Re: troff Memorandum Macros documentation derived from the paper "MM - Memorandum Macros"

[G. Branden Robinson]

At 2021-08-09T23:28:57+1000, Damian McGuckin wrote:
> On Mon, 9 Aug 2021, G. Branden Robinson wrote:
> 
> > There is clear evidence of tmac.s in Version 6 Unix (1975)[1].
> 
> PWB Unix is 1977. But get input from someone who was around at the
> time.

I like documentary sources.  Version 10 Research Unix has a file
"bibliog.a"[1] which documents some even earlier provenance for ms and
mm.

.TL
(1028) Typing Documents on UNIX
.AU
M. Lesk
TM 74-1274-24
A description of TROFF and NROFF macros for
producing papers in normal BTL format.

.TL
(1098) PWB/MM - Programmer's Workbench Memorandum Macros.
.AU
J. R. Mashey
D. W. Smith
March 1, 1976
TM 76-9144-1
This memorandum is a user's guide and reference manual
for PWB/MM, a general purpose package of text formatting
macros for use with the UNIX text formatters NROFF and TROFF.

And here's one I'd be curious to see, but which I wouldn't be surprised
to learn had its last copy wiped off of 9-track tape long ago...

.TL
(1091) A Secretarial Typing System Using UNIX.
.AU
E. F. Engelbert
February 27, 1976
Memorandum For File

I would also note that the version of the ms manual in widest
circulation, Lesk 1978 (appearing in Volume 2 of the Version 7
Manual)[2] explicitly identifies itself as a revision of a 1974
document.  To my regret I've never been able to scare up a copy of that
1974 version, and I can't remember now if I've seen the "Typing
Documents on UNIX and GCOS" version, or only references to it.

> Yes, Sorry. Brain switched off. Nightime here.

It sometimes feels like 10% of my commits to groff Git are typo fixes
for my own earlier work.  No matter how many times I proofread, my brain
refuses to recognize some errors until the whole world can see them.

> > groff mm(7) is already heavily extended; I think we should go ahead
> > and add strings for super- and subscripting unless someone can come
> > up with a really good reason not to.
> 
> Go for it.  My only reason for using eqn(1) is tht 99% of my needs are
> because of mathematics. You examples of footnotes to text and
> chemistry formulae are clear examples where they are not needed. That
> said, I never use superscripts for footnotes and use the 'roman'
> keyword in eqn(1) for chemical formulae.

Yes--outside of a mathematical context, I blanch at the thought of
telling someone they need to set up delimeters and say the word "roman"
just to get a word superscripted.

Regards,
Branden

[1] https://www.tuhs.org/cgi-bin/utree.pl?file=V10/doc/bibliog.a
[2] http://bitsavers.trailing-edge.com/pdf/att/unix/7th_Edition/\
    UNIX_Programmers_Manual_Seventh_Edition_Vol_2_1983.pdf
    (p. 125, PDF page 134)

    This 625-page anthology of articles is a gold mine of practical and
    historical information.  I imagine that it was one of those books
    that got stolen out of university libraries for installation into
    Cool Kids' Clubhouses across the land, whose denizes claimed that
    the man pages were all you needed to learn Unix, and if you couldn't
    master the system from reading those at a terminal, then you were
    intellectually deficient and unworthy of inclusion in the community
    of software sharing whose gates they kept.

signature.asc
Description: PGP signature