Bioperl: Joining up...

[Gunther Birznieks]

On Mon, 25 Jan 1999, Ewan Birney wrote:

[...]
> 
> The main people at bioperl are not computer scientists but self taught
> biologists who learnt perl and programming generally because they had
> problems that they wanted to solve. Some of us have gotten deeper into the
> CS side (to the point of being indistinguishable from them?), but all of
> us are here because we use Perl to solve biological problems and that is
> still our aim.
> 
> We don't mind 'bad' code -
> 
> Some people are very embarrassed about showing their code to others for
> fear of being told that there is a better way of doing it. It is fine
> to have kludges and make hacks - at least if they work - and bioperl has
> a number of wince-making or obtuse bits of code. Eventually someone takes
> them out and (hopefully) fixes it. Nearly all of us have made some real
> howlers in our time (I know I have) so we've all been there.
> 
I have to echo all those sentiments.  

We have a similar network of people developing CGI application-based
modules at Extropia. And EXACTLY the same fears are expressed by people
who learned Perl from the Web Design/Graphics side who think they aren't
up to par with coding "for the public".  We just posted a similar type of
message to people interested in developing for the CGI Application module
library we've been developing.

Anyway, I would have to add that the cleanliness of the code is not the
most important thing. What is important is that people don't have to
duplicate work all over the place. The best way to do that is submit code
as-is that WORKS asap-- its definately worlds better than nothing.  

Plus the likelihood is that someone may take your code and improve it to
give back to you.  And that's a lot better than keeping the code closed
just because it doesn't follow variable naming conventions or use
strict/warnings.

Also, even if the code is not directly useable by me, I find that looking
at what other people are doing helps me figure out the best way to write a
module for myself because then I can keep other's peoples flexibility
/needs in mind as I write code. If other people are not giving out their
code (even if it is not that clean), then when someone else does end up
writing a "clean" module to do a similar function, they may leave out a
critical piece of functionality that your code was doing. By giving out
the code, no matter how dirty, you provide at least some collaborative
communication.

> We aren't a little tight group > 
> We honestly don't mind who does what - the people who have done the
> backbone of the work has changed over time and will continue to change.
> you can quite easily (if you wanted to) become one of the key members
> of bioperl by contributing alot. I wasn't a founder member at all
> for example.
> 
I have lurked on this list for a long time and although there are core
people who have existed from the beginning, I would tend to agree that it
definately seems like people who post and contribute end up being part of
the group through virtue of just doing/posting a lot.

I would definately encourage participation. Just do it.  Even the less
clean code will be a good basis for discussion and other module projects
since it shows what real scientists/programmers are doing.

I am no longer involved with bioinformatics, but I find that
this is one of the few medium-scale object libraries for Perl that has had
an incredible amount of thought put into the infrastructure so that others
can build upon it easily. Thus, I follow this list with great interest to
see what ideas I can steal for building a CGI applications library in a
similar spirit to bioperl.  I love to see participation in bioperl by
novice and expert alike since it means positive things for what I am doing
as well.

[...]

Thanks for BioPerl,
              Gunther


=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc:
http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
====================================================================