We had a meeting of the Mercury group on Fri October 26, 2001.

Attendees:
	Zoltan Somogyi (zs)
	David Overton (dmo)
	Simon Taylor (stayl)
	Ralph Becket (rafe)
	Fergus Henderson (fjh)

Progress reports

	Zoltan Somogyi (zs):
		- Has been working on teaching related tasks.
		- Has collected performance related data for the HLC paper
		he submitted with Fergus last Friday.
		- Posted the HLC paper on the web site.
		- Done some work on the source-source stack-slot
		optimization he mentioned last week.
		- Volunteered to write a sanity checking script for the
		CVS repository to look for corrupted files or files with
		incorrect permissions.
		- Will be interviewing potential summer students early
		next week.

	Fergus Henderson (fjh):
		- Has been working on optimizing data representation for
		the HLC back ends (mainly for .Net and Java) whereby
		nullary constructors can be represented using distinguished
		pointers (e.g. NULL).
		- Encountered an amusing error whereby two different 
		zero-size structures were being malloced at the same address.
		- Has been working on exploiting nested functions under gcc.
		- Submitted the HLC paper with Zoltan last Friday.

	Ralph Becket (rafe):
		- Spent much of Monday and Tuesday setting up the new
		machine (ceres) with David Overton.  The problem with getting
		the network card to run under Linux was solved by Tyson,
		although getting more than 1024x768@60Hz from the video card
		is proving difficult.  Tech. support have been notified.
		- Has been working on the pretty printer (pprint).  A
		performance bug has been fixed (it is now useable on very
		large terms) and a formatting bug has been identified and
		fixed.  A check-in with regression test is imminent.

	Simon Taylor (stayl):
		- Implemented the pragma export declarations.
		- Fixed some bugs in smart recompilation.
		- Has continued with integrating mmake functionality
		into the compiler.
		- Will be moving onto working on the intermodule
		optimization framework next.

	David Overton (dmo):
		- Did a lot of work to get ceres on its feet.
		- Has been working on a mode constraints paper.

Minutes taken by Ralph Becket (rafe@cs.mu.oz.au)