Re: [css-text][css-conditional] language specific support for hyphenation and justification

> On 17 Jun 2015, at 16:33, John Daggett <jdaggett@mozilla.com> wrote:
> 
> This idea sounds like overdesign to me. I think @supports should have a simple implementation as it does now. I absolutely do not think we should turn it into an ill-defined way of querying capabilities, like whatever "nice" hyphenation would be for a given language.
> 
> We really should not be designing extensive hyphenation controls until we have more solid feedback about current implementations. Chrome still lacks any support for hyphenation, which is sort of baffling. But maybe work is underway. Some of the properties Dave points out would be interesting. Not really keen on word spacing, I think that's usually a crutch to work around poor hyphenation algorithms.


I think extending @support is in general an ok thing to do, we added general-enclosed to it  specifically for that purpose. I agree though that the proposals I brought up –hyphenation-language or nice-hyphenation– are probably not worth adding. The first one is really narrow, and as you and Dave said, the second has poorly defined semantics.

However, I do think some amount of language-aware detection of what the browser can do is useful to expose to authors, to let them decide if they want to use justification or not.

I am not a particularly sophisticated with typography, so personally, I'm happy to turn justification on as soon as hyphenation works, in languages that need it.

I appreciate further refinements and controls, but I don't required them before turning on justification. So to me, being able to turn on justification only on languages where it's supported would be nice.

I am still not convinced that's the right solution, but As I mentioned in reply to Xidorn, I think a pseudo class is probably a better approach than @support.

 p { text-align: left; }
 p:supports-hyphenation {
   hyphens: auto;
   text-align: justify;
 }

 - Florian