Hacker News new | past | comments | ask | show | jobs | submit login
HTML 5: Could it kill Flash and Silverlight? (computerworld.com)
21 points by 10ren on June 30, 2009 | hide | past | favorite | 28 comments



How many more threads on this nebulous question?!

http://news.ycombinator.com/item?id=661030 http://news.ycombinator.com/item?id=679886 ...and several other related threads.

Here is the answer: Yes, if enough people adopt it sufficiently fast. But they probably won't. The end.

Here is the meta answer: if sufficient articles are written on the subject, recession-weary programmers will learn the new thing to gain an edge because they're worried about developing for a dying platform, and the more innovation will take place on the new platform than the old. In that regard, the recession is good for HTML5 because it raises the economic stakes for developers.


A big, key component is the creation of proper tools for building HTML 5 apps. If someone could start building an HTML 5 tool that makes simple Flash-style animations as simple to build as Flash itself, that would dramatically increase uptake. As it is right now, you really need to be a programmer to use it.

Microsoft has realized this with Silverlight and has working on these kinds of tools for the next release. I wouldn't be surprised if we start seeing more use of it once those tools are out there (which I'm fine with, since it seems faster than Flash in my experience).


Right now the biggest industry tool for building standard-based websites is still Dreamweaver from Adobe.

Interesting to see if they will be quick to add HTML5 features to Dreamweaver to stay ahead of the curve, or try to protect the Flash platform by being slow to add the features.


Adobe really impress me as a company who has been through that territory a few times, starting with fonts in postscript, way back in the mid 80's I think, only 2-3 years after the company's formation. With their recent(ish) acquisition of Flash (Macromedia), they also opened up the Flash fileformat and protocol (yeah, flash is an open standard), which seems to me to have been intended to deflate nascent open source alternatives. (If you haven't heard of them, it's measure of the success of Adobe's strategy.)

I think Adobe is adding support for HTML5 as we speak (or and perhaps a standalone product); that it will only release it when it is strategically advantageous to Adobe; that it will protect Flash by improving it beyond the features available in HTML5. Note how Photoshop has managed to stay ahead of the Gimp.


But it's still owned by Microsoft, which isn't okay in my experience. ;)


When you say "simple Flash-style animations" did you mean to say "annoying advertisements"?


HTML5 and JS, as-is, don't quite match all the capabilities of Flash(perhaps Silverlight too, but I have less familiarity there). This is not to belittle the fact that the feature gap is definitely narrowing, but as long as the incumbent Adobe can continue adding strong reasons to stay on Flash and Flash supports "the majority" of client platforms, support isn't going to completely crumble.

In Flash 10 the features have included optimizations in the form of typed arrays, some 3D+shader support, and fully client-side file save and load operations. These things are all missing in our present-to-near-future HTML/JS platform, so I think Adobe knows the plot here quite well.


Don't forget Flex/AIR that let you develop hybrids for the browser and desktop.

Flash has slowly matured out of the "punch the monkey"/videoplayer/games niche over the last years and the >90% installed userbase is not going anywhere soon.

I despise adobe and the various problems (linux plugin) like the next guy. But still I'm using it for a lot of frontend work - simply because creating a RIA with HTML and Javascript is like peeling your eyes out with a spoon.

HTML5 won't change that fundamentally, at best it's a babystep into a welcome direction.


Downvotes without comment are just lovely. I realize this is not a popular opinion ("OMG, he defends flash!"). But unless you have worked with both HTML/JS and Flash/Flex it'd be great if you could at least add a comment to your vote.


Agreed. I work with Flash developers, and the way they work is so many light years different from how web programmers work that I wonder how anyone can think it's even comparable to HTML/JS at all.


I think the only thing that can hold up HTML5 - like other web standards - is IE. But as others have noted, IE has tiny market share on mobile devices, and, I suspect, less among young people than old. So here is a winning scenario: mobile web use continues to grow. Sites that target young demographics use HTML5, and their core users are fine with it. As they grow in popularity, IE is forced to support the standard as it becomes obvious that it's "that sucky browser" if it doesn't.


What is this obsession with things killing other things? It doesn't even make sense metaphorically.


It does, as death is basically being obsolete in the technical world.


Why hasn't anyone developed a plugin for Internet Explorer that loads a modern rendering engine?

That way, getting someone to install an HTML5 compliant browser would be as simple as downloading the Flash plugin.


Then you'd have to get IE6 users to install the plugin. The fact that they don't bother to upgrade in the first place tells you something.


But they'll go to the Flash website to install the new plugin when it tells them to.

I think it's the difference between a little dialog helpfully pointing the way to the software they need to use (and the icon that they click on still being the same) and having to actively seek out an upgrade.


flagged as duplicate and dumb.


No. This is the fifth story about HTML 5 that has made it to the front page in the last 24 hours and as much as I want to it to replace Flash and Silverlight realistically speaking it won't happen anytime soon. If anybody really believes that IE will implement HTML 5 anytime soon they seem to have a very short memory. Like it or not unless Microsoft decides to implement HTML 5 it isn't going to take off.


I'm not so sure that IE is that important for forward-looking development. It has a rapidly diminishing marketshare on the desktop, and that trend looks likely to continue. Firefox has a staggering amount of momentum right now, and Safari is gaining as well.

Moreover, IE has no presence on the iPhone, Palm, Symbian, or Android. That's tens of millions of new devices that Microsoft has zero leverage over. Unless they pull a real miracle out of their hat with the upcoming Windows Mobile, Microsoft is going to be a bit player in the uber-important mobile market for the foreseeable future.

IE will likely linger for a long time, but at this point it's looking a lot like Novell NetWare ten years ago: still big, but not driving the market.


I suspect IE's market share will never go below X% (where X is high enough to be significant) as long as it's the only browser bundled with Windows and Windows itself has most of the desktop market.

As long as that's the case, businesses will have to support it. IE may not be driving the market, but can still hold it back.


It'll only hold you back so long as you're wanting to capture 100% of the potential market. Obviously Facebook and Google can't afford to give up any potential customers, but a lot of businesses can.

How many websites have you seen that demand IE? That was acceptable because IE was a majority, and offered certain features like ActiveX. Once IE becomes a minority, isn't it reasonable to demand that users of your HTML 5 application use something else?


My claim is that IE's market share will not become insignificant for a long time. You seem to have read it as "IE will retain nonzero share for a long time".


"IE will likely linger for a long time, but at this point it's looking a lot like Novell NetWare ten years ago: still big, but not driving the market."

Sounds like you forget to include the 100 million enterprise desktops locked down to ie by their employers.


Actually, that's what makes them like NetWare. If IE fails to keep up, those dominos will eventually fall, albeit slowly.

There's still huge NetWare shops (like the California Energy Commission), but they're ever-so-slowly dropping away. Install base doesn't matter. Growth rates do.


IE is the standard where I work. Still lots of people just go ahead an install Firefox.


A lot of people use "HTML5" in a very vague manner.

One of the principle goals of HTML5 is to fully reverse-engineer and document some of Internet Explorer's behaviour so that the other browsers can duplicate it (even in failure scenarios). So a fair bit of HTML5 is already implemented by Microsoft whether they like it or not.

Some other aspects of HTML5 have indeed been adopted by Microsoft and implemented in IE8.

http://a.deveria.com/caniuse/#agents=trident&eras=All...

What's going to "kill" Flash and Silverlight, is specific parts of HTML5 such as audio, video and canvas tags which Microsoft have no interest in supporting as it's against their business interests. Both the need for a Microsoft desktop OS and Silverlight are called into question by useful web standards.

What will (continue to) happen is people using whatever bits of the standards they can and falling back to some mix of Javascript, Proprietary IE features and/or Flash. This is standard operating practice for any serious web development. How do you think Google can use SVG in it's maps when that's not supported by IE?


Falling back can be very expensive though when you start to talk about things like Video & Audio, where the cost to create and maintain different formats of large media assets is high.


Interestingly you can deliver vorbis audio natively, via Java, and via Flash 10 (see https://launchpad.net/fogg), so for audio you only need one format for 99% coverage.

And despite that fact no-one talks about it, Vorbis is actually state of the art, competitive with both AAC and AAC-HE at their respective sweet spots but without the problem of AAC-HE degrading badly on decoders that only support AAC (e.g. Apple stuff).

(Vorbis is also patent clear, used widely by big companies, well supported in hardware, and music seems to have moved away from DRM so it basically trumps every objection made to Theora.)

For video, though people like to talk about H.264 as if it was a single codec, it's actually a family or toolkit designed to go from web video on small portable devices up to Hollywood Blue-Ray on 100" projector televisions. (I've seen people seriously claim Theora is unsuitable as a standard web codec because it doesn't handle film grain as well as H.264. Film grain!) Unfortunately you can't encode one H.264 stream that'll work on everything as encode and decode complexity ramp up harshly as you move up the profiles.

I predict that since the iPhone (and similar) is the hot platform right now, and it doesn't have Flash, you'll see a lot of places following Kroc Camen's lead and encoding twice, one for Theora and one for iPhone compatible H.264 which will then be deliver by Flash for non-Theora platforms.

http://camendesign.com/code/video_for_everybody

The kicker here is that iPhone compatible H.264 is far from the best possible quality you could squeeze out for desktop use, but is more than good enough for most folks so the two videos will be basically equivalent in quality.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: