84% You get pervasive computing, maybe it will be much easier. 02:00:20.960 --> 02:00:23.820 align:middle line:84% IRVING WLADAWSKY-BERGER: But we have. 02:00:23.820 --> 02:00:28.060 align:middle line:84% We're forcing the nurses and others 02:00:28.060 --> 02:00:31.860 align:middle line:84% to learn foundational stuff, and they don't want that. 02:00:31.860 --> 02:00:33.140 align:middle line:90% That's what I'm saying. 02:00:33.140 --> 02:00:36.260 align:middle line:84% By the way, I'm not saying this negatively. 02:00:36.260 --> 02:00:41.900 align:middle line:84% Remember, my point is incredible-- 02:00:41.900 --> 02:00:45.380 align:middle line:84% if the question is why is the internet and all these things 02:00:45.380 --> 02:00:47.820 align:middle line:84% disruptive innovations, it's because I 02:00:47.820 --> 02:00:52.300 align:middle line:84% believe that in the knowledge economy, 02:00:52.300 --> 02:00:59.940 align:middle line:84% they will change all this stuff here in profound ways. 02:00:59.940 --> 02:01:04.900 align:middle line:84% But the key profound way is to totally transform 02:01:04.900 --> 02:01:11.360 align:middle line:84% the way human beings interface with technology by making form-- 02:01:11.360 --> 02:01:18.840 align:middle line:84% I mean, by making them much more appealing to humans. 02:01:18.840 --> 02:01:23.640 align:middle line:84% And appealing to humans is to say, well, but you mean artists? 02:01:23.640 --> 02:01:24.540 align:middle line:90% Well, I don't know. 02:01:24.540 --> 02:01:29.440 align:middle line:90% Is industrial design an art? 02:01:29.440 --> 02:01:35.520 align:middle line:84% I mean, it feels at least at the RCA, that's-- 02:01:35.520 --> 02:01:39.280 align:middle line:84% I don't know if they're artists, but there's 02:01:39.280 --> 02:01:42.140 align:middle line:84% a lot of sophisticated stuff that goes into it. 02:01:42.140 --> 02:01:44.280 align:middle line:84% And if you have more and more tools, 02:01:44.280 --> 02:01:48.960 align:middle line:84% massive simulation, design, showing it to people, 02:01:48.960 --> 02:01:55.100 align:middle line:84% so that part, the design part, needs a massive boost. 02:01:55.100 --> 02:01:57.880 align:middle line:90% That's my feeling. 02:01:57.880 --> 02:02:02.320 align:middle line:84% And more than my feeling, I also believe 02:02:02.320 --> 02:02:05.400 align:middle line:90% that's where the money is. 02:02:05.400 --> 02:02:09.060 align:middle line:90% DRAM is a commodity. 02:02:09.060 --> 02:02:13.390 align:middle line:84% It's hard to make a new innovation in DRAM or even 02:02:13.390 --> 02:02:15.510 align:middle line:90% in relational databases. 02:02:15.510 --> 02:02:19.310 align:middle line:84% Whereas up here, somebody reinvents it, 02:02:19.310 --> 02:02:22.270 align:middle line:90% I believe that's where the big-- 02:02:22.270 --> 02:02:24.550 align:middle line:90% it's a personal belief. 02:02:24.550 --> 02:02:26.500 align:middle line:84% Yes, somebody else had a question and then-- 02:02:26.500 --> 02:02:28.042 align:middle line:84% AUDIENCE: I like your ideas that it's 02:02:28.042 --> 02:02:32.270 align:middle line:84% a very interesting topic about how humans need 02:02:32.270 --> 02:02:34.050 align:middle line:90% to interact with the system. 02:02:34.050 --> 02:02:35.270 align:middle line:90% IRVING WLADAWSKY-BERGER: Yes. 02:02:35.270 --> 02:02:40.190 align:middle line:84% AUDIENCE: Because people always underestimate how much effort 02:02:40.190 --> 02:02:42.170 align:middle line:90% it takes to make that happen. 02:02:42.170 --> 02:02:44.470 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly, that's 02:02:44.470 --> 02:02:46.170 align:middle line:90% why you celebrate good design. 02:02:46.170 --> 02:02:49.970 align:middle line:84% Because good design, I think, means somebody got it right. 02:02:49.970 --> 02:02:52.170 align:middle line:84% AUDIENCE: Right, and usually at the end, 02:02:52.170 --> 02:02:55.590 align:middle line:84% the user can give you feedback, good or bad because they don't 02:02:55.590 --> 02:02:58.790 align:middle line:84% know why is it so simple because of all 02:02:58.790 --> 02:03:01.310 align:middle line:84% the engineering behind thinking about how to make 02:03:01.310 --> 02:03:03.630 align:middle line:90% it as simple as possible. 02:03:03.630 --> 02:03:04.250 align:middle line:90% In fact-- 02:03:04.250 --> 02:03:06.042 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, well, that's 02:03:06.042 --> 02:03:08.710 align:middle line:84% why next week, we'll spend a lot of the class 02:03:08.710 --> 02:03:13.550 align:middle line:84% on Eric von Hippel's theory of end-user innovation, 02:03:13.550 --> 02:03:18.930 align:middle line:84% democratizing innovation because you really 02:03:18.930 --> 02:03:20.890 align:middle line:84% want as much as possible, the end 02:03:20.890 --> 02:03:24.370 align:middle line:90% users involved in this process. 02:03:24.370 --> 02:03:26.290 align:middle line:90% AUDIENCE: And I just, I mean-- 02:03:26.290 --> 02:03:28.998 align:middle line:84% IRVING WLADAWSKY-BERGER: So it's precisely because of your point. 02:03:28.998 --> 02:03:31.170 align:middle line:84% The more you get the end users involved, 02:03:31.170 --> 02:03:34.930 align:middle line:84% the more you have a chance of developing something 02:03:34.930 --> 02:03:38.730 align:middle line:84% that will be appealing to the end user. 02:03:38.730 --> 02:03:41.510 align:middle line:84% AUDIENCE: And I, guess maybe just one last point, I thought, 02:03:41.510 --> 02:03:43.490 align:middle line:84% is [INAUDIBLE] I just read an article. 02:03:43.490 --> 02:03:46.070 align:middle line:90% It's about this Mark Weiser. 02:03:46.070 --> 02:03:48.810 align:middle line:84% He said, the most profound technologies 02:03:48.810 --> 02:03:49.977 align:middle line:90% are the ones that disappear. 02:03:49.977 --> 02:03:51.893 align:middle line:84% IRVING WLADAWSKY-BERGER: Yes, I agree totally. 02:03:51.893 --> 02:03:54.230 align:middle line:84% AUDIENCE: So you don't even need a manual to operate. 02:03:54.230 --> 02:03:55.813 align:middle line:84% IRVING WLADAWSKY-BERGER: That's right. 02:03:55.813 --> 02:03:58.530 align:middle line:84% AUDIENCE: That's when you know it's so robust, so reliable, so 02:03:58.530 --> 02:04:00.130 align:middle line:90% easy to use. 02:04:00.130 --> 02:04:02.910 align:middle line:84% IRVING WLADAWSKY-BERGER: You see, that's design. 02:04:02.910 --> 02:04:03.770 align:middle line:90% Would you agree? 02:04:03.770 --> 02:04:07.570 align:middle line:84% That's a conversation about design. 02:04:07.570 --> 02:04:10.010 align:middle line:84% That's exactly a conversation about design. 02:04:10.010 --> 02:04:11.470 align:middle line:90% I agree totally. 02:04:11.470 --> 02:04:14.070 align:middle line:84% That's exactly a conversation about design. 02:04:14.070 --> 02:04:16.390 align:middle line:84% AUDIENCE: Do you see anything that requires expertise 02:04:16.390 --> 02:04:21.670 align:middle line:84% to run something that is not elegant? 02:04:21.670 --> 02:04:26.110 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, and remember, it's a gradation. 02:04:26.110 --> 02:04:29.910 align:middle line:84% You say, oh, well, so the most elegant design 02:04:29.910 --> 02:04:32.510 align:middle line:84% is any of us can do brain surgery. 02:04:32.510 --> 02:04:38.930 align:middle line:84% We're not taking it to that degree. 02:04:38.930 --> 02:04:42.570 align:middle line:84% And even let's take ERP systems for MIT. 02:04:42.570 --> 02:04:46.270 align:middle line:84% It's OK to say, you don't mean consumers 02:04:46.270 --> 02:04:49.230 align:middle line:90% need to be able to operate it. 02:04:49.230 --> 02:04:52.442 align:middle line:84% So that's why it's so interesting. 02:04:52.442 --> 02:04:54.150 align:middle line:84% And we'll talk more about this next time. 02:04:54.150 --> 02:05:01.590 align:middle line:84% You have to have a pretty good idea who is your end user. 02:05:01.590 --> 02:05:05.870 align:middle line:84% Because let's say in a hospital management system, 02:05:05.870 --> 02:05:08.960 align:middle line:84% the doctors, nurses, administrators 02:05:08.960 --> 02:05:10.480 align:middle line:90% they have to be the end users. 02:05:10.480 --> 02:05:12.060 align:middle line:90% Would we agree with that? 02:05:12.060 --> 02:05:15.600 align:middle line:84% Maybe even the patients have to be the end user. 02:05:15.600 --> 02:05:21.180 align:middle line:84% So you need to at least make sure they are usable by them. 02:05:21.180 --> 02:05:24.840 align:middle line:84% I would say at the very least, by them. 02:05:24.840 --> 02:05:27.380 align:middle line:84% But do they have to be usable by somebody in the street? 02:05:27.380 --> 02:05:29.740 align:middle line:90% Not necessarily. 02:05:29.740 --> 02:05:31.980 align:middle line:84% But when they become patients, those parts 02:05:31.980 --> 02:05:36.620 align:middle line:84% aimed at them, if you really want them to participate, 02:05:36.620 --> 02:05:40.420 align:middle line:90% then it has to be very natural. 02:05:40.420 --> 02:05:42.420 align:middle line:84% AUDIENCE: And also, progressively, 02:05:42.420 --> 02:05:46.700 align:middle line:84% what technology is doing is it's replacing 02:05:46.700 --> 02:05:50.500 align:middle line:90% what experts used to do. 02:05:50.500 --> 02:05:55.420 align:middle line:84% Like pregnancy tests, 100 years ago, probably it 02:05:55.420 --> 02:05:57.120 align:middle line:90% required an expert to do that. 02:05:57.120 --> 02:06:00.433 align:middle line:84% Now, you can do it in five minutes or whatever. 02:06:00.433 --> 02:06:03.100 align:middle line:84% IRVING WLADAWSKY-BERGER: I don't think you have to go 100 years. 02:06:03.100 --> 02:06:03.900 align:middle line:90% [LAUGHTER] 02:06:03.900 --> 02:06:07.300 align:middle line:84% I think it was much less than that. 02:06:07.300 --> 02:06:09.400 align:middle line:90% I agree, I agree totally. 02:06:09.400 --> 02:06:10.102 align:middle line:90% Please. 02:06:10.102 --> 02:06:12.560 align:middle line:84% AUDIENCE: It's interesting that you're talking about design 02:06:12.560 --> 02:06:16.680 align:middle line:84% because on this month's issue of Fast Company, 02:06:16.680 --> 02:06:21.267 align:middle line:84% the front cover is actually [INAUDIBLE] designed. 02:06:21.267 --> 02:06:22.600 align:middle line:90% IRVING WLADAWSKY-BERGER: Mm-hmm. 02:06:22.600 --> 02:06:24.080 align:middle line:84% AUDIENCE: And [INAUDIBLE] there's 02:06:24.080 --> 02:06:28.200 align:middle line:84% an article with Steve Jobs saying that 02:06:28.200 --> 02:06:33.220 align:middle line:84% despite the fact that Apple product, they do crash, 02:06:33.220 --> 02:06:36.640 align:middle line:90% and they have their faults. 02:06:36.640 --> 02:06:39.920 align:middle line:84% And w don;t know anything about that because the design 02:06:39.920 --> 02:06:44.740 align:middle line:84% component is so elegant as far as aesthetic that people forget 02:06:44.740 --> 02:06:45.240 align:middle line:90% that. 02:06:45.240 --> 02:06:46.960 align:middle line:90% Apple crashes also. 02:06:46.960 --> 02:06:52.400 align:middle line:84% They'll say, when you compare that with a normal PC, 02:06:52.400 --> 02:06:54.400 align:middle line:90% and that actually-- 02:06:54.400 --> 02:06:57.977 align:middle line:84% I mean, eliminates the fact of wow, it's the same. 02:06:57.977 --> 02:06:59.560 align:middle line:84% IRVING WLADAWSKY-BERGER: That's right. 02:06:59.560 --> 02:07:00.580 align:middle line:90% AUDIENCE: And crashing Apple. 02:07:00.580 --> 02:07:01.460 align:middle line:84% IRVING WLADAWSKY-BERGER: Yeah, no, that's right. 02:07:01.460 --> 02:07:02.120 align:middle line:90% AUDIENCE: A design component. 02:07:02.120 --> 02:07:03.703 align:middle line:84% IRVING WLADAWSKY-BERGER: That's right. 02:07:03.703 --> 02:07:04.930 align:middle line:90% AUDIENCE: [INAUDIBLE] Apple. 02:07:04.930 --> 02:07:06.230 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, I think, 02:07:06.230 --> 02:07:08.170 align:middle line:84% and then we'll go to this section, that-- 02:07:08.170 --> 02:07:15.590 align:middle line:90% 02:07:15.590 --> 02:07:17.550 align:middle line:90% remember, this is so-- 02:07:17.550 --> 02:07:20.110 align:middle line:84% the reason I showed those slides is 02:07:20.110 --> 02:07:25.350 align:middle line:84% because what we're talking about is a talk of technology becoming 02:07:25.350 --> 02:07:27.430 align:middle line:90% embedded into everything. 02:07:27.430 --> 02:07:30.510 align:middle line:90% So look at the typical home. 02:07:30.510 --> 02:07:33.390 align:middle line:84% The typical home, yeah, typical home 02:07:33.390 --> 02:07:36.790 align:middle line:90% in a poor village in Africa. 02:07:36.790 --> 02:07:40.230 align:middle line:84% Typical, let's say, home in the US. 02:07:40.230 --> 02:07:45.430 align:middle line:84% Look at the number of devices in that home. 02:07:45.430 --> 02:07:48.070 align:middle line:84% And would you agree that in many ways, 02:07:48.070 --> 02:07:52.950 align:middle line:84% we may be in the red zone of complexity 02:07:52.950 --> 02:08:00.470 align:middle line:84% for a lot of those devices, and that breakthroughs may very well 02:08:00.470 --> 02:08:05.290 align:middle line:84% come calm from things for the home that 02:08:05.290 --> 02:08:12.090 align:middle line:84% make it much easier for most people to deal with that stuff, 02:08:12.090 --> 02:08:13.330 align:middle line:90% in whatever way? 02:08:13.330 --> 02:08:17.330 align:middle line:84% That just because somebody steps through, put themselves 02:08:17.330 --> 02:08:21.290 align:middle line:84% in the shoes of people dealing with this stuff, who 02:08:21.290 --> 02:08:26.650 align:middle line:84% are not technology experts, and are now dealing with it. 02:08:26.650 --> 02:08:32.090 align:middle line:84% For example, I view TiVo as a wonderful example of design. 02:08:32.090 --> 02:08:35.890 align:middle line:90% Do you all know what TiVo is? 02:08:35.890 --> 02:08:39.890 align:middle line:84% And what TiVo did is they-- boy, they really 02:08:39.890 --> 02:08:42.610 align:middle line:84% thought through the user interface. 02:08:42.610 --> 02:08:46.810 align:middle line:84% And it's just been wonderful what user interface they've 02:08:46.810 --> 02:08:48.690 align:middle line:90% had compared to most others. 02:08:48.690 --> 02:08:53.350 align:middle line:84% And that's what enabled them to stay in business. 02:08:53.350 --> 02:08:55.930 align:middle line:84% And God knows if they'll continue, 02:08:55.930 --> 02:09:01.240 align:middle line:84% but that's a very good example of design. 02:09:01.240 --> 02:09:05.680 align:middle line:84% And again, the beauty is that we can 02:09:05.680 --> 02:09:10.600 align:middle line:84% have this discussion about almost retail store systems 02:09:10.600 --> 02:09:11.700 align:middle line:90% and telecommunications. 02:09:11.700 --> 02:09:16.840 align:middle line:84% The world is full of systems to go engineer 02:09:16.840 --> 02:09:18.700 align:middle line:90% to make them much prettier. 02:09:18.700 --> 02:09:20.640 align:middle line:84% But now we're talking about making 02:09:20.640 --> 02:09:24.080 align:middle line:84% them both excellent from a bottoms-up 02:09:24.080 --> 02:09:28.540 align:middle line:84% and then attack form and function at the same time. 02:09:28.540 --> 02:09:31.520 align:middle line:84% There is no end in sight, and the vast majority of them 02:09:31.520 --> 02:09:33.180 align:middle line:90% are marketplace facing. 02:09:33.180 --> 02:09:36.640 align:middle line:90% They're not lab-facing systems. 02:09:36.640 --> 02:09:46.560 align:middle line:84% So let's talk about how we dealt with some of these key execution 02:09:46.560 --> 02:09:50.760 align:middle line:84% questions in the world of a business. 02:09:50.760 --> 02:09:55.680 align:middle line:84% And I'll talk about some more internal 02:09:55.680 --> 02:09:57.260 align:middle line:90% organizational questions. 02:09:57.260 --> 02:10:00.020 align:middle line:84% And then next week, we'll talk about the more 02:10:00.020 --> 02:10:01.800 align:middle line:90% market-facing question. 02:10:01.800 --> 02:10:07.180 align:middle line:84% So first, the balance between open and proprietary, 02:10:07.180 --> 02:10:11.220 align:middle line:84% in-house versus partnerships and acquisitions. 02:10:11.220 --> 02:10:14.420 align:middle line:84% How did we focus and organize the offerings? 02:10:14.420 --> 02:10:18.840 align:middle line:84% And then, very important, how did we track progress? 02:10:18.840 --> 02:10:23.400 align:middle line:84% How do you track progress in disruptive innovations? 02:10:23.400 --> 02:10:36.580 align:middle line:90% 02:10:36.580 --> 02:10:46.280 align:middle line:84% There is no question that IBM, before we embraced the internet, 02:10:46.280 --> 02:10:52.740 align:middle line:84% IBM was the bastion of proprietary technologies. 02:10:52.740 --> 02:10:58.410 align:middle line:84% In fact, IBM was the essence of a vertically integrated company. 02:10:58.410 --> 02:11:00.550 align:middle line:84% And that was the mainframes were. 02:11:00.550 --> 02:11:08.010 align:middle line:84% That's what they were because they were designed in the 1960s, 02:11:08.010 --> 02:11:12.810 align:middle line:84% where as we discussed before, you had to be vertically 02:11:12.810 --> 02:11:17.050 align:middle line:84% integrated to get the damn thing out the door and get it to work 02:11:17.050 --> 02:11:18.610 align:middle line:90% at all. 02:11:18.610 --> 02:11:22.290 align:middle line:84% But as we talked in the first class, 02:11:22.290 --> 02:11:25.550 align:middle line:84% we didn't continue changing and adapting. 02:11:25.550 --> 02:11:29.450 align:middle line:84% So we stayed with the vertically integrated model 02:11:29.450 --> 02:11:32.670 align:middle line:84% until it was way past its usefulness. 02:11:32.670 --> 02:11:35.890 align:middle line:90% 02:11:35.890 --> 02:11:39.190 align:middle line:84% Lou Gerstner, I don't know how many of you have read the book, 02:11:39.190 --> 02:11:44.050 align:middle line:84% has some superb comments to make about what 02:11:44.050 --> 02:11:46.730 align:middle line:90% it means to stay beyond. 02:11:46.730 --> 02:11:49.730 align:middle line:90% And he explains the white shirt. 02:11:49.730 --> 02:11:54.230 align:middle line:84% Do you all know that IBM used to pay known for white shirts? 02:11:54.230 --> 02:12:01.870 align:middle line:84% And the reason for that was when IBM started selling tabulating 02:12:01.870 --> 02:12:08.070 align:middle line:84% equipment, now we're talking the 1930s, 1940s, 02:12:08.070 --> 02:12:14.990 align:middle line:84% the people who dealt with the customers with equipment, 02:12:14.990 --> 02:12:17.850 align:middle line:84% they would have been mechanics or something like that. 02:12:17.850 --> 02:12:21.150 align:middle line:84% And IBM wanted an image that these 02:12:21.150 --> 02:12:24.550 align:middle line:90% were two notches above that. 02:12:24.550 --> 02:12:26.310 align:middle line:84% And part of getting that image is 02:12:26.310 --> 02:12:30.790 align:middle line:84% to have the people who dealt with the customers 02:12:30.790 --> 02:12:35.390 align:middle line:84% wear clothes like what the customers wore. 02:12:35.390 --> 02:12:38.850 align:middle line:84% And if the customers were banks, guess what the bankers wore? 02:12:38.850 --> 02:12:41.330 align:middle line:84% They wore dark suits and blue shirts. 02:12:41.330 --> 02:12:43.710 align:middle line:84% So IBM told its people, so if they 02:12:43.710 --> 02:12:45.590 align:middle line:84% wore dark suits and blue shirts, you wore 02:12:45.590 --> 02:12:47.230 align:middle line:90% dark suits and blue shirts. 02:12:47.230 --> 02:12:48.770 align:middle line:90% This is in the '30s. 02:12:48.770 --> 02:12:51.630 align:middle line:84% And then Lou goes on to explain, then 02:12:51.630 --> 02:12:55.940 align:middle line:84% everybody forgot why they did it to begin with. 02:12:55.940 --> 02:12:59.020 align:middle line:84% And so here you are in the 1980s, 02:12:59.020 --> 02:13:02.780 align:middle line:84% when everybody is going around in jeans and so on, except 02:13:02.780 --> 02:13:07.240 align:middle line:84% for the IBM people showing up in dark suits and blue shirts. 02:13:07.240 --> 02:13:12.500 align:middle line:84% So that's a very good example of an action 02:13:12.500 --> 02:13:15.820 align:middle line:84% that once upon a time, made a lot of sense, 02:13:15.820 --> 02:13:17.880 align:middle line:90% totally outlived its usefulness. 02:13:17.880 --> 02:13:20.860 align:middle line:84% But through bureaucracy, people forgot, 02:13:20.860 --> 02:13:23.280 align:middle line:84% and they thought, oh, that's what it means. 02:13:23.280 --> 02:13:26.660 align:middle line:90% 02:13:26.660 --> 02:13:35.060 align:middle line:84% But what was really clear is that if you 02:13:35.060 --> 02:13:38.540 align:middle line:84% wanted to embrace the internet, you 02:13:38.540 --> 02:13:41.280 align:middle line:90% had to embrace open standards. 02:13:41.280 --> 02:13:45.120 align:middle line:90% You couldn't have it both ways. 02:13:45.120 --> 02:13:47.380 align:middle line:84% You had to embrace open standards 02:13:47.380 --> 02:13:50.720 align:middle line:90% to embrace the internet. 02:13:50.720 --> 02:13:52.880 align:middle line:84% You can say, well, how about Microsoft? 02:13:52.880 --> 02:13:55.940 align:middle line:84% I guess the answer is, well, there are multiple models, 02:13:55.940 --> 02:13:59.540 align:middle line:84% but Microsoft has embraced a lot of open standards, 02:13:59.540 --> 02:14:03.040 align:middle line:90% just not Linux and a few others. 02:14:03.040 --> 02:14:09.520 align:middle line:84% And if you said, well, so how do you make money 02:14:09.520 --> 02:14:12.300 align:middle line:90% if you embrace open standards? 02:14:12.300 --> 02:14:15.520 align:middle line:90% 02:14:15.520 --> 02:14:20.680 align:middle line:90% The answer is, very carefully. 02:14:20.680 --> 02:14:26.800 align:middle line:90% So let's take browsers. 02:14:26.800 --> 02:14:29.580 align:middle line:84% And when we first started our internet division, 02:14:29.580 --> 02:14:32.360 align:middle line:90% we had an IBM browser. 02:14:32.360 --> 02:14:35.560 align:middle line:84% And we had an IBM HTTP stack, which 02:14:35.560 --> 02:14:43.160 align:middle line:84% is the basic technology on top of which you build web servers. 02:14:43.160 --> 02:14:46.680 align:middle line:84% And we made a decision to get rid of them 02:14:46.680 --> 02:14:51.030 align:middle line:84% because the browsers-- no, but Netscape 02:14:51.030 --> 02:14:52.670 align:middle line:90% was giving away the browsers. 02:14:52.670 --> 02:14:54.930 align:middle line:84% When Microsoft entered with Internet Explorer, 02:14:54.930 --> 02:14:57.310 align:middle line:84% they were giving away the browser. 02:14:57.310 --> 02:15:01.310 align:middle line:84% So it was silly, we felt, to compete in something 02:15:01.310 --> 02:15:03.670 align:middle line:90% that you give away. 02:15:03.670 --> 02:15:10.550 align:middle line:84% The HTTP stack, which we had one, we also got rid of it 02:15:10.550 --> 02:15:16.950 align:middle line:84% because we noticed that when our IBM people needed 02:15:16.950 --> 02:15:19.990 align:middle line:84% to run a really high-performance web 02:15:19.990 --> 02:15:22.910 align:middle line:84% server like the one for the Olympics, 02:15:22.910 --> 02:15:26.070 align:middle line:84% they used Apache, the open server browser. 02:15:26.070 --> 02:15:27.810 align:middle line:84% And we said, why are you doing that? 02:15:27.810 --> 02:15:29.590 align:middle line:84% And they said, well, it's much better 02:15:29.590 --> 02:15:32.710 align:middle line:90% than what IBM is building. 02:15:32.710 --> 02:15:39.030 align:middle line:84% So the HTTP stack, not only did Apache had over 60% share 02:15:39.030 --> 02:15:45.070 align:middle line:84% and IBM's had 2%, but it was better to boot. 02:15:45.070 --> 02:15:47.130 align:middle line:84% That's the marketplace is sending 02:15:47.130 --> 02:15:49.290 align:middle line:90% you signals when that happens. 02:15:49.290 --> 02:15:52.330 align:middle line:84% So we got rid of IBM's proprietary 02:15:52.330 --> 02:15:56.530 align:middle line:84% and aggressively embraced Apache. 02:15:56.530 --> 02:16:03.970 align:middle line:84% However, on top of Apache, we built web server products 02:16:03.970 --> 02:16:06.690 align:middle line:90% like WebSphere. 02:16:06.690 --> 02:16:10.170 align:middle line:90% And WebSphere is proprietary. 02:16:10.170 --> 02:16:15.050 align:middle line:84% It's built on a foundation that is totally open. 02:16:15.050 --> 02:16:22.850 align:middle line:84% But on top of it, what you built around it is proprietary. 02:16:22.850 --> 02:16:26.690 align:middle line:84% So getting back to this question of balance, 02:16:26.690 --> 02:16:30.970 align:middle line:84% if you think carefully your business models, 02:16:30.970 --> 02:16:34.250 align:middle line:84% you can carefully decide that you're 02:16:34.250 --> 02:16:40.370 align:middle line:84% going to embrace open standards in a whole layer of the stack, 02:16:40.370 --> 02:16:45.110 align:middle line:84% in a whole set of components for your offerings. 02:16:45.110 --> 02:16:49.709 align:middle line:84% But then you see the money you're saving here 02:16:49.709 --> 02:16:51.530 align:middle line:84% and the skills you're saving here, 02:16:51.530 --> 02:16:54.950 align:middle line:84% you're now put to enhance it with 02:16:54.950 --> 02:16:57.629 align:middle line:90% your own proprietary offerings. 02:16:57.629 --> 02:17:04.150 align:middle line:84% And then WebSphere is a huge business for IBM. 02:17:04.150 --> 02:17:07.790 align:middle line:84% Let me give you another example of this balance. 02:17:07.790 --> 02:17:14.270 align:middle line:84% Because once IBM got the open bug, we kept marching through. 02:17:14.270 --> 02:17:18.910 align:middle line:84% And as you know, IBM really embraced Linux. 02:17:18.910 --> 02:17:23.150 align:middle line:84% You can say, well, how much money do you get for Linux? 02:17:23.150 --> 02:17:27.450 align:middle line:84% For Linux, we get almost nothing, maybe nothing. 02:17:27.450 --> 02:17:28.490 align:middle line:90% It's open source. 02:17:28.490 --> 02:17:29.950 align:middle line:90% It's unloaded. 02:17:29.950 --> 02:17:35.910 align:middle line:84% There are companies that you can get Linux from Red Hat, Novell, 02:17:35.910 --> 02:17:37.850 align:middle line:90% but we're not a distributor. 02:17:37.850 --> 02:17:44.920 align:middle line:84% However, IBM has a multi-billion dollar business around Linux. 02:17:44.920 --> 02:17:47.139 align:middle line:90% How so? 02:17:47.139 --> 02:17:52.580 align:middle line:90% Linux drives sales of servers. 02:17:52.580 --> 02:17:55.780 align:middle line:84% So servers running Linux, you obviously 02:17:55.780 --> 02:17:59.260 align:middle line:84% don't charge for Linux, you charge for the servers. 02:17:59.260 --> 02:18:02.379 align:middle line:84% All of IBM's proprietary software 02:18:02.379 --> 02:18:05.620 align:middle line:84% runs on the Linux operating system. 02:18:05.620 --> 02:18:10.180 align:middle line:84% So you sell the software, and you make money from that. 02:18:10.180 --> 02:18:12.980 align:middle line:90% 02:18:12.980 --> 02:18:16.660 align:middle line:84% See, this gets back to business model innovation. 02:18:16.660 --> 02:18:23.740 align:middle line:84% If you're very careful about where the role of the standards 02:18:23.740 --> 02:18:29.620 align:middle line:84% and open source, even, and the role of proprietary you view, 02:18:29.620 --> 02:18:33.660 align:middle line:84% to the discussion we had earlier, the open source 02:18:33.660 --> 02:18:40.219 align:middle line:84% and standards as almost a way to outsource some of your work 02:18:40.219 --> 02:18:44.600 align:middle line:84% to this wonderful community, and then you 02:18:44.600 --> 02:18:48.200 align:middle line:84% can participate with that community 02:18:48.200 --> 02:18:51.959 align:middle line:84% and probably save 90% of what it would have cost if you 02:18:51.959 --> 02:18:54.240 align:middle line:90% had to do it all by yourself. 02:18:54.240 --> 02:18:57.240 align:middle line:90% So that's how we did it here. 02:18:57.240 --> 02:19:00.959 align:middle line:84% So first point is it was a balance. 02:19:00.959 --> 02:19:02.480 align:middle line:84% And it's something we keep looking 02:19:02.480 --> 02:19:09.200 align:middle line:84% over and over and over and over and so do every other company. 02:19:09.200 --> 02:19:11.760 align:middle line:84% Any comments or questions about this point? 02:19:11.760 --> 02:19:14.660 align:middle line:84% It's a very, I think, profound point. 02:19:14.660 --> 02:19:15.952 align:middle line:90% Yes? 02:19:15.952 --> 02:19:18.160 align:middle line:84% AUDIENCE: So when you're basing all your applications 02:19:18.160 --> 02:19:18.910 align:middle line:90% with open source-- 02:19:18.910 --> 02:19:20.368 align:middle line:84% IRVING WLADAWSKY-BERGER: Not all, I 02:19:20.368 --> 02:19:22.920 align:middle line:84% didn't say we're basing all of our-- no, no, this is very. 02:19:22.920 --> 02:19:24.520 align:middle line:90% I didn't say all. 02:19:24.520 --> 02:19:27.600 align:middle line:84% AUDIENCE: OK, so you're basing a significant share 02:19:27.600 --> 02:19:30.219 align:middle line:84% of your systems on an open source, 02:19:30.219 --> 02:19:32.680 align:middle line:84% and that open source keeps changing because it's open, 02:19:32.680 --> 02:19:36.639 align:middle line:84% and everybody's adding stuff, it seems that creates a lot of work 02:19:36.639 --> 02:19:38.453 align:middle line:90% in order to keep up with. 02:19:38.453 --> 02:19:39.870 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, but 02:19:39.870 --> 02:19:43.150 align:middle line:84% the good open-source communities, 02:19:43.150 --> 02:19:46.910 align:middle line:84% like Apache or Linux, have governance that 02:19:46.910 --> 02:19:49.470 align:middle line:90% keep the interfaces standard. 02:19:49.470 --> 02:19:52.290 align:middle line:90% 02:19:52.290 --> 02:19:57.030 align:middle line:84% Exactly, you couldn't possibly keep coming out 02:19:57.030 --> 02:20:02.350 align:middle line:84% with releases of Linux that make all applications on top 02:20:02.350 --> 02:20:03.730 align:middle line:90% of Linux incompatible. 02:20:03.730 --> 02:20:05.390 align:middle line:90% You just couldn't do that. 02:20:05.390 --> 02:20:08.910 align:middle line:84% So there is a very good governance project. 02:20:08.910 --> 02:20:11.750 align:middle line:84% And remember, you don't have to join 02:20:11.750 --> 02:20:13.690 align:middle line:90% every open-source community. 02:20:13.690 --> 02:20:17.470 align:middle line:84% You join the ones that you think make sense. 02:20:17.470 --> 02:20:20.710 align:middle line:84% And so you usually do due diligence. 02:20:20.710 --> 02:20:24.650 align:middle line:84% And you join those communities that have good governance. 02:20:24.650 --> 02:20:29.170 align:middle line:84% And part of good governance is exactly what you're saying. 02:20:29.170 --> 02:20:29.670 align:middle line:90% Please. 02:20:29.670 --> 02:20:31.962 align:middle line:84% AUDIENCE: So what happens if somebody, another company, 02:20:31.962 --> 02:20:35.310 align:middle line:84% comes and releases an open-source version of something 02:20:35.310 --> 02:20:36.830 align:middle line:90% like WebSphere? 02:20:36.830 --> 02:20:38.810 align:middle line:84% IRVING WLADAWSKY-BERGER: Well, they exist. 02:20:38.810 --> 02:20:41.090 align:middle line:90% No, no, they exist. 02:20:41.090 --> 02:20:42.110 align:middle line:90% Hey, what's the open-- 02:20:42.110 --> 02:20:43.450 align:middle line:90% AUDIENCE: [INAUDIBLE] 02:20:43.450 --> 02:20:45.158 align:middle line:84% IRVING WLADAWSKY-BERGER: There was Jboss, 02:20:45.158 --> 02:20:46.570 align:middle line:90% and there were some others. 02:20:46.570 --> 02:20:48.230 align:middle line:84% AUDIENCE: Yeah, so in such a situation, 02:20:48.230 --> 02:20:51.270 align:middle line:84% how would IBM still continue to make money out of WebSphere? 02:20:51.270 --> 02:20:52.687 align:middle line:84% IRVING WLADAWSKY-BERGER: Excellent 02:20:52.687 --> 02:20:56.530 align:middle line:84% question because for the most part, 02:20:56.530 --> 02:21:02.530 align:middle line:84% the open-source communities attack a high-volume space. 02:21:02.530 --> 02:21:06.890 align:middle line:84% And so you then give up that market, 02:21:06.890 --> 02:21:09.970 align:middle line:84% and you go to the more complex sections 02:21:09.970 --> 02:21:14.370 align:middle line:84% where customers are willing to pay for the extra function, 02:21:14.370 --> 02:21:18.970 align:middle line:84% for the testing and the support that you don't get here. 02:21:18.970 --> 02:21:20.290 align:middle line:90% That's what it is. 02:21:20.290 --> 02:21:22.590 align:middle line:84% Now, will that always be the case? 02:21:22.590 --> 02:21:24.950 align:middle line:84% Is it possible this will keep growing up? 02:21:24.950 --> 02:21:26.390 align:middle line:90% It's quite possible. 02:21:26.390 --> 02:21:27.470 align:middle line:90% It probably will happen. 02:21:27.470 --> 02:21:31.290 align:middle line:84% And then you have to keep adding-- 02:21:31.290 --> 02:21:34.570 align:middle line:84% or could it be that after a while, 02:21:34.570 --> 02:21:36.460 align:middle line:90% this is no longer a business? 02:21:36.460 --> 02:21:41.260 align:middle line:84% Well, sure, but you can make a lot of money here, 02:21:41.260 --> 02:21:44.277 align:middle line:84% and eventually, you have to move up to the next level. 02:21:44.277 --> 02:21:46.860 align:middle line:84% AUDIENCE: Yeah, because stuff gets more and more commoditized, 02:21:46.860 --> 02:21:48.020 align:middle line:90% you have to keep on innovating. 02:21:48.020 --> 02:21:49.440 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly, 02:21:49.440 --> 02:21:51.480 align:middle line:90% gets back to innovation. 02:21:51.480 --> 02:21:55.200 align:middle line:84% If you decide I'll make a stand and take them on, 02:21:55.200 --> 02:21:58.000 align:middle line:84% often that means you don't want to innovate. 02:21:58.000 --> 02:22:02.640 align:middle line:84% Now, if you're a monopolist, have 90% share, 02:22:02.640 --> 02:22:09.920 align:middle line:84% that's a hell of a stand because usually, 02:22:09.920 --> 02:22:16.200 align:middle line:84% when you have a huge share, you control the prices and margins, 02:22:16.200 --> 02:22:19.440 align:middle line:90% and you have them very high. 02:22:19.440 --> 02:22:22.540 align:middle line:84% So the new things are really disruptive, 02:22:22.540 --> 02:22:24.760 align:middle line:84% like happened to us in mainframes. 02:22:24.760 --> 02:22:28.740 align:middle line:84% Once you're playing in the marketplace, so screw it. 02:22:28.740 --> 02:22:30.420 align:middle line:84% If you have to eventually get off, 02:22:30.420 --> 02:22:34.500 align:middle line:84% you'll find another set of products to go after. 02:22:34.500 --> 02:22:36.380 align:middle line:84% AUDIENCE: If somebody else comes up 02:22:36.380 --> 02:22:38.620 align:middle line:84% with a free version of WebSphere, 02:22:38.620 --> 02:22:41.180 align:middle line:84% that basically means you have to redefine your business. 02:22:41.180 --> 02:22:41.940 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly. 02:22:41.940 --> 02:22:43.620 align:middle line:84% AUDIENCE: Earlier it was developing 02:22:43.620 --> 02:22:45.080 align:middle line:90% WebSphere-like products. 02:22:45.080 --> 02:22:48.940 align:middle line:84% Now it is something else, like, giving services. 02:22:48.940 --> 02:22:51.980 align:middle line:84% IRVING WLADAWSKY-BERGER: Exactly, exactly, exactly, 02:22:51.980 --> 02:22:56.460 align:middle line:84% or keep doing versions of WebSphere that 02:22:56.460 --> 02:23:00.660 align:middle line:84% go up to our business, providing more and more business 02:23:00.660 --> 02:23:06.140 align:middle line:84% services business performance management, and so on. 02:23:06.140 --> 02:23:10.420 align:middle line:84% How about the questions about in-house versus 02:23:10.420 --> 02:23:13.420 align:middle line:90% partnering and acquisitions? 02:23:13.420 --> 02:23:20.060 align:middle line:84% This is another gigantic cultural change. 02:23:20.060 --> 02:23:23.780 align:middle line:84% One of the most interesting meetings I 02:23:23.780 --> 02:23:27.580 align:middle line:84% ever had with Lou Gerstner and companies, 02:23:27.580 --> 02:23:34.770 align:middle line:84% when we decided in mid-1996 that we would 02:23:34.770 --> 02:23:38.670 align:middle line:90% join Sun and embrace Java. 02:23:38.670 --> 02:23:42.690 align:middle line:90% 02:23:42.690 --> 02:23:45.650 align:middle line:90% Do you all know what Java is? 02:23:45.650 --> 02:23:49.030 align:middle line:84% It's a layer of software that works on all architectures. 02:23:49.030 --> 02:23:52.170 align:middle line:90% It was totally developed by Sun. 02:23:52.170 --> 02:23:56.750 align:middle line:84% Sun was IBM's key competitor in the server space. 02:23:56.750 --> 02:23:58.190 align:middle line:90% They were kicking our ass. 02:23:58.190 --> 02:24:01.650 align:middle line:84% They were just killing us in servers. 02:24:01.650 --> 02:24:07.170 align:middle line:84% But Java really was getting a lot of traction, 02:24:07.170 --> 02:24:11.730 align:middle line:84% especially for in the internet space 02:24:11.730 --> 02:24:19.290 align:middle line:84% and especially for products like WebSphere that we wanted to do. 02:24:19.290 --> 02:24:24.430 align:middle line:84% So we had to go and say, we need to do Java. 02:24:24.430 --> 02:24:25.650 align:middle line:90% We need to embrace Java. 02:24:25.650 --> 02:24:26.930 align:middle line:90% It's what we need. 02:24:26.930 --> 02:24:29.950 align:middle line:84% It cuts across all the platforms. 02:24:29.950 --> 02:24:32.810 align:middle line:84% Remember IBM had multiple architectures. 02:24:32.810 --> 02:24:37.310 align:middle line:84% So a layer like Java was particularly important to work 02:24:37.310 --> 02:24:39.990 align:middle line:90% on multiple architectures. 02:24:39.990 --> 02:24:43.630 align:middle line:84% And yeah, we realized we are working 02:24:43.630 --> 02:24:49.230 align:middle line:84% with perhaps our toughest competitor in a very important 02:24:49.230 --> 02:24:54.150 align:middle line:84% segment, but they have a good thing, so we joined them, 02:24:54.150 --> 02:24:56.950 align:middle line:84% and we worked very closely with them. 02:24:56.950 --> 02:24:59.150 align:middle line:84% And that was a critical decision because it 02:24:59.150 --> 02:25:02.270 align:middle line:84% was embracing Java and something we 02:25:02.270 --> 02:25:10.250 align:middle line:84% did called J2EE, Java Enterprise Edition, is that what it is? 02:25:10.250 --> 02:25:10.843 align:middle line:90% AUDIENCE: Yes. 02:25:10.843 --> 02:25:12.510 align:middle line:84% IRVING WLADAWSKY-BERGER: That it allowed 02:25:12.510 --> 02:25:16.670 align:middle line:84% us to move forward and build things like WebSphere. 02:25:16.670 --> 02:25:22.030 align:middle line:84% Similarly, with SOA, IBM partnered with Microsoft, 02:25:22.030 --> 02:25:26.430 align:middle line:84% and IBM and Microsoft have been the primary companies that 02:25:26.430 --> 02:25:31.160 align:middle line:84% have pushed Service-Oriented Architectures, 02:25:31.160 --> 02:25:35.340 align:middle line:84% and XML and all this stuff around it. 02:25:35.340 --> 02:25:37.520 align:middle line:90% And do we always agree? 02:25:37.520 --> 02:25:41.340 align:middle line:84% Hell no, there are huge battles because everybody is trying 02:25:41.340 --> 02:25:44.860 align:middle line:90% to find its own advantage. 02:25:44.860 --> 02:25:50.860 align:middle line:84% However, see this gets back-- remember these pictures, which 02:25:50.860 --> 02:25:55.880 align:middle line:84% is if you're going after this goal, 02:25:55.880 --> 02:25:58.140 align:middle line:84% and if you spend time in a skirmish 02:25:58.140 --> 02:26:00.840 align:middle line:84% here, that your clients could care less-- 02:26:00.840 --> 02:26:04.020 align:middle line:84% do you think a client cares whether IBM and Microsoft get 02:26:04.020 --> 02:26:05.540 align:middle line:90% along? 02:26:05.540 --> 02:26:08.980 align:middle line:84% Shut up and get your goddamn stuff to work together 02:26:08.980 --> 02:26:12.180 align:middle line:90% and let me run my business. 02:26:12.180 --> 02:26:18.660 align:middle line:84% I'm being very serious --that then drives your actions, 02:26:18.660 --> 02:26:20.940 align:middle line:84% as opposed to going to visit a client 02:26:20.940 --> 02:26:24.100 align:middle line:84% and say, let me tell you why that some people are bad. 02:26:24.100 --> 02:26:29.800 align:middle line:84% You can do that if you are competing for a server sale 02:26:29.800 --> 02:26:33.180 align:middle line:84% because my servers are better than their servers, 02:26:33.180 --> 02:26:36.960 align:middle line:84% that's a different thing, and clients understand that. 02:26:36.960 --> 02:26:40.960 align:middle line:84% But if you're talking about thin clients, one integration, 02:26:40.960 --> 02:26:43.800 align:middle line:84% and the fact that you guys don't get along is in the way, 02:26:43.800 --> 02:26:47.960 align:middle line:84% the clients will punish you by not buying from you 02:26:47.960 --> 02:26:50.240 align:middle line:90% if you don't cooperate. 02:26:50.240 --> 02:26:53.400 align:middle line:90% So we cooperated. 02:26:53.400 --> 02:26:58.620 align:middle line:84% At the same time, there were major areas 02:26:58.620 --> 02:27:03.200 align:middle line:84% we were lousy with our internal products, 02:27:03.200 --> 02:27:05.840 align:middle line:90% and we had to make acquisitions. 02:27:05.840 --> 02:27:09.200 align:middle line:84% And probably the two critical acquisitions 02:27:09.200 --> 02:27:15.800 align:middle line:84% we made were Lotus for office systems and collaboration 02:27:15.800 --> 02:27:18.380 align:middle line:84% and Tivoli for systems management. 02:27:18.380 --> 02:27:21.680 align:middle line:84% Those were critical acquisitions. 02:27:21.680 --> 02:27:30.350 align:middle line:84% And prior to those acquisitions, the mood in IBM 02:27:30.350 --> 02:27:34.610 align:middle line:84% was that making an acquisition was a failure. 02:27:34.610 --> 02:27:40.030 align:middle line:84% That was like, you're not man enough to build it yourself. 02:27:40.030 --> 02:27:41.850 align:middle line:90% I'm not kidding. 02:27:41.850 --> 02:27:43.990 align:middle line:90% I mean it. 02:27:43.990 --> 02:27:47.770 align:middle line:84% If you are good, you would have built it yourself. 02:27:47.770 --> 02:27:54.150 align:middle line:84% So the fact that you keep failing for 10 years in a row 02:27:54.150 --> 02:27:59.250 align:middle line:84% on office systems, aren't you man enough to try the 11th time? 02:27:59.250 --> 02:28:03.830 align:middle line:84% Now, usually if you fail for 10 years, 02:28:03.830 --> 02:28:07.470 align:middle line:84% maybe there is something in your DNA 02:28:07.470 --> 02:28:09.150 align:middle line:84% in the culture of the company that 02:28:09.150 --> 02:28:14.190 align:middle line:84% makes it very difficult to do certain kinds of systems. 02:28:14.190 --> 02:28:20.310 align:middle line:84% So as we talked about, I believe last week, 02:28:20.310 --> 02:28:22.390 align:middle line:90% you are not good at everything. 02:28:22.390 --> 02:28:24.750 align:middle line:84% And if you say, well, but I should be good at that. 02:28:24.750 --> 02:28:27.450 align:middle line:84% Well, maybe you should be good at that. 02:28:27.450 --> 02:28:29.250 align:middle line:90% You are not. 02:28:29.250 --> 02:28:35.170 align:middle line:84% So if you are not, then you need an injection of culture 02:28:35.170 --> 02:28:37.650 align:middle line:90% and talent, obviously. 02:28:37.650 --> 02:28:39.230 align:middle line:90% You don't just buy culture. 02:28:39.230 --> 02:28:43.030 align:middle line:84% A McKinsey study that tells you how to change your culture. 02:28:43.030 --> 02:28:48.650 align:middle line:84% You can do that, too, give them a lot of money, and that and $5 02:28:48.650 --> 02:28:50.350 align:middle line:84% will get you a nice cup of coffee. 02:28:50.350 --> 02:28:53.450 align:middle line:84% But when it comes to culture, there 02:28:53.450 --> 02:28:56.210 align:middle line:84% are many other things that will help you a lot. 02:28:56.210 --> 02:29:00.850 align:middle line:84% But what you really need often, as we discussed last time, 02:29:00.850 --> 02:29:03.730 align:middle line:90% is an injection of new blood. 02:29:03.730 --> 02:29:09.210 align:middle line:84% And these acquisitions were new blood. 02:29:09.210 --> 02:29:14.610 align:middle line:84% And IBM culture has changed radically. 02:29:14.610 --> 02:29:18.730 align:middle line:84% So we've become more open to acquisitions like Cisco has 02:29:18.730 --> 02:29:19.990 align:middle line:90% and other companies. 02:29:19.990 --> 02:29:23.600 align:middle line:84% So often there are companies in the marketplace 02:29:23.600 --> 02:29:26.240 align:middle line:90% that are very successful. 02:29:26.240 --> 02:29:27.660 align:middle line:90% They have done something. 02:29:27.660 --> 02:29:28.700 align:middle line:90% Could we have done it? 02:29:28.700 --> 02:29:33.880 align:middle line:84% Absolutely, we could have done it, but we didn't do it. 02:29:33.880 --> 02:29:36.400 align:middle line:84% It's like asking you, well, why didn't you 02:29:36.400 --> 02:29:39.533 align:middle line:84% buy this stock when it was low, and now it's so high? 02:29:39.533 --> 02:29:40.700 align:middle line:90% Couldn't you have bought it? 02:29:40.700 --> 02:29:42.220 align:middle line:90% Yes, you could have bought it. 02:29:42.220 --> 02:29:43.260 align:middle line:90% Why didn't you buy it? 02:29:43.260 --> 02:29:49.240 align:middle line:84% Well, sometimes you do, sometimes you don't. 02:29:49.240 --> 02:29:52.700 align:middle line:84% But it sounds like I'm being glib, 02:29:52.700 --> 02:30:00.000 align:middle line:84% but it takes culturally you have to think objectively as opposed 02:30:00.000 --> 02:30:02.200 align:middle line:90% to feeling like you failed. 02:30:02.200 --> 02:30:05.480 align:middle line:84% And therefore, because you didn't do it, 02:30:05.480 --> 02:30:08.120 align:middle line:84% you're not going to make the acquisition. 02:30:08.120 --> 02:30:10.040 align:middle line:84% And the net result is you won't be 02:30:10.040 --> 02:30:11.660 align:middle line:90% successful in the marketplace. 02:30:11.660 --> 02:30:12.940 align:middle line:90% That's really stupid. 02:30:12.940 --> 02:30:18.940 align:middle line:84% But we had to get that cultural barrier this pushed us to get. 02:30:18.940 --> 02:30:19.920 align:middle line:90% Questions? 02:30:19.920 --> 02:30:25.820 align:middle line:90% 02:30:25.820 --> 02:30:34.140 align:middle line:84% Now, when we were starting with the business 02:30:34.140 --> 02:30:39.940 align:middle line:84% and organizing our offerings, we had 02:30:39.940 --> 02:30:45.740 align:middle line:84% to pick certain things we would do, 02:30:45.740 --> 02:30:48.100 align:middle line:90% certain things we wouldn't do. 02:30:48.100 --> 02:30:52.560 align:middle line:84% I already told you we said browsers. 02:30:52.560 --> 02:30:55.380 align:middle line:90% 02:30:55.380 --> 02:30:59.900 align:middle line:84% First of all, Microsoft and Netscape were having what 02:30:59.900 --> 02:31:02.120 align:middle line:90% people call the browser wars. 02:31:02.120 --> 02:31:04.700 align:middle line:84% Do you all know what I'm talking about? 02:31:04.700 --> 02:31:11.500 align:middle line:84% So could we have thrown our hat in the ring in those browser 02:31:11.500 --> 02:31:12.040 align:middle line:90% wars? 02:31:12.040 --> 02:31:16.180 align:middle line:84% We could have, but we all knew how 02:31:16.180 --> 02:31:18.440 align:middle line:84% that would have ended that we would 02:31:18.440 --> 02:31:22.680 align:middle line:84% have been crushed between these two people fighting 02:31:22.680 --> 02:31:24.640 align:middle line:90% for their lives. 02:31:24.640 --> 02:31:30.120 align:middle line:84% Microsoft felt that if Netscape established dominance 02:31:30.120 --> 02:31:35.280 align:middle line:84% in their browser, it put their whole Windows franchise 02:31:35.280 --> 02:31:36.420 align:middle line:90% in peril. 02:31:36.420 --> 02:31:40.560 align:middle line:84% So they were going to do whatever it took. 02:31:40.560 --> 02:31:45.000 align:middle line:84% And Netscape was fighting for its life, which eventually it 02:31:45.000 --> 02:31:47.760 align:middle line:90% lost. 02:31:47.760 --> 02:31:50.980 align:middle line:84% And we wouldn't have done anything there. 02:31:50.980 --> 02:31:51.700 align:middle line:90% So we didn't. 02:31:51.700 --> 02:31:54.440 align:middle line:90% We got out of that. 02:31:54.440 --> 02:31:59.820 align:middle line:84% But we found that people who were trying to do e-business, 02:31:59.820 --> 02:32:02.260 align:middle line:84% remember, integrate the web with IT, 02:32:02.260 --> 02:32:06.740 align:middle line:84% our enterprise customers needed a number of important things 02:32:06.740 --> 02:32:08.400 align:middle line:90% we were very good at. 02:32:08.400 --> 02:32:12.300 align:middle line:84% Security, remember now, for the first time, 02:32:12.300 --> 02:32:17.870 align:middle line:84% you were opening your jewels, applications, and data, 02:32:17.870 --> 02:32:19.990 align:middle line:90% to everybody. 02:32:19.990 --> 02:32:22.350 align:middle line:84% And I don't need to tell you, not 02:32:22.350 --> 02:32:27.030 align:middle line:84% everybody out there who came in was a good guy. 02:32:27.030 --> 02:32:30.710 align:middle line:84% So all of a sudden, firewalls and security, 02:32:30.710 --> 02:32:38.150 align:middle line:84% that became a very big business, built security, hosting. 02:32:38.150 --> 02:32:40.510 align:middle line:84% Remember, before the internet, everybody 02:32:40.510 --> 02:32:44.470 align:middle line:84% knew exactly how big a machine they needed, how big everything 02:32:44.470 --> 02:32:47.470 align:middle line:84% they needed because they knew exactly how many people were 02:32:47.470 --> 02:32:48.670 align:middle line:90% going to access it. 02:32:48.670 --> 02:32:52.310 align:middle line:84% With the internet, you had no idea. 02:32:52.310 --> 02:32:55.210 align:middle line:84% So a lot of people said, this is a pain in the ass. 02:32:55.210 --> 02:32:57.750 align:middle line:84% Why don't you host it for me, and you 02:32:57.750 --> 02:33:00.910 align:middle line:84% handle the variable scalability of hosting? 02:33:00.910 --> 02:33:03.630 align:middle line:84% So we developed hosting services. 02:33:03.630 --> 02:33:09.750 align:middle line:84% Backend integration, integrate the legacy backends, a lot of it 02:33:09.750 --> 02:33:15.250 align:middle line:84% was IBM systems, with the new web servers from IBM and others. 02:33:15.250 --> 02:33:22.310 align:middle line:84% So we built a whole set of new offerings from our services, 02:33:22.310 --> 02:33:25.770 align:middle line:84% from our products, that were very, very 02:33:25.770 --> 02:33:31.850 align:middle line:84% valuable in the internet and fit what we were good at. 02:33:31.850 --> 02:33:38.390 align:middle line:84% And then we started to organize the internet offerings. 02:33:38.390 --> 02:33:42.130 align:middle line:90% 02:33:42.130 --> 02:33:49.590 align:middle line:84% These three categories, which we actually came up with in '96, 02:33:49.590 --> 02:33:52.150 align:middle line:84% they continue to make a lot of sense. 02:33:52.150 --> 02:33:55.230 align:middle line:84% And we said, when all is said and done, 02:33:55.230 --> 02:33:58.130 align:middle line:84% there are three classes of applications 02:33:58.130 --> 02:34:04.850 align:middle line:84% in the world of the web content, collaboration, and commerce. 02:34:04.850 --> 02:34:09.270 align:middle line:84% And so we started counting everything, also, our market, 02:34:09.270 --> 02:34:11.840 align:middle line:84% our communications, our education, content, 02:34:11.840 --> 02:34:17.400 align:middle line:84% collaboration and commerce over and over and over and over. 02:34:17.400 --> 02:34:19.220 align:middle line:84% The reason this is very important is we 02:34:19.220 --> 02:34:21.480 align:middle line:84% learned to say, well, God, what the hell 02:34:21.480 --> 02:34:26.480 align:middle line:84% is any business application that is too amorphous a subject? 02:34:26.480 --> 02:34:29.040 align:middle line:84% But if we say, OK, let's talk about content. 02:34:29.040 --> 02:34:30.620 align:middle line:90% What do you mean by content? 02:34:30.620 --> 02:34:31.743 align:middle line:90% Well, how about a website? 02:34:31.743 --> 02:34:32.660 align:middle line:90% Do you have a website? 02:34:32.660 --> 02:34:33.952 align:middle line:90% I don't have a goddamn website. 02:34:33.952 --> 02:34:35.900 align:middle line:84% Well, we can help you build a website. 02:34:35.900 --> 02:34:38.000 align:middle line:84% How do you help me build a website? 02:34:38.000 --> 02:34:41.920 align:middle line:84% So now you can start doing offerings, customer 02:34:41.920 --> 02:34:42.860 align:middle line:90% self-service. 02:34:42.860 --> 02:34:46.440 align:middle line:84% By the way, I think customer self-service, in retrospect, 02:34:46.440 --> 02:34:50.440 align:middle line:84% was the killer app in the world of the web. 02:34:50.440 --> 02:34:54.880 align:middle line:84% Things like FedEx, UPS, and huge number 02:34:54.880 --> 02:35:00.800 align:middle line:84% of others that allow people to do for themselves things 02:35:00.800 --> 02:35:04.840 align:middle line:84% they couldn't do before, get access to information, those 02:35:04.840 --> 02:35:07.060 align:middle line:90% just those exploded. 02:35:07.060 --> 02:35:10.300 align:middle line:84% Those were, I think, the killer app. 02:35:10.300 --> 02:35:17.220 align:middle line:84% And those primarily required backend web integration 02:35:17.220 --> 02:35:18.500 align:middle line:90% and security. 02:35:18.500 --> 02:35:24.080 align:middle line:84% Remember, you open up the data center to everybody, 02:35:24.080 --> 02:35:26.160 align:middle line:84% you have to have very good security. 02:35:26.160 --> 02:35:27.740 align:middle line:90% So we developed that. 02:35:27.740 --> 02:35:32.780 align:middle line:84% Collaboration, played very well because of our Lotus 02:35:32.780 --> 02:35:33.940 align:middle line:90% acquisition. 02:35:33.940 --> 02:35:38.140 align:middle line:84% We positioned Lotus Notes very much as a platform 02:35:38.140 --> 02:35:41.180 align:middle line:84% for collaboration and started building 02:35:41.180 --> 02:35:42.600 align:middle line:90% more and more offerings. 02:35:42.600 --> 02:35:47.020 align:middle line:84% Also, a lot of the intranet applications, 02:35:47.020 --> 02:35:50.780 align:middle line:84% intranet are applications inside the company 02:35:50.780 --> 02:35:54.620 align:middle line:84% to help employees share information and communicate, 02:35:54.620 --> 02:35:57.020 align:middle line:90% were collaboration in nature. 02:35:57.020 --> 02:36:01.100 align:middle line:84% And then commerce, and remember, especially-- 02:36:01.100 --> 02:36:08.290 align:middle line:84% not at the beginning '96, '97 but as the internet got better, 02:36:08.290 --> 02:36:10.610 align:middle line:90% e-commerce just took off. 02:36:10.610 --> 02:36:15.570 align:middle line:84% And there was B2C and B2B, and you'll hear, in two weeks, 02:36:15.570 --> 02:36:18.730 align:middle line:84% from Chris McCormick of L.L.Bean, 02:36:18.730 --> 02:36:21.370 align:middle line:84% who was actually our launch partner 02:36:21.370 --> 02:36:23.730 align:middle line:90% for our e-commerce platform. 02:36:23.730 --> 02:36:26.870 align:middle line:84% And this kept going up more and more and more and more. 02:36:26.870 --> 02:36:30.490 align:middle line:84% So we started imposing a structure that 02:36:30.490 --> 02:36:34.670 align:middle line:84% proved very beneficial, if nothing else, 02:36:34.670 --> 02:36:37.490 align:middle line:84% because it allowed the salespeople to know 02:36:37.490 --> 02:36:39.810 align:middle line:90% what the hell to talk about. 02:36:39.810 --> 02:36:42.930 align:middle line:84% Otherwise, they say, internet, well, 02:36:42.930 --> 02:36:44.430 align:middle line:90% good, what do you want me to do? 02:36:44.430 --> 02:36:46.790 align:middle line:84% Well, we have to have the internet. 02:36:46.790 --> 02:36:49.912 align:middle line:84% Well, you have to do more than that. 02:36:49.912 --> 02:36:51.370 align:middle line:84% AUDIENCE: So you mentioned content. 02:36:51.370 --> 02:36:56.690 align:middle line:84% Did IBM ever think of doing a search engine portal? 02:36:56.690 --> 02:37:01.850 align:middle line:84% IRVING WLADAWSKY-BERGER: Yes, and we did some, for patents, 02:37:01.850 --> 02:37:03.690 align:middle line:90% we did a superb-- 02:37:03.690 --> 02:37:10.870 align:middle line:84% but it never developed into a business. 02:37:10.870 --> 02:37:14.630 align:middle line:84% The first year I was the general manager 02:37:14.630 --> 02:37:18.070 align:middle line:84% of the internet division, I spent more time 02:37:18.070 --> 02:37:21.710 align:middle line:84% killing offerings we had started in the past 02:37:21.710 --> 02:37:24.470 align:middle line:84% than building new ones because we had started 02:37:24.470 --> 02:37:32.750 align:middle line:84% a lot of user-oriented activities that we were lousy at 02:37:32.750 --> 02:37:38.270 align:middle line:84% and substituting it for things we were much better at. 02:37:38.270 --> 02:37:44.150 align:middle line:84% Because we just were not good at it, I don't know. 02:37:44.150 --> 02:37:46.730 align:middle line:84% It's almost like if you say, why is again, 02:37:46.730 --> 02:37:50.950 align:middle line:84% Michael Jordan not good at baseball, 02:37:50.950 --> 02:37:54.470 align:middle line:84% not as good in baseball as in basketball, I don't know, 02:37:54.470 --> 02:37:58.290 align:middle line:84% but we were just not as good in those things. 02:37:58.290 --> 02:38:04.510 align:middle line:84% So we tried a lot, and we tried and we killed lots of stuff. 02:38:04.510 --> 02:38:07.860 align:middle line:84% We killed lots and lots of stuff. 02:38:07.860 --> 02:38:11.780 align:middle line:84% Now, the last one is very interesting 02:38:11.780 --> 02:38:16.960 align:middle line:84% because we needed to count something. 02:38:16.960 --> 02:38:20.260 align:middle line:84% Because if you didn't count anything, how the hell 02:38:20.260 --> 02:38:23.380 align:middle line:84% did you know if you were making progress? 02:38:23.380 --> 02:38:29.660 align:middle line:84% Counting revenue and profit didn't make any sense 02:38:29.660 --> 02:38:33.340 align:middle line:84% because this stuff was just ramping up. 02:38:33.340 --> 02:38:39.500 align:middle line:84% So in the first few years, what we counted primarily, 02:38:39.500 --> 02:38:46.880 align:middle line:84% was marquee accounts in each area that we were interested. 02:38:46.880 --> 02:38:50.820 align:middle line:84% So when we had reviews with Gerstner and others-- 02:38:50.820 --> 02:38:58.600 align:middle line:84% so what new clients do we have in content, in collaboration, 02:38:58.600 --> 02:39:04.100 align:middle line:84% in commerce, in finance industry, distribution industry, 02:39:04.100 --> 02:39:06.480 align:middle line:90% transportation industry? 02:39:06.480 --> 02:39:10.600 align:middle line:84% And so we were counting the really 02:39:10.600 --> 02:39:16.600 align:middle line:84% how were we winning in each major area that we were after. 02:39:16.600 --> 02:39:22.040 align:middle line:84% But we were defining winning as these clients were partnering 02:39:22.040 --> 02:39:23.720 align:middle line:90% with us. 02:39:23.720 --> 02:39:28.120 align:middle line:84% The deals were initially not very large. 02:39:28.120 --> 02:39:31.280 align:middle line:84% So from the point of view of the bottom line, 02:39:31.280 --> 02:39:34.520 align:middle line:90% they were minuscule. 02:39:34.520 --> 02:39:38.660 align:middle line:84% But if you didn't do those deals, you didn't exist. 02:39:38.660 --> 02:39:42.880 align:middle line:84% When I say there were minuscule, if you were a brand-new company, 02:39:42.880 --> 02:39:45.140 align:middle line:84% you were very proud of those deals. 02:39:45.140 --> 02:39:51.200 align:middle line:84% Remember the discussions we had that if you're a larger company, 02:39:51.200 --> 02:39:54.800 align:middle line:84% the bar for what catches management attention 02:39:54.800 --> 02:39:59.920 align:middle line:84% is so high that you leave a lot of good ideas on the table. 02:39:59.920 --> 02:40:05.810 align:middle line:84% However, if you count engagements by industry, 02:40:05.810 --> 02:40:08.470 align:middle line:84% by clients, and all of a sudden, you 02:40:08.470 --> 02:40:10.690 align:middle line:84% can start asking questions, well, Jesus, 02:40:10.690 --> 02:40:14.430 align:middle line:84% why did Citibank go with our competitor in this area 02:40:14.430 --> 02:40:15.530 align:middle line:90% instead of with us? 02:40:15.530 --> 02:40:18.190 align:middle line:90% I mean, what's wrong? 02:40:18.190 --> 02:40:20.350 align:middle line:90% Are we upset or not? 02:40:20.350 --> 02:40:21.410 align:middle line:90% Yes, we're very upset. 02:40:21.410 --> 02:40:23.110 align:middle line:90% Well, why did we lose? 02:40:23.110 --> 02:40:28.910 align:middle line:84% So you can start asking questions like that. 02:40:28.910 --> 02:40:32.970 align:middle line:90% We had a few set of projects. 02:40:32.970 --> 02:40:36.510 align:middle line:90% We counted also financials. 02:40:36.510 --> 02:40:39.870 align:middle line:90% But how can I say that? 02:40:39.870 --> 02:40:47.870 align:middle line:84% It wasn't counting like a mature sector doing 02:40:47.870 --> 02:40:51.030 align:middle line:90% through quarterly earnings. 02:40:51.030 --> 02:40:56.790 align:middle line:84% What we were doing, in the early stages, was below the radar 02:40:56.790 --> 02:41:00.090 align:middle line:84% when it comes to quarterly earnings, rounding off, 02:41:00.090 --> 02:41:03.070 align:middle line:84% which is true any time you're starting something new. 02:41:03.070 --> 02:41:08.170 align:middle line:84% However, there were a few key products we were counting. 02:41:08.170 --> 02:41:13.790 align:middle line:84% How are we doing in Lotus products, Domino, in particular? 02:41:13.790 --> 02:41:15.370 align:middle line:90% So we counted those. 02:41:15.370 --> 02:41:18.270 align:middle line:84% How are we doing in WebSphere when it started? 02:41:18.270 --> 02:41:20.010 align:middle line:90% So we were counting those. 02:41:20.010 --> 02:41:21.470 align:middle line:90% How about hosting deals? 02:41:21.470 --> 02:41:23.890 align:middle line:90% We were counting those. 02:41:23.890 --> 02:41:29.170 align:middle line:84% And then we counted subjectively, 02:41:29.170 --> 02:41:36.010 align:middle line:84% how much revenue from the other parts of the company 02:41:36.010 --> 02:41:40.550 align:middle line:84% are being driven as a result of the internet initiative. 02:41:40.550 --> 02:41:46.730 align:middle line:84% So if a client bought extra mainframes because 02:41:46.730 --> 02:41:49.510 align:middle line:84% of this new e-business application, 02:41:49.510 --> 02:41:52.090 align:middle line:84% obviously the revenue for the mainframe 02:41:52.090 --> 02:41:57.210 align:middle line:84% went to the systems unit, but we counted that 02:41:57.210 --> 02:42:00.360 align:middle line:90% as indirect revenue. 02:42:00.360 --> 02:42:09.120 align:middle line:84% And after awhile, we had over 20%, 25% of IBM's revenues 02:42:09.120 --> 02:42:11.320 align:middle line:90% were due to the internet. 02:42:11.320 --> 02:42:13.660 align:middle line:90% And by then, the game was over. 02:42:13.660 --> 02:42:15.760 align:middle line:84% When you get to that number, it's 02:42:15.760 --> 02:42:20.560 align:middle line:84% time to go away because now you're a pain to-- 02:42:20.560 --> 02:42:23.000 align:middle line:90% let them do the work themselves. 02:42:23.000 --> 02:42:26.680 align:middle line:84% So we had to count because if you don't have the discipline 02:42:26.680 --> 02:42:29.760 align:middle line:84% of measuring, you don't have any discipline, 02:42:29.760 --> 02:42:32.640 align:middle line:84% and you don't know if you're winning or losing, 02:42:32.640 --> 02:42:38.280 align:middle line:84% but you cannot count what you count in a mature business that 02:42:38.280 --> 02:42:42.800 align:middle line:84% is material to the quarterly earnings. 02:42:42.800 --> 02:42:44.820 align:middle line:90% Any questions or comments? 02:42:44.820 --> 02:42:51.280 align:middle line:84% And we'll review this some more when we start next week. 02:42:51.280 --> 02:42:52.340 align:middle line:90% Thank you very much. 02:42:52.340 --> 02:42:55.630 align:middle line:90% And I'll see you on Thursday. 02:42:55.630 --> 02:42:57.000 align:middle line:90%