[TAP] sync vs async
Paul Harrison
paul.harrison at manchester.ac.uk
Thu Mar 5 09:56:13 PST 2009
On 2009-03 -05, at 15:18, Roy Williams wrote:
>
>
> Paul Harrison wrote:
>> As for whether the asynchronous mode should be mandatory or not - I
>> say yes
>
>
> Paul
> I have been wondering about this word "mandatory". Is it semantic
> matter, as in "you are not a compliant TAP service"? Or, by
> contrast, are you saying that if a publisher puts up a Synchronous
> service with valuable data behind it, then the IVOA should refuse to
> register it or connect tools and services to it?
> Roy
Hi Roy,
I was one of the people who wanted TAP to be a "pure" advanced
service, by which I mean async/query language only - with the
synchronous parameter model based service being a "cone search" 2.0,
so that there is a clear semantic/use case difference between the
services. As I have said before on this list, having a TAP service
with various optional parts makes life much more difficult for the
client - first they have to find the service and then they have to
find out what kind of TAP service it is - that inhibits
interoperability and makes it more difficult for the client to make VO-
wide queries on many services.
As far as someone putting up a non-compliant service with valuable
data behind it - of course people will use it if the data are good,
but that his the situation we have today, and the VO will have made no
difference.
Paul.
JBCA, Manchester University
http://www.manchester.ac.uk/jodrellbank
More information about the dal
mailing list