- From: CSS Meeting Bot via GitHub <noreply@w3.org>
- Date: Wed, 17 Dec 2025 16:27:09 +0000
- To: public-css-archive@w3.org
The CSS Working Group just discussed `[css-grid-3] Make align-self/justify-self align including adjacent extra space`, and agreed to the following: * `RESOLVED: Find a solution to this` * `RESOLVED: use align-self and align-items` * `RESOLVED: define normal resolves to start when stacking forwards and end when stacking backwards` <details><summary>The full IRC log of that discussion</summary> <TabAtkins> For a simple example, the image in https://github.com/w3c/csswg-drafts/issues/10275#issuecomment-3169522037<br> <Kurt> fantasai: there was a question of what should we do with alignment in the stacking axis. For the align-content property on the container, we move the entire stack of items without moving them. There was some discussion of distributing space, but it is hard to reconcile with spanning, so we didn't go down that direction<br> <fantasai> https://github.com/w3c/csswg-drafts/issues/10275#issuecomment-2100950232<br> <Kurt> ...there was an idea of using aling-self or justify-self to handle gaps in layout. If you have a spanning item, see illustrations on the 8th, you would be able to control the item that is adjacenbt to the empty space up or down or stretch to create slightly different layouts. Stretch probably most popular. Discussion about do we want to, does it<br> <Kurt> make sense. Alison has been digging into more detailed option. Is this a model that's appropriate for alignment.<br> <Kurt> alisonmaher: From an implementation perspective, looking at the item above the spanner makes things a lot simpler, would push for. Stretch is the main case that I think would be useful. We can discuss other options too. Start, end, center are also other options worth doing.<br> <Kurt> astearns: first question: do we want these affordances at all<br> <TabAtkins> +1 to doing this behavior, and weakly support using align-self for it<br> <Kurt> ...do we need to resolve on solving this problem and then work on details later?<br> <Kurt> alisonmaher: seems like a good start<br> <Kurt> astearns: any concerns?<br> <Kurt> ...or doesn't care?<br> <Kurt> ...proposed resolution - will create some affordance for aligning within possible empty space in grid-lanes layouts<br> <Kurt> ...objections?<br> <Kurt> RESOLVED: Find a solution to this<br> <Kurt> ...should we resolve on a method like align-self or take back to the issue?<br> <TabAtkins> yes, align-items works as normal<br> <Kurt> miriam: if it's align-self, align items should set the default?<br> <Kurt> ...sounds good to me<br> <Kurt> fantasai: main downside to align-items is awkwardness - can't control first row independently of last row. Could expand at some point for all systems if we want to<br> <Kurt> miriam: one use case is setting everything to stretch<br> <Kurt> ...would be easier to do with align-items than selecting everything<br> <Kurt> ...and doing align-self<br> <Kurt> astearns: does align-self have the right default?<br> <Kurt> tabatkins: deafult doesn't do anything useful<br> <Kurt> fantasai: start alignment is the most reasonable default<br> <Kurt> astearns: proposed resolution: we will use align-self and align-items to control this behavior<br> <Kurt> ...objections?<br> <fantasai> Actually, if we add wrap-reverse, maybe we need a track-start/track-end pair kinda like flex-start/flex-end ...<br> <Kurt> RESOLVED: use align-self and align-items<br> <fantasai> or maybe we just need to make normal do that and don't give people end<br> <Kurt> fantasai: follow-up question about defaults, resolving normal to start makes sense if we stack forward. If we reverse direction and are stacking from bottom, we probably need normal to behave as end. We need to change initial value to behave differently.<br> <Kurt> astearns: couldn't reverse change start?<br> <Kurt> TabAtkins: no, but it can if there's no normal<br> <Kurt> fantasai: Proposed: initial value resolves to start when stacking forwards and end when stacking backwards<br> <Kurt> astearns: concerns?<br> <Kurt> ...objects to proposed default behavior<br> <Kurt> RESOLVED: define normal resolves to start when stacking forwards and end when stacking backwards<br> </details> -- GitHub Notification of comment by css-meeting-bot Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/10275#issuecomment-3666150085 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Wednesday, 17 December 2025 16:27:09 UTC