On 8 November 2017 at 15:36, Thomas A. Schmitz wrote:
On 11/08/2017 11:34 AM, Mojca Miklavec wrote:
I'm still not arguing that this is the most brilliant idea, but I can totally imagine a Serbian professor wanting to "auto-generate" a Cyrillic version of his book on top of the Latin edition with close-to-zero extra effort.
Ok, I can see that this may be a convenient way of producing different output from the same source; I wasn't aware of this (and I was somewhat provocative about Latex, of course :-) From a conceptional point of view, it still feels a bit hackish to do these things on the font level, because they are not/should not be tied to specific fonts - you'd have to rewrite your features or goodies or whatever they are called now for every font you want to use (and you may run into a number of funny inconsistencies in character names or even unicode slots).
Now for a bit of off-topic-ness. Trivia. (Ignoring the attempts to make our own national keyboard) we are using "Croatian" keyboard (which is probably the same as Serbian layout) which has all the relevant-for-TeX keys ({}[]\) on the third plane (alt-gr+<something>), but I learnt computer programming on an US keyboard and preferred using US layout to those strange keys in the third plane. In computer programming there's basically never the need to use non-ascii characters. And in writing texts in native language there's no need to use those strange backslashes, so life was mostly good until I started using TeX in UTF-8. Back then I was basically switching the keyboard a couple of times per sentence (if not per word) and somewhat hated typing any TeX in native language for that reason. Then I switched to Dvorak and made myself a special layout. Now I have all the special keys from US keyboard easily accessible and all those strange non-ascii character on the third plane (alt-gr-C to get "Č"). That works much better for me now. So at least I know the pain of constant need of switching the layouts. Nevertheless I would still say that it makes more sense to put some effort to get nice UTF-8 documents. (Except, again, giving Serbian a bit of an exception due to the fact that the document would still be valid and perfectly readable in its Latin form.) One could argue in the other direction as well. It should be pretty straightforward to "transliterate" all ConTeXt commands into Cyrillic (ok, I have no clue what people usually do with q, x, y, w, ... but I'm sure there's a solution for that as well) and simply use English commands in Cyrillic script to simplify typing :) :) :) Mojca