As I prepare for a long tour of Europe for various EU project evaluation, I would like to provide some snippets and comments on some debates and discussions for which maybe a long post would be not appropriate, starting of course with the ping-pong like debate between Matt Asay and Tarus Balog on what constitutes an OSS company. I already have wrote too much about it, but I would like to point out two errors (at least in my view) in Matt Asay post, in particular in the phrase:
Red Hat is an example of “free done right,” following analysis from TechDirt. We’ve moved beyond the business models that insist that every line of software be open source: they couldn’t scale and tended to treat openness as an end in and of itself, rather than as a means to an end.
The first point is that RedHat is a perfect example of an OSS company under almost any definition of the terms. There is little or no code that is unreleased, and actually most of those cases are for code that was recently acquired, and thus still not vetted for release. I know that Matt disagrees with this (because the service contract is more restrictive), but the point is that you acquire all source code and after removal of the trademarked logos you are allowed to do whatever you want with it. If you don’t like RedHat services you can go with Oracle, or the many companies that provide additional support contracts. If Matt has a substantial example of withheld code that is sold for a proprietary license, I would happily apologize; until that moment, I stand my case.
The second point is related to the fact that OSS companies are unable to scale. This is something that I already discussed in our study on business models, and it is mainly an organizational problem: the reality is the correct phrase should be “small companies’ service based models do not scale”, as there are several excellent examples of service based companies that are very large (Accenture, IBM global services, HP services, CapGemini, Fujitsu, BT, and many others) and that are human capital intensive. The critical point is that to scale it is necessary to change internal structure, and become organized in a more efficient and “industrial” way; there is in this no difference between OSS and non-OSS companies.
On a totally different field, it was interesting to notice the great amount of interest for Android-based netbooks. Many claimed this combination to be the real alternative to XP netbooks, or in the near future to Windows 7 netbooks. The reality is that Android as a system does not have magical properties; the underlying kernel is still Linux, and having a set of customized interface reduces greatly the memory consumption, but does not provide any significant improvement when compared with lean, netbook-optimized linux distribution. On the contrary, the user interface (designed for one-app-per-screen, and use with imprecise controls like touch screens and trackballs) is not exactly ideal for something like a netbook, that does have a keyboard and a large-enough screen. In this sense, I would say that Moblin may constitute a much better environment for this kind of applications.
MID (Mobile Internet Devices) like the Nokia N810, Archos devices and many more are clearly a better match for Android; while other environments (like Maemo) do exist and are stable, the App Store and the enlarging software ecosystem may become a differentiating element. As I already wrote in the past, I believe that VoIP applications (and in a more general sense, social and interaction applications) will become the real differentiating element in the future. I believe that as 3.5G wireless data contracts start to become common, the phone will become more and more a “modem” for connecting with a separate and non-carrier-controlled MID, where VoIP and IM applications will join the web browser as the most frequently used applications.