ADASS discussions

[Guy Rixon]

On Tue, 23 Sep 2003, Chenzhou Cui wrote:

>
> Guy Rixon wrote:
>
> > On Mon, 22 Sep 2003, Alasdair Allan wrote:
> >
> > ...
> > Recently, I'm beginning to wonder if OGSI is worth the bother.  Here in Java
> > land, we're stuck with GT3 as the only implementation, which makes development
> > slow and difficult.  Parastatidis et al. have questioned the usefulness of
> > OGSI as a standard, proposing an alternative that matches better to web
> > services standards.  If we don't do OGSI now, I'm wondering if it will have
> > become obsolete by the time we need to exploit it.
> >
> >...
>
> I think it is suitable for a small VO project, such as China-VO, to
> focus its effort on one choice or just a few choices. We can adopt other
> technologies after they have been implemented successfully in other VO
> projects.

China-VO can do whatever works best internally, of course.  We wouldn't dream
to trying to stop you using OGSI.  However, we need to decide whether OGSI
services become _mandatory_ for any part of IVOA.  I.e., to connect a
hypothetical Xxx-VO to the global Virtual Observatory, do the Xxx-VO people
_have_ to add some OGSI services, or can they do it all with "standard" web
services?

> What's more, the flag of Grid can absorb many eyes from IT community.
> It's very useful for VO development.

I see your point, but actually I doubt if we'll get that many eyes on OGSI.
It can go two ways. Firstly, all of e-Science (outside our VO movement) picks
up on OGSI, and a part of the IT industry too.  That way, we get scrutiny by
many eyes and lots of product to use.  Secondly, only the super-computer
community continues with OGSI (because they use it to build compute grids),
other scientists don't adopt it, and the industry forgets it.  That way, we
get no proper product and no eyes on the residual code-base.

I don't think OGSI, in its current form, is going to become common and
long-lived in science.  I think that industry will side-step it and that later
science projects (e.g. biologists) will use the industry-standard techniques.
Sad, but true IMHO.

Remember, if the only eyes on OGSI are inside the VO movement, then we're
paying heavily to support aomebody else's ideas and code.

Guy Rixon 				        gtr at ast.cam.ac.uk
Institute of Astronomy   	                Tel: +44-1223-337542
Madingley Road, Cambridge, UK, CB3 0HA		Fax: +44-1223-337523