[glue-wg] Adopting a GLUE2 XML Schema?

[JP Navarro]

Greetings Laurence,

I would like to see us in the TeraGrid contribute to the development
of the official OGF GLUE2 XML rendering (or renderings).  Please let
Warren and I know what we can do in that respect.

Thanks,

JP
On Oct 2, 2009, at 9:29 AM, Laurence Field wrote:

> I would like to second Stevens comment. Those that need XML should try
> to define an agreed rendering.  We should probably start with the
> Teragrid one as a baseline and go from there. If anyone thinks that
> there are any issues that need discussing we can always collect them  
> and
> have a phone conference to sort them out. For the LDAP rendering it  
> only
> took a could of calls. Is there anyone who is motivated enough and has
> enough time to lead this effort?
>
> Laurence
>
>
> stephen.burke at stfc.ac.uk wrote:
>> glue-wg-bounces at ogf.org
>>
>>> [mailto:glue-wg-bounces at ogf.org] On Behalf Of Warren Smith said:
>>> Of course, the TeraGrid GLUE 2 XML Schema isn't a standard
>>> and I can't say how similar it will be to an official OGF GLUE 2 XML
>>>
>> Schema.
>>
>> Well, this is supposed to be a community effort so it should be  
>> possible
>> for you to contribute to the definition of the official XML rendering
>> ... however personally I'm not involved in that so I can't comment
>> directly.
>>
>>
>>> In
>>> fact, one warning I'd give about the TeraGrid schema is that
>>> we chose a
>>> flat hierarchy of XML elements in that schema (e.g.
>>> ComputeActivity is a
>>> sibling of ComputeShare instead of a child of ComputeShare). The  
>>> last
>>> discussions I remember seeing on the GLUE working group list
>>> seemed to
>>> have more people favoring a deeper hierarchy...
>>>
>>
>> I think Nordugrid favoured that. However, for the LDAP rendering we
>> tried to keep the abstract schema properties as much as possible -  
>> LDAP
>> obviously forces you to have a tree, but the objects are all linked  
>> via
>> foreign keys and unqiue IDs so you don't need the tree to navigate.  
>> In
>> that case you could potentially restructure the tree without changing
>> any queries. If XML follows the same kind of route you could even do
>> joint queries across representations.
>>
>> Stephen
>>
>
> _______________________________________________
> glue-wg mailing list
> glue-wg at ogf.org
> http://www.ogf.org/mailman/listinfo/glue-wg