VOEvent II summary, part 1

[Rob Seaman]

Hola,

I've been under the weather the last few days and writing this has  
been moving like molasses.  It seems best to send what I have at the  
moment, mostly Monday and Tuesday and send a separate message  
discussing Wednesday's mini-interop in the next couple of days.   
Don't wait for me, however.

Bottom line:  great workshop!  Thanks to everybody for making it so.   
Credit for the excellent choice of food, etc., goes to Kathy  
Glockner.  Thanks to NVO, LSST and NOAO for paying for it.  Pete  
Marenfeld of the NOAO photo shop made the cool poster.  VOEvent  
remains one of the VO's hot tickets.  Current thinking is to hold  
VOEvent III in June, with the venue shifting a few hundred miles  
further east.

Capsule comments follow on the speakers.  Presentations and links  
(and the workshop photo) are available from the twiki:

     http://www.ivoa.net/twiki/bin/view/IVOA/VoeventWorkshop2

Roy Williams opened the workshop with comments and intros.  The basic  
status is that a dozen more people would have attended except for  
various conflicts.  Several groups are pursuing active development  
centered on VOEvent.  These include not only traditional VO partners,  
but traditional "real" observatories like UKIRT/JCMT, Gemini and CTIO.

Managed to deliver my slides on the state of the specification in  
about 20 minutes.  Took me an hour and a half the week before at a  
staff meeting. Lots of good work has been done on the spec.  Lots  
remains to do.

Alasdair Allan covered a vast amount of territory related to our  
baseline use case of rapid robotic response.  Much discussion  
resulted and we spent close to an hour thrashing things about with  
great abandon.  Robert White filled in some more details of the TALON/ 
eSTAR collaboration.  Seems to be quite healthy.  More on various  
related issues when the Wednesday mini-interop is discussed.

We decided to ditch the demonstrations from the schedule.  Various  
folks did include demos in their presentations.  Phil Warner of NOAO  
was busy at the far end of the table doing battle with JMS and Java.   
You may have heard Bugs Bunny coming from my end of the table during  
one of the breaks on Tuesday - that was a VOEvent packet arriving via  
Phil's server.

Scott Barthelmy brought us up-to-date with GCN activities.  GCN-2 is  
to be layered directly upon VOEvent.  Scott later rescheduled his  
flight to be able to attend the Wednesday session.  It can't be  
overstated how important GCN has been to the success of VOEvent.

Chris Smith provided the first talk focused on science.  I hope  
science always remains an emphasis of even highly "technical"  
discussions in the VO.  Chris talked about SNe from the NOAO ESSENCE,  
SuperMACHO and DES surveys as well as microlensing events from SM.   
In late breaking news, these alerts should indeed be hitting the  
VOEvent pavement as a result of the deeply appreciated support of an  
NVO research initiative grant submitted by Phil, Chris and me.

The original agenda included a half dozen discussion sessions on  
specific topics.  The discussions were very successful throughout the  
three days, but were connected to the agenda in only the loosest  
possible fashion.

Frossie Economou picked up after lunch with a repeat of her  
successful comments from the ADASS VOEvent BoF session.  She  
delivered her talk remotely via polycom from Hilo.  The system for  
rapid response to alerts at UKIRT and JCMT provides a very  
interesting use case separate from robotic follow-up.  Kim Gillies  
had talked about the rapid response queue at Gemini at ADASS, but  
wasn't able to attend this workshop.  Gemini may already be  
implementing VOEvent based infrastructure in support of their queue.   
It should be a priority to continue this collaboration with the  
classically scheduled O/IR (and sub-mm) large glass (and aluminum or  
steel).

I had contacted folks from many of the observatories that are local  
to Tucson.  This included Richard Green, now at LBT.  Mark Wagner  
delivered a great overview of how operations at LBT might connect to  
VOEvent.  It certainly is a fascinating telescope and I was left  
wondering if there might be some follow-up use cases specific to  
optical interferometry or to having duplicate instrumentation on dual  
eight meter telescopes.

Arnold Rots discussed the current status of STC.  We continued the  
discussion at various points over the remainder of the workshop.  It  
is critical that we reach a consensus on space-time coordinates.  It  
sounds like a possible consensus is in sight through the use of XLink  
and embedding single coordinate elements within a <value2>.

Frank Hill delivered another excellent talk focused on science,  
specifically solar astronomy and the Virtual Solar Observatory.  Have  
to believe that both solar and nighttime events should be supported  
by the same semantic and transport mechanism.  In general, any  
opportunity for the IVO and the VSO to work together should be seized.

LSST was the third sponsor (with NVO and NOAO) and Kem Cook continued  
the discussion of microlensing events opened by Chris Smith.  GCN is  
bootstrapping the VOEvent standard and network.  LSST provides a  
target for our vision of where we should be in 5-10 years.  Events -  
interesting events - are central to the concept of LSST.

Excellent dinner and company at "Cuvee World Bistro".  Enjoyed coffee  
afterward - even had a chance to sit down after a while.

A thought at the end of day one:  pawns are the soul of chess -  
follow-up observations are the soul of VOEvent.  The semantics of  
describing the observation of a transient are interesting by  
themselves, of course - but there is no reason except for follow-up  
that such shouldn't simply be stored in and retrieved from a more  
traditional archive or database.

The second day opened with a session exploring the breadth of  
VOEvent, with Robyn Allsman giving an update on LSST's plans for  
generating transient alerts.  Nat Butler followed with a discussion  
of GRB alerts from HETE, and Ashish Mahabal talked about near Earth  
asteroid alerts from NEAT.  The semantics of the various event types  
are very rich.  The quantities of packets to be generated within a  
few years is very large.  It is a safe bet that VOEvent will remain a  
priority in both the virtual and real worlds.

After the break, Matthew Graham and Andrew Drake filied us in  
regarding VOEventNet and such technology as Jabber. Alasdair  
responded with happenings in the Heterogeneous Telescope Network.  I  
won't belabor these robust projects since others will be reporting  
back regularly on their progress.

Finally, I had contacted Dan Green of the Central Bureau of  
Astronomical Telegrams as well as his colleagues at the Minor Planet  
Center and the International Comet Quarterly.  These organizations  
have long experience alerting the community to transient events as  
well as serving as the official IAU naming bodies.  Over the past few  
decades the meaning of the word "timely" as in "timely alerts" has  
changed drastically.  A mechanism such as the various IAU circulars -  
whether delivered electronically or via physical telegrams - that  
relies on a model of human peer review is simply not appropriate for  
many of the science drivers of VOEvent such as GRB follow-up.  On the  
other hand, we have a lot of work to do before a VOEvent packet can  
comfortably represent the rich semantics of the natural language  
circulars.

More later...

Rob



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ivoa.net/pipermail/voevent/attachments/20051212/58d563b3/attachment-0001.html>