Notes on the HTML 5 video codec debate

HTML video tag

You may or may not have heard about the disagreement over HTML 5’s video codec support. Let’s ignore the patent arguments completely, and the licensing fees (which I’m sympathetic to) — I’m not a lawyer, so I can’t bring anything useful to the table.

Basically, the choice is between H.264 and Ogg/Theora, and both Google and Apple think Theora sucks — for hardware support, and in general. (By the way, I’ll apologise now for my inconsistent references to Ogg and Theora; for reference, Ogg is the umbrella project, Theora is the video codec, and Vorbis is the audio codec.)

Let’s examine video quality. Noses were put out of joint by this comment from a Google guy:

If yt were to switch to theora and maintain even a semblance of the current youtube quality it would take up most available bandwidth across the internet.

Clearly, that’s hyperbole. Equally clearly, he doesn’t think much of Theora’s video quality. (Bear in mind, this is coming from the side that’s shipping support for both H.264 and Theora in Chrome; he has fewer reasons for bias than, say, an Apple guy who has some investment in H.264’s success.) This prompted a comparison from Greg Maxwell at Xiph, the people behind Ogg. I’ll summarise my criticisms of that comparison in Handy List Format:

  • Only the first pictures compare H.264 and Theora. The second set compare H.263, the legacy YouTube codec, and Theora, which is not even slightly relevant to HTML 5 (and only relevant to the quotation above if we’re being slightly pedantic).
  • YouTube’s H.264 files aren’t anywhere near the best quality that H.264 can deliver. They’re designed for mobile playback, and as such are designed to be playable by common, low-cost mobile hardware. Files designed for desktop playback — even GPU-accelerated playback — can have much more complex features enabled, and deliver vastly better quality at the same file size. The Ogg files, to my knowledge, are the best that the codec can deliver, without regard for the power of the hardware needed to play it back.
  • Even this H.264 file clearly clearly beats Theora. Note especially the grainy banding around the water lilies, which is often indicative of a codec that’s struggling slightly at the specified bitrate.
  • The Theora files were encoded by an expert, but the H.264 files were encoded with catch-all settings by an automated process. Anyone vaguely serious about video will tell you that different clips need different treatment, and a human who knows what he’s doing will totally outclass a machine which applies the same settings to every clip.
  • H.264 is famously useful at all kinds of bitrates and resolutions. Most codecs are only suitable for a limited range of values, so even if H.264 can be outclassed under specific circumstances (and, again, I see no evidence here that Theora’s going to do that), it wins for its versatility.

So, you’ll gather I’m not a giant fan of Ogg either. From my lengthy use of H.264, my (admittedly limited) exposure to Theora, and considering the opinions of those I trust, I can’t convince myself that Ogg is anything other than inferior. This is not to say that it’s irrelevant, or unimpressive; just less desirable in this context.

The lack of hardware support, though, is inescapable. Lots of video cards have H.264 decoding capabilities, which allow the playback of compatible video without stressing the computer. This is a Good Thing™. There aren’t any such decoder chips for Theora (or if there are, they’re extremely rare), which means decoding would have to be done on the CPU. While that’s generally fine for desktops (if inefficient, even a possible problem for HD video), it completely rules out mobile devices; iPods, iPhones et al rely on hardware decoding to play back video, as their CPUs are about an order of magnitude too weak. So, this alone is probably a deal-killer for anyone interested in mobile video… as both Google and Apple are.

Think of the situation with Flash video. That is largely CPU-grunt-based, and runs bearably on the desktop. When it comes to mobiles, it’s like running through treacle, and really not a viable option. Apple and Google even went to the trouble of creating an H.264-based alternative — which can utilise hardware acceleration — to YouTube for the iPhone. We’ve got essentially the same choice here: either go with Ogg/Theora, which will exclude mobile browsers completely and perpetuate exactly the same bloody mess of incompatibility that HTML 5 is supposed to help resolve, or go with a solution which can easily be supported on a wide gamut of platforms.

So why don’t Apple follow Google, preferring H.264 but implementing both? They’ve actually, of course, taken a quintessentially Jobsian position: a crappy implementation is worse than none at all. (Especially if Theora’s adoption might harm the chances of something better coming along.)

So, what to do? Ian Hickson wrote a remarkably even-handed review of the situation and concluded that any requirement for codec compatibility had to be dropped, to prevent an unworkable specification.

His conclusion rather surprises me though. What’s wrong with requiring at least Ogg or H.264 support of compatible browsers? Sure, it’s just codifying the status quo, and it means that there’s no single file that can be guaranteed to play in every HTML 5-capable browser. It does, though, protect against any further fragmentation in support.

Needing two files might be inconvenient. But it beats 3. Or 4. Or 5…

15 Responses to “Notes on the HTML 5 video codec debate”


  • You’re somewhat misrepresenting the ‘choice’ here: The W3C’s rules preclude specifying a format in the standard which requires patent royalties. So the choice is between recommending a baseline which vendors should implement in addition to whatever other codecs they support, so that web authors will have at least one option they know will work everywhere that supports the video tag, or making no recommendation at all.

    Theora is one of the only possible options for the baseline, and it’s believed by virtually everyone to be the best of those. (Dirac, for example, is another alternative but it performs pretty poorly at these rates)

    Apple has changed their argument point a number of times. I believe the most relevant fact is that Apple has never shipped a patent-unencumbered highly compressed media format. For example, for lossless audio rather than use FLAC they invented ALAC which is both slower and achieves worse compression ratios according to the published comparisons. (Fortunately lossless formats are easy to compare)

    The comparison page I created is quite specific that it’s a comparison with YouTube and not other things, and that it was created specifically to address the quoted hyperbole.

    Accordingly, many of your bullet points were actually raised by me in the comparison.

    The H.263 is what youtube still provides by default to me, Chris had indicated that he (might?) provide the breakdown by format. Since he seemed to be suggesting that YT couldn’t even offer Theora as an alternative format (a fallback), it seemed relevant to me.

    Your expert encoding comment is, however, off the mark:

    There was absolutely no special treatment provided in the Theora case. Unlike other video encoders the reference Theora encoder doesn’t expose the user to an “space ship console” of knobs. There are basically two user accessible settings: Bitrate (or quality) and keyframe interval (which controls seeking granularity). No expertise required. If you’re an expert you can modify the software, but I did not do so for this comparison.

    Hardware support is less of an issue for Theora than it is for H.264: As a simpler format Theora uses less CPU. On the CPU used in the first generation ipod touch the Theora video file in that YT comparison requires less than half the CPU than the YT H.264 in a pure-software decode. Also, on many devices (like GPU assisted decode, and devices like the Palm-pre) the ‘hardware’ support is really just embedded software (on a DSP in palm-pre’s case). Theora can be equally provided there.

    The reason that “just provide both formats” isn’t a universally good recommendation, beyond the fact that the rules preclude it, is because H.264 doesn’t just have implications of browser vendors: There are also per-use fees required for anyone who distributes H.264 video. These fees are waved until 2010 for web use, on something of a “your first hit is free, kid” basis. Nothing should stop anyone who offers H.264 from also offering a Theora copy, but the other way around is not true. (For example, Wikipedia offers only Theora).

    But *where licensing costs or other factors do not preclude H.264 entirely* I completely agree with your overall position that it would often be preferred— It’s a newer, more advanced (and thus CPU hungry) specification and with a good encoder it will achieve better quality for a given bitrate. It also will have better battery life on some devices (ones with specialized H.264 support) Fortunately the language in the standard always allowed you to offer both (or more) and prefer whatever works best.

    If Apple offered both, the standard would recommend that everyone at least offer Theora and web providers would have something to include in addition to their preferred format so they could be sure video will work, and all of the people advocating a royalty free baseline in the standard could declare victory and go home. I don’t think anyone involved imagined that Apple wouldn’t also back their own horse. A baseline doesn’t need to be preferred to do its job, but it does need to be (near-)universally available.

    Because of the licensing, even if you never intend on using Theora, you should hope that it ends up being mandated as the baseline, or at least effectively mandated by market pressure. If everyone is free to use Theora exclusively, if they want, then it will be more difficult to charge high rates for H.264.

    Cheers.

  • Luke Mildenhall-Ward

    I do very much believe H.264 is a superior quality codec when fully utilized.

    I think the fundamental idea Gregory has put forward above (if I understand it correctly) – to have ogg Theora as the base video option and H.264 as sort of a ‘premium’ video option – would be a good idea. This would allow a commercial site like YouTube to include better quality H.264 videos if the user had a browser that supported it, whereas someone who just wanted to put a quick video on their personal website would be able to freely use ogg Theora.

    I see no reason Apple shouldn’t do as Google and include support for both. If users knew they’d get higher quality videos if using Safari to watch YouTube it could easily encourage them to switch browsers, but at the moment it means a trade-off in losing ogg support, which is a problem. I think it’s difficult for Apple as they’ve never officially supported ogg formats in OS X, Quicktime, iTunes, etc. but that can be changed.

  • Gregory, thanks for taking the time to add your input here. It’s most welcome.

    Per-use licensing is something I didn’t bring up, partly because I can’t actually conceive of a model that actually makes this workable. I can see that it’s a concern, albeit one for content distributors rather than browser makers; but I can’t imagine it being truly enforceable and therefore quietly expect it to be essentially dropped. I’ll admit to pure speculation and a decent chance of being wrong on that one though.

    As for the W3C’s rules on patents, I’ll offer only two observations. Firstly, that it didn’t seem to be Mr Hickson’s primary concern, and secondly that perhaps over recent years (particularly with HTML5) the W3C has learned that what it says and delivers isn’t always what people actually want.

    You’re also clearly right in that whatever reasons Apple gives (and however valid they may be), the simple fact is they clearly don’t want to ship ogg support. Apple isn’t exactly allergic to open source – I mean, for all their diplomatic clumsiness at times, WebKit is proof of that – so I’m not sure it’s anything more sinister than a typically-geeky “I’ve picked my favourite, so you’re all automatically wrong and I have to convert you” stance. They’d look better if they were honest rather than providing vague justifications, though… Again, pure speculation.

    It occurs to me that I’ve come across as rather harsher towards Theora than I intended, and for that I apologise. Theora would be a big improvement over the most commonly delivered types of Flash video. It also has, of course, the giant advantage of its openness; the reason a collection of arguments against Theora is presented, is because individually they’re less important than this. I should probably have opened the post with this, but frankly I was only expecting to write a short paragraph.

    I think my real “fear” (and here I’m addressing Luke, too) is that a baseline often hasn’t proved to be something that people build up from, but something people sit on. The continued prevalence of crappy-quality Flash video, lingering table-based web design, divx-in-avi, the continued dominance of Internet Explorer… These things and countless others indicate that if something is seen to work with minimum effort, it’ll be used even when there’s a better alternative. Universality is good, but if we’re going to undertake the monumental task of getting content providers to adopt a new method of delivery I really feel we should be pushing for the state-of-the-art. If there’s a fallback, I’m worried that we’ll never actually move beyond it. (Or at least, not until the next revolution.)

    Theora looks good compared to today’s norm, but gets beaten by state-of-the-art. What happens in, say, 5 to 10 years’ time? The proprietary plugin du jour (please let it not still be Flash) will likely provide everybody’s favourite codec, and boast massively superior video quality. If we’ve started from less than the best available, the difference will be more pronounced, and the temptation to jump ship will be greater.

    For that matter, when Flash already provides h.264 streaming, how many developers will switch to HTML5, drawn by the promise of video quality almost as good? For its lack of ubiquity, HTML5 has to offer something special to get its <video> tag adopted widely. (And, indeed, a superb <video> experience could be a massive force for its adoption.) Maybe that’s against the spirit of the WHATWG, which has certainly put a healthy cupful of pragmatism into the HTML5 specification – and it’s easy to appreciate that, until it comes to something you care passionately about.

    This does seem to be largely academic, as the arguments have been made many times on either side now. HTML5, bar some miracle, will not have any codec support specified, and I think that’s sad. Hopefully, the status quo won’t deteriorate, and we’ll be able to get away with just two different formats for the foreseeable future.

  • Hawkman, I think you’re missing a key point in your “we need state of the art to get people to move from flash to HTML5 <video>” assertion.

    If you really want HTML5’s <video> support to succeed, you first need lots and lots of people using a <video>-capable browser.

    There are 4 browsers shipping HTML5 <vide> now or in the immediate future, Firefox, Safari, Chrome, and Opera.

    (Firefox _cannot_ ship h.264. Apple will not ship Theora. Google is shipping both Theora and h.264. Opera chooses not to ship h.264.)

    Firefox has about 300 million users, or ~25% of Web usage. Safari claims 70 million users accounting for about 6% of Web usage. Chrome claims 30 million users and has about 2-3% of Web usage. Opera claims 30 million users and probably also has about 2-3% of Web usage.

    Of the four browsers that care about HTML 5 and making video a first-class citizen on the Web, only one, Firefox, has enough users to _maybe_ convince content producers to move away from Flash video to a new technology (HTML5 <video>) and that one browser _cannot_ ship h.264 because of both Mozilla’s and MPEG-LA’s licensing regimes.

    Safari, with its 70 million users, even when added to the 30 million Chrome users, just don’t come close to comparing with Firefox’s user base and together those two don’t even reach 10% of Web users across the globe.

    The h.264 browsers cannot move the Web away from Flash. They just don’t have the reach.

    Firefox and Opera and Chrome combined, on the other hand, account for 360 million users and over 30% of Web usage. That’s starting to look like to enough reach to convince Web developers and content producers to move off of Flash or at least start supplementing it with HTML5 <video>.

    All other things being equal, sure, a superior codec is more likely to lure content creators and Web developers to adopt <video> on the Web and begin moving away from Flash. But all other things are not equal. Less than 10% of the Web will have h.264 browsers at the end of this year. Maybe 12% by the end of 2010. More than 30% of the Web will have Theora browsers at the end of this year and more than a third of the Web will have Theora by the end of 2010.

    If we’re going to see video become a real HTML element and get it out of the plug-in prison, I think we’ve got a better shot with the codec deployed to a much larger audience than the codec deployed to the much smaller audience. We’d have an even better chance if Safari, the lone Theora holdout, would join with the other modern browsers and take that number up another 70-80 million users.

    Because in the end, this isn’t really about h.264 vs Theora. They can co-exist on the Web just fine. What this is about is real Web video vs Flash plug-in video. Disrupting the Flash monopoly will require both solid technology (in terms of codecs and DOM media APIs) and large scale deployment. Right now, I think the proper trade-off for breaking the Flash stranglehold is the “good enough” and getting much better Theora codec, a kick-ass and very compatible across video-capable browsers DOM media API, and as much user coverage as we can possibly manage.

    It sounds like you’re arguing that the best codec trumps those other considerations. Is that a fair characterization?

    – A

  • Watch as Microsoft implements with support for H.264 (and WMV) by default and no Theora, rendering Asa’s marketshare argument completely irrelevant.

    The problem with crappy baseline standards is that they have a nasty habit of never being upgraded. Just look at how there’s no decent lossy codec for . Instead, we have JPEG, made of stone-age technology worse than Theora that’s outperfomed by well over 10x with modern tech. And if not for Unisys spurring development of PNG, we’d still be stuck with GIF and its palette of 256 colors. PNG would never have become prevalent on the web otherwise despite it being technologically much superior to GIF.

    If Theora became the baseline and people actually used it, we’d be as stuck with it for the next 15 years as we’re stuck with JPEG now. What makes this particularly galling is that we already have a vastly superior already becoming the de-facto standard for internet video, and forcing Theora sets us back a decade.

    Ironically, this is one strength of Flash – only one company controls the implementation so they can upgrade codecs without all this worry of getting a multitude of browser vendors to support it.

  • “we have JPEG, made of stone-age technology worse than Theora that’s outperfomed by well over 10x with modern tech”

    JPEG has been and is serving the Web quite well. I don’t see large numbers of users clamoring for something better. Heck, I don’t see large numbers of content producers clamoring for something better. Good enough is often good enough.

    I think that the YouTube explosion of 2005 is relevant here.

    When YouTube launched, it didn’t have the best video technology. It had, arguably, some of the worst, with Flash 6’s Sorenson Spark and mono mp3 audio and at god-awful bitrates. They opted for the technology equivalent of what was shipping in RealPlayer way back in 1997 and in QuickTime in 1998.

    But what it didn’t have in great video quality it more than made up for with Flash’s ubiquity. YouTube videos were pretty crap, but “everyone” could view them without downloading a plug-in or waiting very long for the content to buffer.

    And YouTube videos continued to be crap for several years because download wait time and player ubiquity trumped everything. And users loved it!

    There were higher quality solutions available when YouTube launched. QuickTime had shipped h.264 in v.7 that spring but YouTube didn’t use it. They opted for the decade old tech, but more widely deployed solution. Microsoft had shipped WMP v.10 with the WMV9 codec almost a year before YouTube launched but they didn’t use that either. Again, they opted for the much worse and much older tech. Both Apple’s and Microsoft’s solutions were far superior to Flash’s video capability in every way but neither had the installed base of Flash. So, given a choice between the best available technology a 10 year old codec that was more widely deployed, they went with the 10 year old codec.

    Sometimes good enough is actually good enough. Sometimes worse is even better.

    As a baseline codec, Theora could serve the Web well. It could provide an experience that is more than sufficient for all but the most discriminating users. And for those users, no one is trying to deny them H.264 or H.265 or whatever is the next cutting edge solution.

  • Luke Mildenhall-Ward

    Asa, the problem with that I see with your argument, is that it will be content providers (significantly, YouTube) who will ultimately determine what people will use to view video on the web. If a sizable number of content producers choose H.264 as their choice codec, favoring the additional quality and bandwidth savings, it could shift the majority of web users to switch to a H.264 browser, especially if YouTube followed suit (which seems likely.) The dominance of Flash was originally rooted by the content providers, and it proved users will download additional software if they want the content enough. If YouTube visitors were one day greeted by the message, “Use YouTube in higher quality, more efficiently, and much faster by downloading Chrome”, we would easily see a huge proportion of web users switch browsers overnight, let alone the number who would switch if YouTube dropped Flash support altogether, making a browser switch a requirement.

    If there’s going to be a war, it’ll end up badly for everyone. But if there was going to be a war, the winner would be determined by the content providers.

    I think it’s also important to consider the impending mobile internet boom. And we all know who’s the dominant player in that sector…

  • Luke said “it could shift the majority of web users to switch to a H.264 browser”

    So you think that content producers are going to adopt the video tag and a codec that’s only available for less than 10% of the Web and that’s going to cause those sub-10% browsers to grow their share dramatically?

    That sounds like a pretty unlikely future to me.

    “The dominance of Flash was originally rooted by the content providers, and it proved users will download additional software if they want the content enough.”

    Not really. Flash was ubiquitous because Microsoft shipped it on XP and where they didn’t ship it, OEMs added it. It was everywhere and that’s why content producers used it. Not the other way around.

    “Use YouTube in higher quality, more efficiently, and much faster by downloading Chrome”, we would easily see a huge proportion of web users switch browsers overnight,”

    YouTube and Google have already done this with banner ads on their sites advertising Chrome for several months now and as best I can tell it took their share of Web browser usage from ~1.5 to ~2.0%. I think that’s far from a “huge proportion of web users”.

    I think you think the Web moves much faster and more dramatically than it actually does.

  • Asa Dotzler, on July 14, 2009 at 11:39 pm, said:

    It sounds like you’re arguing that the best codec trumps those other considerations. Is that a fair characterization?

    Yeah, I think that’s fair. I’m well aware of the problems, I just place different emphasis on aspects of the equation. I’m also pretty confident that if h.264 becomes a common format on the web, Mozilla will find a way to support it. Perhaps because by then certain licensing restrictions would almost certainly have already had to be overcome.

    Asa Dotzler, on July 15, 2009 at 1:37 am, said:

    When YouTube launched, it didn’t have the best video technology. It had, arguably, some of the worst, with Flash 6’s Sorenson Spark and mono mp3 audio and at god-awful bitrates. They opted for the technology equivalent of what was shipping in RealPlayer way back in 1997 and in QuickTime in 1998.

    And look what we’re stuck with. Video so poor it’s left in the dust by h.264 and Theora! Not a happy place to be.

    If you really want HTML5’s <video> support to succeed, you first need lots and lots of people using a <video>-capable browser.

    Maybe. Or, to get lots of people switching to HTML5-capable browsers, you need some kickass <video> implementations to give them a reason to switch. I’m not sure we’re in a position to say which is true just yet.

    Firefox and Opera and Chrome combined, on the other hand, account for 360 million users and over 30% of Web usage. That’s starting to look like to enough reach to convince Web developers and content producers to move off of Flash or at least start supplementing it with HTML5 <video>.

    I’m not exactly sure where your numbers are from, but Net Applications – which tends to overestimate the market share of decent browsers like FF, Opera et al – would put those combined figures at about 25%, and Chrome + Safari at 10%. Either grouping would be the underdog.

    I can’t help but think people underestimate Google on this one. Their brand rather crosses boundaries, in a way that Safari (mainly Mac users) and Firefox (largely enlightened Windows users and Linux types) just can’t; and they own YouTube. If they were to push Chrome on YouTube, everybody’s mother and aunt Flossie would trust them – because they do what they’re told and Google must know best, right? It ties into what Luke says, too:

    Luke Mildenhall-Ward, on July 15, 2009 at 1:44 pm, said:

    If there’s going to be a war, it’ll end up badly for everyone. But if there was going to be a war, the winner would be determined by the content providers.

    As he points out, Google seem to like h.264; and YouTube already has a giant library of h.264 content for compatibility with mobile devices. They’ll clearly make that available, and probably push it hard. Theora, on the other hand, involves transcoding their entire library… I kind of suspect, too, that many content providers will follow YouTube if they’re successful in whatever <video> solution they provide. People ape successful businesses, even if the success isn’t really due to the exact implementation details.

    John Michael, on July 15, 2009 at 12:11 am, said:

    Ironically, this is one strength of Flash – only one company controls the implementation so they can upgrade codecs without all this worry of getting a multitude of browser vendors to support it.

    It’s interesting, isn’t it? Closed platforms can be great for consumers in some ways, and death in others. The real trouble seems to come when you get a successful platform that doesn’t share your priorities; like, say, “not crashing my web browser all the sodding time”, or “allowing copy and paste”…

    I certainly don’t want to lose sight of the fact that whatever happens, it beats today’s norm of dubious-quality plugins. I trust that we’ll have a clear winner, or at least a clearly winning solution (which is frankly in sight, with the excellent Video for Everybody), and that all sides will pull together to evangelise it.

  • Asa Dotzler, on July 15, 2009 at 1:37 am, said:

    “I don’t see large numbers of users clamoring for something better. Heck, I don’t see large numbers of content producers clamoring for something better.

    Most people don’t realize that you can get 10x better than JPEG with modern codecs. And those that do also realize that also realize that the img tag implementations are frozen to JPEG, GIF, and PNG for all eternity. And on the topic of content providers, both Google and Facebook at least are very interested in saving image bandwidth. There just isn’t anything they can do but use JPEG.

    As for the rise of Flash, you’re right in that it was because it was the only thing that worked everywhere and had *nothing* to do with how good the technology was. This is the exact same reason that if Theora becomes the standard for the video tag, then that’s what the video tag will be used for. And given the history of the img tag, that’s all it will be used for for the next decade and a half.

    Given Thusnelda still adds quite visible artifacts even at max quality on half the videos I’ve tried in it (and that’s ignoring how badly it massacres just about any HD video I feed it), any situation that involves using Theora that long is deplorable given that vastly better codecs already exist and are already being widely adopted as the new standard for web video.

    And if the video tag was really going to be a threat against Flash, it would have supported H.264 so that websites could have a painless transition from Flash instead of having to go back and transcode their whole video library and have a long period with them serving up two formats.

    It’s just sad that Mozilla and Xiph don’t care about video technology and would rather keep web video as seas of blocks, rather than push forward with better technology. If Xiph had done *something* to further open video codecs other than sit on their asses for years after slapping a few unused features onto VP3 and calling it Theora, this whole situation wouldn’t be so pathetic.

  • “Most people don’t realize that you can get 10x better than JPEG with modern codecs.”

    So there are codecs that give 10x the compression for the same quality as today’s JPEGs? If it’s 10 times better, surely Nikon and Canon are shipping it in all their most popular cameras. Right?

    “It’s just sad that Mozilla and Xiph don’t care about video technology”

    I think I’m in a better position than you to speak to what Mozilla does and doesn’t care about. Anyway, as you’re now into reading and reporting on the hearts and minds of people you don’t even know, I think I’m through with responding to you.

  • “If they were to push Chrome on YouTube, everybody’s mother and aunt Flossie would trust them – because they do what they’re told and Google must know best, right? “

    Google’s been pushing Chrome not just on YouTube, but with a giant and first of a kind banner ad on the front page of Google.com for months now. The banner ad has a big Chrome log, a big “Install Google Chrome” button and the message “A faster way to browse the web”. It’s had only a marginal impact on Chrome’s growth and couldn’t even lift Chrome up to Firefox’s early growth rate.

    I think you wildly over-estimate the general Web population’s willingness to download, install, and use software replacements. We’ve been fighting this complacency for a lot longer than Google and it’s just not an easy task to get people to change.

    “Or, to get lots of people switching to HTML5-capable browsers, you need some kickass <video> implementations to give them a reason to switch. I’m not sure we’re in a position to say which is true just yet.”

    I’ve yet to see content driving browser adoption in any significant numbers anywhere since the first progressive images hit the Web. I’ve been working on this problem for a decade with Mozilla and new and awesome content just doesn’t cause rapid migration to new software clients. Even plug-ins like Flash didn’t get very far until Microsoft bundled it with Windows ensuring pretty much everyone got it. And it’s even worse because there are very few people or companies willing to invest in new and awesome content that is only viewable by small fractions of the Web population.

    The Web browsing population evolves slowly. That’s just a fact. There are a lot of chicken-and-egg problems and most of the ones I’ve seen become successful were the result of a broad platform, not a superior kind of content.

  • Asa Dotzler, on July 15, 2009 at 5:23 pm, said:

    So there are codecs that give 10x the compression for the same quality as today’s JPEGs? If it’s 10 times better, surely Nikon and Canon are shipping it in all their most popular cameras. Right?

    Why would they when anything but JPEG would likely frustrate the average consumer? It’s the exact same situation with the img tag – in order for it to be widely implemented there has to be demand from people using it, but people won’t use it until there’s widespread support.

    And professional photographers don’t want lossy compression anywhere near their cameras.

    Asa Dotzler, on July 15, 2009 at 5:23 pm, said:

    I think I’m in a better position than you to speak to what Mozilla does and doesn’t care about. Anyway, as you’re now into reading and reporting on the hearts and minds of people you don’t even know, I think I’m through with responding to you.

    Actions speak louder than words, and words speak louder than thoughts. And from Mozilla/Xiph’s actions and words, the only thing they’ve shown themselves to care about with regard to video codecs is patents.

    Though that was a bit harsh on Mozilla; after all they did pay Xiph to get their butts in gear and upgrade Theora from pathetic last generation tech to mere last generation tech.

  • “Or, to get lots of people switching to HTML5-capable browsers, you need some kickass implementations to give them a reason to switch. I’m not sure we’re in a position to say which is true just yet.”

    Sorry, I read that too quickly. You were talking about a browser implementation and not a content implementation, I think.

    Sure. We probably don’t know which is true yet. I’m leaning pretty far to one side in my predictions, though.

    I don’t think anything gets lots of people switching to HTML5-capable browsers. There are no quick wins when it comes to moving the 1.whatever billion web users in a new direction unless you own the Operating system that most of them use. Even then it’s years before you get them moved to your latest version.

    It’s a long slog and we’ve got some more years before the modern browsers, in total, pass IE’s best offering globally. (note, in some places in the world, that’s already happened. but it’s got a ways to go before it’s solidified across a range of geographies.)

    I’m of the opinion that a kick-ass <video> implementation isn’t going to have _any_ measurable impact on migration from other browsers. I’ll re-visit this post in a couple of years and admit if I’m wrong. But I don’t think I will be.

  • Asa Dotzler, on July 15, 2009 at 6:08 pm, said:

    Sorry, I read that too quickly. You were talking about a browser implementation and not a content implementation, I think.

    A bit of both.

    Asa Dotzler, on July 15, 2009 at 5:39 pm, said:

    Google’s been pushing Chrome not just on YouTube, but with a giant and first of a kind banner ad on the front page of Google.com for months now. The banner ad has a big Chrome log, a big “Install Google Chrome” button and the message “A faster way to browse the web”. It’s had only a marginal impact on Chrome’s growth and couldn’t even lift Chrome up to Firefox’s early growth rate.

    I’ve seen it when testing in IE6. Very, uh, subtle. But given the difference even one word can make in a call to action, I’m optimistic about a “make your video suck less” banner’s appeal – because it can promise a tangible difference to something the user cares about, right now. Still, selling a product to someone who has no idea they need it isn’t easy, and it’s a situation you’ve obviously had more experience of than me; so I’ll concede that such optimism is probably misplaced.

Comments are currently closed.