The HTML5 boom is coming. Fast | Hacker News Reader

The HTML5 boom is coming. Fast

gigaom.com

86 pointsekm73 comments

'HTML5' is losing all meaning and just becoming a buzzword.

"HTML5 is not just going to be big, it’s going to be huge — and it’s coming fast.

More than 2.1 billion mobile devices will have HTML5 browsers by 2016, up from just 109 million in 2010"

No, more than 2.1 billion mobile devices will have web browsers. Yes, they'll use HTML5 but the figure is significant because of users gaining access to the 'full web' on fast 3G connections- that has nothing to do with HTML5.

"Much of this growth will be thanks to Apple’s massive support for the HTML5 platform, according to the study. And Apple is also likely to be one of the biggest beneficiaries of the technology’s wide scale adoption. "

Actually I'd argue that Apple has done some of the biggest damage in harming HTML5 adoption. Their refusal to make HTML5 a first-class citizen in the app-writing world (and Android is equally as guilty of this) means that people are still being funneled through Objective-C pathways, locking their apps into the iOS platform.

"Because Apple has so much control over its software and devices, it will be most poised to take full advantage of HTML features as they emerge in the coming years."

Or, they'll be poised to take full advantage of their market position and make it very difficult to write any kind of cross-platform app. We'll see.

While I don't disagree that the term "HTML5" is overused... I thought it was obvious that the expansion of "HTML5 browsers" was in reference to well performing and compliant WebKit browsers that can leverage the richer and more featureful... features of HTML5.

I also don't see how the native platforms in iOS or Android impede anyone's ability to use purely web technologies. At all. I find that to be a disingenuous or at least off topic insult.

> Or, they'll be poised to take full advantage of their market position and make it very difficult to write any kind of cross-platform app. We'll see.

Are you trying to imply that Apple is going to disregard the HTML5 specs or fork it in some way that makes web apps for Android and iOS incompatible? I guess I don't understand where you're coming from... at all.

>At all. I find that to be a disingenuous or at least off topic insult.

Is it even possible to insult a multinational company?

My point is that the article suggests that Apple is a big friend to HTML5, and that the growth of HTML5 will be in large part due to Apple's "massive support". I do not believe that Apple is a "massive" supporter of HTML5.

Apple's first priority is securing their own walled garden, not HTML5. HTML5 and its associated technologies are capable of creating fully-featured, offline-capable web apps that work on a variety of smartphone platforms with minimal changes. Apple will not allow such submissions into the App Store, and instead channels users towards an Objective-C, Apple-only path. They are pro-HTML5 when it suits them (fighting against Adobe) but when it threatens their walled garden, they discard it.

It isn't my intention to single them out in this- Google is just a bad with Android, as is MS with Windows Phone. The only reason I'm focusing on Apple is because the original article made them out to be a big flag-waver for HTML5, and I think that is disingenuous.

Again, how is having a native platform in any way shape or form hostile towards HTML5? The only possible way I could understand your paranoia was before Apple turned on Nitro for their embedded web views. Now that they've done that, there's simply no reason to believe that they would deprecate their browser or fork HTML5.

Besides the article doesn't say that they're a big supporter (though I think they are: WebKit, Safari, Mobile WebKit, millions of iOS devices, etc) of HTML5... it simply says they are putting capable browsers in the hands of users.

>Apple will not allow such submissions into the App Store

A false accusation, and more importantly... HTML5 applications using those technologies would not even [need to or benefit from] be submitted through the App Store. That's the entire point. What are you even talking about? The only way your argument makes any sense is if you're implying that Apple scrap Mobile Webkit, remove Mobile Safari, or intentionally REGRESS their own mobile browser to prevent native-app-like-features from HTML5... which, I'm sorry, but I find to be a ludicrous assertion.

Who isn't bad in your scenarios? Every platform in existence has a native layer in it that you could "cite". If it weren't for Google (Android, Chrome, WebKit), Apple (WebKit, Safari, Mobile Safari) and Microsoft (IE9, Mobile IE9), we wouldn't even be speaking hypothetically about web apps as the future, as they'd be impossible!

At this point I really don't know what to reply- I really don't understand how on earth you'd read the comment I posted and interpret it as "Apple should scrap Mobile Webkit". You seem to have interpreted my comments to be the exact opposite of what I was actually saying. How does "I think Apple are holding back HTML5" (my original comment) end up meaning "I want Apple to remove HTML5 features from their phones"?

There is no reason why App Store apps could not be written with HTML5. It works offline. It has local storage. Apps written in such a way could work on iOS, Android, WM, WebOS... the whole lot. Yet they are not allowed in the App Store (or Android Market, etc. etc.)- this holds back HTML5. (and of course they would benefit from it- users go straight to the App Store to download apps. Offline web sites just don't have the same understanding)

I am not suggesting that the manufacturers throw out their existing native layers- there are times (3D games, etc) where they are entirely appropriate. But while they are the only option, developers are forced into walled gardens when writing apps. I can't work out how anyone would perceive that as a good thing.

I really don't mean to sound rude, but you're not reading what I'm writing and I'm honestly not sure that you understand HTML5 or the capabilities that current mobile devices have in the HTML5 front.

You can absolutely publish fully functional HTML5 apps right now and any user on any mobile platform can use them. Even more, you can make simple Web View wrapper and publish them to Market, App Store, etc. Surely, surely, you're not really sitting there saying they have to build some new manifest and packaging format to support web apps... in their NATIVE application store, right? The entire point of web apps is that you don't need the concept of an "app". The app is the webpage as it's displayed in the browser. If you want to avoid that perception issue, then use an embedded web view wrapper.

My point was that all of these functionalities and abilities are in existence RIGHT NOW in (at least) the WebKit browsers. You're acting like it's not possible now, or that it won't be in the future. My point was that your accusations imply that Apple/Google/etc will, at some point in the future, go back and remove the Location API or Local Storage APIs or future Device APIs from their browsers... (since that's the only way your accusations make any sense).

You continue to act like the native platform or native app store somehow impedes the ability to use the browser or web views, which is just either above my head or just plain wrong.

"My point was that your accusations imply that Apple/Google/etc will, at some point in the future, go back and remove the Location API or Local Storage APIs or future Device APIs from their browsers..."

Throughout this entire discourse (and I'd challenge the claim that you've "tried to be nice", unless you just didn't try very hard) you've thrown out various accusations of what I think, then dismantled them- despite me having never suggesting anything of the sort in the first place.

Why on earth would I accuse Apple/Google of planning to remove HTML5 APIs? I genuinely have no idea why you are projecting these ideas. My point is that HTML5 could be a first-class citizen when it comes to app making. It is not. And the fault for that lies with the phone manufacturers. Therefore, in my mind, Apple does not have "massive support" for HTML5.

But hey, let's leave it here, while you're trying to be nice. I'm not sure I'm interested in finding out how condescending you'll be when you aren't.

It's amazing how - I think at least - the web is moving away from framework based technologies to HTML, CSS and Javacsript as primary sources. Indeed, Windows 8 is pushing support for such technologies and actually supporting a move away from framework based technologies to the HCJ combo. Even now with the popularity of HTML5 API's and NodeJS - the push away from traditional frameworks is becomming evident. Just look at Silverlight/Flash as sources of this - MSFT touted Silverlight as the "new future" and now its all but giving it the big middle finger.

Is this a good thing ? I think that Google and Apple are intent on killing Microsoft and Adobe respectively and they believe the best way is to strike at their respective hearts is by reducing their core platform reliance. I'm not necessarily of the view this is the best way forward, and data security and privacy laws having a long way to go before we all start using "Chrome Books". I just sometimes wonder whether Google/Apple self-interest in pushing ahead HTML5 is, and will, benefit the developer community as a whole in development of web technologies or rather create some sort of HCJ hybrid "gadget" '@web application@' community.

Further, I'm not always convinced that writing lines of HTML5 and Javascript will replace the eons of work [flagrant exaggeration intended] that has gone into traditional languages. Just like the HTML4 spec, it will be a while before the HTML5 spec is "truly adopted". A simple DocType switch and you're "technically" in HTML5.

"The HTML5 boom is coming. Fast" - maybe it is, but I don't know whether that represents the "super exciting" future we all dream of - or whether it represents a flood of new poor constructed web technologies aiming to be the next "Facebook". Call me a cynic - I just think we need to preserve our roots as much as we do need to forge a better web future.

What do you mean by 'framework based technologies'?

There are many kinds of frameworks in many languages and technologies.

I believe he meant technologies that have the bulk of their logic server side, as opposed to on the client side (ie, old Twitter vs. new Twitter).

The iOS and Android browsers have laughable HTML5 support. On paper they may seem OK, but implementations are highly buggy at best. (Speaking from the perspective as an HTML5 game dev.)

WP7 might be the first to have a good default browser for HTML5, and that kind of dumbfounds me. Get on the ball, Apple and Google. I know it's in your best interests to keep the more powerful stuff out of your mobile browsers, but the sooner the better.

Totally possible that when you get down and dirty with games they are not great, but from the perspective of a general web dev they are awesome and let you do any of the standard things you'd want to on the desktop side with ease (again, talking about regular web apps here, not games).

True. But the buggy stuff for games is still stuff the mobile web apps can use to enhance themselves, and bring them closer to native apps in features and even performance.

Are you talking about the un-released version of WP7, Mango? Because as far as I know HTML5 support is missing from WP7 currently, with its IE7 mobile browser.

If you're going to compare the IE9 mobile browser, compare it with iOS 5.0 browser and Android 4.0 browser (also coming out this fall).

Yeah, I was talking about Mango. I don't think Android 4.0 has been announced, but I will look into iOS 5. But if Mango hits before iOS 5, even if the iOS 5 browser fixes all the bugs and adds some HTML5 features, it will be hard to stack up to the desktop-class IE9.

This is an odd disparity between desktop and mobile because IE9 is definitely the worst HTML5-supporting desktop browser. But on mobile it's going to be top dog.

Yeah, for all the HTML5 webapp ferver, if you've actually designed/developed one for mobile devices you know how bad it is. Mobile devices are simply too underpowered at the moment to run huge amounts of interpreted code without struggling.

> Mobile devices are simply too underpowered at the moment to run huge amounts of interpreted code without struggling.

I don't really think that's established yet -- Android's default browser doesn't have the same engine that chrome does. It's obviously going to be slower than desktop, but it might still be improved to the point of being fast enough.

Also, see this post about code generation on ARM: http://news.ycombinator.com/item?id=2805825

It's every established. Go build even a basic list with a slide transition, wrap it in Phonegap, and deploy it to an HTC device from the last year running Android 2.3. It won't work well on 90% of the smartphones out there. Simply put, even the latest devices struggle with basic CSS animations/transitions.

Hell, scrolling isn't even smooth on most devices for even a fairly simple DOM.

You misunderstood my point -- I was not talking about the current situation, but what today's devices will be capable of with software from a year in the future.

Even now, I believe Mobile Firefox is much faster than the stock webkit browser for javascript processing.

What relevance does iOS's HTML 5 support have when comparing to flash? (Or did I miss flash arriving on the iPhone?)

On android, you can use alternate browsers. If google doesn't improve the built-in browser, something like Firefox can displace it.

This is true and welcome. But the stock browsers will be how the majority access web apps (until there's an absolutely huge gap in features and performance like IE when FF hit the scene, but that will be years in the making.) The point is that Google shouldn't rest on their laurels because alternatives exist. If that were the case, we wouldn't have any Google products.

Youtube's HTML5 player is still very alpha and poorly designed (still no fullscreen?). It has hardly changed in over a year since it was released.

I question Google/Youtube's commitment to moving to HTML5.

AFAIK, it isn't possible to watch fullscreen video in HTML5. At least, not in the way you with Flash.

Go to http://teevox.com and press F11. Works fine. Youtube could easily have a "full-page" button that also asks you to press F11.

It's really not an ideal solution. Does F11 even work on the Mac? I F11 in Chrome, I get no (ahem) chrome, but if I F11 in Firefox and IE, my tabs are still there.

(EDIT: just realised they disappear after a little while. Interesting.)

In any case, the fact that there is nothing clickable is baffling. I really don't understand why no-one thought to tackle this when discussing the <video> tag.

I love this behavior actually. Besides the fact that Flash has had years to get fullscreen right, I've yet to have it worked 100% as expected and I might as well light myself on fire as try to use it in Linux.

(With HTML5 <video>) I can easily get the video to go fullscreen in the window (as is the behavior of "fullscreening" video in Chrome) which is nice because I can have a video play and not have to have my browser window at the normal size... and if I want, I can press F11 and have it go full screen (without borking my computer in the process). (Who cares what the button is... is the clipboard going to fail because it's Cmd+C in OS X and Ctrl+C in Windows?)

edit: I can't reply any further, but this provides context for why this decision was made. When considering mobile devices, this seems like an even more prudent choice: http://stackoverflow.com/questions/1055214/is-there-a-way-to...

It's possible in Safari. Mozilla and Opera have said they'll support it in the future. Here's an example in Safari:

http://videojs.com/

New mac user here. How do you do ANYTHING full screen in safari? The green plus button just makes a big window, not a full-screen one.

Not to mention that pausing still puts it in a weird state where hitting play doesn't make it pick up again. You have to slightly move the knob.

If I recall it correctly, flash versions also have this problem (intermittently).

Yea, I get this all the time. Probably happens about 10% of the time I pause the video.

I've always questioned it. Reason: Webcam support.

YouTube was founded on the idea that people could record via webcam and post videos immediately with no knowledge of video files, etc.

How is HTML5 solving this problem for YouTube?

HTML5's answer is still in unimplemented draft form: http://www.w3.org/TR/2010/WD-html-media-capture-20100928/

I thought I read somewhere last year that Flash was dead. Yet, Flash is still here.

I think everyone can agree that we all understand that Flash is on its way out as a tool for websites. Those stupid intros and Flash menus are already dying out thanks to javascript. Nothing to do with HTML5 at all.

So HTML5 will offer us similar capabilities to Flash using Canvas. Does that mean in a few years we can start hating on Canvas because it'll do all the same things we hate Flash for? Banner ads in Canvas anyone?

Flash video you say? Not until some form of DRM is in place. Content providers probably wouldn't care for that right-click "Save Video As..." choice we would have. But, most likely, it'll eventually happen.

So, can silly articles like this just stop already? If all you know is Flash on web pages then, yes, Flash is indeed dying. Screw that, as a website tool Flash is already dead. But, if you can think beyond that, Flash is turning into something else entirely. If you pay attention to things Adobe is doing or has done, such as embracing HTML5 with their related technologies, then you shall see that Flash dying on websites is far beyond Adobe's concern.

Agreed. Copyrighted content delivered solely via HTML5 has a long way to go. Proprietary runtimes controlled by large companies with DRM technology are currently the safest method (for copyright holders) to deliver that content to a paying audience.

    “I think the disappearance of Flash is closer than 
    people think,” ABI senior analyst Mark Beccue said 
    in a press release accompanying the data.

This is from the mouth of someone who doesn't really understand what Flash is, apparently. HTML5 may dethrone Flash as the defacto video delivery system. But, to say HTML5 completely obliterates Flash is...kinda hyperbolic in a way.

So what reasons are there to use flash over HTML5, I ask myself and the HN crowd?

  -Video. (Mwoah, only non-WebM video, can't really count this one)
  -Webcam
  (thanks garethsprice)
  -Simultaneous sounds (thanks AndyJPartridge)
  -Cross-browser support (browsers interpreting HTML5 differently) 
  -???

Non competitive advantages:

  -Games (see Angry Birds)
  -3D hardware acceleration (both have the same security issues exposing shaders)

Competitive advantages HTML5 over flash:

  -2D Hardware acceleration
  -Open
  -Cross-browser support (flash has it's update/versioning problems)

Cross-browser support. Getting much better, but HTML5 is still a pain to test/debug cross-browser (or, "in IE and in everything else", which is mostly what CBT comes down to)

For me, the biggest problem I have trying to use HTML5/Javascript/CSS for a game is sound.

I can't get background music and sound effects working simultaneously under iOS using just those standards.

is that a deficiency of the iOS browser or the HTML5 standard?

iOS mainly: Sounds need to be triggered by touch events, which of course for gaming seriously restricts what you can do.

EG: No sound effect from an alien shooting at you.

There's a pretty massive software ecosystem surrounding Flash. Tons and tons of mature libraries for Flash. Not so much for HTML5 (though that will change with mass adoption and time).

I think video will remain one of the last applications of Flash on the web, because there is no way to implement DRM in HTML5. What's interesting is that even the latest video chat applications by Facebook and Google do not use Flash but their own proprietary plugins (though Google aims to replace this by standard web technologies).

What other use cases are there for Flash (let's say in one year, when HTML5 capable browsers will be the default for everyone and their grandma)?

DRM will be obsolete, because there's no consumer demand for it.

There's tons of demand from consumers, if the consumers of your product happen to be movie studios.

end-consumers.

There's plenty of content-provider demand, though. :)

But there's consumer demand for content. By your logic, the web will be advertisement free in a few months.

You're trying to dilute Flash into a few "kinda necessary" features.

  * The kind of 3D coming from Flash Player 11 makes WebGL look bad.
  * Audio API?
  * How about recording via webcam (kinda important to YouTube).
  * Not sure how HTML5 stacks up against Flash in terms of handling advertising 
  (also important to YouTube), but I think Flash wins here.
  * Flex, Adobe AIR?

Oh yeah, technology also evolves.

As much as I love the concept of open web technologies, a revolving truth has become painfully clear: What you might be able to do in HTML5 tomorrow, you can do in Flash today.

I'd phrase it slightly different: What you will be able to do in HTML5 everywhere tomorrow, you can do in Flash today on desktops (assuming your browser does not crash and you can live with the roaring sound of your cpu fan).

Or even better: What you might be able to do (pending royalties, patents) tomorrow with HTML5, you can do with Flash today without worry of being pursued by a consortium representing a specific codec you may or may not be allowed to use.

The "my fan turns on when I use Flash" is weak. I use a MBA and I can tell you just about anything involving video or any sort of high resolution graphics gets "the fan spinning".

I love the "you can't play Flash video for more than 6 hours on mobile devices!" argument, too. You know what else I can't play for 6 hours? Angry Birds, Netflix, Sonic All Stars Racing or even TuneWiki (audio) or any of the other semi-demanding apps I use. So Flash is supposed to be better than all that?

Yes, HTML5 is coming fast, and I think we actually have Microsoft to thank because previously their lack of support was hindering adoption. However, they are making incredible strides in IE9 and IE10.

http://html5test.com/results.html

Now, if they would only adopt WebGL.

As a Web Developer I've been doing HTML5 for a few months now. I just use JS libs to fix old browsers. I've produced 20 or so websites already all using HTML5 and none have had "ZOMG YOU AREN'T SUPPORTING MY BROWSER" complaints. (Daily traffic to these 20 is about 3-5k uniques each, it's not a massive test base so take it as you will.)

Is it really AAPL that is giving massive boost? How about GOOG?

Ohh..c'mon! Feels like I'm reading AAPL fanboi blog.

Well, before Google started with Chrome you could say that was true, they were doing impressive things with WebKit. I am not counting commits and their weight now, so hard to tell how it is today.

I'm fairly certain the point is about getting good (WebKit) browsers into the hands of users... not the attribution of features and commits to WebKit.

The only thing that will keep Flash alive even once HTML5 is pretty much universally supported will be the content control Flash has. With the <video> or <audio> tag, all you have to do is view the page source and you can download anything. I think companies like Vevo will fight hard against that.

I see your point and I agree that big companies will probably want an offering which provides some content control, but I don't think that flash (or anything else) does or can do a good job at this. It's hard to have open standards for DRM since most content protection relies heavily on security through obscurity.

Yeah, I mean, it's not like there are any Flash download tools now or anything.

Or even digging through various directories to find where your video's been downloaded to.

For a while there, Flash was insisting on deleting its files (the directory entry is removed but the handle and content remain) in Linux. You can still find the filehandle through /proc and restore it.

Content control doesn't work so long as users own their systems and you've got to distribute it in some means or another.

Browser trends from a fairly-well-trafficked site, this month: 31% IE8, 18% Firefox, 16% Safari, 11% Chrome, 11% IE7, 10% IE9, 3% Other (IE6, mostly).

IE8 has dropped 12% share in a year. Firefox has dropped 2% in a year. Safari has climbed 12% in a year (iPad and iPhone usage). Chrome has climbed 6% in a year. IE7 has dropped 7%. IE9 obviously has climbed 10% share.

That's 55% of audience that has an HTML5-capable browser. Enough for us to put in CSS3 tweaks, selectors, etc., but not enough to abandon the metric ton of legacy Javascript and div soup that we have. At its current rate, IE8 is going to be around for a long, long, long time, much like IE6 was.

Mobile Safari is not the same thing as desktop Safari, just like Android browser is not Chrome, Opera Mobile is not desktop Opera, and mobile Firefox is not desktop Firefox.

It's completely misleading to suddenly add iOS browser market share to desktop Safari market share and go like "Wow. Safari market share grew 12% - in a year!" - when Mobile Safari has grown at a steady pace since 2007 and has never been considered part of desktop Safari's market share.

They may share code with their desktop counter-parts, but they are not the same, and if you are going to lump the desktop version with the mobile version together, then at least do it for all browsers.

See http://www.brucelawson.co.uk/2011/html5-notes-for-analysts-a...

Where do Flash games come into this? I'm not very familiar with the development of a flash game but some guy in this article claims "Flash will disappear" or something. I assume this means whatever source the games have will just be compiled into some other sort of file compatible with HTML5? Or am I missing something completely?]

He is implying that new content (including games) will be created in HTML5 using Canvas and JS instead of in Flash. While google has made a tool that can convert some content from Flash to HTML5 automatically (Swiffy), it only supports very old Flash content.

I've always wondered how games written with javascript on canvas would handle cheating since you can type javascript code into the web page you are currently looking at.

Adobe also has a tool for converting Flash to Canvas: http://labs.adobe.com/technologies/wallaby/

If you write the javascript code inside an anonymous function, outside code by the user can't interact with it.

However, the user can still use javascript to manipulate the dom, and hence the canvas (I think).

It says the HTML5 standard is not slated to be complete until 2020. That seems ridiculous. Why so long?

Because by definition it requires three browser wendors with _complete_ implementation of the standart. Looking at it this way we have no version of HTML complete.

Because they can't even agree on a database schema. HTML 5 is a total disaster zone.

Unless things have changed recently Mozilla and Microsoft are refusing to support the SQL lite definition. MS avoiding the open source in favor of there own SQL implementation and Mozilla going to for a super cookie hash table nosql thing.

Video is a disaster, audio the same, I'm sure theres more. I've given up and am only supporting iOS HTML 5 for the foreseeable future.

so fast that gigaom missed it, apparently.

I DON'T KNOW WHAT HTML5 IS, BUT SIGN ME UP FOR SOME. I'LL TAKE 13!!! BUY NOW!!! CAN WE INVEST? IS IT PATENTABLE??? DO IT NOWWWWW.

"The HTML5 boom is coming. Fast."

That's what she said?

Reply on news.ycombinator.com