Dear Maude Abusers,

A new alpha release can be accessed at SRI-CSL in:

~eker/public_html/Maude/Alpha81/

or downloaded from:

http://www.csl.sri.com/~eker/Maude/Alpha81/

Alpha release site authentication:

User Name: maudeabuser
Password: bughunter

This version reports itself as Maude 2.0.1 and will be released as such
if no show stoppers are found.

Bug fixes:
(1) A critical bug in the AC/ACU matcher (reported by Paco) triggered
    by the following example:

fmod AC-BUG is
  inc NAT .
  sorts NatSet State .
  subsort Nat < NatSet .
  op i : Nat -> Nat .
  op g : NatSet NatSet -> NatSet [assoc comm] .
  op f : NatSet NatSet -> State .
  var X : Nat .
  vars S T : NatSet .
endfm

match f(g(X, S), g(i(X), T)) <=?
f(g(0, 1),
  g(g(i(1), i(2), i(3), i(4), i(5), i(6), i(7), i(8)), i(9))
) .


(2) A bug where Maude would go into infinite recursion when a unary
    operator had empty syntax (reported by "Sylvan S. Pinsky"
    <pinsky@thematrix.ncsc.mil>):

fmod BUG is
  sorts T A .
  subsorts T < A .

  op __ : A A -> A .
  ops pl(_) +(_) : T -> A .

  vars g h : A .
endfm

red h g .


Other Changes
==============
The big change is that the builld system now uses autoconf/automake
for easier porting. There are also many portability fixes.
There are ready made binaries for:

maude.darwin	    Darwin (Mac OSX), dynamically linked
maude.freeBSD	    i86 Free BSD, dynamically linked
maude.linux	    i86 Linux, statically linked
maude.macLinux	    Mac Linux, statically linked
maude.alpha	    DEC Alpha OSF, statically linked
maude.solaris	    Sparc Solaris, dynamically linked

So far I've failed to make it build on AIX (insufficient memory to
link), HPUX (math library issues) and SGI/IRIX (g++/libstdc++
issues).

There are now --help and --version command line flags.

The internal ordering on Qids is now alphabetical at Manuels request.
Previously it was based on the order the underlying identifiers had
been seen which caused reproducibility problems for proof scripts.

I will be out of town and off line for the next 2 weeks, though I may
be able to read email sent to steveneker@att.net (no large files!).

Steven