Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!gatech!emory!dtscp1!scott
From: scott@dtscp1.UUCP (Scott Barman)
Newsgroups: comp.text
Subject: Re: Troff Macro Programming A: Part 1 [repost]
Message-ID: <741@dtscp1.UUCP>
Date: 2 Jun 89 16:02:45 GMT
References: <18@nx32s> <2610025@hpsal2>
Reply-To: scott@dtscp1.UUCP (Scott Barman)
Organization: Digital Transmission Systems (a subsidiary of DCA), Duluth, GA
Lines: 25

In article <2610025@hpsal2> morrell@hpsal2.HP.COM (Michael Morrell) writes:
>/ hpsal2:comp.text / scott@dtscp1.UUCP (Scott Barman) / 12:24 pm  May 30, 1989 /
>Just as a suggestion, one of the things I hacked into ditroff (many
>years ago) was to accept "legal" \s but added the ability to do a \s(xx
>where the xx can be any point size.
>----------
>This feature is present in ELAN's Eroff.  I don't know why the original troff
>decided to make \s the exception in not requiring the "(" when giving a 2
>digit argument.

You have to look at the characteristics of the C/A/T to understand:
That wonderful device ( :-) ) can only do fonts of size 6-12, 14, 16, 18, 20,
24, 28, and 36.  Therefore, it was safe (at the time) to assume that if
it encountered a 1, 2, or 3 after the \s, then it had to be a two digit
font because the C/A/T does not have a point size 1, 2, or 3.  Yes, it
is not the way things are now, but troff was written to be device dependent
at that time.

A better question is why is \f(xx allowed (C/A/T troff) when the C/A/T had
only four font positions and the fonts R, I, B, and S?
Just something to ponder....

-- 
scott barman
{gatech, emory}!dtscp1!scott