Re: [w3ctag/design-reviews] Design Review: Speculation Rules (Prefetch) (Issue #721)

> we're noting a lack of multi-stakeholder interest in this. Do you have any info on this can you can share?

Nothing beyond the standards positions linked in the original post, sorry!

In general we've found that second and third implementers often take a while to follow on these sort of progressive enhancement/for-performance features.

> We're concerned about developer complexity when it comes to this feature, especially considering the need for a new HTTP header that requires server configuration. Is there an alternative design that wouldn't require as much complexity?

It depends on what you mean. Fundamentally, an opt-in is needed, for security reasons. I don't think that opt-in is very _complex_; it's a single HTTP header, and things don't get much simpler than that.

It's possible that you're referring to the _difficulty_ of configuring HTTP headers, which is e.g. impossible on some older static hosts like GitHub pages, and thus requires the use of other free hosting like Netlify/CloudFlare Pages/etc. (Or to use a non-free host.) We could support even those older static hosts by working on [this future extension which the explainer explained](https://github.com/WICG/nav-speculation/blob/main/opt-in.md#an-in-markup-version), i.e. `<meta http-equiv="supports-loading-mode">`. Arguably, this adds a good bit of _complexity_, as in-markup versions come with a lot of restrictions around parsing, appearing within the first few thousand bytes, etc. But it does make the feature _easier_, at least for those stuck on such older static hosts.

Is that what you were referring to, or was there a different meaning of complexity that I missed?

---

We have a number of other small enhancements to speculation rules/prefetching/prerendering coming up, e.g. support for customizing the referrer policy used. We're planning to continue pinging this thread with small summaries like I did previously, but if you'd prefer us to hold off (e.g. until the base feature gains more implementers) or start new threads, please let us know!

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/721#issuecomment-1336649923
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/721/1336649923@github.com>