Sorry to jump into this thread so late. However, I'd like to take a moment to remind everyone that some time ago I put a concrete proposal for thread-local variables on the table.
http://article.gmane.org/gmane.comp.lang.haskell.cafe/11010 I believe this proposal addresses the initialization issues that Einar has been discussing. In my proposal, thread-local variables always have some defined value, and they obtain their values at well-defined points. The liked message also gives several use cases that I felt motivated the proposal. -- Rob Dockins Talk softly and drive a Sherman tank. Laugh hard, it's a long way to the bank. -- TMBG _______________________________________________ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell