Open Source Maturity
By webmink on Dec 20, 2008
There's an interesting comment from 451 Group's Matthew Aslett on the five stages of community open source engagement that's worth reading. I've been using a model something like this for quite a while too. It's interesting to visit companies and see their reach and spread on this model, and to spot the distinctive signs of change that result.
It is resonant with a much older piece of research on the psychology of human belief systems (which works beyond just the religious since capitalism, for example, is also a belief system) by James Fowler in his book Stages of Faith from 1981. The two aren't identical but they suggest there's an underlying maturity model at work.
Aslett asks where Microsoft is placed in the model. Cal Evans responds
Having just returned from 3 days in Redmond discussing PHP on Windows with Microsoft, I agree that Microsoft cannot be pinned to a single stage on this chart. However, realistically, even their forward thinking divisions are no farther along than contribute. I would spread them between denial and contribute with denial being the majority and contribute being the long tail.I'd agree that every organisation shows this characteristic. Even at Sun, where the software development groups are deeply engaged at stage 5, I still find myself sometimes with staff who are at earlier stages. This is as one would expect from Fowler's work, since every individual will reflect a different stage of their evolution.
They will likely operate at different levels of maturity in different areas of their life as well. This is not a bad thing, and one of the risks of using a "maturity model" is the temptation to treat the later stages as "better". When we do this, instead of valuing and supporting people in the lower-numbered stages, we treat them as "in need of growth".
Maturity Spread and Reach
It's easy to forget that corporations (and indeed large non-profits) are not people, but are rather a vehicle for the collective expression of the vision of many individuals. Things happen not because a faceless corporation somehow chooses to act, but because of the persuasive decisions of actual people, acting within their belief systems. Every good - and bad - decision ultimately goes back to an individual somewhere.
In any collective group, there will be a maturity reach and spread. The "reach" is the furthest stage the most pioneering individual of any influence has been able to take a team. The "spread" will be the range of stages the collective grouping is willing to tolerate existing within itself (there may - will - be isolated earlier and later stages too). Thus in the example Aslett seeks of Microsoft, the reach is to stage 3 on the data in the comments (I would actually suggest recent staff changes may even push that to stage 4) and the spread is back to stage 1 with the weighted centre around stage 1. In Sun, the reach is stage 5 and the spread goes back to stage 1 and is weighted around the stage 3-4 boundary. This is also a fractal effect, exhibited within the groups that comprise each organisation.
That's why I expect Sun, Microsoft, IBM and other corporations to be inconsistent in their approach to open source. I also expect to find that in large groups like Apache and Eclipse there will be a good deal of inconsistency of view. I get very suspicious of attempts to make open source engagement appear uniform across such a wide spread of activities. Both the model Aslett reports on and the work Fowler did in 1981 suggest to me that real engagement will be diverse. And that's actually a good thing.