[groff] 03/03: NEWS, ChangeLog: Clarify the history of \sN.

[G. Branden Robinson]

gbranden pushed a commit to branch master
in repository groff.

commit 5cc546e12ff17ee276bb8671b1d325bed17139e2
Author: G. Branden Robinson <g.branden.robinson@gmail.com>
AuthorDate: Thu Sep 3 22:30:27 2020 +1000

    NEWS, ChangeLog: Clarify the history of \sN.
    
    The weirdness of legacy \sN parsing was in fact documented some years
    before 1992, but unfortunately not in a place where users might think to
    look for it: background material in CSTR #97, the paper describing
    Kernighan's device-independent refactor of AT&T troff.
    
    In this document, dated as "Revised, March 1982", Kernighan writes:
    
    "TROFF turns out to be surprisingly dependent on the Graphic Systems CAT,
    not just in the code but in its design.
    
    Some of the design dependencies are pretty obvious.  For example, the
    CAT provides four fonts (of 102 characters each) and 15 sizes.  The
    specific sizes are wired into the syntax of the language: since the CAT
    has no sizes larger than 36 points, \s46 can be uniquely decoded as a
    4-point 6, while \s36 is simply a switch into size 36."
    
    It will require grayer beards than mine to opine on what proportion of
    troff users had on-demand access to CSTR #97 in the 1980s; being of
    1982 vintage, I surmise that this paper came too late to be included in
    the commercially-sold volumes of Version 7 Research Unix manuals.
---
 ChangeLog | 7 ++++---
 NEWS      | 4 ++--
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 1616621..436ee4d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1007,9 +1007,10 @@
        To anticipate objections: Why not throw only a warning?  Because
        there isn't a warning category for supported but ambiguous
        syntax (this behavior of AT&T troff dates back to the mid-1970s
-       but apparently was not documented until 1992).  Why not throw
-       the error outside of compatibility mode too?  Because outside of
-       compatibility mode we (now) have an unambiguous parse.
+       but was not documented in the Troff User's Manual until 1992).
+       Why not throw the error outside of compatibility mode too?
+       Because outside of compatibility mode we (now) have an
+       unambiguous parse.
 
        Background: The Graphic Systems C/A/T phototypesetter (the
        original device target for AT&T troff) only supported a few
diff --git a/NEWS b/NEWS
index f36585d..7ab198b 100644
--- a/NEWS
+++ b/NEWS
@@ -24,8 +24,8 @@ o Point-size escapes of the form '\sNN', where NN is in the 
range 10-39,
   interpreted as setting the point size to the single digit value N,
   which ends the escape.  This change eliminates a quirk in the language
   grammar that dates back to the mid-1970s (AT&T troff by Ossanna) but
-  apparently was not documented until 1992 when Kernighan updated CSTR
-  #54 for device-independent AT&T troff.
+  was not documented in the Troff User's Manual until 1992 when
+  Kernighan updated CSTR #54 for device-independent AT&T troff.
 
   The form '\s(NN' is accepted for two-digit point sizes in all known
   troffs.  The form '\s[NNN]' accepts a numeric expression of variable