Re: [rfc-i] No, constraining to a custom SVG profile is not trivial

[Carsten Bormann <[e-mail]>]

On 2020-01-22, at 04:31, Brian E Carpenter <[e-mail]> wrote:
> 
> I agree that the lack of coloured shading is annoying for many people, but *using* shading is annoying for people with limited colour vision.

In Germany, we have colored traffic lights, even if the coloring is less useful for people with limited color vision.

Accessibility is important, and faithfully reproducing colors is near impossible, but still many archival journals manage to allow color in diagrams.

It seems we are making the same mistake here as with the expansion beyond ASCII:  by adopting a weird, idiosyncratic subset of the stable state of the art (UTF-8, SVG) we are incurring most of the problems (as well as some additional ones specific to our idiosyncrasies) while enjoying a small part of the benefit.

Instead of coming up with more weird technical rules, we should focus on handling potential problems in the editorial process.  Guidelines can help here (say, go for more than 60 % contrast in the Lab color model, don’t make semantics depend on non-Latin characters outside Math), but not all will be enforceable by tools, just as with other editorial guidelines (e.g., go for stable references).

Grüße, Carsten