Why? You make two very sweeping statements there, with no underlying arguments at all. Later versions obviously have bugfixes, obviously have new features, but the stability of the versions doesn't necessarily change from release to release.

Perl 5.004, which we (mainly) use, can handle the 100,000 rows of data we throw at it, and output the reports we need quickly and efficiently, without crashing. That's our requirement for stability - and 5.004 meets it.

In my mind there's very little substance in a "update, because a later version's there" mentality. We don't need bleeding-edge, we need mature, stable code and mature, stable scripts. I'm not saying that later versions of Perl aren't mature and stable (completely the opposite, I'm sure they are), but from my perspective, and the point of view of the other people in the tech group where I work, there's very little potential benefit in spending a weekend upgrading a system and ensuring that everything we have still works. Ultimately, the system we have isn't broken, so why fix it?

When it comes to mission critical systems, I think it's better to play it safe, and stick/cope with a solution that's guaranteed to work for at least a short while longer, rather than perform an unnecessary upgrade and risk things failing.

-- Foxcub
A friend is someone who can see straight through you, yet still enjoy the view. (Anon)