- From: Tab Atkins Jr. via GitHub <noreply@w3.org>
- Date: Mon, 16 Jun 2025 22:42:32 +0000
- To: public-css-archive@w3.org
Any time we consider adding a new construct to the grammar, the question that needs to be asked is always "how much does this save us versus just writing it out"? There's only a handful of us editors that work with the grammar, and every novel ability we add is something that people reading the grammar probably don't know and will have to learn. Grammars are already really hard for many people to understand (but worth it for their conciseness), so we need to really lean towards ease of reading over ease of writing here. The boolean construction pays for itself, I think - doing it by hand is complex and easy to get wrong, and we use it in enough places that had to copy-paste things that having a concise, obvious construction worked out. It's also pretty clear what it means in the first place - "the argument, combined with boolean operations". I don't think this idea pays for itself. We don't do this sort of repetitive pattern very often in the first place, and when we do, I think it's a lot clearer to see each specific grammar written out normally. The options and variation that we have to account for when we *do* specify this sort of repetition also isn't super well-established, so I don't think we should try to bake it in by creating a new grammar construct right now. Overall I don't think any of these examples would be more readable or understandable in a more generic form. We don't need 100% machine-readability in all of our grammars; we need them to be understandable by humans (and ideally, by website authors). -- GitHub Notification of comment by tabatkins Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/12332#issuecomment-2978389268 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Monday, 16 June 2025 22:42:33 UTC