Paul Strassmann
Former CIO for Kraft Foods Inc., Xerox Corp., U.S. Department of Defense, and NASA
Paul Strassmann is currently a Distinguished Professor of Information Sciences at the George Mason School of Information Technology. Previously he was the acting Chief Information Officer of NASA, with direct responsibility and accountability for the NASA computing and telecommunication information infrastructure. Before that, he was the Director of Defense Information, where he was responsible for organizing and managing the corporate information management (CIM) program across the U.S. Department of Defense, and where he had policy oversight for the Defense Department’s information technology expenditures.
Mr. Strassmann’s earlier career includes the Xerox Corporation, where he began as director of administration and information systems, with worldwide responsibility for all internal Xerox computer activities, and from which he retired as Vice President of Strategic Planning for the Information Products Group. He also held the job of Corporate Information Officer for General Foods, and Chief Information Officer for Kraft Foods Inc. He is also the author of more than 250 articles and nine books on various aspects of information technology.
Ed Yourdon: One of the things I want to ask you about, because of your position, is your opinions about some of the trends you have seen over the years as [a] CIO.
Paul Strassmann: Yeah, there are clearly trends. You must understand that I’ve been a CIO since ’61. And it’s all I’ve done. In other words, there are very few people that I know of who have been in the CIO position as long as I have. Because even today I am basically a professor teaching CIOs.
Yourdon: Fascinating.
Strassmann: And there are very few people who made it over two generations. In other words, they used to fall out and fall off at the end of a generation. And maybe they fell off after the second generation, but the life expectancy of a CIO has always been very short.
Yourdon: Yes. That definitely is true.
Strassmann: And I’ve been pulling this stunt now for a long time.
Yourdon: [laughter] Ahh… that obviously will give you a better perspective.
Strassmann: If you’re looking for a catchphrase for the book, for whatever it’s worth, you could say you talked to the oldest holder of the CIO title in, perhaps in the world. Perhaps. You have to validate that.
Yourdon: Well, why don’t I start on something that I gather is part of what you’re lecturing on, about new trends that might influence the industry in general, whether it’s the military or any other industry. Are you spending a lot of time worrying and thinking about things like virtualization and cloud computing and so forth?
Strassmann: Oh absolutely.
Yourdon: Or are there, if you’re looking further into the future, are there other things?
Strassmann: Yes. By the way, my course, that I’m starting … has 13 lectures. And they are three-hour lectures, and two of those lectures are on virtualization and cloud computing.
Yourdon: Okay, well, that obviously gives it some great significance.
Strassmann: And the significance is really driven by the economics, the shifting economics of how do you equip an enterprise with information technology that’s long-lasting? And so you have to go towards cloud computing.
Yourdon: The less exotic sort of form or sister of that is the virtualization approach, which also seems like a clear-cut economic issue for any large organization. They would have thousands of servers on all over the place.
Strassmann: You know, so-called cloud computing, which means lots of things, is really an extension, a progression from virtualization. The foundations of cloud computing were really laid by a Stanford professor by the name of Mendel—who created the idea of a universal virtualization with capability. That firm became VMware.
Yourdon: Ah, okay.
Strassmann: VMware has now about an 80 percent market share of virtualization. And the virtualization then became the [springboard] for going to the next generation—namely, if you can virtualize all these things, that means you can suddenly draw in huge complexes.
You know, we’re talking about billion-dollar data centers. The economics will drive you. Economy of scale will drive you to a huge scale, namely, a cloud. But then you have to do something with software.
Yourdon: Right.
Strassmann: So, I’ve been following VMware, I’m a shareholder and my son is one of the ringleaders. It’s looking at the whole issue of virtualization and cloud computing as a transformation, a whole shifting of the cost structure. And what is really happening with all of this environment is the shifting from a dominant part of the IT budget, which is hardware, to less than 10 percent.
Yourdon: Okay.
Strassmann: And you achieve that through cloud computing and virtualization. And then all of the other stuff is then devoted to other objectives. But when I got started in computing, 90 percent of my budget was hardware. You know, IBM manpower didn’t matter.
Yourdon: Right. [laughter]
Strassmann: Manpower was cheap. Today manpower is extremely expensive.
Yourdon: That’s right.
Strassmann: I don’t know where you are leading with this, but I just want to comment that the whole underlying issue is one of economics.
Yourdon: Well, certainly, I can appreciate that that’s the driver, although it’s interesting that I participated in a cloud computing conference in Rome a couple of months ago, where you could see there’s still enormous resistance—showing that the barrier, the obstacle, is not economics but rather the more familiar things like security and privacy.
Strassmann: Oh, yeah, sure. Sure, that’s the usual stuff. That’s what’s called “friction.”
Yourdon: Meaning what, you expect it to fade away?
Strassmann: Oh yeah, it will all be swept away because the economics, the power of the economics will totally, totally dominate, see?
Yourdon: Aha. Interesting. You know, the kind of separation that’s more readily apparent right now in the industry is the big, regulated companies versus the smaller companies that have no regulation—where there’s less concern about security and privacy and more of a willingness to really take advantage of the economics, versus the more established firms.
Strassmann: Let me give it to you: those are not inconsistent. Security and economics are not inconsistent. Security, depending on how you do the architecture, can be very cheap. The question is, where are you going to put this security?
Yourdon: Ah, okay.
Strassmann: See, you’ve got lots of places where you can put your first line of defenses.
Yourdon: Right.
Strassmann: Right now everybody’s loading their power with firewalls and antiviruses and so forth. You are really like the guy in the cottage 200 years ago—going to Kansas and building a house out of local wood, making his own latrine, and buying a can of kerosene. Now the thing has shifted. Now we are in an urban area now, you are a New Yorker, and the economics of you as contrasted with the farmer in Kansas is totally different. And in the old days, the farmer in Kansas had his own Winchester. That was the security.
Yourdon: Okay.
Strassmann: Well, now we have a different security, but the economics of the police department in New York is totally different to the economics of the guy with the gun in Kansas.
Yourdon: Okay.
Strassmann: So security is a thing that is thrown out without any really thought given to the whole question of what the security issues are and where do you spend your money on security. The Department of Defense now spends over 50 percent of its budget on security. Most of it is wasted.
Yourdon: Because they put it in the wrong place?
Strassmann: Too many places.
Yourdon: Too many places. Oh, okay.
Strassmann: Many places.
Yourdon: Ahh, interesting. And is your argument that because the economics are shifting so rapidly … the security friction will be taken care of because there’s so much more money available?
Strassmann: Yeah, well, one of the ways of dealing with the problem of security is to look at the cost of protection, and what do you protect against? Are you protecting yourself against infiltration? And you know, there’s all kinds of infiltration. Or are you protecting yourself against exfiltration?
Yourdon: Stuff leaking out?
Strassmann: Yeah. You know, I’ve been involved with Citicorp, where money went out because of insiders. So every bank today and every financial institution has a problem with exfiltration. And certainly the Department of Defense has a big problem with exfiltration.
Yourdon: Sure.
Strassmann: So then the question is, if you are confronted with infiltration, exfiltration, what are you going to do about it? And how are you going to protect against it?
Yourdon: Okay, interesting. So those technologies are big ones in terms of your vision of the future. What I see of virtualization is that it’s no longer a leading-edge or early adopter stuff. It’s becoming mainstream.
Strassmann: Yes, it is.
Yourdon: Cloud computing is a little further out.
Strassmann: Well, it depends. You know, I have a list of cloud computing companies. I don’t know if you’ve looked at the list. They’re global now. I looked at companies that provide servers that have over 100,000 servers in one building.
Yourdon: Wow.
Strassmann: So a huge amount of business is now being channeled to cloud computing. One of the intriguing things is that many of the startups are experiments—in other words, if you are in a given corporation, and you want to experiment with something, and they don’t want to let it into the protected area, you just go out to Amazon and you buy yourself a server, for 25 cents a minute. And you buy the server, a big Dell server, for 25 cents a minute.
Yourdon: Yeah.
Strassmann: So you can feed in an application, a complicated application, and for less than $14, get it done. Pay with a credit card, and you’re done. It’s thoroughly accepted today.
Yourdon: Yeah, I agree. And then you can try out your idea, and if it fails, you’ve got no capital.
Strassmann: Well, sure, no problem. And some of the stuff that is available as a service like Google Apps. There are hundreds of thousands of people using Google Apps now.
Yourdon: You’re right. A lot of the articles that I’ve been seeing lately about government agencies switching to Google Apps and Google Mail1 is driven by the economics, so that if there are concerns about privacy or security, those are dwarfed by the savings.
Strassmann: No, no, no, no. No, they are not dwarfed. They are being taken care of.
Yourdon: Ah, okay.
Strassmann: In other words, you cannot just go and do this without taking care of security. My point is that the security on Amazon EC2 is cheap. They give you very good security. They may actually give you, for the same bucks, 25 cents an hour, a security that’s better than anything that you have right now in your data center.
Yourdon: That’s a very good point. I hadn’t really thought about that: that for a lot of companies, if they do make a comparison, they would have to acknowledge that their existing security is not very good to begin with.
Strassmann: Yeah, the existing security, particularly because of legacy instances, implementation. There are too many damn servers. I’m looking at a population of approximately a quarter of a million servers.
__________
1 Gmail.
Strassmann: Doing 7,000 major applications and untold number of local homegrown fixes. You know, I’m looking at 750 data centers just in the Department of Defense. You try to tell me that they have security? I mean, you must be kidding.
Yourdon: I remember you giving me similar numbers back in the early ’90s about the number of payroll systems you had seen in data centers all over the Defense Department.
Strassmann: Oh, yeah. Oh, yeah.
Yourdon: Well, aside from virtualization and cloud computing, what other key technology trends do you see on the horizon that we should be thinking about or planning for? Particularly if you have a longer horizon than the rest of the CIOs that I’m talking to.
Strassmann: Well, the heart of an enterprise is data. The heart of the enterprise is software.
Yourdon: True.
Strassmann: When you look at the budget of a typical installation, you’ll find that the bulk of the people are not in the computer room, but are running around fixing things, downloading things, improving things, and most importantly, doing bridges between system A and system B.
Strassmann: Interoperability. So when you look at your budget, you will discover that you may tie up almost all of your available resources on maintenance. It’s a killer, absolute killer. And, and in addition to this, what you find is that there’s always somebody who needs something—an operator, a financial analyst, a venture capitalist. You know, “I want to do something different. You know, I want to have millisecond response.” So they just go and buy something, and they buy it, put it in, and use it. But then somebody either merges it, or combines it, or needs feeds, and that’s when they start hiring somebody to do feeds. Protecting the feeds and so forth.
Yourdon: Aha.
Strassmann: So you are really looking at an environment where the technology is largely misused because the integration isn’t there. So the big question of integration is, what kinds of applications and through what protocols are you going to be writing those applications? Are web pages to be interoperable?
Yourdon: Okay.
Strassmann: So to answer your question of what is the technology that I view in the next 20, 30 years, it really deals with the subject of integration and service-oriented architecture, where you have a billion-dollar network called the Internet and inside the network survive thousands, sometimes millions of applications that depend on each other. And you don’t have the human intervention labor to do all of that. You have to automate the interoperability between data and applications and then have a layer which then will allow a new application to come in, a new app—or 99 cents.
Yourdon: Right.
Strassmann: A button. And that button will have the protocols, so it will go to discover where the pieces are. And from multiple databases it will do what’s called a mash-up and bring back an answer to you in the protocol that deals with a certain human interface. So the future is not virtualization, the future is not cloud. The future is really in how a CIO will create an environment which will provide his customers with the ability to operate in an extremely information-rich environment that is extremely complex and that must be secure.
Yourdon: Are you essentially talking about the buzzword that for years and years has been known as enterprise architecture integration?
Strassmann: Sure. Well, this idea has been around for a long time.
Yourdon: And I guess that’s my concern. It’s been around and it hasn’t happened yet.
Strassmann: Well, it hasn’t happened for a number of reasons. First, the technology has not been mature. The emphasis on people building their own shanties in Kansas, digging their own latrine, and having a can of kerosene has been very strong.
Yourdon: Right.
Strassmann: There is a whole cohort of contractors who live like that. I mean, the place is full of people who do maintenance programming. That’s their business. They would be all blown away.
Yourdon: You’ve mentioned another aspect of this that I wanted to touch on, and that is the transition from a CIO environment where the enterprise owns and controls the technology versus the world that we’re beginning to see more and more of now, of employee-owned technology.
Strassmann: No, no, no, no, no, no. No, the employee owns the buttons, the apps. The infrastructure —which may be either local for a number of reasons or part of a cloud environment—it’s somebody else’s.
Yourdon: Oh, okay.
Strassmann: So the employee still owns … you know, when you sit down at a Mac, you don’t own its code. You don’t even know what the code is. So if you think you control technology, don’t kid yourself.
Yourdon: Well, you control it in terms of whether you buy a Mac or a Windows machine.
Strassmann: Well, fine, but technology is not the issue anymore. See, for people to go in the way I did it 20, 30 years ago. When I wanted an application, I went in, got a bunch of programmers, we designed it, got a computer and debugged it and did all of the other stuff. But that was home-cooked, that was home-improvised; it was all put together by a bunch of amateurs.
Yourdon: Sure.
Strassmann: Now you have a bunch of people like my son who are really deeply embedded in real fancy software. I mean, really fancy software. And they’re doing all of that. And it’s all available for you by pushing a button.
Yourdon: Okay. One of the battles that I see going on in companies all over the place is the CIO who says, “You can’t have one of these. You’ve got to have a BlackBerry. And I don’t like the fact that there are 300,000 apps and you can choose. I want to choose for you.”
Strassmann: Well, yes and no. It’s an issue because behind all of that thing is the interoperability. Because if I give you an iPod, I have cut a link with my interoperability and my security. I cannot allow that.
Yourdon: Okay. So that’s where the real control comes in; it’s the interoperability.
Strassmann: It’s a question of whether the CIO of the future will really be a sort of master orchestrator, like a conductor—although it’s a poor analogy. But it’s more in that direction than the CIO of the past.
Yourdon: Okay, interesting. Well, the related question I had is: What was the most significant change or development that you’ve seen now that your career is, as you say, much longer than anyone else’s? Two generations of a career, looking back over 40 or 50 years, what would you say have been the most significant changes?
Strassmann: Maybe a few things. Now you buy everything. And the entire progression since ’53 or ’54. In other words, in ’54 I actually plug-wired the plugboards. Nobody does that anymore. So now you are in a much higher level. The technology is in the wall, and now the issues are much bigger. They have to do with the issue of how the computing environment is going to be harnessed to support your business. And so you’ve gone away from technology. You still have to understand how to orchestrate the technology.
Yourdon: Mm-hmm.
Strassmann: But then you are moving to a much higher level. And one of the most interesting things that happened in the U.S. Navy, which I consider one of the leading organizations. Last March they decided to abolish the function of information technology as a separate function.
Yourdon: Really?
Strassmann: They created information technology and intelligence into one unit. Because they basically decided information technology is really an arm of intelligence. Totally brought in, totally new character of people, including some technologists. But when you look at the staffing in the organization, it’s an intelligence organization.
Yourdon: Interesting.
Strassmann: Because with all these Predators running around doing color video data and stuff like that. The issue is what do you do with this massive, gigabytes-per-second downloads in Afghanistan? That is not an IT problem anymore.
Yourdon: [laughter] Of course not. I could go off on that tangent, but let me go back to this question of what has been significant over the last 40 or 50 years. You had mentioned earlier an obvious transition from the era where hardware was expensive and people were cheap, compared to where we are now. What about kind of an even broader version of that? Certainly I remember when I started, because hardware was so expensive, it was a controlled commodity. We were not allowed into the computer room or even the Xerox room.
Strassmann: Look, I built the Xerox video center that ran the United States. I built it in 1971. It was a block building with a deep cellar for tapes and what-have-you with a nuclear resistor.
Yourdon: [laughter] So that the tapes would survive.
Strassmann: And it had a wire fence, and it was called “Strassmann’s concentration camp.” And it had guards at the beginning. In other words, the whole thing, I built a data center. And its own uninterrupted power supply, and nobody could come in without permission and clearance and God knows what.
Yourdon: So now 40 years later, we’re in a world where the hardware is cheap and is a commodity and available to everybody in some form.
Strassmann: No, not cheap. It’s in block houses. You must understand. You can go to Secaucus and see buildings that look like block houses. Those are data centers, and you can’t get in. No way.
Yourdon: That’s true.
Strassmann: And those block houses, you go in, and there’s a row of cages, and in those cages are either dedicated servers or utility servers. And you may have forty, fifty thousand servers in that room. And, by the way, the cost of electricity and air conditioning costs more than the servers. It’s just totally different.
Yourdon: Yeah. Certainly, that’s an interesting transition when you see the effort that Amazon or Google are making in terms of supporting conservation and, finding sources of energy.
Strassmann: What choice do they have?
Yourdon: Yeah. That’s amazing. Well, on some level, though, wouldn’t you agree that there are many forms of technology that are almost free for the individual in terms of being able to go down to the local store and buy it and use it?
Strassmann: Oh yeah.
Yourdon: Versus a generation ago.
Strassmann: You know, anybody can go to Radio Shack and buy some fantastic technology.
Yourdon: Right.
Strassmann: The question is, if you have an organization, how do you bring the technology? I have no problem with people going to Radio Shack and buying some widget. My problem arises when they come with technology and want to draw on data that I own. Under what conditions will they be allowed to do that?
Yourdon: What if they create their own data and communicate it with their friends, sort of outside your borders entirely?
Strassmann: Oh, if they want to do social computing—and by the way, we have that problem. In the Navy, 60 percent of transmission today, particularly on ships, you know, the ships go out for 8 months. The guys get bored silly. They do social computing.
Yourdon: [laughter] So it would seem!
Strassmann: And so the real issue is, how do you make sure—because they are using the same transmission circuits, which [are] satellites—how do you make sure that they are not used for infiltration and exfiltration?
Yourdon: Right.
Strassmann: That becomes a very dicey issue. And it’s a very detailed issue. It has to do with the design of the software. It’s the design of the client or how much do you permit an edition of a browser to be in that machine?
Yourdon: It’s obviously a very serious issue within the military. Do you see that elsewhere?
Strassmann: No, it’s also in every financial organization.
Yourdon: Well, that’s where I was going. Do you see it gravitating, that financial organizations will be the next one?
Strassmann: All commercial. You know, right now, you have this big thing in France because in Peugeot, much of the drawings of their electrical car leaked out to the Chinese.
Yourdon: Ahh, I didn’t know that.
Strassmann: Oh yeah. Everybody’s got the same problem. Some people talk about it; some don’t, but there’s a disgruntled employee sitting in a cubicle, with access and password access. And then, then you have things like intelligence like [Aldrich] Ames. Ames who was in counterintelligence. And he rode through CIA and accessed files, which he had no business accessing, but nobody knew.
Yourdon: Is he the one that provided it to the WikiLeaks guy, to [Julian] Assange? Or is that someone else?
Strassmann: No, Ames provided to the Russians.
Yourdon: Yeah. I thought he was the more traditional one. Who is the guy who gave all of the Wiki stuff?
Strassmann: Oh, you know, a low-level sergeant.
Yourdon: But the same problem?
Strassmann: Same problem, exfiltration. And by the way, the stuff that is being reported is defective. The stuff that is reported that he downloaded, he cut a CD, and sent out by mail for WikiLeaks.
Yourdon: Really?
Strassmann: So now you are dealing with a problem. I have several trays in there [pointing to his desktop computer] with CDs. Now, are you going to permit people to burn their own CDs? And, you know, today on a DVD, you can put a big database. Right now, WikiLeaks has a list of all the people who had accounts in the Cayman Islands. You know, that’s a three-million download.
Yourdon: Right. [laughter] Amazing. Any other significant changes and developments that you think have really been worth mentioning over your career?
Strassmann: Yeah, the awareness of money. You see, it used to be that everybody was budgeting the cost of money. Much of the cost used to be capital purchase. So everybody, the controllers, were watching capital purchases and head count, and that was it. And it was head count in the data shop. Well, I beat this thing very quickly by deciding to move my output out to the user.
Yourdon: Ahh.
Strassmann: And the clerks loved to be computer operators, so I put the printer out with the users. I reduced head count in my bursting decollection room, which was enormous. Where I showed that it was someone else’s payroll. I’m just giving you little tricks you can play, you see? It’s a CIO against a moving . . . it’s a shell game. So, the real issue is that when you start looking at these big trading rules and you say, “What are these people doing?” Much of what they are doing is actually data processing.
Yourdon: Okay. So that has been a shift.
Strassmann: So you are suddenly seeing a shift which is externalization of IT from an enclave, which was identified as a capital cost enclave to something which is now part of overhead. It’s part of the overhead cost. It’s a totally different way of looking at it. In other words, the people who are sitting in the room are actually information processors. They do very little trading.
Yourdon: Right.
Strassmann: So, you know, I’ve been tracking the costs of overhead in the United States, and one of the most interesting cases comes from General Motors. General Motors bought EDS.
Strassmann: I was very much involved in that. What really basically happened is that it showed up as an acquisition capital cost, but the operating cost moved into the overhead of General Motors. So as General Motors started outsourcing … by the way, General Motors, at the worst time, 87 percent of their cost were purchases of parts.
Yourdon: Wow. Really?
Strassmann: The rest of it was overhead. Information processing. So the ratio of the truth-to-tell ratio went the wrong way. So, coming back to your question, what I do see is recognition that today the bulk of the population of America [are] information processors. Hardly anybody makes anything.
Yourdon: That certainly is true.
Strassmann: So one of perhaps the major insights for your book is that when I look at America and where we are going, I’m seeing the dominant occupation is information processing jobs. Which is not just people in jobs labeled as IT, but people who do information processing.
Yourdon: Across the board.
Strassmann: And, basically we don’t have secretaries anymore. So everybody got upgraded. The women became administrators. The women love computers because it allows them to upgrade themselves from a secretary position to an administrator.
Yourdon: Aha. Interesting. If there was a list of really significant developments over the last 40 or 50 years, would you include the Internet or the Web or Google or those things?
Strassmann: Oh yeah, yeah. Sure, the Internet is very important.
Yourdon: Okay.
Strassmann: Sure. Very important and growing in importance. There are going to be billions of nodes on the Internet, and it’s growing in complexity and there are issues of cost. There are some deep issues of cost—who is actually paying for Internet?
Yourdon: Right.
Strassmann: Who is making a profit out of the Internet? There is increased demand for high bandwidth, so you’re talking about 10- to 30-gigabyte circuits. There’s also the question of computing at the edge rather than computing in the center in order to reduce latency because if you are in a certain situation, particularly in Wall Street. Some of the trading is now, people are fighting nanosecond delay in differences.
All I’m saying is there’s some deep, deep issues here of latency, availability, up time. In other words, is 99.999 percent an acceptable level of reliability? Very difficult to do.
Yourdon: Yeah.
Strassmann: And also that with technical issues, where do you put your redundancies? Because there is no way of improving reliability on a single server. No way, can’t be done. I mean, it goes only so far and then you’re dealing against the inexorable problem of electrical or mechanical failure. So then you have to start working on the redundancies. So the issues are enormous.
Yourdon: I certainly agree. Let me jump on to an area we’ve not talked about and I’m certain you have some strong opinions. The whole generational issue. Do you see significant differences in the behavior or attitude of the generation of workers, IT workers, in particular, coming out of college today as opposed to a generation or two generations ago?
Strassmann: Well, it varies. The funny thing is, at the elite end—I have a grandson at Carnegie-Mellon. He’s taking a degree in software engineering. You know, this guy is going to be a hotshot.
Yourdon: Right.
Strassmann: And I have another one at Virginia Tech, also doing electrical engineering. So, you know, I’m keeping everybody on the narrow. Those guys are going to be hotshots and they’re going to do very well, unless they do something crazy. My son has a PhD from MIT. He’s doing extremely well. He’s a programmer. Nothing wrong about that, my friend.
Yourdon: Right.
Strassmann: He doesn’t want to be a big boss, because that is not where the fun is and where the money is. They pay him extremely well.
Yourdon: But as you say, those are all examples of the elite.
Strassmann: It’s the elite. And by the way, that’s what the Chinese are going into. You see, the big problem we have is when you look at the Silicon Valley software factories, half of them are not Americans. The class I’m going to be teaching, half of my class will not be Americans.
Yourdon: I think that’s only a problem if we make them so miserable that they want to go back or we make it so difficult to get in that they can’t get in. I think it’s wonderful if we attract them in and they stay here.
Strassmann: Well, it varies, and we cannot get into this. The big problem I have is with the young generation. They’re no good, they’re really no good. They are superficial. You know, I’m also an educator, so I have them at my class. They don’t bore down; they are not engineers.
Yourdon: Ahh, okay.
Strassmann: Their mental state is behavioral, PowerPoint slides, the gloss. And I’m talking about fairly high-level people. I’m talking about rear admirals and up. They’re an outgrowth of an education that is not rigorous, that is not science-oriented. I really despair of the current generation. The bulk of it is not good, not suitable.
Yourdon: And you think this has been going on for a generation?
Strassmann: Oh, this has been creeping up since the ’60s.
Yourdon: Certainly you hear a lot of people complaining about the level of superficiality of the younger generation, that they don’t bore down.
Strassmann: They don’t bore down. In other words, you cannot have an in-depth conversation; they will just glide away. They don’t have the mental mechanics of an engineer, a scientist to start going into the details and starting to examine how the thing works.
Yourdon: Do you know why that is true? Or how that has happened?
Strassmann: I absolutely blame grade schools, and from grade schools up. It’s all this grade school, high school stuff. You know, I have seven grandchildren who have gone through the public school system, so I watch these kids. And some of them are doing very well, and I have one grandchild who is a marvelous kid. He’s in California, he’s in Davis, the University of California in Davis. He’s a nice kid, a very nice personal kid, but he’s a fluff. By the way, he’s a better personality than my geeks.
Yourdon: [laughter] Well, all of these kids, obviously, are coming out of high school or college with an exposure to technology and information.
Strassmann: Oh they think they know technology. They think they know computers.
Yourdon: Okay.
Strassmann: I mean, please, give me a break. They have learned how to operate toys.
Yourdon: True. And I’m not just talking about the ones that are going to go into a formal career of IT but, generally speaking, the white-collar workers who come into the workforce with their toys. And, as you say, they don’t have the ability to bore down into any problem. Do you see them making bad use of their toys or superficial use of their toys?
Strassmann: Well, it varies. They just can’t cope with it, so you hire more people to do less and less. And then you always find a contractor. And then you hire a contractor to do the contract. And then the contractor provides the brains, the analysis, but the contractor will make sure that whatever the contractor does for you is not fungible, which means that the contractor will stay there for a long time.
Yourdon: Or come back periodically to tweak it?
Strassmann: Yeah, yeah. And, so, I’m very unhappy with the modern generation. They are just not . . . this is not an information society.
Yourdon: Could it also possibly be, at least partly, that they have a whole different focus instead of the priorities their parents’ generation did? Let me give you an example. I went out to California in the fairly early days of the whole Web 2.0 movement and talked to one guy in a start-up company who said, “Today’s colleges have never seen Microsoft Outlook. And if they did, they would be horrified, not just by the ugly user interface, but by its task orientation.” Which appeals to me, because I get up in the morning and my first question is, “What tasks do I have today?” which is often a function of the e-mail messages I’ve gotten overnight, with people demanding this or that and everything.
Strassmann: Yeah.
Yourdon: Well, this fellow in California said the average college kid gets up and his first question is, “Where are my friends and what are they doing? And how can I find them?”
Strassmann: It’s social. It’s social.
Yourdon: So they need tools, calendars, and things of that sort that cater to that orientation. And they find the older generation of tools completely alien from that perspective.
Strassmann: Let me just say, everything was fine until three years ago, when things went to hell.
Yourdon: What happened then?
Strassmann: We have now $15 trillion’s worth of debt and going up.
Yourdon: [laughter] Okay.
Strassmann: So everybody was fiddling and singing while it looked like everything was just great. Suddenly, the young people have 20 percent or more unemployment. Can’t get jobs. There are no jobs there, because the manufacturing jobs are gone. The information jobs are not available because I need now less and less information people, and I will only hire the Carnegie-Mellon guys who can really do the job for me.
Yourdon: Yeah.
Strassmann: So we’ve got a problem. We’ve got a social problem. Now in Tunisia, you know, there was 30 percent unemployment for people under 25. Well, they’re going to have a revolution. You are going to have a social problem in the United States with a large mass of unemployed, young, college-educated people who think they’re entitled to a job of consequence.
Yourdon: Aha.
Strassmann: So the growth in America among the intellectuals has been in government-supported activities. Enormous growth.
Yourdon: Interesting. Well, does all of this provide any guidance or recommendations for the CIO? What should you do with this wave of graduates coming in now?
Strassmann: Do exactly what I’ve always done.
Yourdon: Which is what?
Strassmann: I’ve done this consistently. What I’ve done starting at General Foods, I made sure that I gave summer jobs to people that I could handpick.
Yourdon: Okay.
Strassmann: So you handpick them. Now, half of them will be gone. But you watch them, you give them a summer job, which is better than any other summer job they can get.
Yourdon: Absolutely. I’m one of those people. [laughter]
Strassmann: Yeah, and, and some of these guys actually went to grad school and were still working and then they got full jobs. And then when I moved to Xerox, they moved with me.
Yourdon: Oh, that’s right. You started that way, too, when you were at the Sloan School.
Strassmann: So, my advice to the CIOs today is what the baseball people do. Have a training farm somewhere. You just train them and then you get the good guys.
Yourdon: So that’s how he or she can groom their own IT staff. What do they do with the hundreds or thousands of people they see being hired into the marketing department or finance department, all of these other people?
Strassmann: You can’t do that. You’re a CIO. You have a given power position. You have a budget.
Yourdon: Okay.
Strassmann: Everybody has a budget, okay? So the question is what are you going to do? You’re asking me what practical recommendation do I have for CIOs, and my answer is, grow the young guys and nurture them. And bring them along.
Yourdon: But, as for all those other people? Because part of the problem is that they’re bringing their own toys with them and their own expectations about the technology they’ll use.
Strassmann: Yeah, but up to a point. They can bring their own toys, but you cannot allow those toys to come into the database. You must understand, when all is said and done, the CIO is a guy with a big stick. He has a budget, he has a big stick, and he has power.
Yourdon: Right.
Strassmann: Now, if you don’t have power and you don’t have budget, you are nothing. So, let’s talk about practical politics, the Machiavellian view of the situation. As a CIO, you must maintain a position of power. Otherwise, you’re gonna find yourself out of a job.
Yourdon: And how do you do that? How do you maintain this position of power?
Strassmann: Well, it varies. At General Foods I created an alliance with the CFO, who really then was willing . . . that was the first time ever. You must understand, the battle between the CFO and the CIO is as old as … this is classic.
Yourdon: Sure.
Strassmann: The CFO always used to own this thing. And also sucked off most of the output out of the IT. And this lasted for a long time, for some organizations.
Yourdon: Probably a generation or more.
Strassmann: Oh, almost a generation. It varies from place to place. What happened was that at General Foods, the controller—a guy by the name of McDade—used to be MacArthur’s intelligence officer.
Strassmann: Yes. And so McDade and I go out for lunch, and McDade is just trying to size me up. I’m the new guy on the block. And I said, “You know, the problem is that you control it all now, and you think you have control, but these marketing guys are really on their way out, doing their own thing.” And the big issue in those days was advertising. And the money that Finance would spend and control and the stuff they were controlling at the factory was this much [gesturing with thumb and index finger close together]. The big money at General Foods was for advertising. And I said, “Well, Tom, you now, these guys go out to BBDO2 . . .”
Yourdon: Right.
Strassmann: “And renting 7090s and really playing the big game, and this is getting away from you.” So McDade looks at me. He said, “I understand now.” Now, I was just a low guy when a week later I become the director of information for General Foods.
Yourdon: [laughter] Aha.
Strassmann: And so that’s when the game played, and I had an alliance with McDade; then McDade decided to retire and I was his retirement play. I mean, he was an elderly gentleman, a very fine, educated, Harvard kind of person that used to gravitate toward CIA. Yale and Harvard people used to go to the CIA.
Yourdon: That’s right.
Strassmann: See, that’s the way it used to be done. The moment McDade retired, of course, all the knives were out. The positions of all the controllers was, “We’re going to cut his throat.” Well, as they were ready to cut my throat, I get a phone call from the controller of Kraft: “You know, we hear good things about you. Would you like to come over as director of information for all Kraft worldwide?” And I said, “Kraft? You know, I’m not sure, you know, as compared with General Foods.” He said, “Well, you know what? Why don’t you have a look at our financial statement?” It turns out Kraft was much bigger than General Foods.
Yourdon: [laughter]
Strassmann: And there was a raise. So I just inherited one controller and went to work for another controller. So you create alliances. When I got into the Department of Defense, the issue was one of alliances. I came to the Department of Defense when the Soviet Union gave up.
__________
2 A global advertising agency.
Yourdon: That’s right.
Strassmann: Congress decided to declare a peace dividend: $74 billion. And Congress can just take the money, yank the money out, and it’s gone. So as they yank it out, they say, “Oh, by the way, we’re going to leave you three billion to do efficiency. You’re going to make it up with efficiency.” Cheney turns around and brings in the vice chairman of General Motors as his deputy secretary. Now if you understand politics, the secretary is Mr. Outside, the deputy secretary is Mr. Inside.
Yourdon: Right.
Strassmann: Okay. The deputy secretary, an outstanding MIT engineer, a spectacular engineer, had enough of General Motors and Mr. Perot, so he was ready to roll over. He comes into the Pentagon, looks around, and well, months later, he says, “I’m going to start a corporate information management initiative. And I know just the bastard who knows how to do it.”
Yourdon: [laughter]
Strassmann: Because I did a hatchet man’s job at General Motors. I took about $300 million out of Mr. Perot. So he brought me in. So I was Atwood’s man, and everyone knew in the building that I saw Atwood once a week. You know, the rest of it is just details.
Yourdon: Yeah.
Strassmann: So from a political standpoint, if you want to be a CIO, you must have an alliance with a source of power. Otherwise you can’t play.
Yourdon: Interesting. Well, that plays into my next question, and, in fact, you may have given the answer. Obviously if you’re a CIO and you’ve got lots of people below you, you can usually spot one or two aspiring CIOs—the person who says, “Some day I want his job.”
Strassmann: Oh sure.
Yourdon: What’s the key advice that you would give that person?
Strassmann: Give them hell. Just give them as much work as they can carry, just overload them, just break them.
Yourdon: Break who? This aspiring CIO?
Strassmann: Break them with work.
Strassmann: Just see how much they can do. A good man can do an infinite amount of work.
Yourdon: So your advice to aspiring CIOs would be: show your ability to take on an enormous amount of work and succeed with it?
Strassmann: Oh yeah. And they may stay with you, they may go somewhere else, who knows? But you’re going to get a couple of years of good work out of them.
Yourdon: Oh, okay.
Strassmann: But some of these are loyalties, which then persist. So when you move from Kraft to Xerox, there’s a whole bunch of people who come over to Xerox.
Yourdon: Interesting. That actually is part of a related question, which I hadn’t really thought of when I started this whole thing— and that is, CIOs generally have a whole team of immediate subordinates that help them get their job done, and the question here is: as a CIO, what key qualities or characteristics do you look for among the people that form that team that you depend on?
Strassmann: Hard work. Long hours.
Yourdon: Okay.
Strassmann: When there is a crisis, they’re always available.
Yourdon: Ahh, okay. Something I’ve heard that I didn’t expect is CIOs telling me, “I want people on my team that I can get along with, because if you’re going to be working hard, 12 hours a day, you want to have people you like.”
Strassmann: Oh, you have to get along, but you don’t have to be buddies, everybody’s buddies. I don’t believe in that.
Yourdon: Okay.
Strassmann: Everybody knew I was fair, but I was a bastard. I really was. I was driving to consolidate data centers at Xerox in order to come into a company that only 67 percent of its bills got out on time. Inheriting a data center run by local controllers and mash it all together in two years. You know, that took some muscle.
Yourdon: And in terms of looking for the people who were going to help you do that, then, you were mostly looking for people who would work really hard, be equally tough?
Strassmann: Yeah, you needed people who really go and work.
Yourdon: And who would be there if there was a crisis?
Strassmann: Yeah, you know, this is the military view. You must understand that if you were a commanding officer. You must understand that I am a military man. You know my military service? My father was a military man, one of the few Jews who had a command of a regiment in World War I.
Yourdon: Ahh, okay.
Strassmann: So, I go back. And in the military, when you go out there and there’s a bunch of bad guys out there, you know the team spirit and the dedication and the commitment and the loyalties are very important.
Yourdon: That certainly does make sense. But let’s go further with that. If you consider that sort of thing a priority, have you seen situations where some of the other business executives might misunderstand or feel differently?
Strassmann: Oh, absolutely. You know, there is an absolute classic thing against a controller and that never goes away. As a matter of fact, Harvard Business School has a case study of a division controller at Xerox versus Strassmann. There is such a case study.
Yourdon: Aha.
Strassmann: And this guy, who hoped to be the president at Xerox, his name was Engelman. He decides he’s going to attack the IT establishment when I was ready to take computers away from these little places where they were accumulating their own enclaves of power. And I had none of that. Well, you know, it’s nasty stuff, nasty stuff.
Yourdon: And so that just gets to be a power struggle?
Strassmann: Well, it has to be done. First, it has to be done in a very gentlemanly way. Never personalities.
Yourdon: Okay.
Strassmann: You never use personalities. You have to make sure you tow the line on HR-related issues in tough situations so that you are bulletproof if things get ugly. And there have been issues, including legal cases.
Yourdon: Mm-hmm.
Strassmann: And then you have to spend time with lawyers, because you have to spend time finding out what is feasible and what can be done and what accommodations can be made. This is a tough business. To be a CIO is a tough, tough business.
Yourdon: Yes. Certainly, I am coming to appreciate that more and more as I talk to people. Actually, that leads us to the next question, which is: it’s a tough business, but if you succeed, then what? You know, you say you’re still a CIO in a sense, and you’ve gone on obviously to kind of an advisory, mentoring, professorship role.
Strassmann: Well, what happened was I did my bit, 18 years at Xerox. And Xerox is going down the hill also.
Yourdon: Right.
Strassmann: Okay? So there’s always a time to bail out. And then when you bail out, do you look for another CIO job or do you become a consultant? You know, the usual stuff? To become a consultant is death, I mean, that’s death. A CIO who becomes a consultant, it hardly ever works. And I know lots of CIOs who have tried this thing and who just couldn’t engage.
Yourdon: Well, but it sounds in your case that there were a couple of situations where you weren’t a consultant in the traditional definition, but you contracted in to spend a certain amount of time doing consulting work.
Strassmann: Yeah, I spent a year with AT&T. The vice president of IT for AT&T hired me to do a specific job, and I spent a year there.
Yourdon: In a sense, General Motors did a similar thing.
Strassmann: I have never done what’s called “spot consulting.” Here and there I give a speech, or what-have-you, but I’ve never done what’s called “taking consulting jobs.” I look at the things. And this Defense Department thing just came over the transit. Now, I’ve done lots of things for the Department of Defense before. In other words, my approach is usually suitable for lots of things for lots of people.
Sometimes for no compensation. And sometimes it pays off and sometimes it doesn’t. So, I did a number of things for General Motors and then I end up in the Department of Defense, and I stay in the Department of Defense. My biggest confrontation in the Department of Defense was with the controller of the Department of Defense because the moment I walked into the Department of Defense, I said, “We cannot leave IT under the controller.”
[joint laughter]
Strassmann: I’ve done the same thing over and over again. There’s a procedure for doing this thing, and there were a number of reasons to yank it out from the controller, because the controller didn’t really control anything. The controller just controlled capital budget, which was not the issue. So, the thing is yanked out and put into a separate organization. It gets a very high position, by the way. You know, I don’t even want to brag about the kind of position I had. So I do my Department of Defense job. And then, of course, we were going to do it for eight years, but time runs out because of the election. So Clinton comes in—I will not work for Clinton or any of his appointees, which by the way turned out to be a disaster anyway. So, I go off and then look around, and then some people say, “You know what? We would like you to do some teaching for us.” So I start liking the academic thing. The academic thing gives you a tremendous amount of freedom.
Yourdon: Right.
Strassmann: You go somewhere with a card, “I’m a consultant.” You know, that’s terrible. If you come in as a distinguished professor, it’s, “Okay, well.” It’s the same person, but it’s different. So I do this, and then one day I get a phone call from NASA. “Umm, we have a problem with NASA. We have a new administrator.” “Who is the new administrator?” Well, the controller from the Department of Defense now becomes the administrator of NASA because NASA has deep financial problems, so they yanked out the financial guy and put him in charge of NASA. Stupid thing. “And [laughter] he thinks very highly of you. Would you come down and just review the situation?”
So I go down to NASA, arrive early in the morning. . . I know that these people always work early in the morning, so I arrive early in the morning. Very important thing, by the way. The key levels in the government, people start working at 7:00 or 6:45.
Yourdon: Ahh.
Strassmann: Oh yeah. They show up early. They show me the whole thing. Then I have lunch. After lunch, they say, “The administrator would like to see you.” So I go in and there is my friend, Sean O’Keefe, the administrator: “Oh, so nice to see you. I’m so glad you came. Please tell me what you have seen.” And I just dump it on him. I said, “You guys are so screwed up. You guys don’t know what you’re doing. You are wasting money. You are spending money with Peat, Marwick, [Mitchell & Co.], a billion dollars on something that could be done on a small computer.” They had a big SAP program, which can suck in an enormous amount of money without ever showing any results.
Strassmann: So, Sean O’Keefe sits there and all of his guys, they’re all lots and lots of smiling. And I just say, “Look guys. I don’t know what I can do for you guys because you have made a commitment, you have a contract, you are going to be spending a billion dollars on something totally useless. The various centers are totally independent computers. I don’t know what to do for you.” So after I finish this thing, O’Keefe turns around and says, “Well, Strassmann, you come here and you straighten it out.”
I said, “Well, you know, guys, I just finished six years in Washington. You don’t want me to come to Washington.”
“Well, we’re going to make it attractive for you.”
I said, “Well, you know, I will have to talk to my wife. Let me go home and think about it.”
So O’Keefe, who is a really sharp diplomat says, “Well, if you go home, you’ll never come back. You know what? I’m going to make a bet with you. I’m going to offer you the job of CIO of NASA, and if you accept, I’m going to have a badge from NASA for you by five o’clock this afternoon.”
And this is like three o’clock in the afternoon. Now, I know enough about Washington to know that that can’t be done.
Yourdon: Right.
Strassmann: You know, you have civil servants and on and on. It’s what’s called a “senior executive position.”
Yourdon: Right.
Strassmann: And I’m stupid. I’m really stupid. This is a smart Irish man. And he says, “I’m going to put you on my payroll.” And so I said, “Okay, sure. If you can do it by five o’clock, I’ll take the job,” sort of laughing, like, “No way.” Sean O’Keefe picks up the phone. Lawyers, personnel people start walking into the room. They all said they were just waiting. The whole thing was a setup! At five o’clock, I had a badge as the CIO of NASA!
Yourdon: That’s amazing.
Strassmann: Now, you know, and this is my clear adversary, who I’ve been fighting tooth and nail. He lost the battle, because we took IT out of the controller’s department in the Department of Defense. You know, that’s a 30-billion-dollar piece of money.
Yourdon: Yes.
Strassmann: So then I’m stuck for a year. I say, “Okay, Sean, I’m going to give you a year.” I’m just giving you vignettes of what … it’s a game.
Yourdon: Is it a game you would recommend to other successful CIOs when they feel that they’ve reached the end of their journey?
Strassmann: Of course. Absolutely.
Yourdon: So never become a consultant, but look for interesting assignments.
Strassmann: Look for assignments that you can leverage. And there are several assignments that can be leveraged. And never work for Peat, Marwick & Mitchell,3 or Deloitte. They burn you up.
Yourdon: Have you seen CIOs trying to move up to the next level, [to] CEOs?
Strassmann: Oh yes, oh yeah.
Yourdon: Are there many success stories?
Strassmann: Oh yes, there are many, quite a few. And the reason is that these guys are inside the power structure, and they create alliances and they are acceptable. And they have become politically acceptable. They also know a great deal about the company, because as a CIO you really learn the company, particularly if you do it the right way—namely, you bore down and you see who connects, who talks to who.
Yourdon: And, obviously, more and more companies are beginning to realize that they live or die with their information.
Strassmann: Oh yeah. And so the CIO who is doing the right political moves and is doing the right homework, understanding the infrastructure and the way the blood vessels function, the nervous system … this guy, after three or four years, knows a great deal of how the place operates because all the other guys are functional. They are marketing or they are lawyers or somebody else.
Yourdon: It was probably 20 years ago, I think, that John Reed became the CEO of Citibank and he came out of its IT department. I don’t know if he had been CIO.
Strassmann: Yes, he was. I know John Reed. Oh, John Reed came to see me.
__________
3 Peat Marwick International merged with Klynveld Main Goerdeler to become KPMG in 1987.
Yourdon: He came out of the Sloan School.
Strassmann: Oh yeah. John Reed hired me.
Yourdon: Ahh, okay. I was just saying that he was the only example that I remember off-hand of somebody rising up through IT. There are probably lots more.
Strassmann: There are others, yeah, there are others.
Yourdon: There are probably an equal number of situations where the CIO feels just completely burned out because he’s been fighting all those battles and maybe wasn’t prepared for.
Strassmann: Oh, you don’t fight battles. The answer is, you always pick your battles, and pick very few. And you always pick the battles that you know you cannot lose.
Yourdon: Oh, okay.
Strassmann: Because if you lose, you have to leave. The worst thing is for a CIO to lose a battle and stay around. He’s a living dead guy.
Yourdon: Aha, okay. Well, for a long time, as you know, the life expectancy of a CIO has been about two years or even a little bit less.
Strassmann: Oh yes.
Yourdon: Now is that because they’ve lost battles?
Strassmann: No, no, no. There are lots of reasons, but let me tell you basically what the flaw is. The people who are hiring the CIO don’t know what they want the guy to do. They haven’t cleaned up the power structure. So they hire somebody with a totally mistaken idea of what this guy’s supposed to do.
Yourdon: From the outside, usually?
Strassmann: Oh, from the outside. This is the two-year CIO. So they come in, the guy sits down, gets an office, and then he has to decide what to do. And then he has to scramble for everything. He does not know what his budget is; he doesn’t know who’s his rabbi. You know, you always need a rabbi.
Yourdon: Yes, yes.
Strassmann: So, after a year, the top executives look and say, “Well, we really didn’t make the right choice. He’s the wrong guy.” He’s not the wrong guy. The guys who picked him were the wrong guys because they’ve never done the work. They’ve never done . . . one of the things that your book can do is to bring into a CFO/CEO position a thinking of what it takes to make a CIO a success.
Yourdon: Okay.
Strassmann: You see, the CIO cannot be a success unless somebody wants him to be a success.
Yourdon: But that requires a fairly clear understanding of what the role needs to be and what that role needs to be.
Strassmann: Yeah. And then you hire the right person for that role.
Yourdon: Do you see many situations where the CIO rises up through the ranks, as opposed to being chosen and brought in from the outside?
Strassmann: Very rarely. It does happen. There are some exceptions. I had people working for me at Xerox because I had a real mean training school in the data center, and some of them worked their way up from shift supervisor to CIO.
Yourdon: Wow.
Strassmann: But they were exceptional. I mean, there are such cases.
Yourdon: Actually, there’s a variation on that that I have now begun to see, and interesting that you should mention Citibank. Citibank is one of many large companies that has multiple CIOs. I have been trying to contact one particular CIO. Well, it turns out that he’s not the CIO. He’s one of ten.
Strassmann: Always, there are lots of CIOs around.
Yourdon: And especially with the big multinational companies these days. So maybe they’re at the sub-CIO level and there’s more of this rising up through the ranks. But if not, the question becomes, where do CIOs come from? They’re not born into that role, so they have to make a jump. Maybe you jump from position X in one company to CIO in another company? Is that the more common way of doing things?
Strassmann: You know, it’s very hard to generalize. I’ve jumped. You know, I’ve been with General Foods, Kraft, and Xerox. The Department of Defense and NASA. Five. Okay? So I’ve been in this business since, depending on how you count it, I’ve been in this business for 60 years. So I had 5 jobs in 60 years. On the average, you know, that’s …
Yourdon: That’s a lot longer than average.
Strassmann: The question is, are these jobs additive? Or just musical chairs? One advice I would like to give to everybody, very important: When you are a CIO, you are also a programmer.
Yourdon: Aha.
Strassmann: You never let go of the craft. You never let go of it . . . I’ve published now 300 papers. I usually write papers because there’s something I don’t know about. And the best way to learn something is write a paper about it, as you undoubtedly know. The only person who ever learns anything is the professor, by the way.
Yourdon: [laughter]
Strassmann: The professor learns much more than the student. Guess why I’m teaching a course on cyber-operations?
Yourdon: So you can learn about it.
Strassmann: I am learning this stuff! I’ve spent the last two and a half months learning this stuff. So, so the craft . . . you must understand there’s a fundamental craft that you have to nurture. You cannot let go of that.
Yourdon: Now, once you got into the business world, were there business mentors or role models that kind of gave you a sense of how to behave or, what was important versus not important?
Strassmann: I would say one of the great experiences was when I arrived at Xerox. I arrived on a Monday, reported to work at 8:30, and at 10:30 a gray-haired gentleman walked into my room and says, “I’m Joe Wilson [the CEO of Xerox].”
Yourdon: “Just call me Joe.” I read that part of your book.
Strassmann: Yes, “just call me Joe.” And it is that humility. See, there is a certain sense of decency and ethics which is very important. You know, it’s power struggles, but it has been always decent, it always has to be ethical, it always has to be done with recognition of people. Now, I violated that maybe once or twice, and I’m really sorry about it—when I sort of acted harshly, retaliated against certain individuals. But that’s not the way to operate.
Yourdon: Do you think that kind of behavior has largely disappeared in American business today?
Strassmann: Oh, American business, so far as I know, is extremely confrontational now. Too confrontational.
Yourdon: And certainly the kind of behavior you read about and hear about in Silicon Valley is kind of, you know, the epitome of the IT world. It seems like a lot of backstabbing and confrontational behavior.
Strassmann: Well, you know, Citicorp was just infested with that. I mean, it was a sick, sick situation.
John Reed hired three outsiders to come in and have a look at his place. We went around, visited lots of Citi locations. And we came back with a remark that the issue was not technology, because the technology was good, but it was the confrontation and the continual conflict which was tearing the place apart.
Yourdon: How long ago was this?
Strassmann: [1995], ’96.
Yourdon: Interesting, interesting. One of the other things, of course, that is often true in the Silicon Valley kind of IT world is companies being run by very young and often somewhat inexperienced and somewhat immature people, who may not have learned proper behavior, by being part of an organization where they could watch others.
Strassmann: Oh, you learn your behavior from your parents, my friend. I mean, don’t wait for corporations to straighten out your behavior.
Yourdon: Well, parents first and foremost, of course, but also your entire upbringing and schools and so on. You’re right; it’s not just in the business world. Although I think that people will often see a lot of bad examples in the business world. If you were brought up in a certain way and suddenly find yourself in a backstabbing culture, that’s got to be a culture shock.
Strassmann: Yeah. My years with Kraft, by the way, were very nice. These were hard-working people. They were all milkmen starting local routes. Hard-working, hard-drinking, very nice people. It wasn’t much money, but it was pleasant. It was very genteel.
Yourdon: Hmm, interesting. How is it by contrast in the military? I mean, the people who end up doing the computer stuff . . . some of them may have come into it right from university, but don’t some of them come from a more traditional, West Point kind of culture?
Strassmann: Yeah, but let me tell you something. I find a larger collection of gentlemen of substance in the military than in corporate America. Particularly in the Navy.
Yourdon: Really?
Strassmann: Oh, yes. I mean, the rough kind of in-your-face approach is very much scoffed at. These guys will never make a promotion, because you‘ve got to go through lots of steps before you get up to admiral. These are all gentlemen, every one of them. They may have different opinions, they may be backwards, they are obsolete, but they are gentlemen.
Yourdon: Now does this start at Annapolis and carry them all the way through?
Strassmann: They don’t even get into Annapolis.
Yourdon: Unless they have that foundation. Oh, okay.
Strassmann: By the way, my grandson got a congressional nomination for Annapolis.
Yourdon: Congratulations!
Strassmann: But they decided he’s a geek and he would only make ship captain, so they sent him to Carnegie-Mellon, and they are paying for him.
Yourdon: One last technical question and then I will leave you alone. One of the popular IT things these days is the agile development approach, which seems to also have some impact on project management. It’s not just a technical way of developing systems. But have you seen any advantages or disadvantages of agile systems development from a management perspective?
Strassmann: [laughter] Let me say that the word “agile”—lots of things get peddled under that thing.
Yourdon: True.
Strassmann: The more I look at things, particularly in the last 20 years, the more I’m concerned not about the development, but the architecture, the design, you know, how are the relationships set today? Who’s really deciding what kind of databases, what kind of metadata to put into place?
Yourdon: Okay.
Strassmann: You can do lots of agile development extremely inefficiently if the thing is not put together with any kind of a sense. Lots of little ants are moving all kinds of twigs all over the place.
Yourdon: [laughter] That’s a nice metaphor. Okay.
Strassmann: I think that’s great. But I have not spent time on that. I really have not spent time on that. I’m more concerned about the things which are not being done to make it even possible for people to do development.
Yourdon: Well, you’re looking at things, as you said earlier, in terms of a 40-year or 50-year lifecycle. So it’s one thing to move twigs around, but the twigs will get blown away in a windstorm.
Strassmann: See, one of the most important issues is, what will be the life of the code and how maintainable the code will be. Has this code been agilely developed for a one-year lifetime and then be junked and thrown away?
Yourdon: Some things are, but, of course, you and I are old enough to know that some things we thought would be thrown away in a year have lived 25 or 30 years, so that’s the big surprise.
Strassmann: And the difference between what gets thrown away and what doesn’t get thrown away: Has it been embedded into a framework? See, I’m very much concerned about data.
Strassmann: And have these people really answered the question of how well this particular application gets the data securely? Now, the code itself—particularly if I have an infrastructure in place—infrastructure is a service, in other words, a cloud service which provides an infrastructure. The infrastructure, I don’t have to do. If the front end you throw away every three months, I don’t care.
Yourdon: Okay.
Strassmann: The problem is to stack. And the stack is data, communication infrastructure, the coding infrastructure, the interoperability, metadata, connectivity, the protocol. I mean, there’s layers in a stack.
Yourdon: Right.
Strassmann: I don’t mind if you throw away the top 5 percent every time you want. It doesn’t matter. It’s a button, just throw it away. Ninety-nine cents. That’s all it was worth. The problem if you go with an application that will build an entire stack, or you go in with a big project that’ll take a year and it will cost $10 to $50 million. That’s not the way to do things.
Yourdon: So for things of that sort, do you feel we shouldn’t be using the popular forms of agile at all?
Strassmann: Oh, you can use agile. And decide on what layer you’re going to put it in.
Strassmann: You don’t want to do agile on the infrastructure, for instance. You don’t want to do agile on data. You just don’t touch it.
Yourdon: Okay.
Strassmann: In fact, you have a totally different environment set up for control of data. And infrastructure in particular. You know, in the infrastructure you may have things starting on the Internet. And then because of latency, you may decide to put the data center capability, what’s called “on the edge,” to get you down to milliseconds.
Yourdon: Right.
Strassmann: You should be able to slide out of the stack the infrastructure, slide it on a different . . . without the upper layer or lower layer being affected. That’s where the costs come in.
Yourdon: Right. Interesting. Well, that’s, as you know, an area where there are huge battles being fought today, as was the case 20 years ago when my friends and I were introducing the structured methodologies.
Okay, well, I think I have covered just about all of the questions I had on my list, which I very much appreciate your taking the time to talk to me about.