S4 LIGO-GEO Burst Review Committee Meeting Wednesday 21 November 2007 12:00 Eastern
Minutes: Wednesday 21 November 2007 10:00 Eastern
Agenda and Contact Info
- Status of action items from previous telecon.
- List of actions needed for completing review:
- code review
- paper review
InterCall telecon service: USA (toll-free): 1-866-616-1738 UK (toll-free): 0800 073 8914 Italy (toll-free): 800-906-494 Germany (toll-free): 0800-1014-907 participant code: 251 288 9495#
- Status of action items from previous telecon
(Klimenko) Update cWB efficiencies for resampled MDCs.
PENDING - working on it with Yakushin, Hughey. Yakushin gave Hughey triggers for resampling.
(Heng) Present WB+CP efficiencies for resampled MDCs.
PENDING - Cadonati to make efficiency curves.
Check on the tag versions of WB/cWB used for S4-LIGO-GEO.
DONE - Yakushin put cWB version for S4 into CVS, WB+CP version already in CVS and same as for S4 LIGO-only search.
Check segment lists, DQ flags for cWB.
DONE - Segment lists with category 1 ("pre-ETG") and 2 flags are in CVS. Both applied at run time. PENDING - Yakushin to send links. QUESTION - Were category 3 flags (upper limit flags) used at any point, since no ULs were computed?
(Heng) Check segment lists, DQ flags for WP+CP vs. cWB
- (Klimenko) Update cWB efficiencies for resampled MDCs.
List of actions needed for completing review
In previous telecon Riles reported that past reviews of WB had run into problems with livetime accounting.
Klimenko: Same code but separate executable used to run over the output ROOT files to compute the livetimes. This executable, "MRF", is separate from WAT. Pointed out that since we do not report an upper limit, an error of even a few percent in the livetime will have negligible effect on the stated results.
Action item: reviewers will look at paper and decide if the effect of a plausible error in the livetime (order a few percent) is significant enough to warrant review and testing of the livetime calculation.
Zanolin: In his opinion, the livetime calculation is the only 3-detector vs. 4-detector code issue.
Riles: There are 2 scripts that differ between the three and four-detector cases.
Action item: Klimenko will send links to these scripts and we will review them at next week's telecon.
- In previous telecon Riles reported that past reviews of WB had run into problems with livetime accounting.