Future development of GDE

Steven Smith smith at phylo.life.uiuc.edu
Thu Jun 3 18:22:17 EST 1993


Notice to all GDE users:

After a brief break, I am back to doing development on GDE.  I hope to
get version 2.2 out shortly, and then begin working toward 3.0.  I would
like to start soliciting ideas about the 3.0 version.  The one key issue
to be addressed is whether or not it should be moved from XView (openlook)
to Motif.  The reason for this discussion comes from Sun's official adoption
of Motif as part of the COSE agreement with IBM, HP, and USL.  The benefits
of such a move would be portability, and long term(?) compatability to other
software running under X.  The down side would be that this would take a
tremendous amount of effort (read time) when no new functionality would be
added.  It might even be the case that some functionality would be lost
at least for the immediate future.  Sun has insured support of XView for the
forseeable future, and is taking steps to simplify it's operation under mwm.

The second issue to be raised is that of joint development.  Thanks to the
efforts of Tim Littlejohn at UMontreal, now seems like the perfect time to
solicit assistance from the community in implementing some of the new 
features that people would like added to GDE.  There is an ongoing discussion
on the MegaGopher at U. Montreal about features that people would like to
see added to GDE.  I am aware of many places that have already made additions
through the .GDEmenus file, and external functions.  I also have submissions
for modifications to the core editor that are folded in for the 2.2 release.
Tim and I would like to propose a standard method of submitting additions
to GDE to the mega-gopher for eventual inclusion in the released version.

External functions (in whatever language) would be submitted in source
form, along with a Makefile for compilation, a GDEmenus addition file, and
a help file.  In this manner, a directory of additions can be maintained
which people can pick and choose as to whether or not they want a given
addition in their own installation.  As time progresses, additions would
migrate into the release.  A perfect example of such a submission would be
Don Gilbert's latest version of ReadSeq.

Source code additions to the core tools should be submitted as diff files
from the original, latest source code tree.  This would involve a bit more
manipulation to work into the local versions.  After a sufficient period
of testing, the new additions would be folded into the released version.

By doing things this way, people can benefit from developments as fast as
possible.  Yet there will be time for testing to insure reliability of
additions, and development would not be under the oppressive control of
one flakey software engineer (namely me).

Comments are always appreciated.  Email can be send to me at 
smith at nucleus.harvard.edu (dont know how for how long), or
smiths at megasun.bch.umontreal.ca.  Please Cc: all suggestions to
gde at megasun.bch.umontreal.ca.

I would like to thank everyone for their support and suggestions, and
I hope that we can all help GDE to grow in the necessary directions
for the future.

Sincerely,
Steven Smith




More information about the Bio-soft mailing list