Karl Rupp <[email protected]> writes:
> Cool! For packaging, we should either integrate it completely (i.e. make 
> it part of the ViennaCL namespace), or keep it entirely separate, 
> similar to how e.g. amgcl handles things. Packaging the two together 
> would be a nightmare for packagers (Toby, please correct me if I'm wrong).

Yeah, you're right: distribution packagers prefer to keep things at a
sensible granularity; to avoid dependency hell and a security nightmare,
distributions prefer to 'unbundle' bundled software, so that separate
pieces can be upgraded or patched as needed and independently.

> Yep, that's exactly why I've spent quite a bunch of effort on the 
> 'infrastructure' rather than adding more and more implementations of 
> algorithms and preconditioners recently. Having a full-fledged API for 
> all three backends now makes algorithm development a lot more fun :-)

To me, the ViennaCL infrastructure is one of its most exciting
properties :) And, as an aside, I'd be generally in favour of
integrating the nonlinear minimisation library into ViennaCL (or at
least into pyViennaCL), since I'm starting a course on nonlinear
optimisation in October, and so Philippe's work is quite interesting to
me.


Best,

Toby


------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
_______________________________________________
ViennaCL-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/viennacl-devel

Reply via email to