- From: Bruno Racineux <bruno@hexanet.net>
- Date: Sun, 26 Jan 2014 19:20:58 -0800
- To: Qebui Nehebkau <qebui.nehebkau+whatwg@gmail.com>
- Cc: WHATWG <whatwg@lists.whatwg.org>
On 1/26/14 3:39 PM, "Qebui Nehebkau" <qebui.nehebkau+whatwg@gmail.com> wrote: >On Sat, Jan 25, 2014 at 4:13 AM, Bruno Racineux <bruno@hexanet.net> wrote: >> What exactly do you find misguided, can you be more specific? > >Basically, - and I'm trying not to over-elaborate here, since my >opinion isn't really very important - I just mean that I don't think >there should be any guarantees about how (or whether) browsers will >preload, nor any specific means of controlling this, because the way >resources get loaded is not really any of the author's business. I couldn't disagree more. It is my business to know how the browser loads assets in order to better assist the best performance for my users. Don't want to over-elaborate either but I'll leave you with a relevant quote: "The good news is all of these optimizations are done automatically on our behalf and often lead to hundreds of milliseconds of saved network latency. Having said that, *it is important to understand how and why these optimizations work under the hood*, because we *can* assist the browser and help it do an even better job at accelerating our applications." Ilya Grigorik - High Performance Browser Networking http://chimera.labs.oreilly.com/books/1230000000545/ch10.html#BROWSER_OPTIM IZATION There is a balance between putting DOM loading on auto-pilot, blindly assuming the preloader is your friend and can do no wrong, and the reality that the preloader can't ultimately do the best job on its own. It needs to be assisted with resource priority attributes (or preferably css) for better control. Without them, the preloader is no more than a 'seen-first priority' faster loader, with a priority agnostic performance shortfall. >the argument that preloaders >can't consider CSS isn't compelling to me, because a browser's choice >to preload an image or not isn't important enough (or, I think it >shouldn't be) to justify entrenching in a specification. If not for specifications, we certainly need 'preload opt-out' tools to better manage loading priorities. Whether it be 'postpone', 'lazyload' or said implications of 'hidden', and some documentation as to what to expect from preloaders.
Received on Monday, 27 January 2014 03:21:34 UTC