sebastiangreger / kirby3-commentions

A versatile comment system and integrated Webmention endpoint for Kirby 3.
MIT License
58 stars 7 forks source link

Final task list for 1.0 #55

Closed sebastiangreger closed 4 years ago

sebastiangreger commented 4 years ago

Optional/tbd:

Before launch:

Things/ideas moved to possible future updates:

fabianmichael commented 4 years ago

@sebastiangreger I published the sibling plugin at https://github.com/avoskitchen/kirby-sanitizer, please give it a try and tell me if it works for you.

sebastiangreger commented 4 years ago

I published the sibling plugin at https://github.com/avoskitchen/kirby-sanitizer, please give it a try and tell me if it works for you.

Works like a charm; I updated the Readme accordingly: https://github.com/sebastiangreger/kirby3-commentions/commit/b039c1fb3e524cdc0fee6f787d96d6394be8c992

sebastiangreger commented 4 years ago

Convert semantic inline markup (em, strong, del) to presentational equivalents (i, b, s), analogue to transformation bevavior of headlines.

This is an interesting one. I've been thinking about it back and forth - there are good arguments for and against. Since comments presented are merely "citations", I too believe that the original semantics are not so important. And we indeed already "loose" semantics as we render headlines as paragraphs. On the other hand, there is not really much of a benefit in changing them, as visually they will be the same, and for assistive technology they would simply be described more "visually" rather than based on their meaning.

Since this is not a critical feature, I would suggest to leave this as is for now - unless you have strong feelings about it and can spare the time; no objections from my side, of course.

fabianmichael commented 4 years ago

@sebastiangreger No, no strong feelings, just thought it would lead to better consistency across the plugin. But in practice, it should not have any effect whatsoever. As far as I know, most screen reader (for example) are ignoring <em>, <strong> etc. and do not pronounce their contents differently.

I’ve been using <em> and <strong> exclusively when XHTML was still a thing and found it to be some kind of relief, when the definitions of <b>, <i> etc. where updated for HTML5 and the whole thing finally made sense. However, this is probably irrelevant for most use-cases and I bet that these are not used correctly in most cases either. Reminds me a bit of <acronym> vs <abbr> or the <q> element. Unless you are dealing with scientific documents, written by experienced authors, you cannot be sure that people really understand the differences between all of these. Maybe it’s just fine to keep tags as is and jsut do sanitization.

I just checked the list again and added a transformer for turning <strike> elements into <s>, because the former is obsolete in the HTML5 standard.

sebastiangreger commented 4 years ago

Great, thank you for those recent commits! :bowing_man:

Now we are really close. I did another editing round on the readme, fixed some small issues I discovered as I tested with "real-life" data, and unified the version number to be 1.0.0-rc.1 - sounds nicer than the eternal "beta" this had been in for a year now ;)