[ANN] Closure Compiler

Cees de Groot cg at cdegroot.com
Wed Mar 26 13:29:54 UTC 2003


On Wed, 2003-03-26 at 13:55, goran.hultgren at bluefish.se wrote:
> > Squeak's license is irrevocable(*). So there is no risk at all that
> 
> How sure are you of that? I have been trying to find info on
> irrevocability but had a hard time finding stuff.
> 
Gut feeling. IOW, quite sure ;-). The stuff is published, and
publication typically is an act you cannot revoked. Furthermore, the
stuff is published with a license. Now, it is only reasonable to assume
that the term of the license is the term of the publication, which is
open-ended.

You can't publish a book and then 5 years later ask people to destroy
all their copies...

> Eh, IIRC Andrew even got to the conclusion that you *must* share it even
> if you are not distributing it! Probably an unintended flaw in the
> license, but anyway.
> 
Who am I to disagree with Andrew, but I'm going to do so nevertheless
;-). "If the Modified Software contains modifications [...], then for so
long as the Modified Software is distributed or sublicensed to others,
such [...] portions of the Modified Software must be made publicly
available. [...]". QED.

> This is the interesting question. So you mean that we should allow
> multiple licenses in "official Squeak core"?
> 
Well, I'm arguing we shouldn't absolutely forbid it.

> Wouldn't this create a big messy pile of licenses that possibly
> conflict? I mean, who am I/you to say that MIT/BSD doesn't conflict with
> Squeak-L? I agree - it looks like they don't but hey... :-)
> 
Who are you to say that the SqueakL is binding, valid, and doesn't
conflict with operation of the VM on a computer?

The analysis to be made is will it impact someone who wants to use
Squeak and doesn't want to wade even through one license, let alone n
licenses? Roughly speaking, as I indicated, you can list the rights,
limitations and obligations of both licenses. If the set of rights of
the alternative license is a (strict) superset of the set of rights of
SqueakL, and the set of limitations+obligations is a (strict) subset of
the set of rights of SqueakL, then someone just reading the SqueakL is
informed of all rights, limitations and obligations that apply to
her(*). Now, for the legal paranoids we may want to include a full
disclosure of the packages included and their licenses, like:
"This software includes the following packages:
- The X11 font distribution, under the following license:...
- SmaCC, under the following license:...
- RB, under the following license:...
We believe, after careful analysis, that these licenses in no way limit
the rights granted to you by the Squeak License nor that they put
additional burdens on you relative to the Squeak License. However, the
Squeak Community is not (yet) a law firm, so this doesn't constitute
legal advice yaddayadda"

As we like to look at Debian (for good reasons!), it is mostly the same
situation. All the packages residing in Debian have had their licenses
analyzed mostly by laymen; Debian in effect declares "here's a bunch of
stuff that we think all deserves to be entitled free software, which
means that it grants you these and these rights. However, we're not
lawyers so please consult one for legal advice if your life depends on
it". Heck, I bet that Debian even distributes non-GPL kernel modules (as
far as these exist). There's no single license on the distribution, just
a general notion (ok, a fairly precise one put forward in the DFSG and
similar documents) of 'free software'. 

So, to repeat ad nauseam (luckily for the readership, I still am able to
find new twists to present the same argument): a 'monolithic' SqueakL'ed
image is a nice *means* of keeping things simple for users (and the
other two reasons I presented), but it should never be an *end*.
Certainly not if it sits squarely in the way with necessary cleanups
(SmaCC, for example) and mainstreamization (replacement of browser by
RB, for example).


(*) outcome of my gender-dice. 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20030326/1f0bbc2a/attachment.pgp


More information about the Squeak-dev mailing list