|
 |
|
|
Open Source
I attended an MIT Enterprise Forum discussion last night on the topic of business issues surrounding Open Source software. The discussion was great, and the moderator, Dan Woods (CTO, O'Reilly Publishing), was exceptionally good. So, what about Open Source? I think a lot of my hesitation about Open Source software comes from the following implications:
A lot of the energy behind Open Source comes from folks who hate Microsoft. Whatever else you might think about Open Source, the anti-Microsoft jihad aspect is real. Since I don't hate Microsoft, I can't get all excited about undermining their business model. Microsoft is a commercial entity that has its own interests at heart, and should be assumed to be as such. Reading through some of the items on opensource.org, and following some of the writings of Lawrence Lessig, I realized, really for the first time, that the Open Source folks see their movement as a fight for democracy itself. Code is Law assumes that Code enforces Law by preventing things outside the boundaries of itself; having democratic control of code that, by its nature, enables liberty (for example, the Net) is critical if democracy itself is to be preserved. Open Source is all about the grassroots: no top-down control is allowed. If code itself is the last bastion of democracy, and if Microsoft is attempting to create code that will one day impose the fascistic One Microsoft Way on all human beings, then I suppose it makes sense that I'd spend my evenings sequestered in my office voluntarily improving my favorite Open Source project. Back to my hesitations. First one: Volunteer programmers will improve the code. Ok, fair enough, if someone wants to volunteer to do anything in this world, by all means, go for it. However, if I decide to make use of something that is a software artifact produced by some entity, I do not want the implicit quid pro quo that's part of Open Source culture. I realize that there are no explicit strings attached, but as an able developer, if I were to find a bug or think of a feature, I'd sometimes rather go shopping than become part of a "community." My Open Source colleagues will argue that there is, in fact, no obligation implied in Open Source. Read about Robert Cialdini's Cues of influence as persuasion techniques; these cues were developed by observing sales groups, cops working for the bunco squad, and fund-raisers, to identify the real-world persuasion techniques that work in many different situations. Read this and tell me there is no implicit quid pro quo in Open Source. Second one: Usage of the software must be free. To me, this is a huge source of discomfort. In order for the Open Source model to work, my software company must not have a licensed-software fee structure. Partly, this is the nature of the Open Source process itself; without free usage, why in the world would I volunteer to help YOU make a profit? This bugs me because it forces me as a software company to choose a different profit model, if I choose to go Open Source. I cannot profit on the code itself without having some non-open upgrade. Last: Nobody can collect a royalty on the software. This is pretty similar to the second hesitation, but with a subtle difference. My software company (by the way, at this point I do not actually have a software company) cannot focus on selling code, and I cannot partner with VARs (Value Added Resellers) who support and implement my code in this model. Nobody can profit from the software itself; this means that I have to be a VAR. It forces me to structure my business around the services side of the fence, and if it were actually my company's Open Source code, then other VARs out there might be tempted to fork my code or create some other software-based value to compete against me with my own product. "It's better for the consumers" may be true, but consumers are not the only stakeholders here. What if I don't want to manage a Process or build a Community, even if I acknowledge some real advantages? The problem is not in the source code licensing model or development process. The problem is that code is built on assumptions about reality that don't hold up to scrutiny. So to me, the argument of Open Source or not misses a more fundamental argument about symbolic modeling of reality or not. If you are attempting to create quality models of reality, it does make sense to have a strong peer review process, which academics and credible journals do. So, the Open Source model for software development also makes a great deal of sense in the context of building better models. Software, as it's practiced now, is all about codifying an optimal model of reality (business, processes, information, and things, optimal in some context). So, sharing the work of creation and review with many other people seems like a good way to go with this problem, IF you get a quorum of other people. If you don't have a community that really represents every stakeholder in the modeling process, then a profit-motivated company can easily out-model your team, and will have the sufficient incentive to do so.
|
