A bug has been fixed in the code used for FSA minimisation. The bug could cause axiom checking for coset systems to fail erroneously, or the code which generates subgroup presentations to fail.
Version 2.1 of MAF has two new utilities. The more important one, gpxlatwa, allows a word-acceptor for one generating set to be translated into a word-acceptor for any other generating set. This is useful if the desired generating set does not give rise to a natural automatic structure, or if the structure is very difficult to compute.
MAF has recently been updated to fix some bugs in the code used to generate subgroup presentations. Previously Version 2.0.4 fixed bugs in the low-index subgroups program gpsublowindex.
V2.0.3 has identical executable files to V2.0.2. A few minor typographical errors in the documentation have been corrected and there is a little more information about the status messages MAF produces while it is working.
V2.0.1 of MAF is now available. This features a few minor bug fixes, most notably one that fixes a problem that stopped automata from completing when a finite index subgroup was processed without -detect_finite_index.
The example input files have been cleaned up somewhat.
I have uploaded new source code and binaries for MAF. This version features many new utilities implementing procedures such as coset enumeration and low index subgroup discovery. It also has extensively rewritten documentation and the 64-bit version uses much less memory.
This release is not quite complete as yet. I intend to change the method used to ensure enough overlaps have been considered when a finite index subgroup is detected by automata, and the documentation probably needs some more corrections. However, I hope it will prove reasonably usable.
The GAP interface for MAF has been on hold for months but I hope to return to this in a few weeks time.