Stupidity aside, I've played about two thirds of the game now. It looks like a 15-20 hour affair in total, which is just about right in my book. I'm definitely enjoying myself.
I think I can safely say that the game didn't blow all its creativity during the first third, and that it's a fairly even distribution with nice surprises all along the way. That's a big relief, and I hope it keeps it up to the end.
Not having to go through random or recurring battles is pretty darn cool. I do miss the action sometimes, though.
It looks like it will be easy to dump the script. The translation side looks like cake so far, and if reinsertion and battle engine/menu hacking isn't a pain, then this could be a relatively simple project to get through.
On the other hand, when it comes to priority, it's hard to say whether this one should be at the top. There's lots of demand, yes, but that's maybe the strongest reason it's got going for it. I hate to say it, but you're really not missing much by playing this via a walkthrough even if you can't read Japanese.
The two other PCE games I'd be most interested in getting scripts for are Xanadu 1 and Tenshi no Uta 2. In Xanadu 1's case, a translation would be extremely useful, because the game is next to unplayable in an unfamiliar language unless a super-detailed walkthrough holds your hand every step of the way. Not to mention, if anyone does Xanadu 1, it makes sense for it to be Esperknight and me.
From what I've seen (roughly 3 hours) and read about Tenshi no Uta 2, it's heavy on drama between characters, and the dialogue itself is much more important. It's also worth mentioning that Japanese PCE fans rate it about as highly as any other RPG on the whole system. I'm thinking of finishing it after I play AFS.
I'll have to get Esperknight's input on the hacking sides of each of these, but what do you guys think?