[ptx] On the future of libpano and panotools

[dmg]

As I think about the future of panotools I am trying to put together a
"one year plan" that justifies the split. This, in my opinion, are the
goals that can be achieved in the next 12 months that will make
libpano and panotools a better piece of software:

----------------------------------------------------------------------
* Clean up libpano

1.  Libpano should export _only_ functions that are relevant to
  panorama making. We need not to export any function that is not for
  this purpose. For instance, without the baggage of the old
  binaries we should not export myfree, mymalloc, GetFullPath, etc.

2.   Rename the functions with a prefix "pano_" just like any other
     library does (see libtiff, libjpeg, for example).

3.   Document the libraries. We have no documentation of the
     libraries. We need to create, at the very least, man pages.

4.   Perhaps create a libpsd library for the PSD format. We have a lot
     of code that supports it and it should be made available to other
     projects who might be interested in using them.

* Documentation

1. We need to document each of the programs that we are planning to
   support. One of the things that we truly need to document is the
   script format, particularly as we keep improving it and forking
   different versions (think the hugin .pto format and how
   different/similar it is to PTstitcher, and then to PTmender's).

* PTMender

1. Implement the following missing features of PTstitcher:

   - Feathering
   - Colour correction, Brightness correction (not currently
     implemented) and potentially implement it for 16 bit images.

2. Add support for EXIF and colour profiles

* Test suite

1. Further improve a test suite. We need at the very least have a test
   that verifies that PTmender does what it is supposed to do. Without
   a reliable test suite development is painful.

2. No new features should be added to panotools unless they include a
   test suite (and update documentation)

* Move to SVN

1. I think we should move to subversion. It is the perfect time to do
   it.

----------------------------------------------------------------------

These tasks will probably take time and effort to succeed, and we will
need to do them as a team.  But I think that at the other end we will
finish with a better product, and it will be fun.

daniel



--
Daniel M. German                  "In other studies you go as far as 
                                   others have gone before you,
    ->                             and"
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with @ and (dot) with .