Front page | perl.perl5.porters |
Postings from May 2013
On Fri, May 24, 2013 at 10:56:52AM +0200, demerphq wrote: > > But I respect your point here. Generally we do discuss first and > generally that is the right thing to do. But I dont think Dave is > really out of line either. Its not that hard to revert or modify such > things after all. > When it comes to style - the issue is not as clear cut. When one commits code that turns out to be broken later on - it is easy to reach a consensus to revert the change. In cases like this, however, there is no consensus by definition, and a revert of Dave's change would imply "You are wrong because I am right". Many do not have the political capital and/or will to win an argument on these terms alone. A personal example - couple of month ago the test coverage of Data::Dumper was severely reduced*. Yet the reduction was subtle enough, that I could not adequately explain what the problem was. The result of the short thread that ensued was "If Peter doesn't like things as they currently are - he can supply patches". Which left me in a tough position - do I write a patch that effectively reverts the work of a prolific contributor, knowing full-well that he disagres with me? I opted to drop the issue, not because I do not believe it to be important, but because I was constrained by time and energy I could devote to the problem. My 2c on why "throw it in, we can back it out later" is a poor strategy. Cheers * Jim, I do not claim you made the changes in question out of malice or ignorance. Stuff gets overlooked when one does a lot of things, and you do a *lot* of awesome things. What I continue to have an issue with is how my legitimate concerns were handled at the time.Thread Previous | Thread Next