[texhax] inconsistency between pdftex and tex

Ian Collier imc at comlab.ox.ac.uk
Mon Oct 20 16:48:37 CEST 2003

"Philip TAYLOR [PC87S-O/XP]" told texhax:
[snip comments from section 2 of tex.web]

>I would argue (indeed, I do!) that the key phrase here is
>"software systems that are fully compatible with each other";
>by deliberately interpreting a leading comment as a pragmat,
>the software earlier cited is behaving in a manner which is
>/not/ compatible with Knuth's reference version, and therefore
>should not be called "TeX".

Personally I think that this `extension' defies neither the
spirit nor the letter of the law and I am somewhat curious
as to why this exercises you so much given that you aren't
actually the trademark holder and also (I'm guessing) that
this difference has almost certainly never affected your use
of TeX and probably never will.

Knuth says nothing about what format TeX loads when you start
it up and very little about what operating system command you
must type in order to get it all started - though he assumes
that there will be an easy way to start TeX with the plain format 
(see the last paragraph on page 25 of the TeXbook).  Therefore,
the fact that TeX might process your input file with latex.fmt
on one system and plain.fmt on another is /not/ an issue of
software system compatibility.  What is at issue is whether
"TeX with the plain format preloaded" will produce identical
DVI files on different systems.

Note that even with "%&"-interpretation built in and turned on
there are at least two ways of processing your file with the
default format, one of which is the official method of invocation
described on page 25 of the TeXbook (namely, typing "tex" then
entering your file name at the prompt; the other way is to say
"tex \&plain filename".  Being able to type "tex filename" is
a shortcut that /might/ work on your computer (see the top of
page 26).

(And I think that will be my last word on the subject.)
---- Ian Collier : imc at comlab.ox.ac.uk : WWW page below
------ http://users.comlab.ox.ac.uk/ian.collier/imc.shtml

More information about the texhax mailing list