Agenda

Peter - issue #2488 reported by Frank Loffler, not present
Steve - problem with svn.einsteingtoolkit.org too old to work. What is needed to make old thorns to work, is (1) update svn server, or (2) port the thorns. Upgrading svn is the easier to do.
Roland - proposes rewrite of the thornlists from the old repository, to correct them to use the git checkout and point to the correct repository in github, and turn off svn server.

Peter - Kuibit toolkit announcement email
Roland - Python package for post-processing simulation data, will be an upcoming presentation
Peter- delegates person to invite Gabriele Bozzola, the package developer.

Peter - SPEC benchmark contribution thorn choice status
Steve - administrative issues with SPEC contribution: who receives the money, how to pay a student, what code to be proposed to include
Roland: proposed to be received as a gift to an account of the University. Choice of thorn in work, compiling McLaughlin, if not will send a survey with the 3 codes suggested, one will be proposed for inclusion in the SPEC benchmark

Frank - online now, explains the issue, suggests alternatives.
Roland - reiterates that is not a simple search and replace. Thorns were moved between branches and repositories.
Frank - proposes to put on a student on it, and have it tested that the old versions work.
Peter - student in mind? No suggestion.
Steve - why don't just submit it as a patch if the fix is easy.
Frank - more complicated, should effort be put in upgrading the server, or instead upgrading the old thorn list, and let server die.
Steve - mailing list will have to suffer if we don't maintain the old server
Frank - less likely to break than the svn stuff on the old server. Would be nice to clean it up and get rid of svn.

Peter - unanswered questions on the mailing list, error messages in the makefile
Roland: fix email question empty repository
Peter - ETK 2010 download error
Frank - first release after moving to github. Fix: point GetComponents correctly, remove http links
Peter - who will send the email? move to next - instability par file
Roland - we do not have it, need to write code to produce ID
Steve: reply email no one has such thing

Terrence new thorn Maxwell equations 2 thorns: ID and evolution ADM like and BSSN-like. generate thorns and put them in WVUThorns arrangement. CarpetX testing. Between Scalar Wave and BSSN.
Update from Ken - parses BSSN equations, still need to double check with BSSN eqs not parsed.

Steve: has a pair of scripts (1) check what Cactus needs, (2) check the SPEC configurations, should be in utilities directory.

Roland: - deliberate: reports as 8 cores, but they are not physical cores? Not limited by that choice and ignore the hyperthread?
Steve - procs is nr. processes, it will be divided by the num threads?

Maria

_____________