You're exactly right that they're different ways of representing numbers in other languages. If you'd like to see an example of what such a set of numbers might look like, try here (chosen at random). The digits are in the 5th column from the left, labelled 104, in rows 0 through 9.

Whether or not having \d match 178 different characters is a good thing depends on the situation. I've been treating the Unicode situation somewhat similar to Y2K -- it's overhyped, but you still need to worry a bit. Any code that might at some point need to be internationalized should be thought through, and idioms like tr/0-9//c discarded.

Of course, I don't turn utf8 on yet, because the support for Unicode is still immature and shaky, and I'd hate to have a random string be validated as a number just because it contained two bytes next to each other that happened to be 0x1048. Line disciplines will solve that, eventually.

In sum: I would certainly urge Monks to be early adopters, or at least stay aware of Unicode issues, if for no other reason than to avoid subtle bugs in the future.

-dlc