philosophical question about VOEvent

[Roy Williams]

I am running into a difficulty with the VOEvent model of reporting a 
transient. The problem is that the schema is ostensibly divided into 
"philosophical" sections: Who, What, Wherewhen, Why etc, but there are 
also "structural" sections, where we ask who is making the data model. 
For example, the Wherewhen section has a data model that the IVOA has 
defined, to make positions and time quite rigorous; but the What section 
has a data model defined by the author of the event, who invents and 
defines parameters that have meaning in relation to a specific 
instrument or project. From a structural point of view, it seems that 
What is where author-defined data model goes, and the other sections 
have an IVOA-defined data model.

So here is the question. Suppose events from a certain stream come with 
classification, and the author of those events wishes to express the 
quality of the classification as some sort of goodness-of-fit statistic 
on the classification, let's call it X. On the "philosophical" 
interpretation of VOEvent, X should be part of the Why section because 
it is "hypothesized underlying cause or effect" (quoting VOEvent spec), 
but unfortuantely the Why section has IVOA-defined syntax, and so there 
is no place in its data model for X (it only allows a percentage measure 
of fit). Structurally, X goes most easily into the What section, because 
the author can describe and define its meaning there.

Should we try to put X into the Why section of the VOEvent (how?), or 
simply put it in the What section, claiming that X is a measurement?

Your opinions welcome!
Roy

-- 

California Institute of Technology
626 395 3670