- From: Domenic Denicola <notifications@github.com>
- Date: Wed, 22 Jan 2025 19:16:48 -0800
- To: w3c/ServiceWorker <ServiceWorker@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
- Message-ID: <w3c/ServiceWorker/issues/1746/2608759152@github.com>
@sisidovski asked me for my perspective on how this works in the web platform in general. The main guidance is given in https://infra.spec.whatwg.org/#algorithm-limits , which advises that by default there should be no limits. But since we have a good reason to have limits here, the ideal is to make them interoperable. So the discussion of leaving flexibility to implementations for the upper limit is a bit scary from that perspective, as it could cause interop issues. If Chrome supports 256 and Safari/Firefox support 512, then it is pretty easy to make sites that work only in Safari/Firefox and not Chrome. So I think the ideal would be if all implementations were willing to pick a specific limit, and support everything in the range [0, thatLimit]. For both number of rules and depth of rules. If implementers have strongly different ideas on what the upper limits should be, and don't want to converge, then the second-best option would be a defined lower limit, and leave the upper limit implementation-defined. This doesn't solve the interop concern in general, as it would still be possible for interop issues to arise between [specifiedLowerLimit, maxUpperLimitFromAllBrowserEngines]. But it at least gives web developers who read the docs a target: don't exceed specifiedLowerLimit, and your code will work in all browsers. -- Reply to this email directly or view it on GitHub: https://github.com/w3c/ServiceWorker/issues/1746#issuecomment-2608759152 You are receiving this because you are subscribed to this thread. Message ID: <w3c/ServiceWorker/issues/1746/2608759152@github.com>
Received on Thursday, 23 January 2025 03:16:52 UTC