A much better suggestion would be to fix the code. My completely unhelpful suspicion is that it's PHP's fault. But I'm a perl bigot, so take it for what it's worth. I'm just not surprised we're having problems.

Right suggestion, wrong reason. I rather suspected PHP as well, but now that I've lived and breathed the code for a week or so, I'm quite statisfied that PHP is more than fast enough. The present code, on the other hand, with all due respect and admiration for Magnus's tireless efforts, is a mess. It has no architecture or design to speak of, and does lots of unnecessary work, including many redundant database fetches.

I'm working on it. I'll release working snapshots of my version as I complete things that the other developers can test, but I'd also like to see our develop-test-install process cleaned up before we jump headlong into anything. 0