Front page | perl.perl5.porters |
Postings from May 2013
On Fri, May 24, 2013 at 12:05:51PM +0200, demerphq wrote: > On 24 May 2013 11:34, Peter Rabbitson <rabbit-p5p@rabbit.us> wrote: > > Whereas the s/80/100/ patch says: "Ah sod it - it doesn't matter. If > > someone wants to have this - they can fix it all in one step at some > > future point. But for the time being I refuse to crowdsource work which > > I find useless anyway". > > IMO its not crowdsourcing, it is misusing the most scarce resource we > have (competent C programmers who know the core) for something that > others can do just as well. As one of the scarse resources I agree the initial reasoning that it's not the best use of time. But I think that taking this approach would be flawed. The problem is that it relies on having other *volunteers* to clean up. "Hey, come volunteer to help with the Perl core. Do you know C? Oh, OK, have fun with all these menial tasks". A volunteer project is completing on the basis of "fun", and deciding that new recruits and "unskilled" labour gets all the unfun tasks is going to result in many of the other volunteers stopping. Which in the end means that the C programmers *also* have to do all the other jobs, because the two-tier system ends up with the other people leaving the project. If we had a ready supply of people keen to do these sort of tasks (and actually *do* them, not just say they were interested and then wander off), then it would be different. But we don't, and I don't think that this is going to change. Also, I don't like it because it isn't leading from the front. It isn't "walking the walk". It's "do as I say, not as I do". The bigger still problem that I don't have an answer for is I can't see how to make *any* sort of "volunteering on the core" more fun. A lot of the usual "fun" reasons just don't work. For example, it's great that v5.18.0 shipped, but it's completely outside of our control as to when anyone will deploy it. And if no-one uses it, where's the fun in that? Nicholas ClarkThread Previous | Thread Next