> Nothing personal with Perl. Just doing stats. Very poor stats if you're extrapolating the time a market participant will be reduced to zero by looking at rel

> Nothing personal with Perl. Just doing stats. Very poor stats if you're extrapolating the time a market participant will be reduced to zero by looking at relative market position without also examining absolute usage. That is, it's possible for something to have less relative market position but be increasing or static in total usage if the total market is increasing fast enough. That doesn't indicate something is dying, but it looks the same in graphs, and extrapolating from those lines won't show you what you think it is. Is Perl in that situation? Probably not that much (but somewhat, I would argue), but if you're going to make a case on statistics, don't do a crappy job of it...