At 22:45 +0100 2001/02/11, Frank Mittelbach wrote:
> > Meta reasoning from the impression of TeX I get, developed as a series
> > smart, but quick features lacking true generality and depth, I surmise that
>that's a bold statement from you, given that TeX though being more than 20
>years old has still to find a opponent program that can produce even equal
>quality in its domain.
I don't see that it is any bold in it; if one ever has tried to do
something of generality with TeX (have you? :-)), it is sort of obvious.
>you might argue that Don as a language designer isn't really strong but
>only a part of the story (though with most of your comments my feeling is that
>you think it is all that is about anything)
I don't think he ever tried to design a language with great generality; he
had his scope, and as far as he could tell resulted TeX would suffice for
The scope has since changed, though.
> > any attempt in developing a TeX successor by enhancing it will fail.
>possibly but not for the reasons you claim
The very heart of TeX is too limited.
-- Otherwise, part of the reason that it is so difficult providing a
successor to TeX is poor documentation of its internals, and the sources
seems to not be very accessible. So I figure that those start with
enhancing TeX, just inherits its problems. Possibly, if one could first
make a clone (entirely new sources but with the same specs written with
more modern tools such as Bison & C++), one could get an idea of how to
write a new version.