[bug #62494] [grotty] Remap ~ and ^ to their ASCII equivalents

[Ingo Schwarze]

Follow-up Comment #1, bug #62494 (project groff):

Indeed.  I said before, and will say again in this context, that i would
prefer keeping these mappings in manual pages (i won't type any punctuation
characters here for fear that the weird Savannah web interface might mangle
stuff):

input -> output

ASCII 0x27 -> U+0027 apostrophe
ASCII 0x2d -> U+002D hyphen-minus
ASCII 0x60 -> U+0060 grave
ASCII 0x7e -> U+007E tilde
ASCII 0x5e -> U+005E circumflex

and *not* U+2018, U+2010, U+2019, U+02DC, U+02C6
like in general-purpose roff typesetting.

At the risk of sounding like a broken record:
The reason simply is that this is what manual page authors and maintainers
have become used to during the last two decades and that very large numbers of
manual pages are written expecting this.  So "fixing" this to conform to
general-purpose roff typesetting conventions unavoidably breaks huge amounts
of real-world manual pages.  Some large corpusses (e.g. the OpenBSD base
system) will eventually get fixed (even though that amounts to a horrific and
not very useful make-work project for myself), but i expect in most other code
bases, this will remain broken for many years to come, if not for decades.  It
also puts additional stress on manual page writers, who are often programmers
with no particular interest in typesetting and who will have to learn this as
a new and rather counter-intuitive rules, which will add to the dislike that
some feel for editing documentation and keeping it up to date in the first
place.

Unfortunately, when i said this in the past, Branden strongly opposed my
argument and not many others spoke up at all.


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?62494>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/