- From: CSS Meeting Bot via GitHub <noreply@w3.org>
- Date: Wed, 07 Jan 2026 17:25:36 +0000
- To: public-css-archive@w3.org
The CSS Working Group just discussed `[css-anchor-position] Aliasing names of span-* values to center-span-* for clarity`. <details><summary>The full IRC log of that discussion</summary> <JoshT> Saron: there has been past discussion for span values<br> <JoshT> ... in spring time, I looked at span-right and span-left, I didn't know what was going on. found it confusing. once I learnt what the values meant, it was fine, but I didn't like the initial friction. if I was a new dev, I wouldn't know what was going on<br> <JoshT> ... thought it would be good to alias it to give clarity<br> <JoshT> ... would be center-span-right<br> <JoshT> ... you start at the center and span to the right<br> <JoshT> ... what do you think?<br> <JoshT> astearns: aliasing sounds fine, but this is a new feature so could we rename?<br> <bramus> q+<br> <JoshT> Saron: I think alias is better. I did poll and asked people what they thought. Views were mixed. Some liked conciseness of span-right<br> <JoshT> ... others wanted more clarity<br> <JoshT> ... I found mixed results<br> <JoshT> ... aliasing gives us both<br> <astearns> ack bramus<br> <JoshT> bramus: looking at the span-left span-right, it does seem to say center space span right<br> <JoshT> ... I understand it that first keyword defines where you start and the next one says where to span to<br> <JoshT> ... so I don't think the aliases make it easier or harder to understand<br> <fantasai> I think Bramus was mentioning 'center span-right' and 'center-span-right'<br> <JoshT> ... does this mean the space gutter gets replaced by the dash?<br> <JoshT> fantasai: center-span-right is only one of the axis<br> <JoshT> ... so there's a difference<br> <fantasai> 'center span-right' sets both<br> <miriam> q+<br> <astearns> ack dbaron<br> <JoshT> bramus: thanks for clarification<br> <JoshT> dbaron: is there a reason not for span-right?<br> <JoshT> Saron: means I'm starting at the center and spanning to the right<br> <dbaron> s/not for span-right/for center-span-right rather than span-center-right/<br> <JoshT> ... so it's to be more sequential<br> <JoshT> fantasai: it groups them together and makes it easier for the author to understand<br> <astearns> ack miriam<br> <JoshT> miriam: I think I would have a preference for span-center-right<br> <fantasai> s/together/together in autocomplete and reference lists/<br> <kizu> q+<br> <JoshT> ... my concern is I don't like span-center-right and center-span-right are different things that read the same<br> <fantasai> s/span-center-right/center span-right/<br> <JoshT> Saron: I think you're right. more distinctions between the values makes sense<br> <astearns> ack kizu<br> <JoshT> kizu: alias could be ok but 2 concerns<br> <JoshT> ... 1 is it could be harder to remember a longer name<br> <bradk> span-center-right sounds like you are spanning to the center right<br> <JoshT> ... we already mentioned span center right, span center, so on<br> <fantasai> +1 bradk good point<br> <JoshT> ... there are many permutations you could have and it might not be logical for the author that you start in the center<br> <JoshT> ... 2nd is that if we are aliasing this, it allows authors to choose what they want, when you read the spec it can be more confusing<br> <romain> +1, I also prefer not to have two keywords that do the same thing<br> <JoshT> ... once you learn it you will understand. those in the issue who understood how it works seemed OK after that<br> <JoshT> ... an alternative that is still small but more understandable would be great<br> <JoshT> Saron: I agree it's learnable. it's about reducing friction<br> <JoshT> ... I think it is helpful for devs to see the values without needing to look it up<br> <dholbert> q+<br> <JoshT> ... with your 1st point, miriam's suggestion to start with span might address the first concern<br> <JoshT> ... when doing a talk last spring, I wanted it to go across. if the first word is 'span', that captures the intuition<br> <JoshT> ... it maps to the intent of the dev. and gives a list of options that say what will happen without memorising it<br> <bramus> But we have `span-all` for going across, no?<br> <JoshT> dholbert: looking at the value space, I'm concerned about values that are already compound words like y-start<br> <JoshT> ... we'd also get span-center, which might be a lot to parse for someone<br> <JoshT> ... I guess depending on whether we're planning on adding it for ???<br> <florian> q?<br> <dholbert> q-<br> <dholbert> s/span-center/span-center-y-start/<br> <JoshT> astearns: if you're doing autocomplete for all of the options, you might get confused by all of the options<br> <dholbert> (er "...we'd also get span-center-y-start")<br> <JoshT> ... we can bring it back to the issue and see if it's something worth adding or not useful for helping understanding<br> <JoshT> fantasai: if you put span at the front, it's hard to break apart the more complex values<br> <bradk> If there are aliases, minifyers will choose the smaller one, and that will show up in browser developer tools.<br> <JoshT> ... where do you break that into the two axies<br> <bradk> Aka inspector<br> <florian> q+<br> <JoshT> miriam: if there is not an alias, if it's just the center-span-right and there's no span on its own, then there's no confusion<br> <astearns> ack fantasai<br> <dholbert> (agree with fantasai that e.. "center-span-y-start" is a bit easier to visually-parse than "span-center-y-start")<br> <astearns> ack florian<br> <JoshT> florian: just thinking if we break away from this problem we could have functional notation<br> <kizu> +1 to functional, wanted to propose the same, it also makes it configurable: `span(center, var(—side))`<br> <JoshT> astearns: sorry it took so long for us to get to this Saron but thank you for your work on this<br> </details> -- GitHub Notification of comment by css-meeting-bot Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/12683#issuecomment-3719935783 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Wednesday, 7 January 2026 17:25:37 UTC