- From: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
- Date: Mon, 27 Nov 2017 21:09:31 +0100
- To: "www-style@w3.org" <www-style@w3.org>
I would like to come back to some of our shorthands. Let's take for example the 'border' shorthand. Section 4.4 of Backgrounds & Borders says it resets the 'border-image' property and the reason given by the prose is clear and valid. But I am seeing a painful side effect in my Wysiwyg editor BlueGriffon, and that side effect is triggering MASSIVE feedback from BlueGriffon users. To understand it, use Firefox (preferred) or Chrome to browse the following URL: http://glazman.org/tmp/border-shorthand.html In that test, I'm using the CSS OM to add 'border-top-color: blue;' to the declarations of existing style rule #a { border: thin solid red; } Since we had a 'border' shorthand, doing that makes a 'border-image: initial' appear. For us spec freaks, that's normal and understandable. If I believe BlueGriffon user feedback, it seems far less normal for many web authors... Since the values of 'border' can't set the border-image, they just don't get it and find the spec, I quote, bizarre, weird, strange, complicated, cryptic, bad, geeky, end of quotes, when I send them a link. Only one replied "ah, ok, I understand". So I really wonder if our choice here of allowing some shorthands to reset properties they cannot set was reasonable and if we've not sacrificed easiness of authoring for an edge case that was not worth it, and if a note in the spec saying a 'border' property will apply the desired styles only if 'border-image' is also reset to its initial value was not a much better, because more coherent, design. Our current design is understandable, but weird. </Daniel>
Received on Monday, 27 November 2017 20:10:06 UTC