On Apr 16, 11:57 am, Sergey Bochkanov <[email protected]>
wrote:
> Hello, Bill.
>
> You wrote 16 апреля 2010 г., 12:23:43:
>
> > What you suggest is to effectively maintain two versions of MPIR, one
> > version 2.1 the other 3.
>
> No, quite the opposite. I am talking about plugin-based framework
> where framework itself is 2.1, and some plugins are 3.0, some are 2.1
> too. Here 'plugin' means either implementation of mpn-function or one
> of the higher layer functions.
Hi Sergey,
Unfortunately, its only notionally modular and, is structured
internally rather like a complex wiring loom with a mass of inetrnal
connections and no circuit diagram.
There is, in my view, no practical way of maintaining a v2+ and a v3+
library from a single code base unless we put a truly massive effort
into restructuring the code
> I don't know MPIR's internals good enough, but I thought that library
> is modular: i.e. you can change mpn internals without affecting higher
> layers and you can change implementation of some function without
> affecting another functions.
>
> > We did consider such an option, but it is much harder than it
> > seems, and we simply don't have sufficiently many contributors to
> > manage that,
>
> Could you point me out one or two of the difficulties? May be there is
> a way to workaround them...
>
> > there seems to be a lot of interest in contributing to a BSD
> > licensed library. Another individual I asked about this said they
> > thought it was a great idea and that one of the main things that
> > puts them off currently is the LGPL.
>
> What is wrong with LGPL as long as it is 2.1? Currently I see only one
> drawback - it can't be used in a BSD open source project. However, I
> know of no BSD open source project which may be interested in multiple
> precision.
In my view the main problem with the LGPL is that it comes from an
organisation - the FSF - that seeks to suggest that there is something
ethically wrong with commercial closed source software.
The LGPL is not even liked that much by the FSF because it allows such
use and they hence constantly push people towards a GPL license.
Which is fine if you are in this camp.
But, as Bill has said, this makes commercial users uneasy. In any
event I dislike the license because I am in a completely different
group in that I _want_ my published code to be used commercially and I
want a license that does not only grudgingly allow such use but
encourages this by making it explicit that this is allowed.
And this makes BSD much better from my point of view than LGPL.
best regards,
Brian Gladman
--
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en.
