On Wed, Feb 23, 2011 at 05:18:05PM -0000, Phil Holmes wrote: > TBH, there are so many changes on the official regtest page > (http://lilypond.org/test/v2.13.51-1/compare-v2.13.50-1/index.html) > that it's not feasible to do a good comparison.
I count approximately 150 changed regtests -- and don't forget that they're ordered by "amount of change" (using a vague measure of this). I mean, jump to the bottom of the page. There's basically nothing interesting about ottava.ly, so I'd check that off in 2 seconds. The top item on the page (phrase-slur-dash.ly) will take longer to compare. It's now spaced on two lines, but if you compare it bar-by-bar, with say 2 seconds for each bar, then it'll be around 15 seconds to verify. Every so often there'll be a more difficult one, of course... let's say that the average time is 20 seconds per regtest. That gives 50 minutes for the entire comparison. That's too much to expect you to do, but if the bug squad was functioning smoothly and everybody had a few minutes free, we could divide it up between them. Of course, there's always an overhead to any organizational thing, so maybe that's not the best way to function. Maybe recruit somebody on -user to do a one-time 1-hour task? Call it 10 minutes to writing a recruiting email plus giving them a basic overview of how to look at the comparison? Of course, it would be easier if the memory info was on a different page, but that's not going to change in the very near future. (I think that 20 seconds per regtests is a overly generous estimate... I mean, jumping about 10% down the page, look at spacing-section.ly. There's only a very slight amount of green in the right-hand image, and the whole thing is pretty simple, so I personally would spend less than a second glancing at that image before moving on.) Cheers, - Graham _______________________________________________ lilypond-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/lilypond-devel
