Full text of YASC (nroff/troff question)

[Spyridon Triantafyllopoulos]

Doug Gwyn writes on my comment on inches/#characters on n/troff:

[original article (mine :-)):

> I have been using Honeywell's "runoff" text   
> processing system on a 68/80 Multics for a long time and all arguments
> are in number of characters or line. Same thing on the WORDIX word processing
> system on the IBM PC/XT. The documentation IS cryptic, and 
> No one says that you have to use the little "i". I got it all OK 
> now, but I guess I'm going to implement macros for Multics runoff.
> It's a pity, n/troff has such a potential but screwy docs....

[Doug's answer]

  Uh-huh.  How big IS a "character", anyway?  My printer and typesetters
  have several character widths, even at the same point size.  I bet those
  Brand-X word processors you like so much would have fun with these devices.

[Me again]

Well, I don't use a typesetter anyway. We have QMS laser printers that
take care of all these little bittle stuff. The good thing is that it is
much simpler to use Honeywell's runoff + laser.ec (a laser script) compared
to anything I have seen on the troff documentation (or whatever they call it).
WORDIX is also very good. It is almost compatible with runoff and gives
very good results on our TI 855 printers. 

geoff at desint.UUCP (Geoff Kuenning) writes:

  Uh, gee, have you tried ".ll 19c"?  T/nroff knows about a number of different
  units:  inches, centimeters, characters (ems and ens), and points are a few.

A saving opinion follows from Kentucky:

  Actually .ll is NOT on the list of nroff commands that work with -ms.
  So it won't work.  What will happen is that the next PARAGRAPH will have
  that page width set, and the rest of the document will have the default.

  David Herron;  ARPA-> "ukma!david"@ANL-MCS or david%ukma.uucp at anl-mcs.arpa
 
[me again]

Specifying character # is much more convenient if you work with standard
printers (no typesetters). But the convenience of typesetting is not 
overlooked. I guess the president of DEC was right about the 5 feet of
VMS manuals after all.... (a little philosophy: I guess we have to live
with the documentation. A while ago I was using the Multics Relational
Data Store, a DBMS for the Multics system. There was a lot of screwy 
documentation over there too. An inquiry to the local guru faculty 
brought: "with only 28 Multics sites around, what do you expect!!!")

-- Spiros

Spiros Triantafyllopoulos  <> USENET {ut-sally, akgua}!usl!sigma
Computer Science Dept, USL <> CSNet  TriantafyllopoulosS%usl at csnet-relay.ARPA

             "This file contains no opinions whatsoever"