VODataService: simple web resource service

[Ray Plante]

Through what I suspect was a typo (or "clicko"?), Guy's response to this 
thread wound up on the gws list.  I have appended it below along with my 
response.

I'm happy with Guy's interpretation using ParamHTTP to describe simple 
static URLs for accessing data, so there is no need to pursue my 
suggestion for adding the "WebResource" interface type.  So unless someone 
wants to resurrect the idea, I will drop it.

cheers,
Ray


---------- Forwarded message ----------
Date: Tue, 27 Oct 2009 09:55:10 +0000
From: Guy Rixon <gtr at ast.cam.ac.uk>
To: Ray Plante <rplante at ncsa.uiuc.edu>
Cc: grid at ivoa.net
Subject: Re: VOSI: capabilities

Ray,

is there any strong reason why this should not work:

<capability standardID="ivo://org.astrogrid/std/VOSI/v0.3#availability">
     <interface xsi:type="vs:ParamHTTP">
        <accessURL use="full">http://casu.ast.cam.ac.uk/ag/iphas-dsa/IDR/vosi/availability
</accessURL>
        <queryType>GET</queryType>
        <resultType>application/xml</resultType>
     </interface>
</capability>

However, I would be happy with a "web resource" interface-type in
VoDataService - that would be useful.

Regards,
Guy

---------- Forwarded message ----------
Date: Tue, 27 Oct 2009 08:04:28 -0500 (CDT)
From: Ray Plante <rplante at ncsa.uiuc.edu>
To: Guy Rixon <gtr at ast.cam.ac.uk>
Cc: grid at ivoa.net
Subject: Re: VOSI: capabilities

Hi Guy,

On Tue, 27 Oct 2009, Guy Rixon wrote:
> is there any strong reason why [ParamHTTP] should not work:

I think I'm on the fence.  I was thinking about the definition:

   It describes a service interface that is invoke over HTTP via a GET or a
   POST [HTTP] in which the inputs are parameters encoded as name=value
   pairs, delimited by ampersands (&) and URL-encoded [URI]. When the
   service is invoked as a GET service, this input list is appended to a
   base URL (where the result must form a legal URL).

I guess I was thinking that since VOSI does not require parameters, this 
definition doesn't fit.  However, you example illustrates a good argument 
for the VOSI interface simply being a degenerate form of ParamHTTP.

I actually would prefer *not* to introduce my suggested change at this 
stage of the game, so if we are happy about with your interpretation of 
the definition, let's not change things.  Any other opinions?

thanks!
Ray

> <capability standardID="ivo://org.astrogrid/std/VOSI/v0.3#availability">
>  <interface xsi:type="vs:ParamHTTP">
>     <accessURL
> 
use="full">http://casu.ast.cam.ac.uk/ag/iphas-dsa/IDR/vosi/availability</acce$
>     <queryType>GET</queryType>
>     <resultType>application/xml</resultType>
>  </interface>
> </capability>