Sponsors of FOSDEM 2009
Summary: A few notes and analysis of Mono news, with increased emphasis on why it is perfectly acceptable to say "no" to Mono
JASON RECENTLY published this good piece titled "Rejecting Mono". In it, Jason attempts to explain what opposition to Mono is all about because Mono sceptics like ourselves have had our views and our opposition mischaracterised, even distorted. It's intended to daemonise and thus dismiss valid critics. Jason concludes by stating the obvious and asking this question:
So, are the Linux/Mac gamers here that don’t want to use mono “zealots” and “freetards”? Of course not. In the context of cross-platform capability, developers might like the idea of Mono because of what it promises – but users don’t like Mono because of what it delivers.
You can only get the full “benefits” of .NET if you are on the Microsoft Windows platform. This is by design. We have the memos – court evidence – where Microsoft lays out again and again the strategy of releasing a small sub-set of .NET, but reserving the “good bits” for Windows. We see this all the time in the real world (ala Moonlight being useless on any site that actually matters). This documented strategy absolutely destroys the notion of .NET or Mono being a “cross-platform” framework. It can not and will not be, because Microsoft will not allow it.
And that is granting the rather large assumptions that .NET is well-designed enough to do so and developers want a Microsoft-controlled framework, even if Redmond would allow it!
Despite
all that we know about Mono after years of research, some developers insist that Mono is good for GNU/Linux and for software freedom. Well, the main thing Mono is truly good for is promotion and spreading of .NET. Is that a commendable goal? Really?
Well, FOSDEM 2011 is already in the planning and
Mono developers/advocates (usually the same group for natural reasons) will have a room allocated to them. Notice how they are very much isolated not based on modularity and separation of a larger system (e.g. kernel versus development environment and bootloaders) but based on the language they use. There is something wrong here, no? Anyway, this room allocation is nothing too shocking anymore; it's an annual thing and not different from previous years:
For the second year in a row, there will be a Mono Developer Room at FOSDEM. We had a very nice edition last year and we are looking to improve on that.
Mono developers have thankfully lost ground in the Ubuntu distribution of GNU/Linux, especially because f-spot was replaced by shotwell. Some people
want to reverse this choice, but they do not take the ramifications of Mono into account. The
CTO of Canonical apparently does. Meanwhile, Canonical controversially develops Unity, which is based on Vala (we are not going to start a patent discussion about it because it's already in today's IRC logs). The developer of Unity is the man who brought us the Mono-based Gnome-Do, which we view as more of a curse than a blessing although some people in today's news
seem to like it based on purely technical grounds.
OMG!Ubuntu is again
implicitly promoting Novell's Banshee, which is a Mono application that we know for sure is a patent problem due to
limitations in the MCP.
Our posts about Mono are not attacks on applications, vendors, and developers. They are intended just to remind people what depends on Mono and
what substitutes are available (see this list if you haven't). Refusing to use Mono is no "zealotry"; it is a rational choice not to be ashamed of. If you dislike Mono,
you appear to be in the majority.
⬆
Comments
Agent_Smith
2010-10-30 00:22:38
I'm amazed, on how this humble comment is, TRUE, and SANE, and, too bad for the rest of us, may Saint Ignutius protect us from american US patent system, or else, we're all @#$@#$#@$@#$. ;-)
Dr. Roy Schestowitz
2010-10-30 00:26:53
Agent_Smith
2010-10-30 00:31:46
Dr. Roy Schestowitz
2010-10-30 00:35:54
eean
2010-10-29 22:16:35
For instance you complain about Banshee, due to the aforementioned patents. What is the chance that Rhythmbox does not violate another US patent? For instance I recall that Microsoft sued Apple over patents it had relating to the iTunes UI. And I'm going to just guess that Apple has its own patents over that UI.
What if tomorrow someone discovered that such a patent existed. Would you then stop using Rhythmbox?
What if some patent was found "violated" by Linux (of course that's probably less of a hypothetical). Would you then start a "Anti-Linux" talk at FOSDEM? What about GCC?
Its really absurd how some free software advocates are sounding pretty much like Microsoft patent attorneys!
The fact of the matter is that you can't write a hello world application without violating some patent, or at least you can't be sure that you aren't.
Dr. Roy Schestowitz
2010-10-29 22:20:18
dyfet
2010-10-28 20:26:04
Vala also has the problem of being an "unfinished" thing without a sufficiently rigorous design. Hence new releases still tend to break existing code from what I have seen of it. I also never really liked the idea of transcoding languages to C and compiling that way. It makes true source debugging impossible.
From a purely technical perspective the one really interesting thing about Vala is that it directly uses the gobject model. I am sure this makes it much easier to write d-bus applications, since d-bus is explicitly tied to gobject and it's object model, and this is what seems to makes it so difficult to map d-bus objects into C++. Python and the way it uses d-bus seemed more palatable to me than the efforts of doing so in C++.
But what of the threat to freedom represented in Vala? This is not something I had considered before because I was not using Vala. I suspect if the issue is purely on language features, most prior art goes to Java, though that too of course is in rather hostile hands currently. I think if Vala truly became more viable and practical I would have to spend time more deeply thinking of this question.