On 1 December 2010 11:36, pang <pablo.ang...@uam.es> wrote: > On 30 nov, 20:50, Robert Bradshaw > >> +1. I have the feeling that people are doing more testing than reading >> of code, which is omitting the most important step, and in particular >> the one that only a human can do. Testing should happen orthogonal to >> someone reading the code and giving it a positive review. > > +1: building and testing eats most of the time and causes most of the > trouble of reviewing. For the easy tickets to review when you're not > knee-deep into Sage, like documentation, or interacts, or docstrings, > or simple new wrappers for previously hidden functionality, it's not > nearly as important as discussing the code with the author, and the > testing process can turn off potential new reviewers.
Why should I waste my time checking the validity of code that the author can't be bothered to check actually works? I feel it's the responsibility of the author to check the code works, not the reviewer. If you submitted a proof to a maths journal, but wrote: "I can't be bothered to check my proof, but I'll address any errors found by the reviewers" then I don't think the journal editor would even bother submitting the paper to reviewers. It would be rejected immediately. Dave -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
