Science Tools Working Group
We last met on May 21; once again this week we will not meet.
Since June 19 the current release version of the Science Tools has been v9r5p6. Here are the differences from v9r5p5. A number of small but important updates have been made, and if you are a regular user of the Science Tools you probably should switch to this version
Data products: No news.
Databases and related utilities
No news
Likelihood analysis
No new news. The bug fix mentioned last week - Jim reported that he "fixed a bug in the convolution code for binned analysis wherein Ping was seeing slight offsets for her high-latitude DM source maps (Likelihood v13r15)" - is in the new release of the Science Tools.
GRB tools
No news
Pulsar tools
From Masa: He and James "are still working on time handling classes. As always, the more we go, the more issues we find. None of them are critical, but we still need to carefully fix them one by one
Observation simulation
Richard has updated the microQuasar source to no longer reset the random number seed; this is in the current Science Tools release.
The bug fix that Jim reported last week - "Johann encountered and fixed a problem in the code that allows one to select submaps when running simulations with the MapSource sources. (celestialSources/genericSources v1r11p3)" - is in the new release, as is another fix by Jim for handling short time intervals in the FitsTransient source.
User interface and infrastructure (& utilities)
Windows users of the Science Tools: Toby reports that he has developed an alternative to the vbs scripts generated by the Release Manager. The approach and installation procedure are described here. Users install CMT and a batch file that Toby has written, in addition to an RM distribution of the Science Tools. The big advantage is that all of the Science Tools are made available in the same terminal window. Also CMT is guaranteed to set the environoment variables correctly.
Eric W. has started work on porting the 'core' (Standard Analysis Environment) of v9r5p5 to the HEASARC hmake build system.
Source Catalog
Met last week. Jean presented an analysis of the detectability of sources in the 'obssim3' data - based on their specifications and locations. He also discovered that the obssim3 data were seriously flawed; owing to frequent resets of the random number seed when the data were generated the Galactic diffuse and extragalactic diffuse emission were 'spiky' - with identical photons tending to pile up in the same directions on the sky. By last Thursday, the obssim3 sky had been regenerated by Nicola and merged back in with the resampled backgrounds. The files are living here but Real Soon Now should be in the Data Catalog and xrootd.
Science Tools Working Group
We last met on May 21; we will not meet this week.
The current release version of the Science Tools is v9r5p5. Here are the differences from v9r5p4. A number of small updates have been made. Jim reports that the principal reason for the new tag is so that for L1 processing the same version of ROOT (v5.18.00c-g11) would be used in the Science Tools and GlastRelease. In terms of changes that you are most likely to notice, gtobssim now enforces a deadtime of 26 microseconds per event.
Data products: No news. The delivery of data products from ASP to the GSSC has been a recent focus. Also, we don't seem to have a universal understanding of the frequency of deliveries of livetime cubes (LS-003).
Databases and related utilities
No news
Likelihood analysis
Jim reports that he "fixed a bug in the convolution code for binned analysis wherein Ping was seeing slight offsets for her high-latitude DM source maps (Likelihood v13r15)" [this is not yet in a release]
GRB tools
No news
Pulsar tools
From Masa: "Nothing exciting in the pulsar tools development. We are still working on the time handling classes, and have almost done what we planned for the classes to change. But, lots of bugs and problems were found in the process, and we are shooting them now. Although I don't think those bugs affect analysis results with the pulsar tools, I think we should fix them now rather than later."
Observation simulation
Richard has updated the microQuasar source to no longer reset the random number seed; this is in the current Science Tools release.
From Jim: "Johann encountered and fixed a problem in the code that allows one to select submaps when running simulations with the MapSource sources. (celestialSources/genericSources v1r11p3)" [not yet in a release]
User interface and infrastructure (& utilities)
Not sure where this belongs, but in case you do not read infrasoftlist, you should see Tony's note about the new version of the GUI code installer. It should be much more robust against network flakiness ("Guaranteed to work even at the cocoa beach holiday inn").
Source Catalog
Did not meet last week. Will meet this week, with one main topic being source detection on the obssim3 data set.