The following problems are found and/or solved:

=========== PROBLEM 001 ===================================================
Description: the name of the AFM file is derived from the absolute filename
	of the type1 font. This does not work when AFM files are kept
	in separate directories (like on CTAN).
Reported and solved by: "Ard Kuijpers" <ard@wfw.wtb.tue.nl> (see PATCH 001)

=========== PROBLEM 002 ===================================================
Description: `checksums = off' in mtpk.cfg does not work as specified in
	mtpk.1
Reported by : Peter Busk Laursen <unipbl@unidhp1.uni-c.dk>
Solution: typo, checksums in the mtpk.1 should be checksum (without last s).

=========== PROBLEM 003 ===================================================
Description: PFA fonts downloaded to MSDOS systems in ASCII mode will
	not be processed correctly by ps2pk.
Solution: download the PFA in binary mode or convert the PFA font to
	PFB format using t1binary from t1utils (Lee Hetherington).

=========== PROBLEM 004 ===================================================
Description: MTPK.EXE on MSDOS will change to the forge directory (%TMP%)
	after a fatal error
Solution: not yet

=========== PROBLEM 005 ===================================================
Description: compilation of mtpk on Ultrix fails
Solution: use `make -f Makefile.unx DEBUG=-g3 EXTRA=-D__POSIX'

=========== PROBLEM 006 ===================================================
Description: when filenames in PSRES.DPR (MSDOS) have a common root this
	root can be specified with
	   \d:\emtex
	while UNIX requires
	   //usr/local/tex
	Although this is not a bug I am not happy with it.
Solution: none yet

=========== PROBLEM 007 ===================================================
Description: core dump on Solaris
Solution: PATCH 007

============ PATCH 001 ====================================================
*** ps2pk.c.orig	Tue Feb 27 15:37:35 1996
--- ps2pk.c	Tue Feb 27 15:38:08 1996
***************
*** 414,420 ****
        psfile = search_file(t1inputs, psname, FATAL);
        if (verbose) msg("Font file <%s>\n", psfile);

!       if (afmname ==  NULL) afmname = newname(psfile, ".afm");
        afmfile = search_file(t1inputs, afmname, FATAL);
        if (verbose) msg("AFM file <%s>\n", afmfile);

--- 414,420 ----
        psfile = search_file(t1inputs, psname, FATAL);
        if (verbose) msg("Font file <%s>\n", psfile);

!       if (afmname ==  NULL) afmname = newname(psname, ".afm");
        afmfile = search_file(t1inputs, afmname, FATAL);
        if (verbose) msg("AFM file <%s>\n", afmfile);

============ PATCH 007 ====================================================
I got a core dump in ps2pk from the recent ps2pk15 release, and found
a fix:  the dump produced this traceback in a debugger:

#0  0xef722e2c in strlen ()
#1  0xef75fbec in _doprnt ()
#2  0xef76b3dc in vsprintf ()
#3  0x16a14 in pkstring (fmt=0x2fe28 "codingscheme=%s") at pkout.c:486
#4  0x16d68 in ps2pk_postamble (fontname=0x4c158 "lbr", encname=0x0, 
    base_res=300, h_res=300, v_res=300, pointsize=13, 
    args=0x43010 "-X300 -Y300 -P13") at pkout.c:565
#5  0x15458 in main (argc=256, argv=0x0) at ps2pk.c:497

The problem is that encname is NULL, rather than "", and an
attempt by _doprnt() to apply strlen() to it produces a segment
violation on Sun Solaris systems. 

The fix I've applied is this:

===================================================================
RCS file: ps2pk.c,v
retrieving revision 1.1
diff -c -r1.1 ps2pk.c
*** ps2pk.c	1996/03/14 15:34:05	1.1
--- ps2pk.c	1996/03/14 15:34:15
***************
*** 235,241 ****
     char *myname = "ps2pk", *psname, *psbasename, *afmname = NULL,
  	*encname = NULL, *psfile = NULL, *psfilebn, pkname[80],
  	*t1inputs, *psrespath, *resourcedb, *AFM_fontname = NULL,
! 	*encodingscheme = NULL;
  
     FontPtr fontptr;
     unsigned char glyphcode[1]; /* must be an array */
--- 235,241 ----
     char *myname = "ps2pk", *psname, *psbasename, *afmname = NULL,
  	*encname = NULL, *psfile = NULL, *psfilebn, pkname[80],
  	*t1inputs, *psrespath, *resourcedb, *AFM_fontname = NULL,
! 	*encodingscheme = "";
  
     FontPtr fontptr;
     unsigned char glyphcode[1]; /* must be an array */

========================================================================
Nelson H. F. Beebe                  Tel: +1 801 581 5254
Center for Scientific Computing     FAX: +1 801 581 4148
Department of Mathematics, 105 JWB  Internet: beebe@math.utah.edu
University of Utah                  URL: http://www.math.utah.edu/~beebe
Salt Lake City, UT 84112, USA
========================================================================