Re: http://www-cs-faculty. stanford.edu/∼uno/programs/tcalc.w.gz

[Andreas Scherer]

Philip,

> Looking at Don's /Fundamental Algorithms /(1968/9), where he introduces
> MIX, it is obvious that at that time he was thinking very much in terms
> of assembly-level programming
Not much has changed in the most recent (and hopefully future) editions
of TAoCP, apart from the underlying processor architecture and the more
sleek assembly language: http://www-cs-faculty.stanford.edu/~knuth/mmix.html

> TeX, it is generally acknowledged,
> is not the easiest program to modify, even though its documentation
> (through the medium of WEB) is exemplary.
I can't say much about the coding quality of TeX, the program, having
personally dealt only with the excellent work by others done already in
Web2c and TeX Live, hiding the nitty-gritty stuff behind a simple
'make', but other Knuthian codes -- including SGB and MMIX, though both
written in CWEB and published in dead tree form -- leave a lot to wish
for in terms of "industrial strength".  ;o)

Cheers,
Andreas