Dear App members,
I arrive late in this discussion just the time for me to catch up with my mail. I find that there are very interesting technical points but in fact the first big question is "do we have to improve Plastic or redefine a new standard ?". If we have a consensus on that, we will progress. Otherwise...
My point of view:
I have to say that I'm a little bit afraid about the consequences of
redefining now a new messaging protocol from scratch. Plastic has
convinced quickly a number of VO developers for adopting it. I would
re-use Doug's arguments for Cone Search standard promotion: Plastic has
been in use for 2 years, with dozen of implementations ;-) I'm pretty
sure than a redefinition instead of an adaptation will refrain
developers for going further for a while (at least 1 year and certainly
more if the consensus is too difficult to reach). I'm not sure that the
benefit of this new standard for IVOA will be higher than the current
Plastic one.
I would really prefer to improve current plastic definition to a first official release 1.0 (trying to incorporate very interesting things such as mtype and ID-ref and removing the java dependancy on the hub).
For low level application interactions that seems to be not possible to implement in Plastic, I wonder wether the work on workflow leaded by the GWS IVOA working group will not be more adapted.
Regards
Pierre Fernique
Received on 2007-04-17Z10:28:58