At 2017-09-04T22:35:40+0200, Werner LEMBERG wrote: > > > Stuff like `this' should be burned with fire and extreme prejudice. > > It's been ugly for over 20 years and is indefensible. > > I disagree. While it shouldn't be displayed after processing with > groff, it is very convenient for input even today.
Ah, yes, I let my passion obscure my context. Yes, I think TeX's convention of quoting ``like this'' is eye-wateringly ugly, but I wouldn't change it. Because it's _input_. What I don't want is computer _output_ that looks like that, for instance in diagnostic messages like the one from Bjarni that I quoted when I spouted off. > On my US keyboard it is still non-trivial to enter ‘ and ’. Yes, I admit I had to look in /usr/share/X11/locale/en_US.UTF-8/Compose for how to do it, and I've gone to the trouble of setting up a Compose key, which many en_US people haven't done. > Additionally, some common fonts like DejaVu have glyph representation > forms for those two glyphs that look almost identical, which is an > invitation for mistyping. >wince< Yes, whatever we do, let's NOT learn from https://websec.github.io/unicode-security-guide/visual-spoofing/ . You think they'd welcome patches? ;-) > Note that using only straight quotes like '...' is not a good > replacement for input IMHO; I much prefer paired quotes. For input, absolutely. Non-identical paired delimiteres are one reason why $() is better than `` for process substitution in the shell. > Note also that I don't object to change `...' to '...' in code > comments or the like. Whatever you do, however, it should be > consistent. Fully agreed. -- Regards, Branden
signature.asc
Description: PGP signature
