Intels reasoning behind this decision is a bit puzzling. Presumably the controllers in the two generations are very similar and ATA TRIM sounds like something that should be fairly easy to implement (compared to scary things like wear leveling or block allocation) so I wouldn't have thought that implementing this and verifying it wouldn't be that expensive. I suspect that the number of G1 owners that get a competitors drive as a result of their annoyance at Intel will far outweigh the numbers who will buy a brand new Intel SSD to get TRIM.
(In my case, this quick abandonment of an expensive product is definately one reason to buy from a competitor, even though Intels SSDs have very impressive benchmark results)
Perhaps someone should remind Intel they don't have a monopoly in SSDs?