Note: I'm currently CTO at latakoo, an enterprise video company that heavily works with broadcast television. This post is my own opinion and nobody else at latakoo was involved in any way.

Last month, Benedict Evans wrote a really interesting post about the future of television which you should read.

This in turn points to another question, and perhaps a slightly subversive one: how do people actually want to watch 'TV' (or whatever we call it)? Hundreds of millions of normal people really do just come home, turn on the TV and watch whatever's on - if you offered something less passive, do we really know how many would do it? That is, the idea that no-one would watch linear broadcast TV if on-demand worked 'properly' (whatever that might mean) is really just an assumption.

The full post is over here.

Over time, TV has transformed into an ad delivery mechanism; the onus has been on producing cheaper content, more fully-laden with commercial sponsorship. Meanwhile, the Internet has turned up on our doorsteps, with far more options for content and entertainment. Entirely new forms are available to us, as well as shows and movies that wouldn't be available over a broadcast signal. Audiences know what they want, and can now grab them quickly using a Bittorrent client if it isn't available legally. I'd go so far as to say that this, right now, is the golden age of television content, even if the broadcast medium is slowly waning.

There's a difference between "television" as a content form, and "television" as a technology involving broadcast towers and cable boxes. As Benedict notes, as entertainment evolves, we'll still be in the market for a "lean-back" experience: there's always going to be demand for a service that lets you turn something on and passively consume content. The question isn't so much the form it'll take - I fully expect to be watching another season of Doctor Who in ten years - as the interface to that content.

I'm confident about the following things:

1. Television will not be broadcast over the air or using traditional cable. (Much of the latter is already sent over IP.)
2. Whereas the channels you could receive used to be limited by physical means - which broadcast signals you were in range of, or which channels were carried by your cable provider - they will, in the future, be limited by artificial scarcity imposed by business considerations, where they are limited at all.
3. Anyone will be able to run their own TV channel.

We've seen glimpses of the future from the services we're already using. Last Olympics, I watched the Paralympics via a live-streaming YouTube channel. Every so often a temporary channel shows up on my Apple TV, for example for music festivals or product announcements. It seems likely to me that we'll be able to treat channels in the same way that we treat apps right now. In fact, the Roku streaming player already does this: its API allows anyone to create a channel and fill it with their own content. I expect Apple and Google to follow suit. (The first Android-powered TVs show up later this year.)

In a lean-back world where anyone can run a channel, you can't expect channels to programme 24 hour days worth of content. That strategy works for major broadcasters, but isn't feasible for smaller companies, startups, or most individuals. So the solution is likely that we'll have a curated mix from content sources that we're interested in. Just as a lot of people follow Andy Baio's linkblog today, we might follow his TV channel in the future, subscribing to content that he thinks is interesting (and in turn is drawn from the channels that he's curated). TV will become more like social media in this way. I would imagine that some sources will come with display advertising from the likes of Adsense, while other content sources - business feeds, for example - will be paid for. They might ask you to pay before you watch an individual segment, or you might have a subscription account. Standard payment options would be available. It might even be something like Paypal, although I hope not.

Imagine a TV that launches into a list of channels, maybe as a grid of icons (a kind of lean-back iPad), or maybe as a TV guide-style menu. In the scenario above, Andy Baio would be a channel, and services would allow me to find new channels based on search terms, content types or similarity to other things I'm interested in. You can imagine Facebook and indieweb integrations: find channels based on the things you typically watch or are interested in.

Oh, and all of this is powered by the web.

It has to be. Television the technology, as it stands today, is not owned by any one entity, just as the web isn't. Companies will certainly try and make land grabs to own the medium, but even if they appear to succeed for a while (eg an Apple offering) they will ultimately be unsuccessful. Market reasons for this include the wide ecosystem of television manufacturers - it's easier for people to integrate with an open standard than a technology that they must license, and as a result, lower-cost, mass-market televisions are more likely to support these. Some TVs directly integrate Netflix and Hulu today, but the behind-closed-doors deals to achieve this are significant. With TV as an open platform that sits on the web, there are no back-end deals. Anyone can make a TV that just works, which is in the interests of manufacturers, content owners and channel curators. Each TV gets to have its own interface, just as they do today, and companies can add value-added services like better search and recording.

How close is this? I think you can almost graph it. The point where bandwidth and server costs have reduced to the point where streaming to each user is cheaper than broadcasting to them. IP multicast is not reliably supported on end-user Internet connections, so I anticipate that Internet broadcasts will remain per-viewer. (By the way, latakoo has some new codecs, developed with the University of Texas, that will make this significantly easier.)

So what about live TV? Up to this point I've been discussing pre-recorded content, but there's certainly a need for live broadcast. In a world where the web powers channels, it's just as easy to programme a live stream as it is to programme a pre-recorded show. (Of course, the technology to live-stream is more complicated, but the actual mechanism is agnostic.) You could even choose to have alerts from, eg, news channels overlaid across your main viewing screen, so that if there was breaking news, you could be alerted and choose to switch. You could choose breaking news providers, of course, just as you could choose every other content source. Television would become a true open market for content.

The other nice thing about the web is that all of your channels and all of your preferences are available on all your devices. The exact same profile can be available on your phone, and you don't have to care about which phone you own, as long as it has a browser that works. You could also choose to have an app on your phone that adheres to the same standards but has a tailored UI and value-added services. You could have browsers that talk to each other and allow you to resume watching a show on a different device, from any content provider. Conversely, providers like Netflix, Hulu, NBC, the BBC and Andy Baio wouldn't need to provide any of that functionality themselves. It would simply be a feature of the platform.

Broadcast channels as we know them would become obsolete, but great content - and content channels - would persist. This future of television would be oriented around viewer preferences and content providers. That's the future television I'd like to be using.