- From: Chris Lilley via GitHub <sysbot+gh@w3.org>
- Date: Wed, 31 Aug 2022 15:17:10 +0000
- To: public-css-archive@w3.org
> I'm not familiar with the patch method and we don't support it, but I see it's also done with HTTP headers and looks to be less dependent on the internal font layout. The patch-subset method requires an intelligent HTTP server which can on-the-fly create font subsets and then compute a binary patch to upgrade one subset to a (larger) subset. And yes, most efficient use of the range-request method needs a re-arranged font as described in the spec while the patch-subset method is unaffected by how the font is laid out. > My question is - under what circumstances would incremental-patch be specified and incremental-range not wanted? Patch-subset is [**vastly** better](https://www.w3.org/TR/PFE-evaluation/#analysis-cjk) than range-request, on 3G and faster networks, in terms of bytes transferred, requests made, and overall network cost.  Stating up-front in the CSS that the server has this capability avoids an initial pre-flight check to ask if the server supports it, so is worth having in the stylesheet. It also allows the range-request version to be served from a different server (any HTTP server) while directing patch-subset request to a specialized font CDN. -- GitHub Notification of comment by svgeesus Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/7665#issuecomment-1233076039 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Wednesday, 31 August 2022 15:17:12 UTC