X-VM-v5-Data: ([nil nil nil nil nil nil nil t nil]
	[nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil])
Date:         Fri, 6 Apr 90 18:03:29 CET
Reply-To: LaTeX-L Mailing list <LATEX-L%DHDURZ1@unido.bitnet>
From: bbeeton <BNB@MATH.AMS.COM>
Subject:      Re: MakeIndex and an apology
To: Rainer Schoepf <BK4@DHDURZ1.uucp>
Status: R
X-Status: 
X-Keywords:                  
X-UID: 70

regarding c vs. pascal, while it's true that knuth didn't specify
that the output of web was to go to a pascal compiler, he did choose
pascal as the basis for his web at least partly because the machine
he was working on didn't have a decent (or maybe not eny) c compiler,
and he believed that pascal was the most widely available of the
other possibly suitable languages.
frank has a very good point though.  most of the ibm world and some
other architectures haven't got any available c compiler, and if a lot
of tex utilities are written only in c, then they are shut out.
(compare oztex, to all reports an attractive implementation, which
can be run, but not modified, by those who don't have a modula-2
compiler, which is a much larger population than those who don't have
c.  sebastian, would you even be in the tex business if knuth had
chosen modula-2?)
sebastian has made a strong argument for pascal, though, in pointing
out that web2c exists.  though it's an extra step, i suggest that this
is the way to go for tex software that is truly meant to be universally
available.
the ibm mainframe market is still the largest in this country, and
a potential growth area for tex.  i think it would be a bad thing to
exclude it categorically.  just gives the present nay-sayers another
reason to say, "oh tex.  that's just an academic exercise."
                                                -- bb
(i apologize for prolonging this conversation, and hope i haven't
contributed too much to the flames.)
-------