[Biojava-l] The open bioinformatics foundation.

[Ewan Birney]

On Mon, 28 Aug 2000, Johann Visagie wrote:

[apologies for the cross posting]

> I also agree about BOSC - the amount of co-operatoin between the Bio*
> projects was astounding!  Which - coupled with the example of Jason's
> cross-posting - brings me to the point:
> 
> Shouldn't there be a general mailing list (moderated?) to cover issues that
> pertain to all the Bio* projects?  ("biostar"? "openbio"?)  This could cover:
> - announcements of interest to all, like Jason's
> - co-operation / code sharing between projects
> - etc.
> 
> (Maybe the bioinformatics.org people can be prevailed upon to host it, since
> they have such a nice domain? :-)

Hmmmm. I had hoped to be able to announce this once this was all tied
down, but I guess sometimes people think ahead of you... ;)


Due to the amount of money that we have to process through BOSC and the
existance corporate sponsors the whole thing, we are going to have
something like

"The Open Bioinformatics Foundation"

which will oversee essentially the finances to the bio* projects. The aim
is to have a board of people to manage this (so it stops being just "chris
and ewan seeing what they can do"). The board will be drawn from people
both from the bio* projects and some vendors and big-pharma people. I
think it will be 6 or 7 in size. At the moment I am putting together the
board by asking people, but trying to be inclusive. I think voting like
debian is way too heavy handed for this. We are not that big a
community... 


The Foundation's accounts will be managed (as long as everything goes ok)
by ICSB which is the umbrella organisation to ISMB and RECOMB, meaning
that we don't have to worry (hopefully) about legal stuff.

This is still somewhat fluid - or at least semi-slushy. If there are
people who have strong views about how this should turn out, drop me a
note. I want to be inclusive here without having a 30 man committee
arguing about web site design.


We have a domain name for this, which is open-bio.org, which has the
additional benefit that it can grow beyond bioinformatics if need be.


Re: bioinformatics.org and sourceforge

I think the projects cvs and bug tracking for all the bio* projects is
best handled through something like source forge. I'm agnostic as to
whether this happens at bioinformatics.org or sourceforge itself. This is
one for the project leaders to decide on.

The bio* server then would focus on web site, with a strong wiki site and
mailing lists, not least because we have them set up.


I storngly hope that bioinformatics.org will be considered as part of the
same community as the bio* projects... which means I guess
bioinformatics.org should be one of the sister sites listed on the bio*
home pages. 


Re: central "coordinating" mailing list

I think the current approach of crossing posting when something "generic"
comes up is the best way.

A central coordinating mailing list suggests, in my view, either

	a) a highly controlling centralised structure, with the "in-crowd"
make broad sweeping decisions. Bad.

	b) a minor mailing list that noone subscribes to because there is
only one post per month and that is "what does this list do?"


Looking at Perl, they have just split the p5 porters list out into sub
lists, each list *having a limited time to run* to prevent (a) happening.
It is an interesting idea.


Currently I think that all the bio projects leaders, being myself -
bioperl, matt - biojava, jeff (andrew) - biopython, brad (bioxml) and
jason (biocorba) I think are subscribed to the other lists. In
addition there are a large number of peole who are cross subscribed. This
means that at least this group of people keeps in touch with each of the
projects, knowing where each of the projects are.

It terms of software coordination of the projects, these should be focused
through bioxml/biocorba which is the right place for this.


I don't want to sound negative about coordination, but I don't want
"coordination" to become a way of having a strong top-down approach to the
bio projects.

Of course, a single mailing list would prevent me fro getting 5 copies of
this mail, but then again, would this mail reach all the people it should
do. 

If you think a single mailing list is a good idea, argue me into it ;)


ewan


-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<birney@ebi.ac.uk>. 
-----------------------------------------------------------------