HTML5 video. Are we getting ahead of ourselves?

One thing I really like about talking to some of the Opera guys I know, is that they will take the time to consider another viewpoint. Whilst they’re all about the open web, they realise they need to listen to people who are making and creating content for clients, but face real world restrictions like time and budget. I know. Ridiculous…

I run a business (well, Pete does most the business end stuff, but you know what I mean) and all this talk about HTML5 <video> tags has kinda riled me a little bit, because it’s getting so much exposure, and to be honest I think it’s already shot itself in the foot.

I emailed Bruce Lawson my points of view after we had a rapid fire exchange via Twitter. I was playing devil’s advocate a little bit by countering OGG/h.264 with a flippant “Flash video FTW” type of comment, but that led to a good exchange, with Bruce citing accessibility options in the new <video> tag as a major bonus. However, Bruce slept on it, and asked me the next morning if he could write it up as a blog post, because in his usual musings he doesn’t get to give my perspective much thought.

Screen shot 2010-02-05 at 20.07.25Currently, the best way to serve video online is via Flash. That’s a given. With the advent of HTML5, a new tag was born to specifically handle video. This would play video natively in a browser, with no need for plug-ins, and allow interaction with Javascript, standard CSS controls and further advancements like annotations etc. It does sound really good, and certainly one of the areas of HTML5 I’m looking forward to exploring. There was quite a bit of coo-ing over this Sublime video page, which is a good indicator of how it’s going, but as Tom Leadbetter points out, it doesn’t work in Firefox. WTF?

This is because there was a small issue of which codec the browser vendors would support for the <video> spec. Apple and Google favoured the h.264 codec, while Opera and Mozilla favoured the open-source OGG video standard.

Of note, Apple faced a similar decision when choosing to create the Apple Lossless codec. They chose not to use FLAC, which is an established lossless format, and opted to make their own. I remember reading at the time that it may be to FLAC not being immune to unforeseen patent issues in the future, which funnily enough is very similar to the OGG video format right now.

Right at the first hurdle, we’re getting shafted, because this means if I want to make video available for all visitors, I need to encode two versions.

If that were for a few videos of my band (what band? I can’t play *anything*) or of my family, it wouldn’t be an issue really, leave ’em encoding over night, upload and use twice as much space, and reference both files in the <video> code.

But for a business, this means I have to charge a client twice for encoding the same piece of video, and if they’re dealing with a lot of video footage, the time plus storage aspects have just been doubled. This is just crazy. A savvy client will decline the niceties of doing the right thing of going with HTML5 and say encode it once and deliver with Flash. Job done.

Oh, but the iPhone and iPad don’t support Flash!

Think around the problem. Flash can stream h.264 .mp4 files, and the very same .mp4 will run in the Quicktime player on the iPhone, so it appears we’re covering a lot of bases with just using h.264.

Furthermore, the h.264 standard is supported by both dedicated hardware encoders and embedded chip sets for decoding, like in set top boxes or mobile handsets. More on the mobile in a bit, but the hardware encoding is the *only* way to go for anyone dealing in large amounts of video. YouTube anybody? Combined with the smaller file sizes and comparable if not better quality, it’s no wonder Google chose h.264. (I was going to link to Chris DiBona’s blog and the relevant article, but it’s a bad UX with the menu *under* the Google Ads box, and I couldn’t be arsed digging. face palm.)

“Oh, but not everyone is Google.” No, no they’re not. But other people use video too. One of our clients who do use a lot of video is D&AD. A quick look through the 2009 Awards submissions shows not only four or five dedicated video sections, but browsing round other areas, such as installations reveals more video. There’s gonna be a lot of media on there, and that’s just 2009… there’s 323 videos for that year alone, with a total of over 1,500 videos for five years of awards. This figure is only gonna go upwards as the 2010 awards are added, and the years before 2005 are added as D&AD work their way back through previous awards for a more complete history.

Going back to mobile again, this is one area of the internet use we’ve seen absolutely massive growth in over the last few years. The market is enormous, and it’s spreading across the globe and bringing access to those without computers or landlines, with enterprising individuals in Africa renting time out on mobiles for internet access.

A growing number of mobile devices are sporting h.264 hardware chipsets as standard, the iPhone, T-Mobile G1 and Android are notable examples. The explosion of mobile browsing is driving the adoption of the h.264 standard as a hardware spec, and I think that’s going to be quite the defining factor.

I would have thought Opera with it’s Mobile and Mini browsers would be an excellent match up for these new devices, but OGG does not reap the benefits of these hardware accelerators. Opera won’t support h.264 due to “obscene patent licensing costs” (which have yet to be levied for content creators, but it’s still $5 million for a browser to include a h.264 decoder. Thank you for the correction, Bruce), yet they themselves will license out their browser technology to companies like Nintendo… I mean, Firefox, fair enough, it’s open source. Obviously, I do not have the faintest grasp of the in’s and out’s of such dealings, or the figures involved. This is almost thinking out loud from me, and can probably be answered quickly by one of the Opera crowd.

Anyway, if you’ve stuck with me this long you deserve a round of applause. Golf clap. The long and short of it is that I can’t see HTML5 video becoming a real-world standard for online video for a while, until these codec issues are sorted out and there is a clear winner. Certainly for me as a professional dealing with clients concerns and costs, we have to be realistic and say at the moment, we’re going to use Flash. And that’s a real shame, as a lot of work has been done to bring HTML5 into the world, and there’s so much to look forward to, but this situation just boils my piss.

Now please excuse me while I go sell that garden ornament Lawson his hub-caps back…