Re: Source code for Gblocks available?
Hi Mark,
[sorry for violating netiquette and answering in public but
I regard your mail on one hand not private and on the other
hand interesting for the Debian Med team.]
On Thu, Mar 06, 2014 at 08:18:33AM -0800, Mark Bartelt wrote:
> Hello ... I was wondering whether you ever received a reply
> to the question you asked in the posting at ...
>
> https://lists.debian.org/debian-med/2011/02/msg00008.html
The short answer is: No. (see below for the long answer)
> The reason I ask is that somebody who recently got an account
> on one of our systems asked us to install a number of packages
> that she needs to use, and Gblocks was one of them.
Please do us (=us in terms of Debian+Ubuntu users including you) a
favour: If you find packages interesting for your colleague which are
not yet on our list of packages
http://blends.debian.org/med/tasks/bio
and thus should be packaged, tell us about it (optionally you can
help us packaging it under kind guidance of our team).
> Unfortunately, the pre-built executable on the Gblocks download
> site is dynamically linked, and it wants a different version of
> glibc from the one that's on the system where Gblocks will get
> used.
>
> We could of course maintain two separate versions of glibc, but
> I'd just as soon not be bothered with that, if we can obtain the
> source code for Gblocks and build our own copy.
>
> I sent a note to Jose Castresana inquiring about this, but thus
> far I haven't heard back. So I'm curious as to whether you had
> better luck.
There are three four basic types of upstream response where we have
different examples for:
1. Real enthusiasm and own commitment (you might like to seek
the list I kept in CC) for the keywords 'blat' and 'phyutility'
to just name two recent examples.
2. Support in case of questions (technical example spades and
licensing example PHYLIP - both can also be found in recent
discussions her on the list).
3. No response because the authors do evaluate the importance
of packaging their software less important than we do.
4. No response because the author does not control the given
e-mail address any more.
It is pure spekulation which type we might see in the Gblocks case. I
personally think that you have given quite valid reasons in your mail
that publishing the source (and thus answering our request) makes sense
but it might be the author has moved on and is not interested any more.
If the author is not really reachable via e-mail (4.) this is quite a
nasty situation and sometimes I spended literally days to detect the
authors of some piece of code to ask for the license and some cases are
open for years which is the most boring and frustrating work in our
field. So if somebody might evaluate my technical skills high enough to
make sure I can spend my spare time fully on this front and wants to
volunteer to care for those connections to upstream to take this burden
from my shoulders this would be *really* welcome help. If you want a
specific example (besides Gblocks) you might like to follow the colt
licensing issue[1] which is affecting BEAST to be non-free for a *very*
stupid reason and in turn makes so boring technical trouble that I have
given up upgrading this software to the latest status.
In short: Please help us contacting upstream which does not require any
technical skills and we do the technical work for you.
> Thanks much ...
Thanks for your interest in Debian Med
Andreas.
> ---------------
>
> Mark Bartelt
> Center for Advanced Computing Research
> California Institute of Technology
> Pasadena, California 91125
>
> 626 395 2522
> 626 584 5917 fax
> 626 628 3994 e-fax
>
> mark@cacr.caltech.edu
>
> http://www.cacr.caltech.edu/~mark
[1] https://lists.alioth.debian.org/pipermail/debian-med-packaging/2014-February/025141.html
--
http://fam-tille.de
Reply to: