Re: removing empty parens after function names

[Bruno Haible]

Karl Berry wrote:
> Personally, I kind of agree with you that in this instance, the parens
> to mark functions is more useful than not.  I don't know why rms felt so
> strongly about it that he put it into standards.texi (umpteen years ago).
> ...
> If you want to ask rms about it, feel free.

Thanks for telling your opinion frankly.

I actually agree with having standards for uniform documentation. And in
the documentation (at least in 2 of the three output formats for documentation:
HTML and PDF) there are means to distinguish a symbol from a normal word,
by use of a different font. So I don't really question RMS' statement about
_documentation_.

But I don't agree with applying the same criteria for source code, because
here we don't have the option of using different fonts to denote symbols,
and as Paul said, putting 'quotes' around every use of a function name in
a code comment would be excessive.

Paul, I know you prefer to have the same conventions for doc and for source
(you said that while discussing "file name" vs. "pathname"). I disagree
here.

So I'll rework your patches to the doc part (that includes the "Description"
field of modules), but ignore the mass of changes to source code, autoconf
macros, test suite, config.guess etc. This reduces the amount of the patch
from 339 KB to 56 KB, which is an acceptable size to me.

Bruno