On Wed, Sep 9, 2009 at 6:07 PM, dmg <dmg@uvic.ca> wrote:
I don't think it is a good idea. libpano is independent of hugin. but
I agree that nobody outside hugin
cares for libpano.

I guess many of us feel that dropping PanoTools-devel is tantamount to saying that libpano no longer has a valid role as a shared platform.  Is that true, or not true?

Though I find it sad, it does seem to me that libpano, having spawned some of the best stitchers around, is nearing the end of its life cycle.  Its progeny are diverging pretty fast, and for the most part do not add new technology via libpano.  However it is still true that PTAssembler, Hugin and PTGui (and perhaps others I don't know about) interoperate with PanoTools to some degree, and all use libpano for at least some things.  So it may be too early to say that libpano has become purely a Hugin library. 

However it does look as though the most likely future for libpano is as a kind of parts bin for Hugin developers.  Some of its better parts can undoubtedly be recycled, along with more useful replacements for some of its not better but nevertheless necessary parts (hmm, I'm starting to sound German...).  And a lot of it can just be scrapped.

IMHO the main thing wrong with libpano is its rather primitive parameter and scripting facilities, and the relative difficulty of improving them within the C language context -- at least by programmers accustomed to C++.  Not that there is any law against adding C++ modules to libpano, but basically it would need a major redesign and reimplementation to turn it into a platform for the best stitchers of the 20-teens.  And it is unlikely the authors of those stitchers would see any advantage in creating a shared platform.

Regards, Tom



On Tue, Sep 8, 2009 at 7:48 PM, Yuval Levy (SF) <sf05@sfina.com> wrote:
> Hi all,
>
> since most (if not all) people here are also on
> <http://groups.google.com/group/hugin-ptx/> and the people there seems
> inclined to accept more dev talk...
>
> and since this list has only the occasional message...
>
> may I suggest that we close/archive this list and continue talking
> panotools development on hugin-ptx?
>
> the advantages:
> - one mailing list less to follow
> - an improved infrastructure
> - more critical mass = more potential contributors
>
> Yuv
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> PanoTools-devel mailing list
> PanoTools-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/panotools-devel
>
>



--
--dmg

---
Daniel M. German
http://turingmachine.org

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
PanoTools-devel mailing list
PanoTools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/panotools-devel