Google and Video

[bryant]

Google hates H.264! H.264 is used almost everywhere, not just for Web video; it’s also the Blu-Ray encoding standard. So this is very exciting.

Despite my knee-jerk pro-Apple response, I believe that Google is correct in stating that WebM is the better political choice for Web standards. It is open in the sense that there’s no licensing fee and Google has no ability to institute one. It is not an open standard insofar as the standard does not belong to an impartial standards body, which is slightly problematic, but practically speaking it’s not a huge deal. H.264 does, FWIW, belong to such a body. But it’s not free to license, and that is again the more important issue.

WebM may not be the better choice from a legal point of view, in that we don’t know if it’s encumbered by patents. It would be nice if Google would indemnify people using WebM from patent lawsuits, but I don’t think it’s reasonable to require them to do so. Google doesn’t have to do your legal work for you, even if it would be nice if they did. Anyhow, I am not competent to have an opinion on the legal issues, so “we don’t know.” If I needed to make a corporate decision about this I’d pay for a lawyer to tell me things.

Technically speaking I don’t care. Web video is not the place where I demand top-notch video quality. H.264 might be better; even if it is, it’s not going to matter 99% of the time.

Now the fun part. Google’s stance, while correct, is in direct conflict with their Flash support. Google’s statement: “Adobe Flash Player is the most widely used web browser plug-in. It enables a wide range of applications and content on the Internet, from games, to video, to enterprise apps.” So, yes, this is true. Likewise, H.264 is the most widely used Web video format, which enables a wide range of video on the Internet. You’re either making decisions based on usage or not.

Which makes me suspect that Google is, with WebM, making the right decision for the wrong reasons. This only makes me about 50% happy.

Edit: this post makes the excellent point that Flash does share one key characteristic with WebM: namely, it’s free to distribute. However, Adobe has not to my knowledge guaranteed this in perpetuity.

Mirrored from Population: One.

Comments

[gentlyepigrams]

It's the trusting Google to make the right decision for end-users when it's so clear they're doing it for reasons that are antithetical to consumers that I have a hard time with. Then I remember that to Google I am the product and not the consumer, and it all makes more sense.

The Flash decision makes much more sense when you realize that for Google, the advertising company, the consumer it supports is advertisers (who run Flash-based ads) and not end-users of Chrome.

[bryant]

Yep. In a small way I feel vindicated -- back at AltaVista, we were always somewhat skeptical of the idea that text-only ads would be a sustainable policy. AdSense, of course, has both image and Flash units available.

I expect Google to win this battle, by the by, since IE9 will support WebM. If Microsoft pulls back from that decision things might change.

[gentlyepigrams]

While I see and agree with your logic, is IE a big player that way any more? I mean, isn't IE the browser of choice for corporate intranets that can't be upgraded? Or is it just that the sites that I read that talk about their stats are unrepresentative that way?

[bryant]

It's between 40 and 50% worldwide, depending on who you listen to. Firefox is #2; Chrome is #4. Safari is small on the desktop, but big on mobile. See here. If Safari is the lone browser not supporting WebM, that's one thing, but if you add IE 9 to that you're definitely at 50%+ of the market that won't play it natively.

[mcroft.livejournal.com]

I think there's another future issue to consider. Mobile vs. Desktop. Currently mobile is about 6% in NA (statcounter). I can't find (and wouldn't trust) a numeric prediction of it, but it's tripled in the last calendar year. Even with H.264, you have to start worrying about profiles at that point.

IE is not a big player in HTML 5 yet, although it is likely to be. Without a more detailed breakdown, it's unclear how much of that 40-50% is never-gonna-upgrade-and-never-gonna-watch- versions of IE. I also wonder how soon IE9 will make that big of an impact, given the slow uptake of the dependent OS. At work, I'm on XP.

statcounter is interesting, but it's never as good as looking at your own stats. I look at w3schools.com browser statistics, which is for obvious reasons tilted towards standards compliant browsers and technology adopters, but within IE they show a 0:3:1:1 ratio of IE 9/8/7/6 (http://w3schools.com/browsers/browsers_explorer.asp)

[bryant]

Yeah, mobile's a big deal -- I think that's why Safari's support is relevant. IE by version is also a big deal. I'm hearing that IE 9 will only support WebM with an additional download? Which is way not as good as built in support.

[kirbyk.livejournal.com]

There's a fine fine line between:

"We do what we must to make the Web the open, magical place that we think it can be, and sometimes that includes throwing our weight around to stop bad ideas."

and

"We do what we must to maximize our value to our stockholders, while making products that are excellent."

Google has inexorably shifted to the latter (where Apple always was). It's a shame, but they resisted it longer than most. But they watch the bottom line now, rather than do what they think is right and wait for the market to agree with them.

Which beats, say, Facebook, which has steered right towards the pile of money and regularly has to have some decency beaten back into them. Google at least has a fond memory of more innocent times to draw on.

I don't think we'll see another industry force that tries to hard to be a white hat as early Google any time soon. We'll be lucky to keep with the modern Google, and Amazon, and Apple, which at least discuss issues about the global tech community, and minimize the Microsoft and Facebook casual disregard for it.

[dr-memory.livejournal.com]

*tries to think what he can say without inviting a lawsuit from his former employer*

Let's just say that if Google's position looks incoherent externally, it's probably a reflection of a not-holding-well detente between a number of factions internally. And until YouTube is consistently and usefully profitable (and not in a "hey, we're cash-positive this quarter if you squint really hard at the numbers" way, but a for-real "we're actually making significant headway against the debt of the last 4 years of insane capex and headcount investment" way), things are not going to get any less confusing.

[bryant]

I am not even remotely surprised at the idea that a company the size of Google has internal debates. Someone somewhere said something about engineering-driven decisions; if I learned that the engineering consensus was to drop Flash but that it was overriden by business necessity, I would believe it. I also might make the same decisions if I were in charge. Or, hey, it could be a completely different conflict.

I mostly wrote this to clarify what I was thinking, because I did have that Apple-centric pissed off kneejerk response and I wanted to figure out if I was even remotely right or not. Google's not evil, Google's just a company.

[dr-memory.livejournal.com]

if I learned that the engineering consensus was to drop Flash but that it was overriden by business necessity, I would believe it.

I suspect it's more that there are two entirely parallel engineering/management consensus tracks at work here. The Chrome team is not answerable to Youtube, nor vice-versa, and it's... culturally unlikely that anyone would force them to harmonize their stories.

[bryant]

Heh.

[mcroft.livejournal.com]

What I wonder is if Google is going to disable H.264 in Android, because it would basically mean that Android tablets and phones wouldn't be able to play video with decent performance and battery life.

Don't get me wrong, I don't have a dog in this hunt, but I don't think the producing and consuming hardware is going to move off H.264. You can't make a movie without H.264 encoding it from the camera. If there were WebM cameras and VP8 hardware accelleration chips in devices, that would be a different story.

[editswlonghair.livejournal.com]

Taking off my Apple-centric hat and just speaking as an editor, H.264 is just an absolute godsend w/r/t image quality and data rate. It has very quickly become a cornerstone of my workflow. How this will impact anything on my end, I dunno and honestly, don't care... I've seen so many flavor of the year codecs come and go over the years.

[dtm]

For obvious reasons I'm hesitant to comment on this, even though I'm far enough away from anyone involved in video that I have no privileged knowledge, so I will just say that it is possible to put forth a coherent stance that one can make decisions about the direction of an emerging standard based on where you want the world to end up, while at the same time making decisions about legacy support based on where the world is currently.

This may be a set of moves you wouldn't have done, and it may or may not pan out ultimately (time will tell), but it's not fundamentally incoherent.