Part 4/5—Interview: co-author Roy Levien; The KeyStone Advantage: “This is an important book that should be read by anyone interested in the dynamics of modern business.” — Bill Gates

Last Wednesday, I began my discussion with Roy Levien. On Thursday and Friday, we continued the interview. Here is part 4.

I would like to hear your opinions in this forum or send me an e-mail at Enjoy! .... From Stephen Ibaraki, I.S.P.

Q: IBM has adopted a Linux strategy. Can you share your views on this?

A: Far more is going on there than meets the eye. This is part of what I alluded to earlier when you asked about innovation and ideas. On the surface, IBM appears to be endorsing Linux as a simple competitive tool: to hit Microsoft where it hurts and to leverage something that is basically free to them to enhance the value of their hardware and, increasingly, services. But what's actually going on is much more subtle and pernicious. IBM is using Linux to establish itself a keystone in the Open Source software ecosystem. By endorsing Linux and Open Source, and by releasing some of their own patents into the ecosystem, as they have done recently, they are doing more than "reduce costs"; they are also undermining the value of intellectual property, on its own, creating an ecosystem where the intellectual property, for the most part, only has value in the context of their hardware and services products, which they can now produce at lower cost.

In this process they are aided by the Open Source ideology. There must be some axiom somewhere that says that any ideology - democracy, socialism, communism, whatever - is ultimately co-opted by established powers. That's the real story of IBM and Linux. The Open Source community, and the anti-intellectual property movement with which it is linked, is so blinded by its ideology that it has become an accomplice to IBM's plans.

This process shifts power away from people (who at least have the potential to file a patent and establish ownership of intellectual property) towards large firms. Firms, after all - if they are serving any purpose whatsoever - are more efficient at doing everything that an individual could do alone, so all that individuals have is their ability to create. When you render that valueless, the result is a geologic shift in power from individuals to firms. That's ironic, even tragic, considering what the Open Source and the anti-patent movements stand for, but it's what's happening.

Q: How do companies go about understanding their dependencies, relationships, identifying critical assets internally and externally, and building strength throughout the business ecosystem?

A: In almost all cases where keystones have managed this, it comes from a deep understanding of how the technology of their network functions. Technology in the broadest sense: how all the pieces fit together, how things interconnect, which parts are essential, which parts are most valuable, where standardization helps, where it is unnecessary, and how these things change over time. IBM, for example, shows that it understood when they "gave away" those patents. Doing so encourages people to use them, to incorporate them into their products, and to accept them as part of the DNA of their own technologies. It also makes it easier for products from other firms to interoperate with IBM's - in short, it draws them into the IBM ecosystem. But you can be sure that IBM has other patents that it will assert against these very companies to keep them at bay. If they know what they're doing, they've taken a hard look at what they have and have said: if we give these away on the one hand, we will greatly enhance the value of these assets on the other. Doing that right requires a pretty good intuition about how everything interacts.

Q: Why should antitrust policies be rethought?

A: Conventional anti-trust thinking still has a role: to protect consumers from unfair pricing or limited choices. Typically that means coming down on firms that are exceptionally large relative to their industries, but in a world where the term industry is hard to define clearly, this becomes problematic, so you have to have a pretty deep understanding of how networks of products and firms fit together before you go mucking around with antitrust remedies. This is something we talk a lot about with respect to Microsoft. Microsoft has a lock on the desktop operating system market; but is that bad? Are they abusing it to harm consumers? Simple formulae no longer apply here because there are significant benefits to connected products hidden, as it were, in the price paid for Windows and in the "cost" of not having an alternative operating system to choose from. Another thing to consider is that if you accept our biological view of business networks, whatever you do to take down the leading player in a such a system will only be temporary: their occurrence is inevitable, so all you get is a costly disruption and an eventual return to the same structure. Different players maybe, but the same pattern. That means policies need to get away from trying to change this structure, and move towards a regime of trying to work with it and the dynamics it creates. Put simply, this boils down to enhancing the things that are good about it (like standardization, predictability, stability, etc.) and confronting the things that are bad about it.

This last part is actually the hardest, mostly because we labor under some false axioms about what's bad: lack of "choice" and the related barrier to entry for new firms in certain domains. We need to get away from thinking that these things are inherently negative. How many digital media formats do I need? Is it okay to have a mediocre web browser that does 90% of what I want, if it means that developers can target a single well-understood platform? There are complex trade-offs here that need to be examined closely, case by case, and not subject to some reflexive application of assumptions about "competition".

Some things about keystones, though, are inherently dangerous, and can only be addressed with some kind of intervention. Wal-Mart is a great example. They create all kinds of efficiencies that benefit many members of their ecosystem; that's their area of expertise. But they are terrible at other things, like architecture, landscaping, and aesthetics. Because of their powerful position, almost every one of us has to deal with their bad taste and incompetence in these areas. Microsoft has a parallel problem; they make great operating systems, great API's, powerful productivity applications and superb tools, but they have some of the worst user interface design in the software universe. Ideally, in both these cases there should be a way to leave the valuable platform aspects of the firms intact, but to intervene to prevent the damage they do in areas where they lack competence.

Comments (1)

  1. Anonymous says:

    Last Wednesday, I began my discussion with Roy Levien. On Thursday, Friday, and this Monday, we continued…

Skip to main content