Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What's new in CSS4

webmonkey.com

Post reply on HN
Print view View original

11–20 of 33 posts

Re: What's new in CSS4

#11

Post by ma2rten »

alexchamberlain wrote: post #7
chime wrote: post #4

Parent selector is cool but I've needed it at most 2-3% of the time when designing a site. As of now, it doesn't appear to support things I really need in almost all mid-large site layouts: variables, mixins/functions, operations, and namespaces. This means even 5 years from now, I'll need http://lesscss.org/ because browsers with CSS4 won't support variables. How does one go about suggesting to W3C that they should…

Why slow down CSS with constructs that can be achieved by prior compilation and a few Makefiles?

You could still compile css to backwards compatible, compact css and test and debug in the browser without effort?

Re: What's new in CSS4

#12

Post by StrawberryFrog »

masklinn wrote: post #5
hpaavola wrote: post #3

I don't need new selectors, I need a way to specify layouts. You know, to do the same things as tables did back in the days, but without all the ugliness.

http://www.w3.org/TR/css3-multicol/ http://www.w3.org/TR/css3-grid-layout/

Are the ms-grids and the multicolumn layouts in CSS4? It doesn't look like it from that draft. Are they just free-floating proposals at this stage?

Re: What's new in CSS4

#13

Post by alexchamberlain »

masklinn wrote: post #9
alexchamberlain wrote: post #7

Earlier quoted context omitted.

Why slow down CSS with constructs that can be achieved by prior compilation and a few Makefiles?

Why require adding a fiddly and complex toolchain making ramp up and usage of these features more complex than it has to be, and harder to troubleshoot, when it's something making production easier to maintain in the long run? Because unless you suggest adding source maps to CSS (even though they still are not in JS), a preprocessor generating CSS makes CSS far more annoying to debug or play around with than native c…

I wouldn't mind a Chrome/Firefox extension to play around with extended CSS, but in production, I'm on the side of higher performance in Web Browser, which comes from simple style syntax.

Re: What's new in CSS4

#15

Post by lukestevens »

masklinn wrote: post #5
hpaavola wrote: post #3

I don't need new selectors, I need a way to specify layouts. You know, to do the same things as tables did back in the days, but without all the ugliness.

http://www.w3.org/TR/css3-multicol/ http://www.w3.org/TR/css3-grid-layout/

Yes, exciting that Gird Layout is shipping in IE10. Hard to believe MS will be the first to ship a true CSS layout solution.

Solutions like display:table enforces source order, which to me is pretty undesirable (e.g. when A/B testing layouts by switching a CSS file).

Re: What's new in CSS4

#16

Post by masklinn »

StrawberryFrog wrote: post #12
masklinn wrote: post #5

Earlier quoted context omitted.

http://www.w3.org/TR/css3-multicol/ http://www.w3.org/TR/css3-grid-layout/

Are the ms-grids and the multicolumn layouts in CSS4? It doesn't look like it from that draft. Are they just free-floating proposals at this stage?

The reality is that there's no such thing as CSS4, and no such thing as CSS3 either: starting with CSS3 the spec was "modularized", meaning it was broken up in a bunch of separate and mostly-independent "modules", which are grouped in a "level" but are specified on their own[0].

So the situation is that TFA talks not about CSS4 but about "Level 4 Selectors" (a draft), talking about "CSS4" is language abuse. The specs I linked are Level 3 specs, but independent of selectors and of one another (columns are a CR, grid is a working draft).

To sum up, your question does not really make sense. Grids and columns are part of "CSS4" insofar as they are part of "CSS3" through being Level 3 specifications, but that's completely meaningless. Level 4 selectors could become recommandations before (level 3) grids reach that state, or not. There's no dependency. (columns are already supported by all modern browsers except MSIE).

I'd recommend using (and abusing) http://caniuse.com/ to see the implementation state of each spec, it's become especially important with modular level 3 specs as implementors are free to add one level 3 spec and ignore an other.

[0] You can see a list of specs at http://www.w3.org/TR/#tr_CSS, note how some Level 3 specs are Recommendations, others are Candidate Recommandations, yet others are Last Calls or Drafts.

Re: What's new in CSS4

#17

Post by lukestevens »

Web Monkey calling this "CSS 4" seems a bit misleading -- it's a spec for Selectors Level 4; I don't think there is a monolithic "CSS 4" in the same way there isn't a monolithic "CSS 3". I assume the wide variety of CSS3 modules will be mostly unaffected.

Re: What's new in CSS4

#18

Post by masklinn »

alexchamberlain wrote: post #13
masklinn wrote: post #9

Earlier quoted context omitted.

Why require adding a fiddly and complex toolchain making ramp up and usage of these features more complex than it has to be, and harder to troubleshoot, when it's something making production easier to maintain in the long run? Because unless you suggest adding source maps to CSS (even though they still are not in JS), a preprocessor generating CSS makes CSS far more annoying to debug or play around with than native c…

I wouldn't mind a Chrome/Firefox extension to play around with extended CSS, but in production, I'm on the side of higher performance in Web Browser, which comes from simple style syntax.

> I'm on the side of higher performance in Web Browser, which comes from simple style syntax.

Constant parsing and replacement would add pretty much no overhead to CSS parsing in the grand scheme of things.

Re: What's new in CSS4

#19

Post by masklinn »

lukestevens wrote: post #17

Web Monkey calling this "CSS 4" seems a bit misleading -- it's a spec for Selectors Level 4 ; I don't think there is a monolithic "CSS 4" in the same way there isn't a monolithic "CSS 3". I assume the wide variety of CSS3 modules will be mostly unaffected.

Correct, as far as I know.
Post reply on HN