Who is this "we" you refer to? [LWN.net]

Who is this "we" you refer to?

Posted Mar 6, 2008 8:00 UTC (Thu) by robla (subscriber, #424)
Parent article: Ryzom returns?

One has to wonder why we seem to be unable to put together a competitive game without relying on a huge infusion of source from the proprietary world.

Why is this framed as an "us" versus "them" sort of thing? Is the Firefox development community part of your "we" or another huge infusion of source from the proprietary world? What about OpenOffice? Is an open source project not valid if it doesn't fit the mythology of a single developer scratching an itch?

This seems to be a variant of "not invented here" to me, and misses the real problem entirely. Companies are now willing to make huge infusions of source code, but hey, how hard can it be to rewrite everything from scratch?

There may be some useful code to be had from the Second Life client...

*cough* um, I sure hope so. The Linux version went into beta (feature complete) after a very long alpha testing phase. There are multiple projects (realXtend, OpenSim) that have done very interesting work by working with this codebase.

Full disclosure: it's my job to stick up for the Second Life code. So, forgive me if I take a little umbrage at the implication that the Second Life codebase is a dead carcass to be picked over for "some useful code".

(Log in to post comments)

Who is this "we" you refer to?

Posted Mar 6, 2008 14:56 UTC (Thu) by corbet (editor, #1) [Link]

You've missed my point entirely. Guess I wasn't writing clearly that day, sorry.

There is nothing inherently wrong with corporate code dumps, though experience shows that they can be a lot harder to turn into free software projects than one would think - a year spent getting the Second Life client into condition for a beta release might be a case in point. What I don't like is the idea that we have to have our projects jump-started in that way. That really plays into the "free software can't actually invent anything" notion that I so strongly disagree with.

So if Ryzom comes along, that's great. But I don't accept that it's beyond our capabilities to build something better than Ryzom on our own.

Meanwhile, I'm most interested in the projects you pointed to which are using the Second Life code base; will investigate further.

"Corporate code dumps"

Posted Mar 8, 2008 17:52 UTC (Sat) by robla (subscriber, #424) [Link]

I think you missed my point about "corporate code dumps". The derogatory way you put it makes
it clear you think corporate contributions are somehow inferior. This sort of attitude is why
free desktop software may remain a niche for a longer than it needs to. The two most
prominent desktop open source projects that I can think of (Firefox and OpenOffice) started
off lives as proprietary.

Regarding the Second Life viewer beta, I'll remind you that "alpha" and "beta" are just
labels, and while we're not ready to take the "beta" label off yet, the software runs as well
or better than just about any other 3D software on Linux. It's just not up to the level of
our Mac and Windows products. Making production quality software available on *any* platform
is hard, and making it on Linux (with OSS/ESD/ALSA/PulseAudio/whatever system comes next) and
video driver issues (which 3D apps are very sensitive to), it's even harder on Linux.

With respect to Ryzom, we might actually be in agreement, but for different reasons, I think.
Ryzom enthusiasts should absolutely be pooling their money to help free the software that
they're so enthusiastic about. However, anyone who isn't already a Ryzom user (like the FSF,
for example) should be very skeptical of investing money in this cause. The free software
community already has plenty of momentum and alternatives in this area. The Ryzom codebase is
now going through its second financial crisis. Banking on Ryzom plays into a far more toxic
stereotype than the one you've identified, which is "open source is the refuge for losers who
can't cut it in the proprietary world". Efforts in the corporate open source world should be
on efforts that have proven revenue models to fuel development. It's one thing to free the
source code, giving developers access to the source code, but it's so much better to then be
able to pay the most productive developers (and QA, and project managers, and documentation
writers, and usability specialists) to improve it for a living.

"Corporate code dumps"

Posted Mar 8, 2008 18:20 UTC (Sat) by corbet (editor, #1) [Link]

No, you are still misunderstanding me.

Remember, on those occasions where I get a chance to do some code work, I do it in the kernel arena - a project dominated by "corporate contributions." I have no problem with them. I will say that large dumps of code from companies can be a mixed blessing; bad practices which would have long been cleaned up in an open project can persist for a long time in a closed code base, leaving big messes to clean up. One may or may not agree with the Mozilla project's decision to, for all practical purposes, start over, but they did have their reasons for going that way.

My point remains entirely about the notion that we need a big injection of code to get a project going in the first place. I do not think that is true. That is different from saying that we have not benefited from such injections in the past and won't in the future, or that they are not welcome.

Who is this "we" you refer to?

Posted Mar 8, 2008 19:44 UTC (Sat) by robla (subscriber, #424) [Link]

Well, maybe you're right about me not understanding you, so let me ask a question.  Who is
"we"?

Who is this "we" you refer to?

Posted Mar 9, 2008 5:55 UTC (Sun) by branden (subscriber, #7029) [Link]

Having read this whole exchange...

If you're curious where the oppositional tone you're inferring from Jon 
Corbet's words is coming from, I suggest you take a look in the mirror.

Who is "we" referring to?

Posted Mar 9, 2008 7:36 UTC (Sun) by robla (subscriber, #424) [Link]

I'm sorry for being way more combative than I needed to be. Looking back through the
exchange, you're right. My last message was meant to be conciliatory, but I just didn't write
enough or in the right way.

I still believe my question about "we" is valid, because it gets to the heart of one of my pet
peeves. The benefits of a good infusion of previously proprietary code from a corporate donor
is a matching infusion of developers who used to be proprietary developers. Saying that "we"
don't need these infusions is saying that "we" don't need the developers to be part of the
"we". Now, I realize that nobody here would actually mean to say that, but that's why I think
being sloppy about the use of "we" is dangerous.

So, that's why I'm asking about "we". I figured rather than assume the worst (as the above
paragraph does), I'd ask point blank what is meant by "we".

I have questions about other phrases, such as what is meant by "free software can't actually
invent anything" (as a stereotype to be feared). Well, as written, that's mostly true. I
think true invention is actually beyond even Eliza, though it wouldn't surprise me if Emacs
has a M-x invent-anything command.

I understand the gist of what is actually meant by that phrase (though a clarification would
help), but I don't understand why it's so worrisome. The bulk of desktop software is
proprietary now; it's better if the existing software is made open rather than rewriting it
from scratch. Anyone so worried about that stereotype shouldn't be out creating yet another
virtual world platform or online gaming system, they should be creating a whole new category
of software.