[MOBY] Re: [MOBY-dev] use case conclusion

[Mark Wilkinson]

On Fri, 2003-01-31 at 18:17, Andrew D. Farmer wrote:

> to
> see if any of the work in this area (web services/semantic web) that others
> have done (and boy is there a lot of it!) 

Geez are you ever right about that!  I am totally humbled by how naiive
my perceptions have been, now that I am immersing myself in the
literature!  I rue the last eight wasted months of my life!

Would you be willing to compile a bibliography of the literature that
you are finding relevant to the project so that we can all read it in
time for the next meeting?  I'm a bit distracted with boning up on my
deeper understanding of SOAP (hokey Dinah what a powerful technology
this is!).  I am also trying to catch up with the other literature but
if you have already weeded through to the most relevant articles it
would really be useful.

As you know, I am still primarily focused, rightly or wrongly, on
cleaning up and implementing the MOBY spec as-is, beauty, flaws and
all.  In this regard, Carole has offered to host me for a week or so in
Manchester to have some face-time with the myGrid developers.  The most
obvious weakness in the current MOBY implementation (IMO) is our dealing
with object/service ontologies.  myGrid has made mind-blowing progress
in this area.  The first link on this page:
http://mygrid.man.ac.uk/rpapers.shtml  details their work in this
regard.  We are infants compared to them, and I want to make sure that
the first "major" MOBY release dovetails as much as possible with their
work - we would be fools not to!  I don't think we can use their
ontology engine lock-stock-and-barrel as it is considerably more complex
(and differently-focused) than the existing MOBY ontologies are, but at
a minimum I want us to be compatible with it (or somehow wrap it and use
it verbatim).  I hope that a few days sitting with the myGrid developers
will make it clear to me how this can be accomplished with minimal
disruption to either project as we follow our independent, but
concerted, efforts.

At the other end of the spectrum, Brian Gilman and I have jointly
decided that the best use of our time at the upcoming BioHackathon is to
unify the I3C's LSID and LSID-resolver with the MOBY-Triple as the base
MOBY object and semantic object meaning.  This is going to be simple in
some ways, and complicated in others - the MOBY Triple and the LSID are
almost identical in their intent, except that the Triple attaches an
instantiation tag to the global identifier.  I'm a bit worried, however,
that some things (e.g. a Blast result) can not be described with an
LSID, since LSID's are more semantically "concrete", so it will be
difficult to 100% replace the MOBY Triple with the LSID...

Anyway, there is a lot of exciting activity all round!  As Damian has
observed, I am a rapid-prototyper, and I find this approach to be the
most useful to me and my "style" of problem-solving (an experimental
bench scientist at heart!), though I admit it makes it hell for the MOBY
newcomers....  You and Damian, on the other hand, are the careful
considerers, and this is wonderful for keeping me (and the other RP's)
in line :-)  I'm very excited about the insights we are finding
together!

Even with the limited deployment we have, plus some experiments I am
doing on my local machine, I am learning a lot of unexpected things
about the behaviour of the current MOBY system that I think will be
useful as we come to make our final decisions on the technology we will
use.

This is SO much fun that it doesn't feel like work!!  8-)  ...and the
solution is so close I can taste it!

best wishes all! 

M

P.S. please excuse the rambling nature of my message - I feel like a T4
phage these days... head-full!  (bioNerd humour ;-))


-- 
=======================================================================
                                    |--==\
Mark Wilkinson                       \==-|       
Bioinformatics Consultant             \=/        0010010010100101110010
Illuminae Media                       /-\        
727 6th Ave. N.                      /-==|       0010100100111101010010
Saskatoon, SK, Canada               |==-/        
S7K 2S8                              \=/         0100100100010010010101
+1 (306) 373 3841                     /\         
markw at illuminae.com                  /=-\        1101001010100101010101
                                    |--==\
=======================================================================