Re: Mostly free software...
On Thu, 25 Mar 1999, David Welton wrote:
>
> > Often there's no particular reason for these in-house apps to be free, but
> > often there is, and there's certainly no harm in it most of the time.
>
> Sure there is, if there is someone else who is paying nothing to use
> the work that you had to pay a developer for.
>
True. As I said my company also has some "secret" code that is more
directly related to what it does.
> > Mozilla,
>
> Started out proprietary, only went open source when it looked like
> they wouldn't make money off it anyway.
>
It's still an example of 100+ people getting paid to work on free
software. These guys are making nice salaries!
> > Red Hat,
>
> Not to knock them, I think they have done some great things for the
> community, but they don't *develop* much at all. I would be surprised
> if they do %1 of the code in their distribution.
>
But again, a bunch of people paid to hack...
> > Cygnus,
>
> Once again, others did the work, and they are making money from it,
> although they are putting a lot back into it. They also have several
> proprietary apps (I don't know when or how they decide to draw the
> line).
>
The founders did some significant gcc work before founding it, and Cygnus
has done most of the recent work on gcc. Again, probably 100+ people paid
to hack free stuff. It's not a "pure" company but they have lots of cool
jobs.
Michael Tiemann's article from "Open Sources" is interesting. It's on the
web somewhere...
> > AbiSource
>
> These guys are interesting. They are the only ones that I know of
> that are writing their own app, from the ground up, and releasing it
> as free software, and at the same time, doing it as a business, and
> not just as a cool project. I really hope they can make a go of it.
>
I do too, I agree it's pretty cool.
It all depends on whether anyone is really willing to pay for support.
In the Cygnus case it seems like they are. However for a word processor?
Time will tell.
> It would be cool if the state of the art were free, and you can't do
> that in your spare time. Universities seem to be one place that can
> produce really good free software, but they are not really profit
> oriented institutions...
>
Yep, that's an option too. Another option is government contract work -
you could maybe sell them on the benefits of free software... they did
fund quite a bit of it in the past one way or another.
> How about "I have this really cool idea for a hot product - how can I
> spend all my time hacking on it, improving it, and making it cool, and
> make it be free software, at the same time?"
>
A good question! I don't think the answer is very simple though, and there
may not be one in many cases. It's just important to keep that in mind, is
my feeling.
> > However I think it would be kind of lame if I tried to say "Since
> > I'm writing proprietary software to make a living the free software
> > community should use my software and not try to create a free clone"
> > (Larry McVoy)
>
> Has he said that? I thought the idea was just that his stuff was
> supposed to be better than CVS.
>
It is better - CVS basically sucks in many ways. But he tries to push a
bunch of ideology too. He has a stated goal of getting 100% of free
software projects to use the software - even though many don't want to use
non-free software. He kind of acts like this is a big donation (even
though the marketing benefits for him are huge, and no free software
project would ever buy a license anyway so he loses nothing).
It's much like Troll Tech: they've figured out that KDE is a huge
marketing stunt and QA team, so they've even hired guys to hack on it.
Of course Larry isn't selling a library so he can't do the viral-license
thing.
I shouldn't pick on him too much, because he certainly *has* contributed
lots of free software. I just think he's misguided on this point.
(FWIW MySQL is a similar example - they get marketing, and they get QA,
and they prevent a truly free project from gathering momentum. Hopefully
BitKeeper won't have a similar dampening effect on PRCS2.)
> Yeah, this is kind of bogus, but I think it varies from case to case.
> In a lot of situations, these things are driven by an individual, a
> small group, a corporation that has already put a lot into a
> particular product. It isn't as if the community will add a great
> deal in the immediate future. See Jamie Zawinski's comments about
> people not helping out with mozilla (and that *is* free software).
>
True. Both Cygnus and Mozilla seem to cite the 80% number - 80% of work
done in-house. However, both seem to be getting very nice, quality
products out of the deal, for whatever reason.
The real gain right now is marketing. I think that's what Sun and Apple
are thinking. It's definitely part of what Troll Tech, BitKeeper, and
MySQL are thinking.
> Let me reiterate that I *do not* advocate relaxing our stance on free
> software. At all. I just don't want to follow RMS' example of giving
> up such things as a family, house, car, etc just because I like to
> write free software;-)
>
Agreed. I think RMS goes beyond liking to write it though, he feels
positively obligated to write it and pours the money he could spend on
those things into the FSF instead, just because he wants to.
Havoc
Reply to: