On Friday 2017-03-31 12:11 +0800, Tommy Kuo wrote: > **Summary** > > I am intent to implement the property `line-height-step`. And it would be > disabled behind the pref `layout.css.line-height-step.enabled` by default. It > is a property to make authors create the content with vertical rhythm easier. > > **Link to standard** > > CSS Rhythmic Sizing > <https://drafts.csswg.org/css-rhythm/>
So in the discussions in the working group, I've been somewhat skeptical that this feature does a good job of addressing the design use cases that it's intended to address. I think there are two issues with it: First, in many cases, the goal of setting up vertical rhythm is to ensure that different (adjacent) pieces of text have the same rhythm, where by different pieces of text I mean things like text on opposite pages, text in different columns, text in a float and in the normal flow, or similar things. 'line-height-step' only addresses this use case if the developer is *extremely* disciplined to ensure that nothing other than line-height (e.g., border, padding, margin) causes vertical spacing. My sense was that the line grid proposals were significantly better at addressing this use case (since they established an overall grid), although they were harder to implement (because they complicate things like sliding an element's position in the block axis). Second, I think the rhythm that line-height-step sets up probably isn't the most desirable one. I'd expect that it is preferable to have the baseline of each line match the rhythm, but line-height-step does not ensure this. For example, if most lines have a line box height that matches the step, but there is something tall on a line (say, an image protruding upwards) that makes its line box height initially (without the adjustment from the step) 2.1 times the step height, then the line with the tall thing on it gets an additional 0.45 multiple of the step height on each side, which means the baseline of the text is 0.45 times the step height above the rhythm, so instead of having an appearance of skipping two lines and then having a baseline at the correct position, you get (ignoring the thing that protrudes above) the appearance of skipping 1.55 lines, having a misaligned baseline (0.45 above the rhythm poisition), and then skipping the remaining 0.45. I'd have thought it would be preferable to put all the extra space above in this case in order to keep the baseline matching the rhythm. (In some cases this would require an additional multiple of the step height.) I might be wrong about this, though; it's worth consulting designers about what behavior is actually desirable. -David -- π L. David Baron http://dbaron.org/ π π’ Mozilla https://www.mozilla.org/ π Before I built a wall I'd ask to know What I was walling in or walling out, And to whom I was like to give offense. - Robert Frost, Mending Wall (1914)
signature.asc
Description: PGP signature
_______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
