Slashdot Mirror
Questioning Mozilla's Plans For HTML5 Video
AberBeta writes with this excerpt from OSNews:
"We're on the verge of a serious evolution on the web. Right now, the common way to include video on the web is by use of Flash, a closed-source technology. The answer to this is the HTML5 video tag, which allows you to embed video into HTML pages without the use of Flash or any other non-HTML technology; combined with open video codecs, this could provide the perfect opportunity to further open up and standardize the web. Sadly, not even Mozilla itself really seems to understand what it is supposed to do with the video tag, and actually advocates the use of JavaScript to implement it. Kroc Camen, OSNews editor, is very involved in making/keeping the web open, and has written an open letter to Mozilla in which he urges them not to use JavaScript for HTML video."
A lot of video producers like to rely on the fact that Flash makes it difficult to download videos to your hard drive. I wonder how they'd react if a major online video provider were to provide its content through a less restrictive method such as the video tag.
"In prison you just have to shut your eyes and take it. Here you have to shut your eyes and give it."
So are we going to require browsers to install with codec packs?
No. The idea is to include the codec in the browser. But to allow that at reasonable conditions, the codec should be Free. The codec proposed for this purpose is Ogg Theora/Vorbis, an OSS codec build specifically trying not to use any patented technology. As you can imagine, Apple, MS and Adobe are not really happy about this, as they obviously would like their patented technology to be used in HTML 5, and because Apple and MS are not only video-codec-makers but browser-makers too, and not small ones, we can not just ignore them and go ahead with Theora. Implement the HTML 5 video tag in Mozilla with Theora looked like a good chance to get the open codec though, but this Javascript stuff post by Mozilla now makes it look like they have other plans.
I just don't trust anything that bleeds for five days and doesn't die.
Flash Video is unbelievably processor intensive (especially given it's pretty crappy quality), surely you've noticed that? Even on modern dual processor machines it can stutter and slow down other processes. If video could settle down like image formats, the web would be a better place for it.
Javascript is a cool language. I've written more than a couple of browser extensions and intranet apps with it in my time and I'm one of those people that disables javascript for browsing the public internet. I consider running random 3rd party code to be an outright security hole, some people are willing to sacrifice security for the bells and whistles but only a complete moron disagrees with the premise.
You can probably guess that I personally am going to disable the HTML5 A/V elements and continue downloading video manually. That aside, browser based audio/video should provide basic playback functionality for the user without javascript enabled. The functionality should also be easily disabled or switched into "prompt to download" mode and finally, I don't want to use GStreamer for anything... ever...
My understanding is that the reason that people use flash and silverlight for video is so that people cannot save, reuse, and redistribute the content.
I've run across very few streamed videos that can't be downloaded. In the olden days I'd use something like WireShark or Network Monitor to get the URL of the content. Nowadays it's much easier with various Firefox extensions.
As far as I know, the reason most sites use Flash or whatnot is because they want the video to be streamable and start more or less instantly. In modern Western society, if you can't start watching the video immediately, how likely are you going to be to remember to watch it after it's downloaded 15-30 minutes (or more) later? The whole (business) idea is to keep peoples' attention, like with television. If they "switch channels", you've lost your advertising opportunity.
"...always new atoms but always doing the same dance, remembering what the dance was yesterday." -Richard Feynman
By default the user needs to initiate playing the video, but there is an optional autoplay attribute which can be used to auto-buffer and auto-play the video.
As opposed to Flash, which is pretty much the ultimate untrusted codec? It's a huge binary blob that has had numerous security problems, and which has a huge attack surface. Even ignoring declared vulnerabilities, Flash allows web pages to do things like access the clipboard and bypass XMLHttpRequest same-origin restrictions. In short, installing flash makes a web browser demonstrably less secure.
It's remarkable, then, that an administrator would be comfortable installing this Trojan octopus of a plugin while ignoring a far simpler open source video codec that he can verify and compile himself.
Really, it just shows that people will trust the familiar without seriously questioning it, at least until a crisis shows up.
It's true, Flash content is choppier in Linux. I blame Adobe and their crappy plugin. If I download an .flv and play it in mplayer the problems disappear.
^^vv<><>BA
That would be the entire point of HTML 5. To bring HTML back to the forefront.
Liberte, Egalite, Fraternite (TM)
Demo of video and SVG support in Firefox 3.5. That's why video being built-in to HTML5 is important.
This has nothing to do with HTML 5 or the video tag. The javascript is used to create a fallback path for users who don't have a particular codec installed. It is not compulsory. Most linux machines install ogg theora with a media player package anyway, it's the rest of the world that need to download it.
Nope, OS X (10.5) on a dual 2.8Ghz iMac. There's no excuse for Flash video's failures.
The point of the video tag is that it should contain some (as yet undecided thanks to infighting) standard codec, in the same way as an img tag should always contain jpeg, png or gif data, a video tag should always contain xyz, abc or nml data. Exactly what xyz, abc or nml should be is yet to be figured out.
Google and apple would like them to be h264/aac, because everyone uses that already, even more recent people using flash.
Mozilla would like them to be ogg theora/vorbis, because they're open, even though nothing actually supports them.
Neither side can agree.
Thusnelda is noticeably better than H.263 (which is what YouTube used to use)
Exactly: used to use. Since then, YouTube serves HQ and HD videos in H.264.
Clipboard.getData().
Only IE, actually.
Flash allows you to request content from sites that would be blocked by XMLHttpRequest. Can you refuse that statement or not?
You inadvertently make a good point. If Linus or RMS had developed flash, its source would have been open sourced, and by now, its capabilities would have been integrated into the browser. We wouldn't talk about what "Flash" can do as distinct from something else, but simply about the abilities of browsers.
That's what the rich media part of HTML5 is all about: doing what Flash can do in a browser.
and actually advocates the use of JavaScript to implement it.
The writer of that linked piece makes it pretty obvious his goal is for the video to work for everyone - and the javascript code is therefore used to basically find a method the current user's browser can support without it being obvious to the user (e.g. not forcing the end user to download the video and view it in a separate player, which the OSNews letter seems to want to push on the user).
In other words, he's thinking about the user's experience first.
The author of the submitted story, on the other hand - as with the one from a few days ago that lamented Chrome's lack of purity regarding HTML5 video support - is more interested in Ogg zealotry. That's fine, if it floats your boat - but let's not dance around and obfuscate this. Make it very clear you want the Ogg format used - and ONLY the Ogg format used. Then the rest of the world (outside of Slashdot) can choose to continue ignoring you, just like it's been doing for the past few years.
#DeleteChrome
Why is anyone worried about the quality of videos your going to watch in your browser? The vast majority of those videos are not going to be interesting enough to want to see them in full HD glory. I would rather see Ogg because it's a free standard, and if we lose quality in order to save bandwidth I don't really care when it comes to the type of videos that I watch via my browser.
"Unfortunately, Theora still needs twice the bitrate as H.264 to deliver the same quality, even with the "Thusnelda" rewrite of the encoder."
Except that statement is provably false if by no other facts than that neither Theora nor H264 quality scales linearly with bitrate.
Beyond the obvious fail in your claim, you're also just wrong.
See this comparison and this comparison to see how Theora compares to the most popular real-world implementations of H264 on the Web.
What you're saying is HTML is going to add the ability to do what people have been doing for 5 yrs with Flash.
No, what we're saying is that video is going to become a first-class Web citizen that can interact with the rest of Web content in ways that Web developers want. Flash's video is locked inside the plug-in prison and cannot be well integrated with non-flash (real Web) content. Bringing video (and audio) to HTML means that real Web content like other HTML, JavaScript, SVG, CSS, etc. can interact with video and improve on what people have been doing for 5 years with Flash.
tepples, you claimed that H.264 requires twice the bitrate to achieve the same quality. d235j responded that it only took15% more bandwidth to achieve the same quality and you failed to respond to that. Also, for the last several years h.263 was "good enough" for billions of video views from hundreds of millions of users. Theora is considerably better than h.263 and very close to h.264. Your claim that it requires twice the bandwidth to math h.264 just doesn't hold water with the overwhelming majority of video content online today. For real-world online video content, both standard and high-def, Theora holds its own against H264. It might not be better, but at comparable bitrates, I'd wager you couldn't find more than 2% of the Web population who can appreciate the differences. Oh, and Theora is getting better with every passing day AND in my experience it beats H264 in decoding CPU usage.