DAL/TAP Response

[Keith Noddle]

Hi Ray,

> Thanks for the nudge.  And just to show I'm not asleep, I'll say that
> I'm not sure I would agree to your framing of the controversy over TAP. 
> Consequently, I think it leads to an incomplete list of options.
I never doubted you were awake :-) And yes, I agree, the list is 
(inevitably) incomplete.

> I feel quite certain that no one is suggesting 2 or 3.  For my own part 
> and on behalf of my project, I've tried to make it quite clear that it 
> has never been our position to attempt to change the agreement to "make 
> parametrised querying mandatory".
Understood - and I hope in turn I have not given the impression that was 
not the case. I have said many times and I'm happy for history to judge 
me on this: all I want is a standard that meets the needs of *everyone* 
in the IVOA, that accurately reflects the priorities and requirements 
the IVOA has decided upon and that is well written and usable in the field.

> In my view, the option that was left out was the one that was in progress:
> (0) To create an integrated spec where ADQL is mandatory and 
> parameterized query is optional.  This was the basic agreement from the 
> IVOA meetings, yes?
> 
> We know WG discussions regularly slow down, and I think in this case of 
> TAP, there's probably a bit of "let's wait and see what happens" going 
> on, and apart from three hearty souls, we may be wondering where to 
> begin.  I would suggest we develop a strawman list of outstanding 
> unresolved and/or controversial issues.  As we write them down, some 
> obvious alternatives could be offered that people can respond to.  I'd 
> be happy to contribute to that list.
The purpose in separating the two elements is simply to expedite the 
finalisation of the part of TAP we are already agreed upon. With TAP 
V1.0 entered into the approval process, I don't see why we cannot set to 
with vigour upon on the remainder and have a TAP V1.1 hot on its heels.

It may well be that agreement on the TAP/Param elements is closer than 
perhaps it appears, but equally, I don't see the impediment to progress 
by separating things this way. If we do stumble finalising TAP V1.1, at 
least we will have something of substance to show at the next Interop; 
if we are successful, we will have completely discharged our obligations.

Having an agreed TAP V1.0 is the minimum required to free up those 
wishing to implement a TAP service but who are waiting upon the spec 
before starting (and there are many). The sooner that starts the sooner 
there are services Chenzhou, for example, can deploy.

Keith.

-- 
Keith Noddle                    Phone:  +44 (0)116 223 1894
AstroGrid Project manager       Fax:    +44 (0)116 252 3311
Dept of Physics & Astronomy     Mobile: +44 (0)7721 926 461
University of Leicester         Skype:  keithnoddle
Leicester                       Email:  ktn at star.le.ac.uk
LE1 7RH, UK                     Web:    http://www.astrogrid.org