Boost logo

Boost :

From: Beman Dawes (bdawes_at_[hidden])
Date: 2001-11-15 22:05:40


At 02:38 PM 11/15/2001, Jens Maurer wrote:
>"Paul A. Bristow" wrote:
>>
>> I think a single licence is highly desirable.
>
>Yes.
>
>I like the approach of having a "default" license lying around
>somewhere in the boost web tree and then just saying
>
>Copyright Jens Maurer 2001. See http://www.boost.org/more/license.html.
>
>in each source file. People checking the license situation
>can then just "grep" for "copyright" (case-insensitive) and need only
>examine those files more closely that don't have the reference
>to more/license.html.
>
>If you want to specify additional terms to more/license.html in your
>particular source code, then you must not refer to more/license.html,
>but copy the license text into your code and make the adjustments you
>deem necessary.
>
>Note that users not looking at more/license.html have no rights
>to use your code whatsoever, because only the license gives them
>the "free use" permissions. The "default" for copyright is
>"nothing allowed".
>
>> Surely for something as simple as this even the lawyers can agree?
>
>Our copyright statements are strikingly similar, and I'm ready to
>drop my "copyright must appear in documentation" requirement from
>Boost.Random.
>
>> 'no other person claiming copyright'.
>> (As I understand the situation, if someone else claimed copyright,
>> they could claim money for infringing it and/or stop others using it).
>
>Not required. The situation is obvious if there's a "copyright"
>notice on it. Nobody else can claim authorship and thus change
>the terms.
>
>Here's a concrete suggestion for more/license.html:
>
>// (C) Copyright David Abrahams 2000. Permission to copy, use,
>// modify, sell and distribute this software is granted provided this
>// copyright notice appears in all copies. This software is provided
>// "as is" without express or implied warranty, and with no claim as
>// to its suitability for any purpose.
>
>
>We would phase this in gradually by encouraging developers to
>switch to this license in their source code. However, authors
>must change their code themselves. There won't be a big, centralized
>change-over.

This really seems like a sensible suggestion to me. We can work on the
details of the wording, but the basic approach of having a single "usual"
license, allowing other licences (assuming they meet our normal criteria),
and having a gradual transition all seem like good ideas to me.

I'm tied up with non-Boost work for a week or so, but after that maybe I
can run the whole thing by a lawyer I know to make sure we aren't doing
something totally stupid in a legal sense.

--Beman


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk