There has been a trend lately of old meta posts brought up to the front page because of retagging or other minor edits. While this happens all the time on the main site and is OK and encouraged in mild doses, the situation in meta is somewhat different. Meta is very slow compared to the main site, by a factor of about 50. (It takes ~7 hours for an inactive post to slide down the main but ~2 weeks in meta.) I would therefore like to bring this up here. If we're OK with this, then we're OK, but if we're not then we need a policy to stick to.
The problem is that bringing an old post to the front page gives the impression that the discussion is still active, and this can be very confusing. Old support posts like this onethis one become meaningless when a tide of changes completely washes over the issue. Old policy discussions (which I won't link to) do not gain anything from the extra attention, and we have seen this have tragic consequences for certain questions.
Mostly, though, I think there's no real reason to be fiddling with the tags on old discussions. They are in the archive and can be found easily enough. They often discuss quite topical issues for the specific time they were asked in and now have only historical value. (A lot of it in some cases.) They do not belong on the front page.
Let me take a stance, then, so people can up/down-vote. Please keep edits of old meta posts to a minimum. Use retag edits on them only when they would be impossible to find otherwise. If you feel a certain discussion should be revived, then give an alternative answer, or better yet ask a new question that's appropriate for the site's current conditions. If you don't have anything to add to such a thread, then maybe it's in the cold archives for a reason.
Some tags deserve special attention. The meta tags status-declined, status-completed and status-bydesign can only be applied by moderators. If you feel a question needs one of these tags and doesn't have it, flag a moderator and tell them that, instead of inventing new tags for it.