On 18 February 2016 at 10:22, Hans Hagen wrote:
this brings up the question: would users (here) start using real math unicode input if we had a monospace math font?
On Mac (TextMate, but I assume other editors would behave the same) the system probably does some character substitution, so as long as I have any font that contains that particular character, I can see that character in the editor. There is no need for a special huge font because the system takes care of it to some extent. This is probably different on Windows and Linux though, so I cannot say that it wouldn't matter, it just wouldn't matter to me as long as I'm using OS X. I have my own keyboard with Greek letters mapped to AltGr+g+"latin equivalent of the letter". So I always use Greek letters rather than \alpha, \beta, ... to typeset symbols. Those are easier to read than \controlsequences. But I probably wouldn't bother entering "unicode math" characters for Greek letters until I would have to deal with frequent mixes of different styles (italic, bold, ...) which would also introduce the need for an easy input method. I ofter use a bunch of other symbols directly (like \sim, logic symbols, ...), but honestly I cannot imagine typesetting math exclusively in Unicode. Mojca