Re: Again: list of advantages of the xGPL

At 07:24 AM 6/21/02 -0400, Roger Fujii wrote:
>Roland <[email protected]> wrote:
>
> > Look, I didn't say that the BSDL hasn't its merits. I just gave an example
> > of a successfull GPL project where there is an BSD equivalent.
>
>Not a great example, given that BSD had a penalized beginning, and 
>considering:
>   SGI has one foot in the grave with Red Hat staring at another 6 foot 
> deep hole.
>   What Sun's entry hasn't been seen yet, and IBM just dropped linux for its
>portables.
>Apple has probably already shipped more OSXs than all the companies you 
>mentioned
>(minus RedHat) has shipped Linux.  Linux *now* has a lead, for sure, but 
>if you
>count the devel dollars for companies porting / developing apps for OSX, I 
>think
>you will probably lose your bet.

I was talking about money invested on BSD, not in apps that are going to 
run under BSD. Furthermore, when you consider OSX, this is an  Apple 
specific OS, it is not freely available. It's a proprietary extension to BSD.

>
> > It may happen that objectively a BSDL would be better for a given project,
> > but in practice it might turn out that only few developers will join
> > because many are biased towards the GPL. So if you choose the GPL you will
> > suddenly attract much more coders.
>
>What evidence have you seen to support what you just said?  Most people don't

I started saying "it MAY happen". Please repeat with me: "it MAY happen". I 
was making a hypothetical example.


>CARE what the license is, so long as they can use it for free.  For most 
>projects,
>any contributions are minor compared to the codebase, so most programmers will
>give it up to whatever the license the project uses, so I'm sure the number
>of X11 (ie I don't care) programmers far exceed the number of GPL only 
>programmers.
>

The recent poll on the wine project proved that a majority voted for the 
LGPL, so what do you say now?

>
> > On the other hand you may choose the BSDL and a company starts using the
> > codebase and giving back many improvements, so the BSDL proves successfull.
> > BUT on the same scenario it may also happen that a company doesn't give
> > back many improvements, hires away the best developers of that project and
> > leaves it to stagnation while the proprietary fork is steadily improving.
>
>You know, I keep hearing about this, but cannot recall *one* example where
>this has ever happened.  Care to name an example?

Well...hmm...I have to think very hard. Ohhh, now I remember...there is a 
project called WINE and a company called TG....
Of course it didn't happen the exact way but similarly: they built Direct3D 
support into WINE but didn't give back the relevant code. So now the free 
project is lagging behind *considerably*. And I think they also hired some 
free Wine developers(I may be wrong on this one).

>
> > So how can you know? Reality is too complex. We see examples of GPL
> > projects that are succesfull(Linux),
>
>Stop using the plural if you are only going to cite ONE example.
>
> > and we see examples of BSD examples that are successfull(Apache, 
> FreeBSD, etc...).
>
>FYI, PHP4 went from GPL->Apache.

And Wine went from xGPL -> BSD -> LGPL

> > Imagine for a minute that Linux where under the BSDL and Apache, FreeBSD
> > where GPLed. I dare to say it wouldn't make much of a difference for those
> > projects. Please prove me wrong...
>
>It would make a difference.  Had apache been GPLed, it's unlikely that 
>Sun, IBM,...
>would financially contribute (you don't see them giving $$$ to FSF, do you?).

Well how comes that IBM is contributing financially to Linux? The same goes 
for RedHat, Lindows, etc...

>Can you name one non-kernel GPL component that can't be relicensed that 
>IBM, SUN,
>etc.,
>has made?

No, but what does that have to do with our discussion?

Roland