[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

ReadSeq (Was: Thoughts on the Vienna RNA package)



On Wed, 14 Nov 2007, Charles Plessy wrote:

- It uses a patched version of Readseq. Maybe we should consider to
 build Vienna on Debian's readseq, possibly by adopting its patches if
 they are not disruptive. By the way, we could incorporate the Debian
 package for readseq in our repository, if its current maintainer
 agrees.

As a former maintainer of readseq I would like to clarify the situation
from my point of view:

  1. When I took over readseq in Nov 2001 the progress of rewriting
     it in Java which leaded to readseq2 [1] was nearly finished.
     The development on readseq1 (as it is packaged in Debian) was
     stopped according to the authors even at this time.

  2. I continued maintaining this upstream orphaned software because
     it was used by some users and a precondition for an other
     package of mine (arb).

  3. If anybody has reasonable patches we are probably free to do
     whatever we want (including becoming "upstream") because it
     is in public domain.  I would regard this as quite reasonable
     but we should discuss this plan with upstream anyway (perhaps
     to place reasonable links for interested users).

  4. I handed over the package to Michael Schmitz <schmitz@debian.org>
     (Changelog entry "took over from Andreas Tille 'cuz readseq is actually
     in use at our site").  I'm a little bit unsure whether this is
     valid until today - at least the package has some lintian
     problems and IMHO definitely needs polishing.  (Michael, are
     you able and willing to work on this package?)

My suggestion is that we immediately move the packaging stuff to Debian-Med
SVN and add Michael and mayelf to Uploaders, while the maintainer should
be the Debian-Med packaging team (Michael, I hope you would not regard this
as piracy ...).

Somebody should discuss with upstream whether they prefer that we
"release new versions" of readseq1 or that we maintain a set of patches
inside Debian.  In any way if there are good reasons to patch readseq
(working nicely with another package would be such a good reason) we
should just do it inside the Debian package.

BTW, I did not packaged the Java version because of the Java issues
(5 Jears ago) and because there was no explicite need for this.  If
somebody wants to build packages for readseq2 - just go for it.

Kind regards

        Andreas.

[1] http://iubio.bio.indiana.edu/soft/molbio/readseq/java/

--
http://fam-tille.de



Reply to: