[Pipet Users] Re: the future of piper (was: semantics, security, ...)

[Luca.Toldo at merck.de]

Dear Jeff,
thankyou so much for your so prompt and detailed reply to my message.
Compliments for your current endorsement in so many activities and good
luck with your Ph.D.
I achieved mine almost 10 years ago and I know how much resource intensive
is to achieve that task.
Good luck with it and take care of your health as it can be very demanding
(also from a psychological point of view).

Thankyou for the hyperlinks you kindly provided (SBW, ISYS, OpenBSA ) and
as well with your opinions on the matter.

>The funny thing is, Bioinformatics.Org had a conference in Arizona, and
almost
>every talk had something to do with "integrating disparate computing
>resources."  I guess this is no longer such a novel idea, probably more
>*timely* than anything.

Well, I certainly agree with you with the fact that "integrating disparate
computing resources"
is not a novel idea. It is a need that the computing community envisaged
since long time before
the term bio was associated with computers or informatics. As you can
notice, the problem has
not yet been satisfactorily resolved yet.

>Personally, I am seriously considering taking what I have written for the
UIL
>(the GUI of Piper) and integrating it with FreeMol, possibly, and having
it
>limited to a single website (integrating tools at one location and
providing a
>Web interface for doing that--I'd have to switch the GNOME canvas for
>who-knows, maybe imagemaps), e.g., at Bioinformatics.Org.

I think is certainly a good thing to get things to a stage were you can
make use of them.
For academics is also important to publish them. When I developed the Java
based Molecular Biology
Workbench in 1996 (http://www.embl-heidelberg.de/JaMBW/) I did not have in
mind large projects or
goals but simply to facilitate the users's daily research life with simple
utilities. The system I
built (and then published in 1997) still runs, and people are using it
thousands of times from allover
the world. Since then, I switched to entirely new fields...

>But, for a distributed system using a traditional GUI, I'll throw my
support
>(whatever it's worth) behind SBW.

I think that "strong believe" is certainly useful to motivate somebody.
However that could also be "risky" as we have all learned one
way or another. I have not yet used SBW although I had a look at it and
seems to me as an extremely early release project. Furthermore,
there are several design flaws in SBW which makes it in my eyes not as
appealing as it might be:
- socket connections
- semantics

The SBW seems to me geared to "within firewall" or "no firewall"
environments...

>My interests in developing an *Internet distributed* system have waned, as
I think leading such a
>project requires a background in CS.

Any time wishes/visions meet reality a clash occurs. This clash shall be
managed in constructive way and it seems to me that you are doing
very fine. However, you might not be aware that the MyGrid
(http://www.mygrid.org.uk/ ) has exactely that goal in mind, it is forged
by CS people
together with biologists and chemists with fun in programming, and has
large founds provided by the ESPRC. You might want to
have a look there for further options...

Luca