[OT] Re: Request: Summary of GPL Problems

Bijan Parsia bparsia at email.unc.edu
Tue Nov 13 23:17:33 UTC 2001


On Tue, 13 Nov 2001, Andrew C. Greenberg wrote:

> 
> On Tuesday, November 13, 2001, at 04:29  PM, Bijan Parsia wrote:
> 
> > On Tue, 13 Nov 2001, Andrew C. Greenberg wrote:
[snip]
> >> writing out of the license the (known to be essential) exception for
> >> independently written programs.  At least not as RMS interprets the GPL
> >> (I do believe there are strong legal arguments to the contrary).  In
> >> part for this reason,

     ^^^^^^^^^^^^^^^^^^^^ what I'm taking as the qualification.

> >> no GPL'd Smalltalk system has been successful to
> >> date, nor can it ever succeed, again, for reasons previously stated.
> > [snip]
> >
> > There is GNU Smalltalk. I don't know if it's "successful", but it's VM 
> > is
> > GPLed, the core class libs are LGPLed and the browser and compiler are
> > GPLed, all, I would imagine, with the blessing of the FSF.
> >
> > 	ftp://alpha.gnu.org/gnu/smalltalk/gst-manual/gst_4.html#SEC4
> 
> That was precisely the example I had in mind when I qualified the term.  
> GNU St hasn't exactly generated a substantial community of developers or 
> obtained substantial support or extensibility, notwithstanding the ready 
> access to the corpus of GPL code.

Erhm. Yes, but even the "in part" isn't reasonably shown, IMHO. It *is*
clear that mere GPLness isn't sufficient for generating a substantial
community etc. Personally, I rather suspect that the value add would go
from Squeak to GPL, not from GPL to Squeak. There don't seem to be
"waiting hordes" of would be Smalltalkers panting for a GPLed
squeak. OTOH, I was chatting with some folks and they wanted to try out
Squeak and found the lack of Debianness a bit of a turn off. <shrug/>
Mostly they just wanted a set of Debian packages which I had no offhand
idea where to find. That seems to be *my* fault :)

Note too that GNU Smalltalk had a *very* long period of stagnation, is
less ported, and started up again about the time of Squeak's surge and
VisualWorks' recovery. There are many reasons why GNU Smalltalk isn't,
afaik, wildly successful.
 
> > I have no idea how this flies. My casual perusal of the uppercased bits 
> > of
> > Andrews messages suggested to me that this wasn't feasible.
> 
> Indeed, it appears that FSF has shown greater flexibility for GNU 
> Smalltalk than he did in Squeak.  Although, it is important to note that 
> Gnu Smalltalk was developed prior to RMS' DLL paranoia, after which he 
> took a substantialy more strident position.

Is this true for the current course of development? I'm pretty sure big
chunks (like the Browser) are quite recent developments. And they're
GPLed.

>  I think, however, that 
> armed with this, I may well go back "to the well," and ask him for an 
> explanation of some of his earlier representations.

Cool. I'll be interested to see the results.

> Perhaps a model of GPL for an underlying VM and LGPL for all Smalltalk 
> image code might not have the inherent deficiencies I noted with the 
> prospect of an All-GPL code.  I regret not considering this possibility 
> earlier.

We forgive you Andrew. At least, *I* do, and who else matters? ;)

> I am curious about the description of "Libraries" -- does GNU St use an 
> image and module concept?

Image yes.

Some sort of namespaceing..."enviroments". Seems a bit like Squeak
enviroments, but heck, what I don't know about modules is pretty close
to what I don't *want* to know about java.

But they don't seem to partition the image even as much as, say,
imagesegments. It's seems to be entirely a name resolution
mechanism. Otherwise, it seems to be a plain jane Smalltalk-80 system,
with block closures.

Oh! WikiWorks is GPLed. I don't know what Cincom thinks of that.

Ah, GNU St has "packages"
	ftp://alpha.gnu.org/gnu/smalltalk/gst-manual/gst_19.html#SEC24

but it just looks like a file-iner/dependancy thing.

But the end result certainly seems to be a monolithic image.

Cheers,
Bijan Parsia.





More information about the Squeak-dev mailing list