- From: John Scharber <jscharber@vidscale.com>
- Date: Mon, 4 May 2015 16:31:40 -0700
- To: Ilya Grigorik <igrigorik@google.com>
- Cc: public-web-perf <public-web-perf@w3.org>
- Message-ID: <CAG52YWE93B26pPHYUgqeU8oVQnyMeEO-X_q7w4MOoa-f12rM2g@mail.gmail.com>
Thanks, I'll take a look at the archive, did a quick glance and didn't see a thread addressing this. We are rewriting the beacon logic in the Pagespeed optimization libraries to be more efficient. The beacon spec looked like a good place to start with some polyfill, and server-side changes. I'll comment back to the list if we learn anything interesting. /jms On Mon, May 4, 2015 at 4:19 PM, Ilya Grigorik <igrigorik@google.com> wrote: > Hi John. > > On Wed, Apr 29, 2015 at 2:25 PM, John Scharber <jscharber@vidscale.com> > wrote: > >> From a mobile device perspective, it would suggest a few additional >> capabilities: >> >> 1) Limit beacon reporting to specific network types, WiFi, Fixed, etc >> 2) For mobile networks, 2/3/4G, limit sending beacons to specific link >> states like active vs. idle >> 3) Aggregate beacons for the same destination >> > > Yup, all great suggestions. If you're interested, check the archives.. we > discussed all of the above while iterating on the spec. Hence the > provisions for "Beacon-Age" header, language for delaying and prioritizing > delivery, and so on. AFAIK, current implementations (FF, Chrome) do take > advantage of some (e.g. prioritization) but not all (e.g. delayed delivery) > of these. That said, nothing stops them from adding these capabilities at > any point either. > > ig > -- John Scharber Vice President Advanced Technology Group VidScale, Inc. +1 (925) 200-7231
Received on Monday, 4 May 2015 23:32:27 UTC