Heh, I've been bouncing this around for a while because it would be something that would be easy to get very wrong by implementing carelessly and without attention to nuance.

Enough people do import from Twitter that I thought it was worth asking, and I did get some developer interest. Without that interest, I don't know that I would have put as much effort into making as detailed a spec as I did. I've had enough inquiries about the specific service I use that it's become apparent that my friend's script is better than any other mass Twitter-import service.

One of DW's core functions does seem to be importing from other places, like LJ, IJ, JF, and making it easy to do so. I don't think we would have the population that we do had we not had that feature. I don't know that we'd necessarily get a lot of fleeing Twitter users, or if Twitter archival is a market that the owners want to court, but it's a feature that existing users want, it's something ought to be seamless but has been consistently hard to get the details of right, and it's something that LJ has and we don't. I'm pretty confident that between DW's reading page filtering and the massive amount of tweaking that semanticist has been doing to the script at my suggestion (it's over two years of fiddling), that it can be done right.