HTML5's Flash and Silverlight 'killer' potential chopped

Much has been made of how HTML5 will "kill" proprietary media tools and players from Adobe Systems and Microsoft. The idea has been partly predicated on the fact those working on HTML5 would enshrine a baseline spec for audio and video codecs everybody could agree on, buy into, and support. It's been helped along by …

[Anonymous Coward]

Silly opinion

This is just a silly opinion!

Why do they need a standard on audio or vidoe codec? Since when is it important ? We can see now images in JPEG and PNG and GIF, There is no STANDARD on that! So incase the media will be in H264 then QuickTime or any OS codec playback engine can take over the HTML 5 video tag. The browser doesn't need to implement that!!! It can rely on OS features. On OSX rely on QuickTime on Windows rely on Windows Media Player etc.... This is just silly opinion, written by somebody that has no understanding of how media technology actually works.

[Anonymous Coward]

I've yet to see a good argument

for the codec being a part of the standard in the first place. Why should the browser not be format agnostic, passing content to the appropriate handlers, as already happens with image formats? Is there something I'm missing here?

[G-HAM 2000]

Idiots

If they'd just write the spec to allow both H.254 and Theora. Although undesirable in comparasion to settling on one codec, it would have satisfied both parties and be far preferable to the free for all it seems as though it will turn out to be. Most browsers would support both codecs eventually.

[Robert Grant]

Three cheers...

...for the first person to get a bit of software out that streams Theora or H.264 depending on the user's browser.

[James Le Cuirot]

Bah

I don't know much about Theora but if the quality isn't up to scratch now yet it has the potential to improve later, they could have at least gone with it anyway. No one said YouTube has to use HTML5. That would have avoided all this patent bollocks.

[h 6]

Just now?

H.264 is an HD quality codec that has been in Apples Quicktime for the past 3 years. And Silverlight is just now getting it?

Typical Micosoft. Always late.

Paris because just now getting it too.

[Anonymous Coward]

Give me anything

and let mplayer handle it. It does a far better job than any browser implementation could, I'm sure.

[Anonymous Coward]

Perhaps some hope...

Someone from Apple was posting on the Xiph (developers of Theora) mailing lists on Saturday asking for more information on the royalty-free nature of Theora, with a view to trying to get Apple to change their mind about including it by default with Safari, so hopefully Apple has realised their stubborn refusal in this matter is hampering the standard.

Since Safari makes use of Quicktime for media playback using the video tag in HTML5 support for Theora can be added simply be installing the XiphQT (http://www.xiph.org/quicktime) plugin, but it would be nice for it to just work out of the box.

[James Henstridge]

Re: Silly opinion

Those image file formats basically form the de-facto baseline spec for image handling in web browsers. There is nothing stopping a browser from implementing more formats, but a browser that omitted support for those formats would not be very useful (even if the standards don't require that support).

As there is no existing body of content being served with the <video> tag, people wanted to have a baseline specified that didn't impose patent royalties on browser makers or content producers (read up on the GIF patent issues from a few years ago if you don't know why this matters).

With that language gone from the spec, it will depend on what browser makers and content producers do. One thing for certain is that Mozilla won't be able to have default support for H.264 while remaining free software though.

[Dan 55]

Video for Everybody

Copy and paste a bit of HTML from...

http://camendesign.com/code/video_for_everybody

Problem solved.

[Anonymous Coward]

@ Silly opinion

Yep, Apple (creators of Quicktime), Google (owners of YouTube), Opera, Mozilla and Microsoft (who've all been writing browsers since pretty much as long as the Web's been around), and the W3C (which is, you know, sort of in charge of setting the standards for the Web) all have absolutely no idea about how media technology works. Good spot. Someone needs to strip these opinionated idiots of their positions of responsibility immediately.

[Mike Kamermans]

To specify:

The intention was to have the specification be format agnostic *unless* all parties agreed on a single, freely implementable, high quality codec.

A consensus could not be reached, and so the specification will be kept to "work" as it does for images right now, meaning that it's up to the manufacturers to add in all-possible-format detection. If your browsers understands that a movie uses [your favourite codec here], and can use the decoder for it, all is well. If it cannot, or doesn't even understand the format definition, then the same happens as when you try to open an image that your browser doesn't understand in current browser: a nice empty or "do not understand!" box on the page.

[Jonathan White]

@Robert Grant

Already been done, in a way

http://camendesign.com/code/video_for_everybody

bit of webcode that will detect which browser you are using and pick the appropriate encoded file to present within the <video> tag. All you have o do is provide the content in H264 MP4 and Ogg theora and it does the rest. Not streaming, but it works.

Jon

[Anonymous Coward]

Man, there're some 'tards on here today

The whole point in having a specific codec in the browser standard is to ensure that third party plug-ins aren't used. We're supposed to be working towards a stable, fast, consistent browsing experience - not one that's dependant on you having the correct version of a relevant third party's module to fire up for video.

And which idiot was comparing image compression with it's long history of pain to the current situation with video? I'm looking at you 14 year old anonymous coward.

[Andrew Ducker]

Canvas?

Surely a large part of the HTML5 spec is the canvas attribute, which will allow for arbitrary 2D interfaces - i.e. what Flash/Silverlight do.

Sur, the video bit is important - but being able to produce shiny applications on a web page in a standard way is a big part of what Flash is used for...

[Anonymous Coward]

Nothing much to agree upon

Commercial players want to use commercially licensed technology to eliminate free alternatives. The only way to gain consensus in favour of mpeg would be to exclude free software developers from the vote.

[Tom 7]

Anybody surprised at this?

Hello and welcome to the standards organisation that's going to put out out of business.

Now if we could just agree on which standard will put you out of business then we'll wave goodbye.

We've had Adobe and MS keeping JavaScript in the 20thC and now its another large part of the browser.they and others have stifled.

Now if someone would just release a browser thats JS2 and any quality open video format then it might just catch on!

[Andrew Hodgkinson]

Mired in politics

Here we are running operating systems such as Linux, OS X and Windows that typically require gigabytes of disc space on installation and almost as much RAM on boot up, yet apparently nobody expects the operating system to supply facilities for image decoding, or video decoding, or audio decoding. People seem to be expecting it to be all built into some humungous great monolithic browser "lump" because "we don't want plugins".

Making use of your host operating system's API to render images, render video, render audio, render fonts, handle colour, do printing etc. etc. is exactly what a browser *should* be doing, all the time. Just because the browser recognises the <video> tag and decides it doesn't need to launch an external plugin - and by the way, it could just as easily do that with HTML 4's <object>, for which HTML 5's <video> is really just a shorthand subset - does NOT mean that the browser has to have an entire media playback engine complete with CODECs all *built into itself*. That would be a preposterous solution.

Since the host OS does the decode, then whatever video the OS can play, the browser ought to be able to play. HTML doesn't specify image formats and shouldn't specify video formats; it doesn't even specify a scripting language! That's why HTML requires "type" (current) or "language" (deprecated) attributes on the <script> tag. As with the likes of Javascript, JPEG and PNG, it's likely that a de facto video standard will emerge based purely on the popularity of a particular container format and audio+video CODEC choice.

So why is a mandatory video format being pressed for? Because big companies are involved and the whole things has become mired in corporate politics. This has nothing to do with the engineering abilities of those companies IMHO.

Incidentally, just because a browser recognises <video> does not mean it avoids plugins just as handling <object> does not mean that a plugin must be launched. The browser may choose to simply show a "play" icon placeholder and when the user activates it, fire up some external helper application or embed a known video handling plugin instead. The helper application option may well be a sensible thing to do for web browsers where screen real estate is limited (e.g. mobile phones), making it easier for the user to control the video and, in particular, show it full screen.

[JC 2]

Dumb To Ask

The problem was asking these companies what they want instead of telling them what they get. While that might cause some of them to not support the standard, it's better to have a full standard that someone refuses to support rather than an incomplete standard that nobody can support because the feature is missing!

Since H.264 is far too expensive, Ogg Theora is the obvious choice if it must be one or the other. Is it too late to write this back into the spec?

Paris, because things go right over her head too.

[Chris Thomas Alpha]

video codec support? have some balls!

Why didnt they just say the following

Ogg will be the default video and audio codec

Then, nothing, the browser writers can fight it out themselves, the spec is the spec, why do we always have to involve companies in these things which dont make sense.

JUST SUPPORT SOMETHING AS THE STANDARD and let the world sort out the bodies, if apple doesnt want to support ogg vorbis, then ok, they dont support it and their code will fail validators that apply the standard.

same with mozilla, microsoft and google, if they want to support it, they will, if they dont, they wont and then what will happen?? nothing as far as I can see

Or if it a case that these people are paid for by these companies and therefore can't make independant decisions??

[Adam 38]

The Point... You Missed It

Software patents on codec algorithms etc... honestly, it's just retarded. We need some big players to take their hands out of their pockets and lobby for patent law change in the US.

[Mark Rendle]

Cave! It's the EU!

If Microsoft tried to build video codecs into IE or Windows they'd get slapped with massive fines by the EU and forced to release an N version without them anyway.

[Anonymous Coward]

Standards? Few bother with 'em...

As I mentioned elsewhere, a lot of websites that meet current standards (XHTML 1.0 and CSS 2.1) mostly look the same. I suppose factoring a common standard for video (which could be a good idea) is to try and make the web more accessible. A lot of declared XHTML doesn't work as proprietary tags have been used and dont get through the W3C 's parser, nor does the CSS get through the CSS parser. Most coders don't really care whether their sites pass the W3C tests, as long as their sites work in IE on a Windows box (the most common combination) and looks good. I suspect most coders don't have the time/money to check if they breach DDA rules on colour schemes (e.g. white text on a yellow background, or more commonly, grey text on a light blue background) and provide a text only alternative, audio transcripts etc. I used to buy a well known magazine that banged on about meeting standards, but the code examples given didn't get through the W3C parser. I stopped buying it, saving myself a small fortune. Building a mainly text based site with images that does parse isn't difficult to do. To illustrate my point, enter a link to your ISP's home page here: http://validator.w3.org/#validate_by_uri and for the CSS: http://jigsaw.w3.org/css-validator/#validate_by_uri My bet is that it won't pass....

[raving angry loony]

kill software patents

Another example of why software patents are such a fucking bad idea.