We had a meeting of the Mercury group on Fri Feb 9, 2001, from 3:15 to 4:00pm. Attendees: Zoltan Somogyi (zs) Thomas Conway (conway) Fergus Henderson (fjh) Mark Brown (dougl) Adrian Pellas-Rice (apell) Ina Cheng (inch) Levi Cameron (lpcam) Simon Mei (xcsm) Sam Thurairatnam (sthur) Julien Fischer (juliensf) David Jeffery (dgj) Simon Taylor (stayl) 1. Progress reports We went around the table, with each person in turn discussing what they had been working on recently and/or what they were planning to work on. Zoltan Somogyi (zs) - Mainly been working with Tom on deep profiling. Thomas Conway (conway) - More work on deep profiling (gave a demonstration today) Fergus Henderson (fjh): - MLDS backend bugs fixed - Testing lcc builds (some still failing, but most pass) - Working on abstract equivalence types for .NET and Java back-ends Mark Brown (dougl) - Mostly working on administrative matters (eg progess report) Adrian Pellas-Rice (apell) - Almost to compile stage - Voiced concerns about need for constructor class to allow generalisation; consensus was that in the past extension limitations have been accepted in an imperfect state if they would prove useful anyway. Future backward compatability loss would probably not be a great concern in this case. Ina Cheng (inch) - Code for web client almost finished - An issue with the problem that the client requires hard coding to support arbitrary methods Levi Cameron (lpcam) - Nondet support finished - Support for type information added - Support for submodules added - Higher order call support added, but still buggy - tests/general pass with the exception of floats, higher order calls and calls involving builting__compare/unify Simon Mei (xcsm) - Finished and posted user's guide for review - Largely finished code; posted for review Sam Thurairatnam (sthur) - Worked on type information database - Has been adding test cases Julien Fischer (juliensf) - Worked on runtime system for Java backend - Most features now working, but currently lacks support for nondet/multidet code, foreign code, higher order predicates and RTTI David Jeffery (dgj) - Fixished a bug fix for polymorphism (involving type var dependency) - Worked on merging HAL branch changes (merge done; still bugs to fix though) Simon Taylor (stayl) - Fixed a bug in intermodule optimisation - Release date still 'soon' 2. Release - Was decided to scrap old release branch and create a new one from the main branch. - Still some low level backend bugs - Still issues with variable liveness, Zoltan will try 3. Housekeeping - NFS mounts are not all consistent; Fergus volunteered to fix them - hydra is somewhat unreliable; David Overton was "volunteered" to move the webserver to venus Minutes taken by Levi Cameron (lpcam@cs.mu.oz.au).