VOConcepts paper

[George Djorgovski]

This is a good initiative.  Let me offer a few random thoughts:

It may be useful to have a group of astronomers with a broad range of  
expertise (perhaps the NVO SSC? augmented slightly?) compile and  
review a good list of terms.  It has to be balanced; you cannot have  
313 terms for stars of various sorts and 3 terms for all AGN, just to  
make up a bogus example.  Any given expert would be biased towards  
the semantic richness and distinctions in their own field, just as  
the proverbial eskimos' 65 words for snow.

There has to be an easy allowance for synonyms, e.g., QSO = quasar.   
There has to be a flexible mechanism for handling classes and  
subclasses, e.g., even though QSO is only one subclass of AGN, it  
should be possible to say QSO.... and not have to say AGN.QSO....

In the time domain, a characterization of an event (and BTW, is an  
event a process or an object? it could be either or both) will evolve  
as the data come in, e.g., transient -> SN -> SNIa -> SNIa 2 days  
before the maximum, etc.  It can also be revised as the better data  
come in (as in "oops, it was really a SNIIpec"), so there has to be  
an easy revisionism and history mechanism.  And what about the events  
where 2 groups give different characterizations (for the same event)?

Cheers, George