Hi,
Here are the meeting notes of today's meeting:
1. Patches
We shortly discussed the two patches which have been proposed to be included in the stable branch. We agreed that both patches are important enough to be included, but need more testing before that happens.
For the GetComponents patch this would mean that it has to be tested to work on all of the machines mentioned in the release (see http://einsteintoolkit.org/toolkit/releases/ET_2010_06_announcement.php).
The GRHydro patch has to be tested by running all testsuites on all of those machines and producing the same results. In addition a new testsuite will check for the patched problem (which was not the case in the released version), and that also has to pass everywhere. Then the code patch and the new testsuite can enter the release branch.
2. Release tags
In order to be able to checkout a particular 'state' of the released branch we will create tags corresponding to each state (e.g. before applying new patches). Those are not meant to be used by regular users (other than bug-tracking maybe).
Erik and me propose to use a tag name which starts with the branch name, adding a string, e.g. ET_2010_06_A, ET_2010_06_1 or ET_2010_06_01. I propose ET_2010_06_ follwed by an upper case letter in alphabetical order, e.g. ET_2010_06_A. This allows for 26 tags, by which we probably should release a new version anyway. The scheme ET_2010_06_ plus one digit would only allow for 9 (or 10 if you include the 0) tags, while two digits could be mistaken to be a day.
Please comment on this scheme if you have an opinion about it. We should come to an agreement soon to create those tags such that the two patches mentioned earlier can be committed.
3. MHD
We didn't talk much about the MHD plans as Christian noted that most of the necessary information is in the wiki (http://docs.einsteintoolkit.org/et-docs/MHD_implementation_details_and_discu...) or the mentioned paper (http://adsabs.harvard.edu/abs/2006ApJ...637..296A) and discussions should happen on the mailing list.
Frank