[csswg-drafts] [css-multicol-2] Is Column-height bigger than height useful? (#12794)

frivoal has just created a new issue for https://github.com/w3c/csswg-drafts:

== [css-multicol-2] Is Column-height bigger than height useful? ==
(This is a spin-off conversation from https://github.com/w3c/csswg-drafts/issues/8721)

It's very possible to set up the `column-height` property on an element to a size that is larger than the actual block-size of the element. Is that useful?

If not, should we cap the used-value of column-height to the block-size of the content-area of the multicolumn container?

A counter argument would be that it's perfectly possible to create a grid that's bigger than its grid container, and since we don't block that, we shouldn't block a multicol that's bigger than it's multicol container either. If it overflows, so be it, and authors who don't want overflow should set things up so that there isn't, and if anyone does want overflow, they'll be happy we're letting them have it.

However, it seems to me of multicol calls for a different conclusion. When you ask for a 50em `column-width` in a element with a `width` of 30em, you get a column which it's 30ems wide, not 50em. Just the same, I think if an author asks for a column-height that's bigger than what can fit, they should get the biggest thing that can fit, and no more.

To me, that is related to the fact that multicolumn fragments its content. This avoids unnecessary overflow from the columns. Either we have sufficiently little content that reducing the column size doesn't lead to overflow, which is great, or there is too much content to fit, and then we trigger fragmentation, which is better than simple geometric overflow.

Although technically distinct, this is conceptually related to the pre-existing spec requirement that: « _If the multi-column container is paginated, the height of each column is constrained by the page […]_ ».

We could have let the column-height exceed (and overflow) the page, but it's a lot more useful not to. I'd argue we should follow the same logic here.

Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/12794 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Monday, 15 September 2025 07:25:09 UTC