Re: [DotGNU]Call for a truce over mono vs pnet

[Paolo Molaro]

On 03/17/02 Barry Fitzgerald wrote:
> Then don't collaborate on the runtime/compiler collection.  There's room
> for competition and room for cooperation -- but, we're both going to have
> to take a step forward and figure out where that is without strangling
> each other.  (Both meaning our two projects)

Agreed, I, too, think that at this point it doesn't make sense to
drop one of the runtimes and the compilers have different goals
(you'll note that I didn't ask for that: I was only replying to Rhys
that suggested mono to drop one of its major components). I'm not
going to ask anyone in DotGNU to throw away his work.

> > cscc is months behind mcs and has different goals, it won't be
> > usable as a component etc, so we can't drop mcs.
> 
> Whoa here -- if we're trying to find a solution let's keep away from
> subjective criticism of the other project.  There's been more than enough
> of that to have it enter this discussion.  Yes, I do view "xyz project is
> months behind abc project" to be a subjective criticism.  It enflames the
> situation and brings bad will.  Keep in mind, there are also things that
> mono can't do that pnet can.

I'm well aware of that. For example you can rightly say that my
interpreter is slow and not thread-safe. I wrote it, that doesn't mean
that it doesn't have problems: it's not a personal issue, but a
technical one. I was only pointing out that cscc can't
compile mcs, so we can't use it. We're not going back to require windows
for development once we can do it all on free systems. If you consider
that argument a personal one or a criticism of DotGNU, I apologize, it
was not intended as such, but only as a technical issue.

> > There may be more room for cooperation in the C# libraries, but here
> > again we can't use your code unless you allow us to, while you can reuse
> > our code. So, it's really up to DotGNU to decide what degree of
> > cooperation they want with mono. Or am I missing something?
> 
> Whoa here as well, the limitations of code copying occurred when Ximian
> changed to the X11 license for the library.  Don't try turning

No, our libraries were LGPL before, so we couldn't have used your GPL
libraries as well before the license change.

> your decisions around on us.  According to the GNU licenses, all you need
> do is ask -- not act like there's some kind of imbalance of power that
> we've caused.  That simply is not so.

I'm not accusing DotGNU of anything: you already can use our code
(and some DotGNU developers already stated they will do so).
If you want to collaborate on libraries it's your choice to share your
code, since our code already can be used by DotGNU.

Some of our developers were wondering about asking DotGNU about some 
code in pnetlib. AFAIK, they didn't ask for it mostly because they were
afraid of bad replys or something like that. What I'd like to happen is
some statement from DotGNU developers about the _possibility_ of
relicensing some library code for use in mono. This way at least
people will not be afraid of actually asking for it, even if for some
reason or for some kind of code DotGNU won't allow that.
I, for one, would never ask for a chunk of code if I'm going to be
accused of wanting to claim it as my own or if I'm going to have
to spend the same time it takes to rewrite it discussing the issue
on a mailing-list.

> And yes, you're missing something.  Please heed some attention to my
> proposed settlement of the library issue.  For now, let's table
> cooperation on the compiler suite and focus on the upper level library
> which, to my estimation, is the place where we have the greatest chance of
> collaboration.  Once we've got agreement on how to handle that, then we
> can talk about licenses and how we're going to handle them.

I think we all agree to collaborate on the libraries (at least, so do
the mono developers I know) or I wouldn't bother to reply to this
thread. How and if that is done depends on the licenses. I see that
can cause problems for some DotGNU developers, but I don't see how
we (mono) developers could solve that, since our code can be already
used by DotGNU. Suggestions welcome.

> Also, the solution can't be "DotGNU will give it's code to mono so that
> mono can claim it" -- that's not a diplomatic solution: it's a stalemate.

mono is not going to claim anybody's code. If some code is relicensed,
it's not going to change attributions or authorship.
You're welcome to reuse our code, we're not going to accuse you of stealing 
it if you do reuse it.

lupus

-- 
-----------------------------------------------------------------
address@hidden                                     debian/rules
address@hidden                             Monkeys do it better