Regions to be cheerful?
The Web has been buzzing with the news that Blink has said “nope” to Adobe’s CSS Regions spec. Here some hand-picked links about that, and the general state of CSS Regions spec.
- Reversing course, Google rejects Adobe Web publishing tech
- Blink-dev discussion on Regions
- CSS Regions Considered Harmful by Opera CTO Håkon Wium Lie. I don’t share all his concerns, although I find the CSS Regions spec “smells funny” (I can’t articulate it any more coherently than this) although I completely agree that the need to be able to fragment text across separate boxes is something that needs solving.
- CSS Regions Matter – counter-argument to Håkon by Sara Soueidan
Other suggestions for text fragmentation are available, such as CSS Overflow Module Level 3, CSS Fragmentation Module Level 3 and (related) CSS Figures.
- HTML is too complex – “unless there is an immediate visual or behavioural benefit to using an element, most people will ignore it”. Some real food for thought in this sentence.
- A World Managed By Apps Is Closed For Those Without A Smartphone – “Every time you make a service or device that can only be managed from an app, you are basically adding to a systematic poor tax. You make it easier for those comfortable, with great smartphones in their hand, to get shit done, while not spreading that benefit to those without the magic box. You deepen economic entrenchment.”
- Why is Progressive Enhancement so unpopular? by fellow old-timer, Drew McLellan
- [HTML Imports]: Sync, async, -ish?. If you’re importing a web component (you hypercool ninja, you) should the whole render be blocked until the component can be drawn? Of course not, argues Jank Archibald persuasively. Elsewhere he writes, “Instead, we should follow the default behaviour of <img>. An <img> doesn’t block parsing or rendering, the image appears when it is loaded. The developer can reserve an area for the image to change the reflowing behaviour.” Not as daft as he looks, that Jake.
- Why does this spec replicate HTML features? – Imagine we have a manifest file for web apps, that pretty much duplicates what HTML can do with <meta> elements already. If both are present, which should “win”?
- Nine Things to Expect from HTTP/2 by Mark Nottingham, chair of the IETF HTTPbis Working Group. So he knows.
- will-change: a CSS hint to browser that you know its appearance will change, so it can make any optimisations, eg paint it to another layer immediately for faster animation. Replaces hacks like
- Pointer Events Progress: Mozilla and Blink Communities Reach a Significant Engineering Milestone – synergies between Microsoft, Blink and Mozilla. Who’s missing, I wonder?
Scampi Bug Yeti
As Mike Taylr points out, “Scampi Bug Yeti” is an anagram of “Spec Ambiguity”. Sometimes, people moan that the specification process of web standards is grindingly slow and laborious with too much detail etc. But the reality is that specs that are amibiguous or not tightly defined cause problems with interoperability. For example,
- When should a StyleSheetList collection update? asks Mike Taylr – “The CSSOM spec isn’t crystal clear on which behavior is correct, so they might both be considered valid.”
- Animating an Offset Value in SVG by Peteypoo Gasston – “The problem is due to a vagueness in the SVG spec, and a differing interpretation by browser vendors.”