StandardInterfaces V0.1

[Wil O'Mullane]

> >>- why restrict this to web services? We should be able to agree more than
> >>one way of implementing services: web service and cgi as a minimum I guess.
> In my mind, I would be very surprise if the future of the astronomy will 
> be only WS. I think that after this fashion time, this technology will 
> take its real place : an open remote procedure call mechanism. Not more, 
> not less. And it certainly won't replace simple CGIs for which there is 
> no real client side.

Yes of course .. I took this out of SkyNode which is SOAP based and it was in the WEb/Grid Servives section. Hence the WS leaning. 

Then I have a question. 
If we have these standard interfaces for any CGI how do we relate that to the
cgi. DO we add a single parameter like SI where
SI=metadata returns metadata
and SI=HarvetLog&fromDate=1-1-2000  returns the log. 
I presume we may still asume XML is returned from these type fo calls.

Or should we assume for each CGI a seperate CGI so if i have
../cgi-bin/X
I should have
../cgi-bin/stdifX
which implenets the standard interfaces..

I rather like option 2 as it means no changes to existing code.


For SOAP services of course the document as stands would be fine. 
XSD may be worked up for the indidvidual return types...


wil


> 
> Regards,
> Pierre Fernique