Firefox 3.5 Beta Boosts Open Video Standard

bmullan writes "Dailymotion, one of the world's largest video sites, announced support for Open Video. They've put out a press release, a blog post on the new Open Video site, and an HTML 5 demo site where you can see some of the things that you can do with open video and Firefox 3.5. (You can get the Firefox 3.5 beta here.) Dailymotion is automatically transcoding all of the content that their users create, and expect to have around 300,000 videos in the open Ogg Theora and Vorbis formats."

Comments

Other sites with support exist as well

[g-to-the-o-to-the-g]

There are some other sites which have had <video> support for a while now, such as omploader. It would be nice if some big sites like youtube get rid of flash too, but I'm not holding my breath.

Disclaimer: it's my site

Linux?

[goldaryn]

Firefox 3.5 Beta Boosts Open Video Standard

Well, bye bye karma... but..

How is this a Linux story/Firefox story? It's a new HTML standard. All browsers will support it, eventually.

Re:Linux?

[Bake]

Really, only Firefox? Because I could SWEAR it was working for me in Safari 4 with Youtube's HTML 5 demo site.

Re:Linux?

[RalphBNumbers]

Safari supports the HTML5 video tag, but doesn't include Theora support because Apple considers it a patent lawsuit magnet.
Thus Safari users are shown better compressed, but definitely patented, h264 streams on those sites.

IIRC some other WebKit browsers use GStreamer as the back end for their video tag support, and thus probably support Theora.

Re:Linux?

[Anonymous Coward]

Try going to http://openvideo.dailymotion.com/ in Safari 4

That landing page requires FireFox, but the actual video pages work fine in Safari.

Re:Linux?

[Anonymous Coward]

Safari supports the HTML5 video tag, but doesn't include Theora support because Apple considers it a patent lawsuit magnet.

Ummm... Apple doesn't include it because it REALLY doesn't want a free video/audio codec becoming widely used.

Re:Linux?

[macslut]

Sorry FlyingBishop, but the original poster was correct. Apple doesn't want to touch Theora, FLAC, Ogg or any other open codec/container because of the patent lawsuit magnet issue. Apple can use h.264 because they can pay to license it and be done with it. There's so much that goes into codecs and container formats that it's absolutely inevitable some patent trolls are going to pounce on any new format. With h.264, Apple is isolated because they've licensed the format. With Theora and the other open formats, any patent troll is going to go after the deeper pockets, which would be Apple, if they supported the format. Apple gets around this, sometimes, by offering a plug-in architecture that allows 3rd parties to develop the compatibility for them.

Yeah, screw you too

[Anonymous Coward]

Opera has supported for a while now. Stupid site says I'm not allowed to open it cause I'm not using Firefox.

Hmm, does this seem familiar to anyone?

Re:Yeah, screw you too

[yincrash]

If the site only needs HTML5 support, you should be allowed to view it with any HTML5 browser, rather than constricting it only to work with Firefox 3.5 beta.

Re:Yeah, screw you too

[cha5on]

If you actually read TFA, you might have noticed

Do other browsers support this HTML tag? Yes, but our code works best on Firefox 3.5 beta and is not yet optimized for other browsers. We would be happy to work more closely with developers from Webkit and Opera.

Considering that the demo is intended to show what an emerging standard can do better than current ones, it's understandable that they want it to look the best it can, which means they're going to want people to watch it using the optimized platform and not something that's barely going to run their demo.

Re:Yeah, screw you too

[Xest]

That's not really the fault of Firefox or HTML5, it's the fault of the site, but really I do think HTML5 is indeed a step backwards.

It reduces separation of the content and presentation layers and it increases parsing ambiguity by relaxing standards. Of course, ambiguity is bound to lead to a performance hit too, albeit perhaps rather small so may not really matter. This is really not great news as far as the web is concerned as it's exactly what we've been fighting against for the last decade with reasonable success - the web is certainly more portable and accessible now than it used to be.

From what I've read previously of the HTML5 spec and comments surrounding it the idea is to make HTML development more accessible, but I'm not sure this is the right way to go about things. If we're going to increase the amount of people who can publish on the internet then a better option seems to be to improve the applications for doing this - whether they're web applications (i.e. Wordpress to Twitter to Facebook to MySpace) or whether we simply make better quality WYSIWYG desktop applications. If we do this on a spec that's better built for the real web developers - those who really need clear separation of concerns to ensure their sites are truly enterprise ready then we'll undoubtedly end up with a much better web.

With tags like and so forth added it's meant to increase clarity, but really it doesn't, because ultimately it will never fulfil everyone's needs, someone will want or so on, this means they're back to something like

meaning half your markup is in the div format and half not, or you could just ignore the feature but then effectively you may as well just carry on using XHTML anyway.

Let web developers develop and let users use applications to publish - it's worked so well as many Web 2.0 successes have shown.

Besides that there's also something that stinks about forcing a standard on the web too - open or not. I think I'd rather have market forces decide a standard over a small clique of people who have their own interests and agendas which may not necessarily be the best for the web overall.

Standards should be lightweight, extensible and well defined, I would argued HTML5 is flawed in all of these areas, whereas with XHTML that is much less the case. HTML5 simply makes worse the very reasons we started to move away from HTML to XHTML in the first place.

Re:Yeah, screw you too

[i.of.the.storm]

That's kind of misleading, since there is no such thing as an HTML5 browser as of yet. All the upcoming versions of browsers that aren't IE are getting support for parts of HTML5, but it would be incorrect to say that they are all equivalent. Especially in the case of the tag, they seem to do different codecs right now. Firefox does Theora, Safari and I think Chrome does h264, and I have no idea what opera does. I'm honestly not sure how this is a better situation than flash video players, at least until everyone decides to standardize on a common format. I guess the idea is that once all the dust settles, we'll have lower CPU usage and maybe nice things like videos being cached and/or easily downloadable, which flv doesn't do easily, but until then this isn't much of an improvement unless you're stuck on, say, 64 bit linux and can't get the flash plugin to work, but that's a really tiny edge case. Last I checked I could play youtube videos under 64 bit linux, so I'm not really sure what the advantages are.

Re:Yeah, screw you too

[sznupi]

So...you're limiting Opera to official/final releases but Firefox Beta is fine?

How cute...

Re:Yeah, screw you too

[sznupi]

Yeah, it's simply exchanging browser monopoly for browser duopoly - previously we've had "best viewed in IE", now it's "...in IE & Firefox". No real progress at all.

Posting from a place where Opera is quite popular (8.5% here, 31.6% in neighbouring country (yeah, more than Gecko - 24.5%); most countries in the region have less than 50% IE usage); trust me, browser-agnostic web is a much better idea.

Re:Yeah, screw you too

[zuperduperman]

> That's kind of misleading .... All the upcoming versions of browsers that aren't IE are getting support for parts of HTML5

Speaking of misleading - IE8 already supports parts of HTML5 and Microsoft have committed to support it "in full" in future versions. Can we tone the bias down a little?

Re:Yeah, screw you too

[ClosedSource]

Actually, no version of HTML is really suited for "web applications" but that's how we roll anyway.

I'd love it if documents and web applications were really treated as "entirely different entities" (i.e. the only thing they'd have in common is the set of transport layers).

Re:Yeah, screw you too

[ClosedSource]

"Dude, HTML 5 is still in the process of being finalized, and it makes a LOT of things much easier to develop."

So, is your point is that we shouldn't complain about HTML 5 because the standard isn't finished, but we can talk about how great it is because the standard .. how does that go again?

Re:Yeah, screw you too

[Locklin]

Considering that the demo is intended to show what an emerging standard can do better than current ones, it's understandable that they want it to look the best it can, which means they're going to want people to watch it using the optimized platform and not something that's barely going to run their demo.

So, they intend to showcase an open standard by publishing something that only works on a single "optimized" platform??

While I understand the pragmatism, it still seems odd.

Re:Yeah, screw you too

[cha5on]

So, they intend to showcase an open standard by publishing something that only works on a single "optimized" platform??

While I understand the pragmatism, it still seems odd.

That the standard is open does not mean that every browser implements the standard properly yet. If you intend to showcase an emerging standard, you want to actually showcase the emerging standard. As this is such a showcase, it's perfectly reasonable to restrict presentation to those browsers capable of displaying the page as intended.

As I quoted earlier FTFA:

We would be happy to work more closely with developers from Webkit and Opera.

Based on that, I expect that we'll see similar demos running on those and other HTML5-capable browsers in the near future.

Re:Yeah, screw you too

[Xest]

Out of interest, do you feel we should have a longer term goal over and above HTML5 that does work to provide a markup language that really is actually good rather than good enough then?

I guess I just follow the school of thought that if you're going to do something properly you should do it right the first time if time permits and I do not personally see any urgency in getting a new spec out there when we have XHTML1 which is extensible enough.

I do not feel HTML5 actually brings anything to the table. HTML to me should define document structure, CSS should define presentation, and Javascript should provide scripting and dynamic content to a site, simply because this is what the 3 things are good at. I do not like the idea of using Javascript for presentation for example, even though it's been used to solve the equal height columnar divs problem in the past and that sort of thing - ultimately it can be solved in CSS and it seems better to do so. This is effectively my main reasoning for disliking HTML5, it seems to be branching away from what HTML is actually good for into areas where we already simply have better solutions cleanly separated as separate tools and standards.

Styling the UI?

[Chris Pimlott]

How does the open video format handle styling the UI? One of the reasons sites love flash for video so much is that it gives them complete control over how the video is presented, e.g. available controls, positions, colors and themes to match the rest of the page, etc. Then you have the more intrusive things, like Youtube's overlay ads, text captions, and suggested videos after playback finishes.

If open video means a widget that site owners have no control over, like Quicktime video embedding, then commercial site operators aren't going to be too keen on it.

Re:Styling the UI?

[db48x]

There's a default UI, but you can turn it off and use whatever HTML/CSS/XML/SVG you care to dream up.

finally

[delirium of disorder]

Thank flying spaghetti monster. Flash is the only proprietary software I use. I can't wait for in browser ogg theora support to take off, and the online video market to embrace it. As soon as I see it working, I'll delete my google video account and self-host all my videos.

Re:finally

[blakedev]

I can has video on FreeBSD plz? One of the major reasons why I don't use FreeBSD as my main os (even though I like it) is the lack of good support for Flash. I feel giddy.

Theora has improved

[tepples]

The h.264 codec that is used to stream their content is far and away better than that Theora garbage format.

The version of Theora that was in ffmpeg2theora 0.19 sucked. But Theora has come a long way since then, coming much closer to x264's fidelity.

Re:Theora has improved

[harryandthehenderson]

But Theora has come a long way since then, coming much closer to x264's fidelity.

You must have missed the retraction that was done when it was shown that they were calculating PSNR wrong for x264. Theora is nowhere near the quality of even a low-range h.264 codec.

'Publishing' the graph like that drew well-deserved scrutiny and unfortunately our own data was also off (although by considerably less). ffmpeg had another bug we didn't know about which caused it to mishandle the colorspace on x264 output, so the x264 PSNR value was too low by 1-4dB. Greg fixed the error in the data collection and immediately set about collecting new measures:

Re:Theora has improved

[tepples]

You must have missed the retraction that was done when it was shown that they were calculating PSNR wrong for x264. Theora is nowhere near the quality of even a low-range h.264 codec.

I read that part. VP3 was garbage, as was the alpha version of Theora included in ffmpeg2theora 0.19. But even the corrected graph shows that Theora has gone from garbage to only 3 dB behind x264.

Sort of related, because a lot of people have DivX on their PC or in their DVD player: how does Xvid (an MPEG-4 ASP video encoder) compare to recent Theora and x264?

Re:Theora has improved

[harryandthehenderson]

I read that part. VP3 was garbage, as was the alpha version of Theora included in ffmpeg2theora 0.19. But even the corrected graph shows that Theora has gone from garbage to only 3 dB behind x264.

Yeah, on one single clip using the default settings of x264 which hardly give the highest attainable quality.

Sort of related, because a lot of people have DivX on their PC or in their DVD player: how does Xvid (an MPEG-4 ASP video encoder) compare to recent Theora and x264?

Theora can't beat XviD and XviD is inferior to x264.

Re:Theora has improved

[BikeHelmet]

Theora can't beat XviD and XviD is inferior to x264.

Nothing beats Xvid for low bitrates. (The bitrates which create ~350MB videos)

It blurs more details than some other codecs, which lets it save on space and put more detail into important stuff.

h.264 gets quite blocky well before Xvid does; mind you, it does a better job preserving details when a higher bitrate is allowable. h.264 is often the preferred format for raw video footage, since at high bitrates it comes remarkably close to totally uncompressed video. (which is usually too big to do anything with)

For static content, VP7/VP8 are quite impressive, but VP3... not so much.

A presentation with slides occasionally changing works wonderfully in VP7. It'll use a couple hundred kilobytes on the first frame, plus any frame where it suddenly changes, but aside from that it won't use much/any bandwidth. The result is a video that looks like a 32bit gif animation(perfect quality).

On2's algorithms for figuring out which pixels changed are quite advanced; a 10 second video showing a slide was a couple kilobytes smaller than a 4 minute video showing the same slide... with x264 and xvid, I didn't get results like that even after spending a day tweaking everything.

On2's encoder also automatically removes minor jitter/angle abnormalities, so if the camera was being held by a person, the difference in size will be even more pronounced.

Too bad last time I checked, their encoder was single-threaded. I'm also betting "Superior codec for lectures." isn't the kind of endorsement they wanted. :P

Re:Theora has improved

[Daniel Phillips]

only?! inst 3db about 50%? huge lag behind.

No, it's about 29%

(1 - (1 / 1.41)) * 100

Why promote an "inferior" product?

[bogaboga]

While I am happy to see that Mozilla and Firefox are setting the standards, let me remind readers that previous evaluations have found the Theora encoders inferior compared to contemporary video codecs. In particular, the reference Theora encoder has inferior picture quality and network frame rate control as of 2008.

Ohh wait a minute...There is a Slashdotter who noted this as well.

Frankly, it bothers me big time. Why not wait until the standard is "up to par" with the likes of Microsoft's Silverlight or Adobe's Flash?

Re:Why promote an "inferior" product?

[siDDis]

Theora is great for embedded devices like cell phones since it is "cheap" when it comes to cpu cycles. For top quality video, Dirac should be used. I wonder when Firefox, Opera or Konqueror will have native support for Dirac.

Re:Why promote an "inferior" product?

[Phroggy]

While I am happy to see that Mozilla and Firefox are setting the standards, let me remind readers that previous evaluations have found the Theora encoders inferior compared to contemporary video codecs. In particular, the reference Theora encoder has inferior picture quality and network frame rate control as of 2008.

The important thing is that we move toward open standards, away from proprietary solutions, because open standards allow us to do more cool stuff with them.

Remember RealPlayer? Remember all the bitching about what a piece of crap it was? People had to have it, even though it sucked, because a lot of content was only available in RealAudio format. Today, RealPlayer is all but gone, and you can play the same type of content using whatever software you like. Why? Because when Apple added Podcast support to iTunes, Podcasts suddenly became hugely popular, and virtually all of the content providers that used to offer only RealAudio now offer Podcasts instead. This means that users are free to choose whatever software they want, and competition will drive the software to improve.

In the same way, if web sites move away from Flash video players to using HTML5's video tag, it will mean users will no longer be dependent on Adobe's plugin to access the content. Unfortunately we still have patent issues to deal with; Ogg is unencumbered, but better quality codecs will be supported by most browsers, and if we can get content providers to get used to the idea of making their video content freely available (instead of wrapping it up in Flash), there can be competition among codecs too.

It's not a perfect world, but it's one step closer.

Re:Why promote an "inferior" product?

[pizzach]

I still do not why slashdoters think Theora has no worth as a baseline free video codec with less legal shackles? H264 is already in the standard. I doubt it's going to disappear. It would be nice if some free (as in beer) software could ship with a working video encoder that isn't illegal in some countries. Just toss the baby out with the bathwater guys...

Re:Why promote an "inferior" product?

[the_other_chewey]

I wonder when Firefox, Opera or Konqueror will have native support for Dirac.

For Gecko (which means Firefox & friends): As soon as libogg supports it, which is pretty much
now. However, it isn't part of the upstream stable libogg yet, so it will not ship with Firefox 3.5,
but very probably show up in the version after that.

Look up the controls attribute

[tepples]

If open video means a widget that site owners have no control over, like Quicktime video embedding, then commercial site operators aren't going to be too keen on it.

HTML 5 Video states that a page can ask the user agent to show a built-in control widget (by providing a controls attribute) or hide it and provide its own widget that controls the video player through its DOM (by omitting the controls attribute).

Re:Question

[Fry-kun]

Yes, because Flash boasts huge market penetration (http://www.adobe.com/products/player_census/flashplayer/). Of course it's Adobe's own market research, but it's probably true that they have most of the market in their grasp.
Add to that the fact that IE still has the largest browser market share.
Those two practically guarantee that Google will stick with Flash for most part. Maybe they will create a dynamic service which would prefer support over flash, but Flash is here to stay for quite a while longer.

Re:Question

[harryandthehenderson]

I would assume that most users would prefer not to have to download Flash plugins..

Most users are probably more inclined to download the flash plugin that happens automatically for them versus downloading a whole new browser to get HTML5 video tags to work.

Re:Other sites with support exist as well

[Anonymous Coward]

http://www.youtube.com/html5

I hope it doesn't a quad core CPU to run

[Krneki]

With Adobe every year my CPU is more loaded when I'm watching Youtube or similar.
While using a different player, the movies uses 10 time less CPU cycles. I can't wait for something to replace that bloat from Adobe.

Sometimes worse is better

[tepples]

In particular, the reference Theora encoder has inferior picture quality and network frame rate control as of 2008.

But as of 2009, Thusnelda is coming soon. The Thusnelda encoder has already fixed some of the problems that Theora inherited from On2's VP3, thanks in part to the flexibility that Xiph added to the Theora bitstream format. Sure, it's still inferior to x264 (50% bigger rate for same distortion as of about a month ago), but it's improving.

Why not wait until the standard is "up to par" with the likes of Microsoft's Silverlight or Adobe's Flash?

Because sometimes worse is better. For example, worse can be better because it's Free and thus more available for deployment on devices other than PCs.

Widespread adoption and annoying ads are over.

[GreatBunzinni]

As soon as major sites such as youtube adopt this standard and drop that PoS adobe flash then flash will be practically relegated to crappy early 90s sites and annoying ads, which means that removing the flash plugin from any system will vastly improve your web experience. Good riddance.

The video tag has a fatal flaw - codecs

[AndrewStephens]

The video tag is great, but it has a fatal flaw (actually two fatal flaws, but one is much more important.) The attempt to standardize on a single codec was correct, but now that it has failed the video tag becomes much less useful. At least with flash you can host a video and be sure that most of your audience will be able to view it. With the video tag, even when browsers that support it become widely available, which codec do you encode the video in? Already the browsers are going in different directions, with Safari using Quicktime to play h.264.

Hopefully it gets sorted out soon. Personally I would like to see h.264 adopted if the licensing issues can be sorted out.

I blogged about this issue a couple of days ago, if anyone is interested in a longer version of this comment.

(The other fatal flaw is that the video tag makes it easy for people to download the original video file - something none of the big content providers want. Yes, everyone knows how to do this with Flash videos, but the illusion of content protection is there.)

Re:The video tag has a fatal flaw - codecs

[Anonymous Coward]

Neither flash nor h264 will be opened unless competing alternatives gets strong enough. Implement html5 tags as options in your site and hope for the best.

Re:The video tag has a fatal flaw - codecs

[aztracker1]

umn, there are several non-gpl implementations of h.264. The issue is h.264 is patent encumbered, and it's use in free systems would have consequences for the likes oif Mozilla. MS is already adding h.264 support to Silverlight, to go along with VC-1.

3 dB

[ivoras]

The meaning of "3 dB" is "twice". Decibels are a logarithmic system, used for two reasons: 1) because for large & weird systems it's easier to say "120 dB" then "a trillion" (of course, this works in certain sciences only), and 2) because our sensitivity to light, sound and probably other sensory input is logarithmic so yes, "3 dB" taken in this context can intentionally be parsed as "small". But for pre-set algorithms (i.e. made to a predefined spec), "two times" is actually a lot of space to fuzz over. You can only do so much before you need to change the very spec that makes Theora - Theora.

Re:3 dB

[yo_tuco]

it's easier to say "120 dB" then "a trillion"... " Then why not say 12 bel instead of that big 120 dB number?

Re:3 dB

[timq]

3 dB is a factor of ~ 1.41 times.

A factor of two is 6 dB.

Re:3 dB

[ceifeira]

Depends on the reference. What is 0 dB? For sound pressure (for example), you're right. However 3 dB is a two-fold increase in watts (power).

Re:3 dB

[master5o1]

I can see your sarcasm, however, it appears to be that the standardised (whether de facto or not) is decibel and not bel. Even the dictionary for Firefox doesn't recognise bel as a word.

Re:3 dB

[Anonymous Coward]

Depends on what your working with. Just straight amplitude, 3dB = double. If you're working with power, then 6dB=double.

But.. audio uses the amplitude scale.

What does dB even mean in this context?

"standard"

[Lord Bitman]

While this new "standard" format is open, it's also something with almost zero support, especially across legacy browsers.
This means Flash is here to stay, even /with/ new javascript capabilities.

Any License that will Prevent Transcoding?

[Anonymous Coward]

I have a really naive question.

Is there any License that will prevent transcoding original video produced by me, to another format, like .flv?

I'd like to make my videos open source only, including the "container".

Eye of the Beholder

[Sloppy]

[smarmy]An objective evaluation of H.264, VP6 and WMV9 show that they are still not as free as Theora. While we hope that these codec's patent holders will continue to work on this defect and catch up, as of 2009 it is still premature to say that any of them will ever be "up to par" with Theora, which totally stomps those other codecs in all freeness tests. Why promote an "inferior" product?[/smarmy]

Now for a little less smarminess: we're talking about interchange formats, used on the fucking internet where you don't know what OSes and archs either side is using. I know Theora is portable to everything and usable by everyone. I don't know about those other codecs. If you want to use WMV9 for your internal security camera, that's totally fine, but on the internet how could something like that be useful? What's the use in serving video in a format that people can't play? Theora doesn't have that problem.

Poor performance of Firefox's audio and video

[Dwedit]

So far, I have been completely and utterly unimpressed by Firefox's built in audio and video features. I'm using 3.5 Beta.

Whenever it plays a WAV file, it plays for a few seconds, then skips audio and runs at 100% CPU usage, then plays again. Sounds like a really bad buffering issue, like they can't get something as basic as buffering correct. Audio which is intended to loop does not. OGG Vorbis files also skip the same was as WAV files.

Video performance is dismal, even worse than Flash player. Videos skip and take more CPU power to play back than other players do. Upscaling the video is done slowly through software, even though Overlays surfaces have been around since 1997 with the NVidia Riva 128.

From what I've seen, in terms of CPU usage, the best video player for the web is Windows Media Player, using non-microsoft video codecs (FFDshow).

Re:Other sites with support exist as well

[TheTurtlesMoves]

Will you pay the 5 Million yearly license fees so firefox/mozillia etc can provide free legal support....

Thought not.

Oh and soon there will be a fee on *providing* content in h.264.

And I left out all the other parts of the license agreement that firefox would be forced to follow before they would be given a license. And this would make it incompatible with GPL or even open source in general.

Re:Major Typo in the Article Title

[Mystra_x64]

Safari and Opera are implementing this too. However article itself is too "Firefox hyped". Opera started playing with long before Firefox, AFAIR.

There's also YouTube's HTML5 demo

[damg]

I'm surprised there was no mention of YouTube's Flashless HTML5 demo page. I think it's a bit more popular than Dailymotion.