[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Groff] Critique this bold-italic private macro for man pages

From: G. Branden Robinson
Subject: Re: [Groff] Critique this bold-italic private macro for man pages
Date: Fri, 5 May 2017 12:26:33 -0400
User-agent: NeoMutt/20170113 (1.7.2)

At 2017-05-03T23:51:05+0200, Carsten Kunze wrote:
> Short answer:  We should differ manpages and other typesetting.

I don't know if this a tenable position.  I agree with several of your
recommendations, but man was never implemented with the capability or
the desire to limit the expressiveness of the input file.  *roff
requests and preprocessors were always available.  That horse was out of
the barn on day one, forty-odd years ago.

> Manpages should use the -man or -mdoc macros itself with as few as
> possible low level requests or escapes.

I agree with this as a goal; I think the .it -> itc change was a good
one precisely because it enables this.  With that change, there is no
font-change-related outcome in a man page that can be achieved with
font escapes that cannot be achieved with man macros--as long as one
does not desire to use a font that man did not historically support.

At the risk of stating the obvious, font changes in man pages are
ubiquitous because they are the primary and nearly exclusive convention
for distinguishing running prose from typed literals and substitutions

Man has 3 typefaces to offer us, and in a command synopsis or function
signature we use all three:

1. bold for literal text to be typed or input as-is;
2. italics for replaceable, user-determined text;
3. roman for non-literal symbols in a mini-language that tells the
   reader which pieces of text are optional, must be selected from a
   set, or may be repeated.

> If there is an ancient system with an old *roff tool, this tool should
> be replaced.

I'm having trouble reconciling your and Ralph's opinions here; I may
have to pick sides.  ;-)

> To answer your question:  Simply use \(lq etc. (in manpages) and
> assume the tool supports it.

I find it hard to agree.  Undefined character escapes expand to nothing
and that's seldom going to be satisfactory to conscientious writers who
care about the result.

Such people are going to want their man pages to render in some
satisfactory way whether the formatter is GNU Troff or Version 7 troff
or something else.

How can we better accommodate these people?


Attachment: signature.asc
Description: PGP signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]