DRobbins departing now, too.

Posted Apr 27, 2004 13:07 UTC (Tue) by Duncan (guest, #6647) [Link]

> BTW, what's happening with Zynot, since you mentioned it?

Well, I wasn't looking for /that/, when I was there, and didn't really
notice. He had up a "why fork" article and a sequel, and I had gotten a
direct link from a newsgroup discussion to the fork article, so I read it
and the sequel, and took a brief look around the site since I'd made up my
mind by then and figured I wouldn't be back, and that was it.

The weird thing about the why fork article from my perspective was how
really out of touch he seemed with what makes many open source developers
tick, and how he seemed to keep attributing to others (DRobbins in
particular) the selfish motives it was quite apparent he himself had.

Keep in mind that as I started looking at Gentoo, coming from Mandrake
(which I'm still using while I experiment with Gentoo), the FIRST thing I
looked at was the social contract and statement of philosophy on the
Gentoo site. Mandrake has always been a strong open source supporter,
but, for instance, I could have never been happy with SuSE, before Novell
bought them and opened up YAST and etc. under the GPL, because, as I've
often put it, if I wanted to work on and support closed source, I wouldn't
have dumped a decade of MSWormOS experience to switch to Linux. I don't
claim to be a good developer and in fact have done NO real development on
Linux at all, yet, anyway. However, I can contribute in other areas and
may or may not eventually BECOME a developer. Anyway, one of the major
things that makes ME tick, is the contribution to the betterment of the
community. I am *NOT* interested in making my work available for some
proprietaryware developer to take and stick behind closed doors. I might
as well find something else to do with my time if that's what it comes to.

Anyway, I was quite impressed with the gentoo social contract and etc.
However, reading about the fork, I knew I had to investigate it and see
what THAT view of Gentoo had to say. Once I got there, as I said, it was
rather surreal watching this guy argue how DRobbins wanted to steal the
profit from Gentoo and all about how he was supposedly going about it,
when it was quite apparent that this guy himself had such motives, and
that was the real disagreement. After all, DRobbins couldn't do anything
about the license of existing Linux software, but he didn't HAVE to set up
Gentoo such that any contributions IT made were GPL licensed. If he was
REALLY intent on taking Gentoo proprietary, GPL licensing Gentoo
contributions was a HUGE mistake, since /anyone/ could fork it if they
didn't like where it was going. What was even MORE ironic about the
entire thing was that the Zynot fork demonstrated this very point! Zynot
was saying how proprietary Gentoo wanted to be, then taking its very GPLed
code to form the fork with!

Time and again, argument after argument, the "why fork" paper demonstrated
not why it was so good, but why it was on the *WRONG* path for an open
source project, as it forked off of Gentoo. This guy had been a closed
source developer for some time, and argued some of the textbook classic
methods for getting the best product possible.. only they applied to
CLOSED source, NOT OPEN source. Among other things, he argued for a set
release timetable and only letting it slip slightly, if at all. The open
source version of that is more or less continuous development, release
early and release often, on the one hand, but version update not according
to a timetable, but rather, "when it is ready". This is only one example.
There were several. It was /quite/ obvious that this guy and me
had /very/ divergent opinions on where the Gentoo code and open source in
general should go. It was ALSO quite obvious that he'd never fit well in
a truly open source environment, and therefore, why he felt compelled to
split from Gentoo, which thanks to DRobbin's evident early lead in the
area, tends to be toward the "open" side, even within the open source
community itself.

Looking at where he wanted to take Zynot.. into the embedded community..
one does see that his approach might work a bit better there. Embedded
code tends to be firmware code, released with the device and often never
updated. Even if it is updated, updates tend to be fewer and skewed
toward the early end of hardware release and customer purchase cycle. In
this environment, code DOES need tested more thoroughly, and the
traditional quality control and release cycles of closed source continue
to make more sense than they do in the general desktop and server
computing environment. As well, his approach to potentially proprietizing
some stuff but on an open base, isn't unknown in the community. I already
mentioned SuSE as a distrib, and there are plenty of proprietary
application vendors. However, the question I kept asking myself as I read
this was how IN THE WORLD he ended up with Gentoo, with its generally open
source emphasis, when his ideas and methods tended to be SO diametrically
opposed to those of Gentoo.

I honestly don't know if it's having any success or not. There are other
already established vendors in the Linux embedded markets, others already
going in Asia, Taiwan/China/Japan/Singapore/HongCong/etc. supported and
encouraged in many cases by their respective state governments. That's
where the embedded market is, and he's going up against existing entities
AND has a language and culture barrier to work against. He sees the
market as wide open to opportunity and it may be. However, Western firms
have always had a hard time breaking into the Eastern cultures, and
there's already state support and encouragement of local Linux based
solutions, more so than in the West, so it's going to be harder than he
believes, I think. As well, once folks get the open source idea, I don't
see them continuing to function well under his limited top-down command
structure. Open source, unlike closed source, simply CANNOT thrive under
those circumstances. I expect he's going to have quite a hard time of it.

That said, again, the culture of embedded systems is much different than
that of general computing. Even if he was quite successful, unless
someone there is making a specific effort to get the PR out to the general
Linux world via LWN or similar, we may not read of it or realize how
successful he may be. From what I know, the world of embedded systems is
like that, and only those in the trade know what's going on in it. As an
example, until recently, how many in the Linux world had any idea how
widespread use of Netfilter was? It's only now, as the netfilter folks
track down GPL violators and force them to release source to comply with
the licence, do we glimpse some of what must have gone on behind the
scenes, the world of embedded code suppliers and the levels it goes thru
before a hardware vendor purchases it, possibly not knowing its open
source origins, and ending up having to go back and track down source to
fulfill obligations they didn't realize they had (due to lack of
disclosure by the middlemen) when they purchased the code in the first
place. It's possible that five years from now, one of those shaddowy code
vendors might be Zynot.

FWIW..

Duncan