parameter default change in CarpetIOASCII

Comments (6)

1. 

   Erik Schnetter

   • removed comment

   A quick note: If a parameter is deprecated, it should not be used if possible, and consequently its default should not be changed either.

   • 2011-11-24T13:58:07+00:00
2. 

   Erik Schnetter

   • removed comment

   Luca, do you want to discuss this on the mailing list? This ticket seems to be ignored, and the changing a default value should be done by consensus.

   • 2011-12-11T20:20:50+00:00
3. 

   Erik Schnetter

   • removed comment

   This issue was raised on the mailing list (October 10, 2011, subject "Questions on CarpetIOASCII"), but no discussion followed. Since changing default values can be inconvenient for many people, I suggest to not change the default unless there is a consensus in a discussion.

   • 2011-12-12T09:15:16+00:00
4. 

   Roland Haas

   • removed comment

   I am fine with a change of default parameters. I am in the fraction that "don't care about the change" (in the sense that I don't mind whether or not it happens) and I'd rather speak up now than complain about it afterwards. I would suspect that this change requires all tests that use CarpetIOASCII's 1d output to be regenerated since the number of points output changes?

   Otherwise: fine by me to apply even if it means allmost all of my parameter files will change behaviour.

   • 2011-12-12T09:22:47+00:00
5. 

   Erik Schnetter

   • changed status to resolved
   • removed comment

   There was no discussion, and my original point stands -- without discussion and consensus I would rather keep the current default values. There are many cases in the Einstein Toolkit where the default values are inconvenient, and are only there for backward compatibility, e.g. most things having to do with coordinates or buffer zones.

   • 2012-05-05T10:13:37+00:00
6. 

   Roland Haas

   • changed status to closed
   • edited description

   • 2019-02-21T19:56:10+00:00
7. Log in to comment