I tried real hard and I can't find a parallel that shows how ill-advised
it is for people to be creative in their licensing. 

1. The fact that the law is bad doesn't imply that what we do doesn't
matter. This is like saying that designing machines and reusing the good
ones is useless because physical law contains indeterminacy. As with
physical law, to cope with the complexity of copyright we should do the
simplest thing possible (use existing, simple, fair licenses), not
depend on the details of licenses, and then hope for the best.
2. I hear people trying to solve all kinds of political problems with
licenses. I want the company to share, I want to know when someone does
this or that. Whatever goals the author has with what code he shares,
including these, are legitimate. And some of these may *require* to use
the awesomely messy, ill-defined, potentially destructive power of
licensing. I don't blame Stallman for writing the GPL, nor you, Alan,
for SqueakL. As you said, it served an important purpose in Disney. But 
most of the reasons would probably be best served by a clear readme, 
without need for inviting the vampire (copyright law) into our home 
(more than needed). If we decide we *need* a non-MIT style license to 
serve our purposes, fine. But lets not shoot ourselves in the foot because 
it scratches.
3. People may ignore it as they wish to, but Smalltalk and Squeak have
languished on the vine because the community *chooses to be ineffective
at sharing*. Huge piles of Tex, Perl, Python are effectively shared by
very large communities, because they have bothered to do the details,
*technical and political*, right. We have better technology, many good
ideas, but even the artefacts most widely shared between dialects
(libraries, Refactoring Browser, camp smalltalk projects) don't reside
in one public repository, with clear permission for all to use. Debians
rules are not just important because they allow Debian inclusion - they
are important because they've thought about the problem of long term
effective software sharing, they're doing it more successfully than we 
are, and we can learn from them.

I have met and spent too much time with stupid people ruining things
completely. Seeing brilliant, good people whose company I enjoy very
much continually waste opportunities is almost as sad. 

Well, enough. To the extent that people care, I have placed what I 
believe is the best route for individuals and the community to take at 
http://swiki.squeakfoundation.org/squeakfoundation/103. Constructing a 
MIT-simple-like license with weak SqueakL style sharing clauses is
probably not hard, though *I* believe that's a crutch Squeak no longer
needs to get contributions.

As for Apple, if they can be convinced to place it even under APSL 2.0,
that would probably be a *very good thing* (... since it seems unlikely
they'll even consider anything better).

Daniel

Alan Kay <Alan.Kay at squeakland.org> wrote:
> I hate to be the kid in the crowd who noticed the Emporer was naked,
> but do the people on this list -- or for that matter, does anyone on
> the planet -- actually know what the law is here? From my
> perspective, having been involved in part of the gradual merging of
> lawyers and computer ideas, there is no "good law" about any of this
> stuff at present. I believe people are getting quite misled by the
> fact that anything, including nonsense, can be written down in
> language. This has nothing much to do with either logic and/or moral
> stances that would be good to take.
> 
> Just to take one of many examples: at present no one really knows
> what "fair use" really means, or even should mean. We know that many
> business interests would like to get rid of most, if not all, of the
> old ideas about "fair use", but regardless, it's not clear what the
> term even means right now.
> 
> Another example is Squeak Central at Disney. You can imagine that I
> got asked quite often (especially for the first 3 years we were
> there) about the IP issues regarding Squeak. I consistently told them
> that they should think of Squeak as an environment of "agencies" that
> are good for making things but that they don't own or control -- and
> to use Java (owned by Sun) as an example. So content creations in
> either system could be owned by them and used by them in commerce,
> but that the foundations of the systems were owned elsewhere (Java by
> Sun, and Squeak by the world). I used the important phrase in the
> Squeak license to explain why we put out to open source all non
> Disney content code that could contribute to making Squeak more
> powerful and useful for everyone. When we left Disney we were very
> careful to leave all Disney content behind -- and this was partly
> done by having a separate Disney update server to keep the public and
> private Squeak stuff separated. Michael Eisner and his staff of
> Disney lawyers, some of the toughest IP hawks you'll ever meet,
> eventually decided that what we were doing was kosher, reasonable and
> a good interpretation of the license we made while at Apple.
> 
> In other words, what Disney actually owned, we were careful to leave
> behind, and Disney owns nothing else. Apple really did own all the
> rights to its early implementation of Smalltalk-80 (as did the other
> first adopters -- but not the later adopters). This is why we were
> very careful to find an original version of this Smalltalk, in which
> the provenance is completely clear and unambiguous, to bootstrap
> Squeak from. And this is also why we were careful to have Apple grant
> a much more free license than (say) GPL (which actually puts really
> stupid restrictions on what others do with the code). I think this
> has all worked out well.
> 
> While all of you are wasting (er, spending) a lot of time on this, it
> might be good to realize that the definitive definitions, suits,
> court cases, and actual laws are all really yet to happen. Given the
> high level of greed and low levels of perspective and intelligence of
> many of the players here, we should hope that all this is delayed as
> long as possible. The chance of actual logic, insight and forward
> thinking having much force here is nil.
> 
> As I said once before: at this point, we need better lines of code
> more than better lines of license! However, I think there are a few
> things in SqueakL -- the Apple license (the only one that obtains in
> my opinion) -- that could be removed to make it smaller and simpler,
> and this might be possible to do.
> 
> Cheers,
> 
> Alan
> 
> At 11:56 AM -0400 8/19/03, Lex Spoon wrote:
> >"Andreas Raab" <andreas.raab at gmx.de> wrote:
> >>  Hi GÐran,
> >>
> >>  > Again, and I am getting somewhat tired to repeat this: I am
> >>  > not talking about the Right or Wrong of the two licenses
> >>  > involved here. I am talking about the very fact that we
> >>  > have TWO licenses currently.
> >>
> >>  Okay, fine. So you talk about the fact. But what's your problem with it? It
> >>  seems to me that you haven't made it quite clear as to what your concerns
> >  > are. (of course, I am assuming that you _do_ have a problem with it or else
> >>  I think you wouldn't talk about it in such length; if you are merely
> >>  rambling please let me know ;-)
> >
> >It bothers me, too.  Having multiple licenses makes things more
> >complicated.  If Squeak-L is okay for VM's, then let's lobby for all the
> >VM's to be put under it.  If there needs to be something more for VM's,
> >then let's at least use the *same* Squeak-VM-L for all VM's to use.
> >
> >And besides that, we *definitely*  need to have accurate descriptions on
> >the download pages.  I hadn't thought about this before, but it's true.
> >For people who do worry about obeying the law, it is important to know
> >what licenses are involved.
> >
> >
> >Lex
> 
> 
> --