Re: [csswg-drafts] Let’s Define CSS 4 (#4770)

I'm not sure I understand this assertion
>Yet I'm not convinced that it will help. The main basis for this idea seems to be the success story of the marketing terms HTML5 and CSS3. Which are undeniable. But that doesn't mean it is repeatable. Those version bumps had incredible tangible value due to a unique moment in the web's history:

in light of this statement. 

>Despite some of my skepticism, I want to express that in recent years, CSS has been extended with truly revolutionary features. Just grids alone is something people 5 years from now will still be exploring and learning, and 10 years from now, people will invent new layouts based on what this foundational technology makes possible. It's that extensive and powerful.

I agree that 2 to 3 was a jump of the magnitude that we may not see again. That doesn't mean versions have no benefit unless they are of that magnitude. It means we shouldn't use them will nilly (hence my off the cuff aversion to yearly numbers... some years might pass with little of significance meeting the combat criterion). 


>We're used to a modular, fast-moving web where browsers auto-update. There is no standardized major version test to measure against, nor do browsers implement based on such grouping of modules. Hence, developers think feature-based, not module or version-based. They use caniuse.com to check what works.

Well, we HAVE to. As I've said upthread, this model puts the onus on devs to keep  up with each and every feature with nothing to differentiate between bigger, more impactful ones and minor, edge case features. You need to parse the firehose to try to get the former and not let the latter distract you. In practice, what I find myself doing is ignoring the firehose of new things and only paying attention to what bubbles up. That, of course, could just be me. 

-- 
GitHub Notification of comment by rickgregory
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/4770#issuecomment-591670791 using your GitHub account

Received on Wednesday, 26 February 2020 22:11:54 UTC