Review Committee Meeting Monday 5 November 2007 08:00 Pacific / 11:00 Eastern
Minutes: Monday 5 November 2007 08:00 Pacific / 11:00 Eastern
Agenda and Contact Info
- Discussion of 070922 Burst Candidate
- Detection Checklist [ HTML ]
- Review of BurstMDC Simulation
- Summary of S5 BurstMDC Simulation [ HTML ]
- Technical Documentation [ HTML ]
- Source Code [ HTML ]
InterCall telecon service:
Phone: 1-866-380-5536, participant code: 435 672 9587 #
International callers ++1-816-249-4731 with same code
International access numbers (~10 cents/min surcharge for LIGO Lab):
Paris 017 080 7156 ;
Lyon 042 603 0036 ;
Germany toll-free 0800 182 1591 ;
Berlin 030 726 167 371 ;
Rome 00645 217 080 ;
UK toll-free 0808 234 7914 ;
London 0203 107 0293
Handy participant command: Press *6 ["*M"] to mute, #6 to unmute
- Discussion of 070921 Burst Candidate
- Detection Checklist: Goal is to reach a stopping point
by end of week. Some things won't have been done. Somethings
are not so important; other things are not appropriate at
this time (e.g., checking if event was seen in other detectors.)
- Things to do in reviewing the Detection Checklist:
- Double check the checks -- e.g., redo them.
- Check the analyses that were used.
- Assess significance -- decide how to apply DQ
flags to background. E.g., cWB gives a false rate
of 1/50yrs -- would this be 1/100yrs with DQ flags?
- When looking at items in checklist, would be a good idea
to get relevant review teams or other committees (e.g.,
detchar) to review each item in list.
- One option would be for burst group to produce a techincal
report that we would review rather than delving into the
checklist right now.
- Will this be an isolated result -- a possible detection
from a high-threshold online search -- or will it be part of
a set of candidates from a larger offline S5 search? This
affects whether this should be a fast-track analysis (like
a GRB) or not. Peter argues that, unlike a GRB, we were not
told when to expect this candidate, so it should be considered
to be part of a longer-term search.
- General consensus (in this meeting) that part of a larger
search is probably the best way to go.
- Assessing significance: possibly need to consider a larger
data set. If this is part of a larger search, false rate
should be assessed for entire search, not just for the day
around the event.
- Different possibilities of what search this is part of
were discussed: A 5-way coincidence search (LGV)? (Seems
artificial.) A 3-way LIGO S5 search? (May not be seen in
triple coincidence by KW/BN because it would be too weak for
H2.) This is an important thing for the burst group to
- Brian: this seems backwards. We should conduct the
searches, open the box, and then follow-up the
- Worry is that tuning on the LIGO-Virgo search hasn't been
done. Now have to be very careful that consideration of this
event does not influence tuning decisions.
- Decision is made to table the review of this candidate
until after we have a set of candidates that are part of a
larger S5 search. Burst group should define what this search
is and then when the box is opened, there will be some number
of candidates to follow-up. (There might be other detection
- Timescale: a few months to tune and complete searches.
Looking to have this done around Jan/Feb. Ambitious goal would
be to have something to present to LSC in Mar.
- Is there anything that must be done before decomissioning?
Only thing burst group thought would be important would be to
perform a hardware injection of the reconstructed waveform.
This was done.
- Review of BurstMDC Simulation
- Q-Pipeline team: David Reitze, Jonah Kanner, and
$Id: minutes-2007-11-05.html,v 1.2 2007/11/05 17:21:35 jolien Exp $