[latexrefman-discuss] \textascenderwordmark and \textcapitalwordmark and ligatures

[Karl Berry]

    \textascenderwordmark and \textcapitalwordmark, which are by default
    undefined in LaTeX.

(Also \textascendercompwordmark.)
They're defined by the T1 and TS1 encodings.  Grep the LaTeX sources.

    the description for \textcompwordmark is quite unclear 

Sure, no doubt it can be improved.

    prevent some ligature

I don't remember the details, but I'm sure they were not added
capriciously.  It may have had more to do with hyphenation patterns than
ligatures.  Perhaps some articles or documentation about the T1 encoding
(e.g., in TUGboat) would explain more.  I can't research it now, sorry.

    - if the input is in UTF-8, then one can also directly use the right
      symbols, œôòó or œôòô for -- or ---. 

Yes, but explaining Unicode is not the job of this manual.  You could
say "or use Unicode character xxx" with virtually every control sequence
that inserts something.  I don't want to.

    - With xelatex engine ligatures are not enabled by default, 

I know, but it is definitely not the job of this manual to explain
XeTeX, and even less so (what you're talking about here) the fontspec
package.  Maybe at some point in the indefinite future when all of LaTeX
has been covered, but let's not go there now.

And then there is LuaTeX, which is (becoming) even crazier ... not our
problem.  We have enough problems.

    - Some fonts do not have the --, ---, and suchlikes ligature, and that
      is the raison d'être of macros like \textemdash.

The manual has the line:
@item \textemdash @r{(or @code{---})}

I don't see that anything more needs to be said.  It's a reference
manual, not a "cover every possible case" manual.  Adding something like
"except when not supported by the current font" would have to be stated
in many many places, and just seems like useless verbiage to me.  -k