On Dec 6, 2010, at 9:24 AM, Markus Hitter wrote: > > Am 06.12.2010 um 16:32 schrieb Stefan Salewski: > >> Sometimes there are some good reasons against code changes: >> >> - huge increase in complexity for minimal gain. gcc 4.x may be an >> example for this -- for some architectures there was not much gain from >> 3.x, for microcontrollers there was some regression. >> >> - sometimes the basic design of software is so bad (spagetti code) that >> each modification will introduce bugs. >> >> - with changes the code will not work any more with old hardware or >> libraries or architectures. >> >> - porting to other languages or hardware can become harder >> >> - licensing may be another issue, BSD/GNU/APACHE... > > At best, these are reasons to ask the commiter to review his code to match > additional criteria. How would he know what traditinal gEDA developers > consider to be well formatted code, a good strategy of conditionals, or what > they consider to be a "huge increase"? In the two months I'm on this list > I've almost never seen such such a request for matching additional criteria, > despite of lots of no-no criticism.
If the bug "fix" merely piles a kludge atop poorly designed code, it should be rejected regardless of style issues. John Doty Noqsi Aerospace, Ltd. http://www.noqsi.com/ [email protected] _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
