Are different versions of the same event unique?

[Roy Williams]

On 3/29/12 9:58 AM, Matthew Graham wrote:
>  Notifications of the
> same astronomical event need to be sent out in both VOEvent 1.1 and
> VOEvent 2.0 forms
> should these have the same identifier or different ones?

(1) For the Skyalert project, the answer is emphatically that we want 
the SAME identifier. We have no interest in writing special code to 
distinguish the exciting coincidences (eg SWIFT and Icecube events at 
the same time) from the procedural coincidences (same event, different 
XML).

(2) Another way to do it is to declare the streams different because of 
the schema change. The IVORNs would be:
ivo://nasa.gsfc.gcn/SWIFT_v1.1#FOM_Obs_34073224
ivo://nasa.gsfc.gcn/SWIFT_v2.0#FOM_Obs_34073224
If there is a solid argument to make the local IVORNs different, I will 
want an explanation for why the stream IVORNs can remain the same.

(3) Once we answer this question of XML representations, we can ask 
about other representations, such as JSON. In Skyalert we have JSON 
outputs: should we then have:
ivo://nasa.gsfc.gcn/SWIFT#FOM_Obs_34073224-501  -- XML version 1.1
ivo://nasa.gsfc.gcn/SWIFT#FOM_Obs_34073224-502  -- XML version 2
ivo://nasa.gsfc.gcn/SWIFT#FOM_Obs_34073224-503  -- JSON version

(4) Next we have the question of *real* changes to the event. For 
example, supporting the CRTS iPhone app uses VOEvents whose content is 
cached at lsst.org, so for example the original event has this:
<Param name="img" value="http://...caltech.edu..."/>
but the cached event has a different URL (but the same content):
<Param name="img" value="http://...lsst.org..."/>

Roy

---
Caltech LIGO
roy at caltech.edu
626 395 3670