Re: New Bound Threads Proposal

I haven't looked at the new proposal closely but I'm very disturbed when Wolfgang reports:

*) It violates requirement #4 "The specification shouldn’t explicitly require lightweight “green” threads to exist."

*) The blocking behaviour of foreign calls is specified with a bias towards GHC's implementation. [...]

If true this will make it harder to implement the proposal in Hugs and NHC. Also, I think the experience of most language designs that I know of is that if your design reflects too much of your current implementation technique, then you're creating problems for yourself in the future. (Examples: ML modules, C typedefs, most C++ features, ...) -- Alastair