Re: [tlug] STM (was: Re: work times & accommodation @tokyo)

["Stephen J. Turnbull" <stephen@example.com>]

Curt Sampson writes:
 > On 2008-07-30 15:36 +0900 (Wed), Stephen J. Turnbull wrote:

 > > The sketch in "Beautiful Concurrency" is nowhere near as detailed as I
 > > remembered it....
 > 
 > I thought it was enough for someone experienced with monads in Haskell
 > to get enough of an idea of what's going on to implement it.

Well, that's not me.  I'm a project leader and sometime debugger, more
than a hacker.  Haskell monads are intellectually interesting, but not
something I've used intensively.  In any case, that route would imply
reconstructing monadic behavior in Java (or C++) which is surely not
straightforward, since you think it's impossible.

 > You can think of the State monad as just another implementation of
 > an alternate "real world" that needs no special support, since it's
 > contained entirely within the language.

By "I/O" I meant to denote interactions with such "real worlds" from
the larger programming language.  Sorry.