Account Options

  1. Sign in
Google Groups Home
« Groups Home
Message from discussion What about allowing the choice of the "Boost Software License"? A promising project could appear...
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 
From:
To:
Cc:
Followup To:
Add Cc | Add Followup-to | Edit Subject
Subject:
Validation:
For verification purposes please type the characters you see in the picture below or the numbers you hear by clicking the accessibility icon. Listen and type the numbers you hear
 
Daniel Berlin  
View profile  
 More options Aug 23 2007, 2:15 pm
From: Daniel Berlin <daniel.ber...@gmail.com>
Date: Thu, 23 Aug 2007 18:15:30 -0000
Local: Thurs, Aug 23 2007 2:15 pm
Subject: Re: What about allowing the choice of the "Boost Software License"? A promising project could appear...

On Aug 21, 2:22 am, "roman.l.s...@gmail.com" <roman.l.s...@gmail.com>
wrote:

> I was wondering whether the choice of license is really strictly
> limited to those found in the dropdown when registering a new project?

It is

> I'm thinking of publishing my "libborg" (code-name, not definitive)
> project on code.google.com. It uses the boost c++ libraries
> extensively and is meant to provide generic APIs for many concepts (i/
> o, fs, markup, gfx, gui, dsp) found redundantly across many frameworks/
> libraries. It allows to write algorithms based on the generic API and
> thus to decouple your collection of algorithms/solutions from the
> particular framework (or even language) used. It will provide proper
> concept checks, archetypes and tests for all the generic interfaces.
> It is an almost header-only library because e.g. the binding of an
> algorithm using the generic XML API to the libxml2 interface is done
> at compile time using template-metaprogramming techniques, which
> results in binary code as fast as calling the libxml2 (C) API
> directly.
> The source tree (and an external project on sourceforge, "jsmm", which
> will eventually be merged in) also contains experiments on language
> interoperability, wrapper generation à la swig, cross-language-
> reflection. involvement of boost::python and cint à la ROOT from CERN.
> A lot of things will be derived/generated from Boost C++ headers and
> from 3rd party libraries with more restrictive licenses like GPL and
> end up in object code together with private or commercial rather than
> open code. However, as the libborg is only a very thin layer of glue
> and a set of tools allowing the user to achieve more decoupling it
> should have a license with the most possible freedom and I'd consider
> the boost license as most appropriate.
> Would the New BSD license probably be an alternative for this?

Yes.  But if you want to be as close to boost as possible, use MIT.

Boost's software license is in fact, a small variation on the MIT
license.
It includes a little more legal language that, IMHO as a lawyer, is
completely unnecessary,
but this is the way new licenses get made.  Lawyers disagree on these
things, and copyright law gives us no answers.  Nobody is willing to
take a stand.

I would encourage you to use the MIT license here.

The reason we are strict about what it is in the dropdown, and what
projects can use is because we are very much against license
proliferation.  There are already too many OSS licenses, and the vast
majority simply do not encompass interesting choices that the ones in
the dropdown don't cover.  Things like the boost software license just
make for a mess in the OSS world.  I understand why they believe they
needed to do it.  I completely disagree that it was necessary.


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.