Subject: Re: [boost] Interest in a Boost.Chrono/Date library
From: Philip Bennefall (philip_at_[hidden])
Date: 2015-11-02 19:17:19
On 11/2/2015 11:46 PM, Ion GaztaÃ±aga wrote:
> On 02/11/2015 1:06, Howard Hinnant wrote:
>> I did not intend for there to be a âbinary clause" for this date/time
>> library. My intent was to make this library have as small a legal
>> footprint as possible.
> Thanks for the details Howard. According to
> it seems to me that MIT License should be acceptable in Boostt... Snip
Personally, even if individual authors who place their code under the
MIT license do or do not intend it to require binary attribution, I have
seen plenty of MIT licensed project authors who do interpret it as
requiring attribution as part of derivative works in binary form - hence
the ambiguity. I come back to this section again"
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
It does not make any distinction between the software as distributed in
source code form, precompiled unmodified form, a derivative work in
source code form , a derivative work in object code form etc. To me it
is obvious that the MIT license does require the copyright notices to
appear in source code copies of the software (whether in its original
form or as part of a derivative work), for the simple reason that this
is the way in which the original software is distributed (at least in
the case of a Boost library). This is of course what the BSL says too.
But it seems to me that the case of attribution in closed source
derivative works is very much a matter of opinion, depending on how you
interpret the words "copies" and "substantial portions of the software".
What constitutes a substantial portion? Am I still distributing "the
software" if I use a library as part of a closed source derivative work?
I should think so, in which case I would assume that the attribution
clause does apply. But I am not sure. These are the questions which I
feel the Boost license answers clearly, and which is where I think the
two licenses are incompatible.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk