Well, several reasons, really.
One, I read a [link|/forums/render/content/show?contentid=121392|really good book] recommended by our friend Mr. Tilly about making products succeed. One of the things proposed was building internationalization in from the very beginning. It's easy to add early on, and successively harder the longer you wait. Products that include it from the beginning have a higher success rate than those that don't.
Two, we have people from other countries here. While you pretty much have to know English to understand the vast majority of the posts here, that won't be the case when the software is (hopefully) being used by other people besides just us here. One of my main goals for the rewrite is making the software stand-alone. The existing stuff is well nigh impossible to get running on another machine at this point. Just ask Yendormike. The new stuff can be installed in a few minutes.
Three, Christian and other Europeans would probably rather see their dates, times, and numbers in a familiar format. And once you're supporting localization, translating the text is not a very big leap from there.
Four, it's good practice for me.
Five, supporting dialects like Nadsat and Guido is easier, yes. ;-)