Post Syndicated from Bradley M. Kuhn original http://ebb.org/bkuhn/blog/2009/10/26/symbian.html

I
agree pretty completely with Harald Welte’s comments regarding
Symbian. I encourage everyone to take a look at his comments.

We are in a very precarious time with regard to the freedom of mobile
devices. We currently have no truly Free Software operating system that
does the job, and there are multiple companies trying to get our
attention with code releases that have some Free Software in them. None
of these companies have pro-software-freedom motives about these issues
(obviously, they are for-profit companies, who focus solely on their own
profits). So, we have to carefully analyze what these proprietary
software companies are up to, why they are releasing some code, and
determine if we’ll be successful forking these platforms to build a
fully software freedom phone platform.

We thus must take care not to burn our developer time on likely
hopeless codebases. I think Harald’s analysis convinces me that Symbian
is such a hopeless codebase. They haven’t released software we can
build for any known phone for sale, and we don’t have a compiler that
can build the stuff. It’s also under a license that isn’t a bad one by
any means, but it is however not a widely used license for operating
system software. Symbian’s release, thus, is purely of academic
interest to historians who might want to study what phone software
looked like at the turn of the millennium before the advent of
Linux-based phones.

Currently, given the demise of mass-market OpenMoko production, our
best hope, in my opinion, is the HTC Dream running a modified version of
Android/Linux. We don’t have 100% Free Software even for that
yet, but
we are
actively working on it, and the list of necessary-to-work
proprietary components is down to two libraries. Plus, the Maemo
software (and the new device it runs on, not even released yet) is the
only other option, and it has quite
an extensive
list of proprietary components. As far as we can tell currently,
the device may even be unusable without a large amount of proprietary
software.

Even so, Android/Linux isn’t a Dream (notwithstanding the name of the
most widely used hardware platform). It’s developed generally by a
closed community, who throw software over the wall when they see fit,
and we’ll have to maintain forks to really make a fully Free Software
version. But this is probably going to be true of any Free Software
phone platform that a company releases anyway.

I’ll keep watching and expect my assessment will change if facts
change. However, unless I see that giant laundry list of proprietary
components in Maemo decreasing quickly, I think I’ll stick with the least
of all these evils, Android/Linux on the HTC Dream. It’s by far the
closest to having a fully free software platform. Since the only way to
get us to freedom is to replace proprietary components one-by-one,
picking the closest is just the best path to freedom. At the very
least, we should eliminate platforms for which the code can’t even be
compiled!

[ PC was kind enough to make
a Belorussian
translation of this blog post. I can’t speak to its accuracy, of
course, since I don’t know the language. 🙂 ]