I think drifting is pretty silly myself. The figure skating thing is a good analogy. There are a lot of people into it though, and if that's their thing then so be it.
The thing is, for it to be worthwhile at all, Shift will have to have a pretty realistic physics simulation. Unless you have to be constantly modulating the throttle and making corrections with the wheel, there's no point.
The developers and physics programmers behind it are easily some of the best around, so they have the ability to make an excellent simulation, but my worry is that EA will be putting pressure on them to make it have a more broader arcade appeal. Obviously, the problem with that is that the both the people who like the more arcade-style games and the people who want a realistic game will probably find driving around a race track in an only superficially realistic game incredibly boring.
Asking developers who have up until now only written PC-only sim-racing games to write a console Need for Speed game is potentially mixing oil and water. Personally, I'm hoping they use all of their skill to put out a genuine racing simulator that could very easily challenge Gran Tourismo and Forza in terms of realism. If the EA suits stuff it up though, as they're so apt to do, it's going to be a huge waste and a failure.
Ian Bell, the head of SMS, has said it's by far their most realistic simulation yet, and I really hope he's not bullshitting. We can be guaranteed that the underlying code is for a very realistic simulation because they've been working on it for years--long before EA came along (10-Tacle going under left them high and dry)--so it's all going to come down to what sort of restrictions EA puts on them. It wouldn't be a conventional Need for Speed game obviously, but it would be something new and potentially excellent. In short, I'm a little nervous that one of my favorite developers has been thrown in the EA profit-only meat grinder.