Re: [svgwg] [svg2] Defs element, misleading wording regarding "accessibility" (#1007)

>  I'm surprised, but curious which reasons you'd still accept.
Accessibility, for example - the matter at hand. Maintainability doesn't apply to live code - then it's not truly minified and sustainable. Let's stick to the topic.

> Your paragraph looks also a bit contradicting to me, but perhaps you've more details.
On re-reading, I realised you might have misunderstood something critical here. There is no contradiction at all, but it might look that way because I didn't spell out why I don't like defs for reusable objects: 
* In an ideal world, when you know you are using an element more than once, you tailor your initial element to be reused so it matches one of the scenarios where you want to use it. You write it as a normal path or whatever, and assign it an ID. Then you use href on it, in exactly the same way as you would with an element not actually displayed on first mention in the defs. 
I should have explained that - without making that crystal clear, everything else is confusing. That's also why there is no rendering benefit to using defs, and a slight performance/sustainability benefit from:
* Reducing the total code transferred, even if the first use is a direct copy of the original from the defs.
* Discouraging the use of transforms, e.g. the element must be drawn the way you intend to ultimately use it in the initial case. 

> ...the spec isn't about the processes. Of course it's good sorting wrong arguments out though.
Right. But if an error in the spec leads to people focusing on the wrong things and unnecessarily bloating their code, that's not pro performance or sustainability.

-- 
GitHub Notification of comment by codewordcreative
Please view or discuss this issue at https://github.com/w3c/svgwg/issues/1007#issuecomment-3286617747 using your GitHub account


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

Received on Friday, 12 September 2025 19:46:27 UTC