- From: Alex Russell <notifications@github.com>
- Date: Tue, 04 Dec 2018 15:51:23 +0000 (UTC)
- To: w3ctag/design-reviews <design-reviews@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
Received on Tuesday, 4 December 2018 15:51:46 UTC
Hey again, So there's something larger happening which this intersects with: we've got a growing number of areas where the default loading behavior needs configuration and which we're putting further out of reach or needing to come up with bespoke one-offs for; think resource priority, CORS, etc. I'm pretty worried that this is another one-off that adds to a pile of other oneoffs. Even if this is the right call for most imports, it (predictably) won't be in some cases and I can easily envision cases where, in working with a partner on perf, I'd like a different policy. How will that policy get added? Should we be working through a more general mechanism for controlling subresource fetch params? @kinu, @kenjibaheux, and I have previously discussed a `PreloadWorklet` type, which could perhaps work for this case too? If so, great! If not, would like to understand why (e.g., we have capability URLs we don't expose from `@import`) and how we might work around that. Anyhow, I think map population is a good thing to do, but the performance impacts will be the primary reason this gets used (or not). Let's make sure it's fit for that purpose. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/w3ctag/design-reviews/issues/213#issuecomment-444149359
Received on Tuesday, 4 December 2018 15:51:46 UTC