From: David Abrahams (dave_at_[hidden])
Date: 2003-09-17 16:43:36
Beman Dawes <bdawes_at_[hidden]> writes:
> At 01:40 PM 9/17/2003, David Abrahams wrote:
> >Beman Dawes <bdawes_at_[hidden]> writes:
> >> Sorry, I haven't been clear. It is a "stable link" to a specific
> >> license that would be a disaster. It would negate our objective of
> >> having a single Boost license for all Boost libraries, because when
> >> the (even though rare) license changes occur, there would be sure to
> >> be some copyright holders who weren't reachable, and thus the licenses
> >> would diverge. Some would be version 1, some version 2, and so
> >> forth.
> >And how will not having a stable link to a specific license prevent
> >> An important advantage of the suggested comment wording is that it
> >> is not version specific. In effect, the copyright holder is saying
> >> "use the version of the Boost license that is current at the time of
> >> release, even if that version changes from release to release."
> >> The safe thing to do is to not mention a version by name.
> >Oh! If that's legal, I'm for it. However, we may have trouble
> >convincing authors to allow their source to be automatically
> >re-licensed under new terms at the whim of whoever's maintaining
> Changes to the license must be very rare and carefully considered. It
> is really like a language standard - stability is necessary to get
> developers on board.
I still question the legality of licensing by reference to text which
may change dynamically. And it may have a chilling effect. It's not
just the authors who might be nervous, but the users as well. If you
were a corporate lawyer, would *you* OK the use of source code whose
license might change at any time without notice? I believe users must
be allowed to continue to use the code under the *original* licensing
terms they agreed to.
-- Dave Abrahams Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk