Google's Effort To Speed Up the Mobile Web (ampproject.org)

An anonymous reader writes: Google has officially taken the wraps off its AMP project — Accelerated Mobile Pages — which aims to speed up the delivery of web content to mobile devices. They say, "We began to experiment with an idea: could we develop a restricted subset of the things we'd use from HTML, that's both fast and expressive, so that documents would always load and render with reliable performance?" That subset is now encapsulated in AMP, their proof-of-concept. They've posted the code to GitHub and they're asking for help from the open source community to flesh it out. Their conclusions are familiar to the Slashdot crowd: "One thing we realized early on is that many performance issues are caused by the integration of multiple JavaScript libraries, tools, embeds, etc. into a page. This isn't saying that JavaScript immediately leads to bad performance, but once arbitrary JavaScript is in play, most bets are off because anything could happen at any time and it is hard to make any type of performance guarantee. With this in mind we made the tough decision that AMP HTML documents would not include any author-written JavaScript, nor any third-party scripts." They're seeing speed boosts anywhere from 15-85%, but they're also looking at pre-rendering options to make some content capable of loading instantaneously. Their FAQ has a few more details.

Comments

I hate mobile web pages

[Anonymous Coward]

They shrink down the content, but keep the ad network bullshit huge in comparison.

Mobile web without that rubbish is almost instant even on a struggling 3G connection. Hell, even regular web pages with junkware scripts blocked are quick-as.

You're looking in the wrong place, google.

Re:I hate mobile web pages

[davester666]

Yes, this is just trying to serve the existing ads and tracking crap faster.

Ad/Tracking Blocking really is the only way forward.

Re:I hate mobile web pages

[Anonymous Coward]

Yep, with JS off, mobile web is really slick. Slashdot redirects to that page complaining I have javscript disabled saying to use the desktop site really fast. Go Mobile web! And you press the desktop site link, and the next internal link you hit puts you back at the stupid page saying to use the desktop version... If you are going to both auto-detect my mobile user agent and my lack of JS support please don't auto redirect to a useless page every time I click any internal link.

I think I should just make my phone always claim to be a desktop browser: at least 3 sites I use have mobile versions that are pretty much unusable with desktop versions that are fine.

Re:I hate mobile web pages

[Anonymous Coward]

This is due to shitty web-devs, brogrammers, etc.

Its entirely possible to make decent pages that don't ask you to rotate your device, and then take up the whole screen with an ad. Its simple, you merely have normal sized banner ads. Oh god, so terrible.

Publishers just do not give a flying shit about mobile page quality. And it shows.

Re: I hate mobile web pages

[Bing Tsher E]

The weird part about Slashdot mobile is how much better it looks if you're not logged in. Also the absence of any form of captcha when you're posting a comment as A..C.

The combination of those things is probably why there are more a.c. Posts now than in the past. It's just not worth logging in to comment and then having to completely shut the browser down to log out.

Re:I hate mobile web pages

[JustAnotherOldGuy]

Yes, this is just trying to serve the existing ads and tracking crap faster.
Ad/Tracking Blocking really is the only way forward.

Yep. I just installed an ad-blocker on my wife's ipad, and *boom* everything is so much faster. It's like night and day, and we're never gonna run without one again.

You hear me, ad companies? WE'RE NEVER GOING TO BROWSE WITHOUT AN AD-BLOCKER AGAIN.

You forced us into this, so screw you, we'll have the last laugh.

Re:I hate mobile web pages

[Anonymous Coward]

Except all the free content you've been looking at, like Slashdot, will become paywalled or start doing shady tactics like being paid posts without notice.

Either that, or you might just want to get a less shitty tablet... All sites work fine for me with or without AdBlock on Firefox on my 2-year old tablet...

Re:I hate mobile web pages

[Anonymous Coward]

I bought a smartphone so I wouldn't go to the mobile web. Now my only portable device option for that is a Microsoft Surface Pro tablet, which is much much much much bigger than a smartphone. I never realized this was a google effort. One more reason to dislike google!

Sorry. Not gonna work

[Anonymous Coward]

I mean, at least I hope it doesn't. Grow the bandwidth, don't restrict developers. Validators that predict performance issues before deployment, which google already provides in some form or another, is a better way to go. Just publicize them more and make them easy to use. How about an official google benchmark suite that works with everything and anything and just emits its opinion of our shitty code?

Re: Sorry. Not gonna work

[nullchar]

Chrome developer tools already give suggestions for improvements. Experienced developers know how to learn and thus improve their sites. I don't see wide adoption here unless built into major content management systems.

Google seems to be avoiding the real problem

[Anonymous Coward]

The performance problem isn't with mobile web, it's with *Android* web. Apple devices are approaching desktop-level performance on js-heavy pages while android devices are getting worse and worse. A considerable contributor to this problem is the trend for android devices to show more cores in them, not faster ones. JS is still single-core after all.

Interesting comparisons here: https://meta.discourse.org/t/the-state-of-javascript-on-android-in-2015-is-poor/33889

Disclaimer: I am an andoid fanboy.

Re:Google seems to be avoiding the real problem

[davester666]

The problem IS mobile web. They actually try to cram more tracking JS crap on pages, with auto-playing video [at least they still generally serve flash to desktop, so blocking Flash works to kill those ads]. It's ridiculous.

Re:Google seems to be avoiding the real problem

[ET3D]

Thanks for the link. I'd mark your comment up as informative if I could. :)

Still, comparing to iOS devices is like saying "there's no performance problem for games on laptops; my Alienware laptop runs them well." Android devices are a lot more varied than iOS ones, and it'd be nice to have web pages work well on a sub-$100 device and not just flagships. I don't know how well old iOS devices cope with pages either. Last I heard a comment from an iPhone 4 user it was that the phone was very sluggish. So just because web pages work well for someone who recently spent $600 and up on a phone doesn't mean that it's not worth optimising them.

Re:Google seems to be avoiding the real problem

[oztiks]

I can say Crosswalk is really the step up that's needed for both iOS and Android and wIth ECMA being enforced with more and more browsers over time. The problem is just that its 40 meg to download.

I can see this AMP project as good pretext for quick and instant functionality, though from downloading the examples and reading a bit about it. It really needs a lot more work before it would become useful to me or any of the projects i'm presently on.

People bitching about mobile websites are in for a real shock when Google Cardboard and Accelerometer tech comes into play for the web. Just imagine looking at a website menu items by simply shifting the pitch and angle of your phone. I'd say 6-12 months before that happens. And with Apple being supported by Cardboard I'm surprised it already hasn't happened! But even without cardboard, Accelerometer navigated website should really be the next thing.

Re:Google seems to be avoiding the real problem

[oztiks]

Probably a good excuse to dump the web as it's just shocking how many JS trackers are getting at your stats with every page visit. Though heading to the App store to serve web content means providers just get more "detailed" stats about their users. So Kobayashi Maru there right.

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

Claiming Safari is outperforming Chrome is a joke you fucktard. Stop posting non-credible references from Reddit and get off your reality distortion device and look at the facts!

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

No, it will suck. No showing the phone to someone else. No using it while walking/traveling (or maybe that's a feature). No adjusting the phone so the sun isn't reflecting in your face. No one with shaky hands. There will be a bug where the shake events are still processed while your phone is locked and in your pocket.

3D mice have existed for decades. They still haven't caught on. You can connect them to your phone. No one uses them.

You know advertisers will jump on the features first. Instead of clicking that tiny X you can shake 3 times, pause, then shake 3 more times to close it. Hell no.

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

The facts you so helpfully provided a link to?

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

Read it and weep motherfucker!

http://www.favbrowser.com/chro...

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

Well, my iPad still struggles with some sites dynamically resizing when I try to zoom, pan or switch screen orientation and losing my place, so the problem with JavaScript isn't just speed on Android. People use JavaScript for formatting, and that's just nuts.

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

I don't get it.

Just reading that post, half the of the posted benchmarks rate the aforementioned S6 as better than everything else (tablets included)...

Just some poorly designed website with (SURPRISE) just an ithing app complains about running slow. lol

Re:Google seems to be avoiding the real problem

[KGIII]

That is 6 year old data.

Re:Google seems to be avoiding the real problem

[Anonymous Coward]

Just imagine looking at a website menu items by simply shifting the pitch and angle of your phone.

I'm not going to navigate a page by waving the phone around. That sounds incredibly annoying, and it is just another feature I will turn off immediately. Just like screen auto-rotation. That's literally the first thing I disable on any new smartphone. I'm usually holding the phone in portrait mode to read a web page, and if I shift body position, the accelerometer detects the movement and the screen will shift to landscape and reformat the text so I lose my place in whatever I was trying to read. I then have to wave it around a bit to shift it back to portrait, or if I am laying on my side it will never stay in portrait unless I force it to. So I always lock the screen orientation unless I specifically want landscape (watching videos ONLY).

Re:Google seems to be avoiding the real problem

[Karlt1]

So just because web pages work well for someone who recently spent $600 and up on a phone doesn't mean that it's not worth optimising them.

It's not just comparing a state of the art iPhone 6s to an old Android phone. The top of the line Galaxy S6 performs worse than the 2011 iPhone 5 when running JavaScript.

https://meta.discourse.org/t/t...

To give you an idea of how divergent it has become, try:

http://emberperf.eviltrout.com...
Complex list
Ember 1.11
This is the benchmark most representative of Discourse performance, and the absolute best known Android score for this benchmark is right at ~400ms on a Samsung Galaxy S6. That doesn't seem too bad until you compare..

iPhone 5 â' 340ms
iPhone 5s â' 175ms
iPhone 6 â' 140ms
iPad Air 2 â' 120ms
iPhone 6s â' 60-70ms
In a nutshell, the fastest known Android device available today -- and there are millions of Android devices much slower than that out there -- performs 5Ã-- slower than a new iPhone 6s, and a little worse than a 2012 era iPhone 5 in Ember. How depressing.

Work on WebAssembly instead.

[Anonymous Coward]

Look Google, if you want to tell us not to use JS after expending all of that effort getting us hooked on your Javascript frameworks, well, I think you'll find some people see it as you giving up and phoning in a lame non-solution. Just work with people who actually think about how to actually migrate the web into the future, like the WebAssembly guys, rather than reinventing it every year with increasingly fragmented "solutions". You've only been causing problems or making them worse by adding your junk to the web. Even WebM was a bust because you decided to drop h264 support, so why the heck would anyone bother with WebM? After all of these wasteful re-engineering efforts, I'm not about to jump ship on scripts just because you tell me to quack. I didn't want them in the first place, but you just said it would be alright, V8 was awesome, so don't bother with noscript tags and graceful degradation, just use Angular or whatever. Now I just don't care anymore. I'm frankly starting to miss the slower web that took years to give us rounded corners, but at least I didn't have to worry about ES6 support dragging out because you want to upsell Dart or something, or not bother with Pointer Events because "fixing our browser to support useful mobile standards is too hard, so we're not bothering and will instead introduce our own half-assed Touch Events spec that we'll backpedal on later".

Re:Work on WebAssembly instead.

[Anonymous Coward]

they don't want you to 'not use JS' they want you to use the particular AMP-js

I mean seriously on "https://www.ampproject.org/how-it-works/" (and amp page btw)
- they say explicitly "For a headline, some text and an image you do not need JS"
- that page fits that description ... BUT shows nothing but a blank page unless you allow javascript from cdn.ampproject.org
for something explicitly aimed at static pages failing progressive enhancement completely is utterly ridiculous

Quite frankly I expect better technical output from Google... or I would if I didn't think this was an attempt at allowing them to track you everywhere through the amp-JS

Re:Work on WebAssembly instead.

[Tallus]

The reason the page didn't work was this: (from the amp sample page)

<style>body {opacity: 0}</style><noscript><style>body {opacity: 1}</style></noscript>

Except that you can now view that page as it no longer has it: meaning either it is, in fact, not valid AMP, or that this tag is not needed and exists only to defeat those who don't/won't their JS.
I suspect the latter: if you go to a page marked up with the latter and use web dev tools to set the opacity to 1 everything works fine.

Bend over

[ADRA]

Its WML time!

Re: Bend over

[MenThal]

WML was apparently inspired by HyperCard. So, hyperlinks and hypertext and tons of hype... The reinvention rolls around again.

Best case scenario; marketing and tracking scripts get a solid broadside across the Web with this. Except Google's tracking, of course...

Won't fly with companies

[Todd Knarr]

This'll be great for individuals, but companies won't accept it. The first problem is that ad networks won't accept the limitation, so any site that shows advertising will have to eschew AMP. The second problem is that companies use Javascript frameworks so heavily in their Web design that they won't be able to just drop it and go back to static HTML/CSS for their sites. If they were willing to, after all, Google wouldn't've seen such a need AMP in the first place.

I think the same results can be achieved by three things:>

1. Strip advertising down. Ads are the biggest things I find slowing mobile Web pages down as the ads do so many things to keep content from being accessed until the ad's been seen and dealt with and fetch so much stuff from so many remote servers. Minimize the number of ads and make them as simple as possible.

2. Stop using images for layout and convert to using CSS instead. Yes you lose the ability to do fancy brand-specific artwork on headers and separators and such, but you know what? Most users don't care about those things.

3. Stop using dynamic layouts that load the entire page and then make changes to it that alter it to it's final layout. Just lay things straight out so the browser can render stuff as it's loaded. Specify sizes for images, drop the "Tap to read the rest" buttons that hide the bulk of the content (but still require it to be loaded before the page can render), that sort of stuff.

Easy way to do this: one of your test machines runs Windows XP on hardware with a 500MHz CPU, 256MB of RAM, an unaccelerated graphics card with 2MB of video memory and a 56K modem (or 115200bps serial link) for network connectivity. If your site performs decently on that, it'll be good on any mobile device.

Re:Won't fly with companies

[Darkling-MHCN]

I'm sure they'll support google advertising

Re:Won't fly with companies

[Ol Olsoc]

This'll be great for individuals, but companies won't accept it.

Then again, they won't accept your solutions either.

The web has a serious problem. The web is broken.

The present day over-commercialization of the web, with tracking and intrusive ads, and making search engines approach worthlessness; and as ad-blockers and script blockers are breaking out of the province of nerds, and onto regular folks machines, is showing the "target's" growing rebellion.

And even though Joe Sixpack might not understand the tracking, he does understand that his webpages are taking a long time to load, and up pops an advertisement for something he already bought. So Joe's geeky friend "who know's this stuff", tells him about the adblocking and noscript add-ons, maybe even installs a hostfile. And Joe is happy. Tells all his friends, and its off to the races.

I know this because I'm the geeky friend, and I've been installing adblockers and noscript, persistent cookie killers and even a few hostsfiles on friend's computers. Happy people, who wern't so much anti-ad, but just wanted acceptable performance

Google's well aware of the brewing web revolution, and trying to figure out what to do about it. I already have.

Re:Won't fly with companies

[Anonymous Coward]

Hi. Experienced web developer here. You're mostly right, but about 7years out of date (you sound like a Drupal guy :)). Your first point handles 80% of performance issues, that is true. You second point... it is true that using images for placeholders makes for poor performance and problematic portability across devices, but nobody does that any more. It kind of went out with tables and rounded corners/raised button effects, although those are trivial to do in CSS3, which is pretty much universal now.

Your third point is just wrong. Explicitly setting image sizes pretty much guarantees that your site will break unless you explicitly set the container width which pretty much guarantees that your site will only work on desktops. What most folks I know of do is set the width in percent and height to auto. If the aspect ratio is important for the page flow, there are CSS tricks to maintain it (see bootstrap's img-responsive or embed-responsive classes). You can also load a properly portioned placeholder (eg, a single 16x9px image everywhere) and lazy load it when it is visible.

Re:Won't fly with companies

[JustAnotherOldGuy]

3. Stop using dynamic layouts that load the entire page and then make changes to it that alter it to it's final layout.

Yes, yes, and YES. I hate pages that resize themselves 3 or 4 times before stabilizing. And some of them continue to resize in a jittery fashion as you scroll or as more remote content (mostly ads) get loaded.

Their are sites I no longer visit because of this behavior and I suspect the owners and developers have no clue as to why people shun their pages.

Too many ads, too much dynamic bullshit, too much load-while-scrolling, too much crap. The web is now a festering pile of ads, with a little content thrown in as an afterthought.

Re:Won't fly with companies

[Todd Knarr]

No, in point #2 I'm not talking about placeholders. I'm talking about eg. using an image for a header so you can put the site's name in it's brand-specific font as opposed to simple setting the background color and using text for the name. Or using images to create a separator between headline and content so you can have one that looks exactly like what you use on your TV shows or in print.

As for the third point, it only breaks things if you set the size to something other than the image size (which your server ought to know since it's the one sending the image). And if you're doing dynamic resizing to fit screens, you're probably one of the people I hate when I start scrolling to see the text on your website and suddenly everything jumps around as your page finishes loading and your JS starts resizing images causing reflows and rerendering of the page. It's not so obvious on a desktop system because things tend to load fast, but on mobile devices it makes people tear their hair out. Which is why I said to test your site on a machine with a deliberately-choked-off network connection: to see how it'll render as it loads through a relatively slow, heavily congested link. I suspect a lot of designers test mobile sites on either emulators or using devices connected through fast WiFi links to fast local servers and never test what happens when it takes 30+ seconds to get all the content through the pipe.

Re:Won't fly with companies

[SoftwareArtist]

The first problem is that ad networks won't accept the limitation, so any site that shows advertising will have to eschew AMP.

Not true: they're very clear in all their documents that having good support for advertising is essential. Remember, it's Google who's doing this! They're basically an advertising company, so they're not going to do anything that hurts that. Here's what they say about how ads work:

We've taken first steps to make ads in AMP HTML better, but we aren't done yet. AMP HTML doesn't allow JavaScript so ads cannot be directly embedded - instead they live in sandboxed iframes with no access to the primary document. Relying on iframes solves some of the worst performance pitfalls with ads, in particular with respect to document.write. We also prioritize ads lower during loading than other content and optimize load timing to avoid jank. Ads in AMP files can still be heavyweight, so there is still a lot of work to do for us.

And here's another essential element of it:

This brings us to the final topic that makes AMP HTML unique: all resource loading is controlled by the AMP library and, more importantly, resources must declare their sizing up-front. Document authors have to state resource sizes explicitly. This doesn't mean that resources can't be responsive - they can be, but their aspect ratio or dimensions needs to be inferable from the HTML alone. This means that after initial layout, an AMP document does not change until user action. An ad at the top of the page can't suddenly say: "I want to be 200 pixels high instead of 50." This dramatically reduces jank and prevents users from losing their place in the document. All custom elements are subject to this restriction. Placement on the screen can be reserved while their implementations download asynchronously. This gets us lazy loading with zero visual jank.

So ads might still take a while to load, but they're trying to make sure that doesn't prevent the rest of the page from loading and becoming usable as quickly as possible.

Re:Won't fly with companies

[Anonymous Coward]

This is designed for advertising. It's by Google.

ads are the problem

[ljw1004]

Ads are what slow down the mobile web. Eliminate them and it runs blazingly fast.

Reckon you can do that, Google?

Re:ads are the problem

[kav2k]

Reckon you can do that, Google?

Of course they can't. Look at their FAQ.

How will advertising work on Accelerated Mobile Pages?

A goal of the Accelerated Mobile Pages Project is to ensure effective ad monetization on the mobile web while embracing a user-centric approach. With that context, the objective is to provide support for a comprehensive range of ad formats, ad networks and technologies in Accelerated Mobile Pages. As part of that, those involved with the project are also engaged in crafting Sustainable Ad Practices to insure [sic] that ads in AMP files are fast, safe, compelling and effective for users.

Re:ads are the problem

[Alumoi]

Not since AdAway. Of course you need to be root, but who in their right mind doesn't root his/her Android phone first thing after unpacking?

Re:ads are the problem

[Anonymous Coward]

You do realize that's essentially painting a giant "HACK ME HERE" sign on the roof of your house and phoning the Chinese with the exact latitude and longitude, along with your recent bank transcripts and a list of hobbies.

Re:ads are the problem

[Alumoi]

What's wrong, Mr. Marketing Droid? Mad I'm blocking even inside apps ads?

Re:ads are the problem

[JustAnotherOldGuy]

A goal of the Accelerated Mobile Pages Project is to ensure effective ad monetization on the mobile web while embracing a user-centric approach.

TRANSLATION: "Forget content and usability, the IMPORTANT thing is to be able to cram the page full of ads."

"embracing a user-centric approach"

FFS, just kill me now.

Why would I help Google with 'open source'?

[hughbar]

Google which dominates the search/ad market can do this by itself, without my help.

Also, looking at the analysis here: http://www.cnet.com/news/world... Open source is simply part of its strategy for distributing software that will help it sell more advertising

This is part of a general trend that I call 'open season', basically big companies persuading naive people to do their work for nothing, under the banner 'open source'.

Re:Why would I help Google with 'open source'?

[hvdh]

Just yesterday I measured data use of Germany's biggest gossip news site (bild.de) on my smartphone (Android 5.1 with stock browser), cached and uncached (browser cache cleared, browser restarted) with and without ad blocking (using AdFree host list). The phone was on Wifi / DSL.
Here's the result:
uncached load (first visit) with ads: 2.4MB data, 26s (!) until the display goes from white to some content
uncached load (first visit) without ads: 1.7MB data, 11s until the display goes from white to some content
cached load (second visit) with ads: 272KB data, 2s
cached load (second visit) without ads: 45KB data, 2s

Re:Why would I help Google with 'open source'?

[speedplane]

This is part of a general trend that I call 'open season', basically big companies persuading naive people to do their work for nothing, under the banner 'open source

While the trend you highlight is disturbing, the alternative is worse: back-room deals between the same "big companies" to stuff this software into your phones and web-apps without your knowledge or say-so.

Re:Why would I help Google with 'open source'?

[Ol Olsoc]

Here's the result: uncached load (first visit) with ads: 2.4MB data, 26s (!) until the display goes from white to some content uncached load (first visit) without ads: 1.7MB data, 11s until the display goes from white to some content cached load (second visit) with ads: 272KB data, 2s cached load (second visit) without ads: 45KB data, 2s

A friend had a similar experience - He opened a webpage both ways, and to see a roughly 500 word page, without blocking he received 40 MBytes of all the other presents they give us.

WAP to the rescue

[Anonymous Coward]

not. google can suck it

Reporting !!!

[Anonymous Coward]

I'm reporting you to TheHaikuLover !!!

Related article

[mork]

NY Times had an article about the cost of mobile ads, the article also had some interesting data about load-times and how much data was javascript, videos, images, embeds etc.
http://www.nytimes.com/interac...
Posting as its related to the efforts described above

Re:Related article

[hvdh]

This article is a good start. There are two things were it could improve.
1) They measured the site's total data size, but not the load time. Load time was estimated = size/bandwidth, ignoring latency, parsing time and data dependencies. Actual load time should be (much?) higher.
2) Strangely, they did not account for http compression, so they cannot have measured actual traffic size. This implies they did only measure uncached loads. Most people visit a news sitte more than once, so cached loads would be more relevant. On a cached load, scripts, design elements and other stuff does not need to be reloaded, only the advertisiing. This makes the overhead for ads relatively much larger. On the German news site bild.de, a cached load is 17% content data and 83% advertising and tracking.

Re:Related article

[Anonymous Coward]

What a terrible article, it skips bbc and includes two bit insignificant pages from detroit etc.

free cache !

[johnjones]

so basically its a set of specs that allow google to cache your webpage and allow you to track it

WHY OH WHY not simply use the standards outlined here : http://www.w3.org/Mobile/

and then produce a validator so that sites can be cached by oh I dont know the network operator or anyone including google after all are they not closer to the endpoint ?

seems very pointless and a reaction to facebook allowing publishers to push articles in their network...

regards

John Jones

Re:free cache !

[sexconker]

You can't cache live, "personalized" content.
More and more of the web is becoming a constant stream of shit with ads and tracking tacked on. You can't cache this.
The solution is to block the ads and trackers. Users WANT the constant stream of shit. The remaining static content is a drop in the bucket, and any decent browser already caches it, nothing special to do on the server/network side.

Just make your website not suck

[0WaitState]

We don't need yet another phone oriented "standard". All that's needed is for sites to clean the stupid cruft out of their web pages, and to not do brain damaged de-contented "mobile" versions. Just clean up your damned code and stop with the stupid shit.

I heard sergey brin fucks gots

[Anonymous Coward]

has anyone else heard this rumor?

Google should focus on speeding up android first!

[zuse]

Jeff Atwood wrote this great article https://meta.discourse.org/t/the-state-of-javascript-on-android-in-2015-is-poor/33889 that shows how far Android has fallen behind.

Wrong direction

[VincenzoRomano]

1. i-Mode failed in a similar trend.
2. We need a different approach for mobile displays, not a different approach to pages. Computer screens are landscape while "mobile stuff" is portrait in most cases
3.We need to close the gap between PC browsers and mobile ones. In either way. As of now the gap is wide large.

I must be a genius

[Anonymous Coward]

I figured all of this out in 1998. No jokes.

You load more crap, the page loads slower. You make it more complicated, it loads slower. Your web developer is an incompetent script kiddie, the page loads slower.

Re:I must be a genius

[Anonymous Coward]

TeX in 1978 Donald Knuth. 37 Years later we are going back to basics. I'm sure Tex would run blazing fast on anything.
Google is not going to do things too radically. But according to NYT, sites like Boston Globe need to be snuffed out until content to ads drops below 50:50. Google would do well to build in an ad bandwidth quota settable by the user, per day, so advertisers have some incentive to compete - or miss out altogether.

They can do this - or adblockers will get smarter and even provide false data. There is no turning back - not when USD $9.78 per month is stolen by so called free content. The target market has woken up and been given adblockers...

adds?

[efezinco]

The problem is not for the big companies , but for personal users who gain from this advertising, and publicity ironically is owned by google ... today news site, trends, technology and video games

Google: We tell you how to make your pages fast

[Anonymous Coward]

so that there's enough time and bandwidth for our ads and analytics. Fucking hypocrites.

Original Design

[Anonymous Coward]

Whatever happened to the original design of the web were web developers only had to describe the content and then the browsers would render it the best way for the device? I liked that. Every step away from that is a step in the wrong direction.

Re:Original Design

[WaffleMonster]

Whatever happened to the original design of the web were web developers only had to describe the content and then the browsers would render it the best way for the device? I liked that. Every step away from that is a step in the wrong direction.

LOL it was replaced by "responsive design".

AMP does *not* disable javascript

[Anonymous Coward]

Why nobody reads? AMP does not disable javascript, because creating elements using "custom elements" and "web components" allow full javascript access.

Javascript inside custom elements causes no problem because doesn't blocks the rendering of the page, just of that element leaving accessibility to javscript e good performance.

Something google is pushing on with their project polymer

How about

[Anonymous Coward]

removing heavy adverts and excessive usage of cross-scripting, to begin with?

The web has outgrown HTML 15 years ago

[Qbertino]

The problem is HTML. HTML is for documents, not the living application-like multimedia canvases we've all been using since 2000.
Flash was pointing in the right direction, but it was proprietary and Adobe screwed it up.

Simply setting up a usefull canvas layout is pure torture in HTML, with tons of libraries, JS and CSS hacks, just to get the thing sort of running.
Ginormous hacks such as Googles Polymer try to pry some sort of sanity from this plattform with a huge effort and enable modern age development, but the simple fact is, HTML is at least 15 years behind what Flash or similar approaches had to offer.
And don't even get me started on building a usefull web-application with useful clientside logic without a bizar convoluted mess of tie-ins and callbacks.

Example: This multimedia website in Flash is 16 years old. That is sixteen years . ... It's from freakin' 1999!!. It's parely possible to make such a thing with todays HTML, without becoming an all-out programming and browser expert and spending a forbidding amout of time getting it right.

HTML, CSS and client side logic - wether with JS or something else - need a massive redesign for modern day multimedia and multi-screen requirements. When that happens, performance will be sane again. I expect web components and web assembly to get us back on track a little, but that's gonna take at least another five years.

Bottom line:
The web is a mess, and frickin' HTML and the ignorant smelly boring nerds that still push it as a cure-all are to blame.

Disclaimer: I'm a senior web-developer with focus on FOSS technologies.

Re:The web has outgrown HTML 15 years ago

[Anonymous Coward]

HTML could be made much more practical if they just added a tiny number of UI tags, eg <menu type="dropdown"><mi><menu ...>...</menu></mi><mi><dialogue>You have selected the second menu item.</dialogue></mi><mi onClick="system:printpage">Print...</mi></menu> (Where MI is "menu item", analogous to LI for list items). OK, so a lot of the MI actions would be onClick events for JavaScript, but that would be a whole massive chunk of JavaScript GUI code sent to the recycle bin of redundant code.

Re:The web has outgrown HTML 15 years ago

[sootman]

> The problem is HTML. HTML is for documents, not the living
> application-like multimedia canvases we've all been using since 2000.

That's half the problem. The other half is people wrapping damn articles in "application-like multimedia canvases".

Why is 75 kB of HTML wrapped up in a 9.5 MB page? That's literally over 100x larger than it needs to be. Hey Google, I think I just discovered a way to speed up mobile browsing 100x. Will AMP do that? If not, call me.

Re:The web has outgrown HTML 15 years ago

[Anonymous Coward]

Agreed with you until you say "Flash was pointing in the right direction"... maybe for "flashy" things but for static content its a mess, overall flash was a big mistake and adoption increased just because "Designers" like the WYSIWYG approach and "scene" based animations plus was already integrated in their "professional" tools.

Disclaimer: I'm also a senior developer with heavy experience on web development not focused on web-design as Parent OP.

Re:The web has outgrown HTML 15 years ago

[cardpuncher]

If you want an app, write an app.

I had a look at your 16-year-old example. Frankly, I'd have quit after the "Now Loading" and "Click to Start" and never got as far as skipping the pointless "intro" as well if I hadn't felt obliged to go the extra step in the interests of exploring your argument.

I just want the information, particularly on a mobile device. I don't care about the "design". I don't care about "immersive". I don't want to waste my time with pointy and clicky things that shoot around the screen for no apparent reason. That's what *you* want. And that's just as bad as what all those ad-merchants want - it's just crap that gets in my way and wastes my time.

Re:The web has outgrown HTML 15 years ago

[Qbertino]

Font integration? - Only since a few years back and only with awkward CSS that doesn't compile fonts and brings along licencing issues.

Relyable Layout behaviour? - Only with hacks and tricks and JS stunts. And not even then do we have the power of a usefull layout engine. ... OK, we've got justify, which no one else offers. ... I remember clearly Flash and the assh*les at Adobe screwing us over with in a fraudulent sale of a fake non-feature in Flash CS 5. In this small detail HTML actually is better than everything else that is out there, admitted.

Intelligent Image handling? - Meeh.

Asnchronous loading of content and further pages? - Yeah, we have Ajax and DOM manipulation which is, like I said, one large tacked on Hack. HTML was clearly not built with that in mind.

Audio? - Laughable. Yeah, it finally works, after 15+years. Sort of. Great. I am over-fucking-welmed.

Video intetration? - Not even today in 2015 are we seeing it as it should be. Instead the HTML Video idiots are wasting our time with DRM and other crap.

Just about all these things can be done better by orders of magnitude. And could have been done better for a long time, for instance with Flash. ... And no, I'm not presenting Flash as the solution to this - I'm giving a living proof that it can be done significantly better, prorietary tech aside.

Has anyone considered that...

[sd4f]

Reading through the comments, many read like as if the commenter thinks that google doesn't know that their ads, scripts and trackers are not the best. I'm quite sure google knows that ads are making the internet, in general, a less than ideal experience. That's why a lot of people are using an adblocker. The thing is, has anyone considered that google may want to speed things up so that they can get away with loading up all these crap scripts, trackers and ads? After all, the less people notice that these things are taking place, the less likely they are blocking them, well that's my conjecture anyway.

Use Ghostery Browser

[bradley13]

Use the Ghostery browser, problem solved.

For those who don't know, Ghostery cannot be offered separately, because Android Apps are not allowed to screw with each other's data. So Ghostery brought out a browser than includes the blocking. The web is a lot saner this way.

Regarding TFA: I am not at all fond of the idea of yet another web standard

"author-written JavaScript"

[Sloppy]

As opposed to the other kind of Javascript.

Re:"author-written JavaScript"

[Anonymous Coward]

Ah pedantry, pedantry. I think you understand perfectly well that they've referring to the author of the page in question. And if you're going to include part of your message in the subject line, you're living in a glass house, and probably shouldn't be throwing stones.

...or we could just totally block Javascript

[Anonymous Coward]

Completely blocking Javascript would probably fix things a whole lot easier.

Problem solved.

If you're on Android...

[Anonymous Coward]

Use Firefox for Android. It's improved drastically over the past 3 or 4 versions, plus you can use most of your favorite extensions. I use it with uBlock and the built-in tracking protection turned on.

Simple fix: stop loading 10 MB of ads per page

[JustAnotherOldGuy]

You want a faster web? Stop loading every page with 5 to 10 MB of ads and trackers.

  A typical page now has 50 ~ 75K of text (content) and 9.5 MB of ads and other horseshit like auto-play video, trackers, remote metrics, etc etc etc.

Seriously, cut that shit out.

Re:Simple fix: stop loading 10 MB of ads per page

[Anonymous Coward]

But also, it's my understanding that the ads are doing real-time bidding on each page load! Seriously, no one, not the page owner, not the ad networks, nobody, knows what ad will display. Not until the auction is complete and the page renders.

They claim it's all done in milliseconds and there's no noticeable loading impact. Yeah right. If that's true then how come so many people report that ad blocking causes dramatic performance improvements? "Oh well that's due to the other guys using inefficient ad design and delivery techniques..."

Huh, somehow it's always someone else's fault, isn't it?

Better Idea: stop the madness

[WaffleMonster]

It shouldn't be surprising to anyone why performance on the web sucks ass. Each site needs to knowingly or unwittingly enlist dozens of global stalking/analytics providers so that no matter where any user goes multiple companies are watching their every move across the web. Apparently it is too difficult to bother running a stats package on your own access logs anymore. I say unwittingly because many of these firms have cross agreements with other stalking companies when you reference their crap references automatically spider out.. Pick a site any site and filter for just DNS queries from browser. The sheer volume of this senseless redundant crap is whacknutz crazy.

Other problems: unnecessary layering, people who don't know what round trip delay is and think piecemeal dynamic loading of content thru xmlhttprequest et al is some kind of badge of honor, jquery gobblygook used to emulate simple hyperlinks, metric shittons of inline javascript and pages that otherwise resemble generated code from Microsoft front page.

Some quotes from amp project:

A goal of the Accelerated Mobile Pages Project is to ensure effective ad monetization on the mobile web while embracing a user-centric approach. With that context, the objective is to provide support for a comprehensive range of ad formats, ad networks and technologies in Accelerated Mobile Pages. As part of that, those involved with the project are also engaged in crafting Sustainable Ad Practices to insure that ads in AMP files are fast, safe, compelling and effective for users.

It is a core objective of the Accelerated Mobile Pages project to support subscriptions and paywalls.

Ensuring publishers are able to get robust analytics insight is a core design goal for the project.

The problem isn't the technology. All resources freed up by consolidating efforts will just be spent on more of the same.

Opera 12.x by site prefs cookies/iframes/script

[Anonymous Coward]

See subject: Globally setting NOT to take them on ALL sites & then making some exception sites that need those to work + hosts & firewalls, I get none of it-> APK Hosts File Engine 9.0++ SR-2 32/64-bit http://start64.com/index.php?o...

FREE & not 'souled-out' to advertisers, + adds speed, security, & reliability, doing FAR more w/ FAR less, more efficiently vs. redundant browser addons & locally installed DNS servers @ home + fixes DNS' many security issues!

It obtains its data vs. online threats & adbanner blocking from 10 reputable sites in the security community!

It SPEEDS YOU UP 2 ways (adblocking + locally cached in RAM favorites placed @ the TOP of hosts for fastest resolution speed), whereas by way of comparison, other "so-called security 'solutions'" SLOW YOU DOWN!

It does all that using something you already have vs. "bolting on browser addons 'MOAR'" in addons that's usermode slower & increases messagepassing, cpu + ram overuse overheads!

* :)

MalwareBytes' hpHosts Admin (MalwareBytes employee) hosts & recommends it -> http://hosts-file.net/?s=Downl... & MalwareBytes = BEST antivirus per this VERY recent testing of them all http://www.av-test.org/en/news...

&

It's safe per it being checked by 57 antivirus programs recently in BOTH its 64-bit model https://www.virustotal.com/en/...

+

In its 32-bit model also https://www.virustotal.com/en/...

---

"The premise is quite simple: Take something designed by nature & reprogram it to make it work for the body rather than against it..." - Dr. Alice Krippen: "I am legend".

APK

P.S.=> By "yours truly" - "The Lord of Hosts" so-to-speak:

PERTINENT QUOTE/EXCERPT:

"The image this title brings to mind is of a mighty military commander, one who can at a mere word summon rank upon rank of protective power" from https://answers.yahoo.com/ques... & THAT WORD = hosts!

(Accept NO substitutes!)

...apk

Benefits the non-mobile web as well

[rnturn]

I strongly suspect that the majority -- maybe the vast majority -- of the data that is being retrieved by my web browser is not related to viewable content at all but is devoted to snazzy menu functions and, well, crap that I might not even choose to view. It like having the entire set of encyclopedias delivered to your door when you want to look up one simple thing. (Yeah... I remember encyclopedias.)

I would welcome eliminating the glitz and dancing bologna that litters most web pages today.

Opera 12.x by site prefs cookies/iframes/script

[Anonymous Coward]

See subject: Globally setting NOT to take them on ALL sites & then making some exception sites that need those to work + firewalls & hosts, I get none of it-> APK Hosts File Engine 9.0++ SR-2 32/64-bit http://start64.com/index.php?o...

FREE & not 'souled-out' to advertisers, + adds speed, security, & reliability, doing FAR more w/ FAR less, more efficiently vs. redundant browser addons & locally installed DNS servers @ home + fixes DNS' many security issues!

It obtains its data vs. online threats & adbanner blocking from 10 reputable sites in the security community!

It SPEEDS YOU UP 2 ways (adblocking + locally cached in RAM favorites placed @ the TOP of hosts for fastest resolution speed), whereas by way of comparison, other "so-called security 'solutions'" SLOW YOU DOWN!

It does all that using something you already have vs. "bolting on browser addons 'MOAR'" in addons that's usermode slower & increases messagepassing, cpu + ram overuse overheads!

* :)

MalwareBytes' hpHosts Admin (MalwareBytes employee) hosts & recommends it -> http://hosts-file.net/?s=Downl... & MalwareBytes = BEST antivirus per this VERY recent testing of them all http://www.av-test.org/en/news...

&

It's safe per it being checked by 57 antivirus programs recently in BOTH its 64-bit model https://www.virustotal.com/en/...

+

In its 32-bit model also https://www.virustotal.com/en/...

---

"The premise is quite simple: Take something designed by nature & reprogram it to make it work for the body rather than against it..." - Dr. Alice Krippen: "I am legend".

APK

P.S.=> By "yours truly" - "The Lord of Hosts" so-to-speak:

PERTINENT QUOTE/EXCERPT:

"The image this title brings to mind is of a mighty military commander, one who can at a mere word summon rank upon rank of protective power" from https://answers.yahoo.com/ques... & THAT WORD = hosts!

(Accept NO substitutes!)

...apk

AMP is not the solution

[CanadianMacFan]

I've seen it with software development and with web development. People get fast computers and fast network connections and they forget that other people don't have those. They forget that not everyone is in their country and has that infrastructure. If you want your pages to load faster on the mobile web then do the following (it will also work on regular sites) and it can save on your bandwidth costs too

1. Clean up your HTML - Many content generators add in a lot of extra space that easily double or triple the size of the file. Get rid of it but still keep it nicely formatted for easy debugging.

2. Optimize your images - A lot of graphics artists will just export their work to a jpg, png, or gif and say it's done. However with by choosing the proper format, compression settings, etc it's possible to get the file size down to a quarter or even a tenth of the original in some cases. Changing the image size for the mobile site also helps by creating a smaller file and eliminates the need for the mobile device to do the resizing.

3. Optimize your code - It wasn't a mobile site but I once was maintaining a site and the developer was using JavaScript to match items from one collection to items in another collection using an O(n squared) algorithm. It became very slow, very fast. As both collections came from a database I matched them up there and got rid of the terrible code. It was still a crap application but at least it was a more responsive crap application.

4. Stop with multitude of external ads/trackers - A couple isn't going to be too bad but it's gotten way out of hand. Every request to a different server slows down the page loading/being displayed. Keep it to a minimum. If you want to know what your users are doing on your site analyze your logs, that's what they are there for.

5. Add dimensions to your HTML - If you know what the dimensions are of any images, cells, or anything else then specify them. The browser can use that information and start rendering faster without waiting to receive that information or having to redraw the page. I hate getting part way down a page and it gets redrawn because the final elements have loaded so I have to find my place again.

6. Serve images from the right place - Don't serve your images from an application server. That's not it's place. Create a graphics server and put your images there. It'll be much faster and take a lot of the load off of your application server. Around 2005 I was with a group that was hosting a Cold Fusion site that was really slow to load. I save the HTML and graphics from the main page and put the images in a temporary directory on a graphics server. I showed the project manager two versions of the HTML. One was pointing to the production server which got the images from Cold Fusion and the other got the images from the graphics server. The production server took about ten seconds to load while the graphics server was almost instantaneously. I've seen the same thing with Java application servers though not the same difference

7. Stop the modal pop-up - This won't speed up your site but please stop the modal pop-up. I'm not going to sign up for your newsletter, especially if I just landed on your site. And especially since you brought a pop-up asking me to.

Re:Enjoy a Haiku

[Anonymous Coward]

Some anon coward
projects insecurity
through crappy haiku

Compact HTML anyone

[neutrino38]

That has been attempted several times. One attempt is compact HTML

http://www.w3.org/TR/1998/NOTE...

Been defending pc users vs. that 4 years via

[Anonymous Coward]

See subject & APK Hosts File Engine 9.0++ SR-2 32/64-bit http://start64.com/index.php?o...

FREE & not 'souled-out' to advertisers + adds speed, security & reliability & does FAR more w/ FAR less more efficiently vs. redundant browser addons & locally installed DNS servers @ home + fixes DNS' many security issues!

It obtains its data vs. many types of online threats & for adbanner blocking from 10 reputable sites in the security community!

It SPEEDS YOU UP 2 ways (adblocking + locally cached in RAM favorites placed @ the TOP of hosts for fastest resolution speed vs. remote DNS) vs. other "so-called security 'solutions'" SLOWING YOU!

It does all that via something you already natively have vs. "bolting on browser addons 'MOAR'" that's usermode slower & increases messagepassing, cpu + ram overheads!

* :)

MalwareBytes' hpHosts Admin (MalwareBytes employee) hosts & recommends it -> http://hosts-file.net/?s=Downl... & MalwareBytes = BEST antivirus per this VERY recent testing of them all http://www.av-test.org/en/news...

&

It's safe proven by 57 antivirus programs recently in BOTH its 64-bit model https://www.virustotal.com/en/...

+

In its 32-bit model too https://www.virustotal.com/en/...

---

"The premise is quite simple: Take something designed by nature & reprogram it to make it work for the body rather than against it..." - Dr. Alice Krippen: "I am legend".

APK

P.S.=> By "yours truly" - "The Lord of Hosts" so-to-speak:

PERTINENT QUOTE/EXCERPT:

"The image this title brings to mind is of a mighty military commander, one who can at a mere word summon rank upon rank of protective power" from https://answers.yahoo.com/ques... & THAT WORD = hosts!

(Accept NO substitutes!)

...apk

A faster, safer, reliable web? Talk to me

[Anonymous Coward]

See subject & APK Hosts File Engine 9.0++ SR-2 32/64-bit http://start64.com/index.php?o...

FREE & not 'souled-out' to advertisers + adds speed, security & reliability & does FAR more w/ FAR less more efficiently vs. redundant browser addons & locally installed DNS servers @ home + fixes DNS' many security issues!

It obtains its data vs. many types of online threats & for adbanner blocking from 10 reputable sites in the security community!

It SPEEDS YOU UP 2 ways (adblocking + locally cached in RAM favorites placed @ the TOP of hosts for fastest resolution speed vs. remote DNS) vs. other "so-called security 'solutions'" SLOWING YOU!

It does all that via something you already natively have vs. "bolting on browser addons 'MOAR'" that's usermode slower & increases messagepassing, cpu + ram overheads!

* :)

MalwareBytes' hpHosts Admin (MalwareBytes employee) hosts & recommends it -> http://hosts-file.net/?s=Downl... & MalwareBytes = BEST antivirus per this VERY recent testing of them all http://www.av-test.org/en/news...

&

It's safe proven by 57 antivirus programs recently in BOTH its 64-bit model https://www.virustotal.com/en/...

+

In its 32-bit model too https://www.virustotal.com/en/...

---

"The premise is quite simple: Take something designed by nature & reprogram it to make it work for the body rather than against it..." - Dr. Alice Krippen: "I am legend".

APK

P.S.=> By "yours truly" - "The Lord of Hosts" so-to-speak:

PERTINENT QUOTE/EXCERPT:

"The image this title brings to mind is of a mighty military commander, one who can at a mere word summon rank upon rank of protective power" from https://answers.yahoo.com/ques... & THAT WORD = hosts!

(Accept NO substitutes!)

...apk

Best adblocker & more (speed/security/reliabil

[Anonymous Coward]

See subject & APK Hosts File Engine 9.0++ SR-2 32/64-bit http://start64.com/index.php?o...

FREE & not 'souled-out' to advertisers, + adds speed, security, & reliability, doing FAR more w/ FAR less more efficiently vs. redundant browser addons & locally installed DNS servers @ home + fixes DNS' many security issues!

It obtains its data vs. online threats & for adbanner blocking from 10 reputable sites in the security community!

It SPEEDS YOU UP 2 ways (adblocking + locally cached in RAM favorites placed @ the TOP of hosts for fastest resolution speed vs. remote DNS) vs. other "so-called security 'solutions'" SLOWING YOU!

It does all that using something you already natively have vs. "bolting on browser addons 'MOAR'" that's usermode slower & increases messagepassing, cpu + ram overheads!

* :)

MalwareBytes' hpHosts Admin (MalwareBytes employee) hosts & recommends it -> http://hosts-file.net/?s=Downl... & MalwareBytes = BEST antivirus per this VERY recent testing of them all http://www.av-test.org/en/news...

&

It's safe proven by 57 antivirus programs recently in BOTH its 64-bit model https://www.virustotal.com/en/...

+

In its 32-bit model too https://www.virustotal.com/en/...

---

"The premise is quite simple: Take something designed by nature & reprogram it to make it work for the body rather than against it..." - Dr. Alice Krippen: "I am legend".

APK

P.S.=> By "yours truly" - "The Lord of Hosts" so-to-speak:

PERTINENT QUOTE/EXCERPT:

"The image this title brings to mind is of a mighty military commander, one who can at a mere word summon rank upon rank of protective power" from https://answers.yahoo.com/ques... & THAT WORD = hosts!

(Accept NO substitutes!)

...apk

No more ad blocking.

[DeVilla]

I'm late to this party, but I'm surprised I haven't any point out the affect on ad blocking.

If the page is pre-rendered on the server, with there be any clues for the browser to guess what are ads?