BitTorrent Live: Cheap, Real-Time P2P Video Streaming That Will Kill TV | Hacker News Reader

BitTorrent Live: Cheap, Real-Time P2P Video Streaming That Will Kill TV

techcrunch.com

158 pointsbrianbreslin50 comments

The problem is, and always has been, on the content side.

Most people in the mainstream still want to spend their time watching mainstream shows - and no mainstream US-based content holder is interested in offering internet-based streaming of their TV channels (as a live stream of the live broadcast). If they were, they would have done so already.

[this just launched today, and read the cavets: http://allthingsd.com/20120213/barry-diller-gets-into-the-co...]

Case in point: There is no Comfast Xfinity or Dish style subscription that is online only that doesn't require a cable box or dish as your primary source of reception. I'm sure everyone on Hacker News would love to subscribe such a service but the rights holders are having none of it.

What's stopping these companies is not that they can't push the bits but that they want to protect their existing investments and not find themselves in a commoditized and highly competitive market with many players.

Comcast in most markets has no competition for cable, Dish has no competition. They don't want to race to the bottom and compete with upstart startups that we might be running - so they're playing hard ball with the rights holders (or they are the rights holders too - Comcast owns NBC) and having none of it.

I like Bram, I like this technology and I think there are lots of awesome things we (away from mainstream) can do with it.

But it won't kill TV because the majority of people want mainstream content. And that's going to sit on TV (in some form) for some time to come.

There is a reason that Apple (not Napster) killed the physical CD. To digitize the TV distribution network we need someone with the legal team to properly negotiate these licenses. BitTorrent is already here, it already distributes TV and the networks will never use it.

The future of TV first involves someone like Apple or Google working with the content providers and the ISPs to provide a solution that everyone can agree with (probably with DRM). This is what happened with music, it will happen again with TV.

What his really provides is a way for two generations from now to steam media, but it's dilusional to think that this technology is what people will be using that far into the future.

"The future of TV first involves someone like Apple or Google working with the content providers and the ISPs to provide a solution"

Won't happen. The music industry handed, well, the music industry to Apple on a plate. The wider media watched, learned and face slapped.

Apple and Google are far too powerful for the media to hand any more of that kind of control to. Watch how Google really is struggling to get industry partnership with Google Music.

With Hulu we saw major players (NBC, ABC and Fox) come together and build out a solution themselves.

My guess is that any future internet-based "TV subscription service" will come from a similar joint venture. The technology isn't that difficult so why hand over the keys to a 3rd party when they can own things themselves?

My personal aspiration is that the %age of the population who want to buy tv subscriptions (by whatever medium - cable, dish, internet) will actually decrease and the power will shift away from the existing incumbents simply providing the same service over a new medium but to actually something radically different.

Google should just put more focus on their artist hub, and promote that. They should promote the decentralized of the music industry and indie artists, and not beg for scraps from the big labels.

As an indie music lover, I totally agree.

As a GOOG holder, I fear that isn't a big enough addressable market to make it worth while.

My personal view on this is that if Google can't get first citizen rights with the music industry like Apple and Amazon have, then they should focus entirely on having a strong music locker service which supports user's "grey" music (ie who knows, who cares where the mp3 came from).

This would then add massive value to owning Android (which I do, and the cloud-based Google Music is a killer app). Google Music is not available on Apple, I don't believe.

And thanks to DRM, the TV industry will also give all the power to Apple or Google, just like the music industry did, and how the book publishers gave it to Amazon.

> Case in point: There is no Comfast Xfinity or Dish style subscription that is online only that doesn't require a cable box or dish as your primary source of reception.

But certain content owners are catering to the internet only crowd. Check out the NHL's Gamecenter Live offering. Sure it's on the pricey side (like most on-demand sporting content), but it actually goes beyond simple streaming and makes use of the richness of the platform with customizable PiP, pause, rewind and a visual break down of the timeline of the game with goals, penalties, other noteworthy events highlighted by a small timeline at the bottom of the viewer. It's pretty slick... when it works (reliability is still an issue) and they went ahead and implemented region based blackouts that really don't translate well to the watch anywhere - anytime pitch of the product.

The NHL is a great example. We carried their games as streaming for years (until Akamai purchased the company that was reselling us), allowing you to listen to every single game, everywhere in North America, every game day. It was like NFL's season tickets on DirectTV, but for hockey. Fans love that.

The content owners who realize they can market mass audience content through traditional "channels" while also marketing niche content at a premium through direct "over the top" streaming, and that the niche offering does not cannibalize the mainstream offering, are showing the way. NHL's less popular games, Champs car racing multiple cameras, Tour de France full coverage -- all these offer the die hard fan something more than they can get on TV, so the fans go for both.

We had fun working in the late 90's with the World Wrestling Federation whose shows were popular enough the cable companies wouldn't drop them even if they were made available as streaming. A given PPV was priced the same on cable and on Internet, meaning the only people who bought it from us were the people who couldn't get it on their cable system. The lift was significant and more than paid for production and bandwidth, back when bandwidth cost 100x what it does today.

Over the top can be a massive revenue driver when used to supplement or support your TV offering. Here's hoping more rights holders at least get the idea they can have tiers of content released appropriately.

The problem is, and always has been, on the client side. It's easy to dream up all sorts of amazing technologies if you're okay with making users download a new browser plugin or app.

Flash sucks, but it's on every single PC. Netflix has the power to get users to install Silverlight -- Microsoft made that pretty easy. Unfortunately you probably don't have their library and will live to regret it[1].

Adobe is also working on this idea with RTMFP (their newest streaming protocol)[2]

In fact few years there's been a company already doing this with a Flash Plugin (Yes dawg, there's a Plugin in my Plugin).[3] CNN use(d) it for quite a while.

I wouldn't bet anything on BitTorrent adding anything interesting here. The technology may be difficult but it's not as difficult as overcoming the huge barriers to entry.

The best we can hope for is that they create an open protocol the way they did with BitTorrent that browsers can implement over the next 5+ years.

1. http://en.wikipedia.org/wiki/Joost#Application_discontinued

2. http://labs.adobe.com/technologies/cirrus/

3. http://www.octoshape.com/addin/about.php

On the other hand, we've seen that people are clearly willing to download bittorrent clients to get their stuff. Add in PEX with DHT and tribler-like functionality (versus having to go find a sufficiently non-seedy website, and not having to put credit card details into netflix, and so on) and you've got a winner.

On the other hand bittorrent live has great distribution channels: The Pirate Bay et al.

There will be many people wanting to "stream bittorrent" and it'll get at least close to critical mass. Whether it can tip over into the mainstream will be the real question.

One example I could see is justin.tv that's betting heavily (with their twitch.tv site) on esports which I'm pretty sure has a good overlap with people who know/use bittorrent.

Another of course will be megaupload/megavideo v2.0.

It's definitely something worth following.

MLG used octoshape for a while, and it was notoriously unreliable. That left a bad taste in the mouth of the exports community when it comes to peer-to-peer distribution.

To me it seems like the rojadirecta.es folks, who are doing pirate live sports streaming, will love this.

But, BitTorrent Live has nothing to do with BitTorrent, besides the inventor and the name. It's a completely new, proprietary protocol. You need to download their proprietary software to view it. You can't view existing torrents with it; so the Pirate Bay doesn't provide it with a distribution channel.

One of the reasons that BitTorrent took off as well as it did was that it was an open protocol and free software. People could write competing BitTorrent clients. They could extend the protocol, adding new features like trackerless torrents and magnet links.

BitTorrent Live has none of that; it's yet another proprietary streaming service, designed to create a walled garden that the content industry can be happy with. It will succeed or fail based on how well BitTorrent Inc can cut content deals, and cut deals with set-top-box manufacturers to include their software. And that's not really a game that you want to be in, unless you have a lot of leverage like Apple or Amazon, or the backing of big content companies like Hulu.

Bittorrent already has a massive amount of users. Couldn't they just bundle their plugin with something like µTorrent?

netflix recently came to ireland. but requiring silverlight rules out linux users.

but if i want to watch irish tv (god forbid, it's usually terrible) i can use http://www.aertv.ie/ with no plugins or downloads beyond flash to watch all the irish channels. not sure what the legality is but it seems to be rebroadcasting all the channels covered under the irish digital channels that are free to air.

many of these channels show u.s. tv shows. wonder what the legality of that is. is aertv blocked outsite ireland?

My quick check, from a US ip address, the page loads but the flash doesn't stream. I'm guessing it's a geoip block.

What you need is a "reverse" version of www.hotspotshield.com

A colleague uses hotspot shield to watch Hulu from Ireland (Hulu is currently US-only).

What ever happened to that Chrome P2P API? http://news.ycombinator.com/item?id=2315908

Think a bittorrent client could be made from this?

It was releazed a few weeks back, and its called the RealTimeCommunication(RTC) API, no longer the P2P API, getting a client from this is possible, especially now that other browsers would be supporting it soon. Windows8's IE10 browser looks promising in this direction. But the API isn't written on stone, which could be a problem in getting a commercial app in the near future.

I have one word for you: websockets. This stuff is going to be baked in your browser very shortly, plugins are yesterday's game. If Cohen can make his stuff work seamlessly with websockets, very interesting things will happen.

It's not clear that we'll get P2P websockets (see WebRTC) or whether websockets are going to be plumbed to <video>.

= What I wished and thought BitTorrent Live was going to be =

Pirated P2P TVOIP with 1 minute delay:

People with REALLY good computers and TV capture cards would record and encode video streams in real time and would seed them out through Bittorent Live. The video might not be in real time but even a 1 minute delay would be fine. Hey live TV and Cable over IP would be awesome. My mom's been wanting to watch the Opera Winfrey Network so badly I'd love to get it working on the laptop. Commercials and all. Hell, the woman in me would love to watch it too.

Yeah it's "pirating", but we do it because content creators don't give us what we want. I don't care about licenses I just want the Operah Winfrey Network, NATGEO, and the Science Channel.

Piracy is the only way to make it work. Television, Film, & Music are stagnant industries that refuse to evolve unless and until they are forced to. And the Pirates have to innovate here, I believe.

It's always funny to read about P2P TV being sold as the cutting edge in the western world, when in fact China has been using P2P TV since 2004. I've used both PPLive and SOPCast to varying degrees of success in the 'states and according to wikipedia (http://en.wikipedia.org/wiki/PPLive) PPLive reported 105 million monthly active users in december 2010, so it's no small phenomenon. The interesting question is which components of the media environment in China allowed for the growth of such technology well ahead of western users.

> “It fits the DNA of what BitTorrent is about because it’s open and free.”

> An SDK to work with the proprietary protocol is in the works.

I'm confused

I think this article is remiss in not mentioning the dozen or so P2P live streaming systems that have been in production for years.

Without any details (really, any at all), it's hard to connect this BitTorrent Live with the idea of killing TV. I agree that "people love what they see on television, but want it accessible from the Web", but one of the main reasons why is that people don't want to watch TV live (other than sports, which has even more geographical licensing problems than regular TV).

TC and their over the top headlines. If we had a nickel for every time they proclaim a killer of any entrenched industry leader, we would all be millionaires.

I wouldn't really blame TC in this case; they're just quoting Bram. (It's an irresistible quote; GigaOM ran with it too.)

I would. They didn't just quote him. Here's the first para:

"Television is going the way of the dinosaur, and the deadly comet is called BitTorrent Live. Today, Bram Cohen, the author of the BitTorrent peer-to-peer sharing protocol, demoed his latest creation at the SF MusicTech Summit."

Nothing indicates that first sentence is a quote.

It isn't, but this is:

""My goal is to kill off television" Cohen said during the SF MusicTech demo session I hosted."

True, but in this case it's Bram Cohen saying it, which lends a lot of weight to the words.

Still, I'd love to read about how it's supposed to work.

It's the distribution method, not the TV per se that is being attacked, otherwise they would articulate the same thing in their Apple TV (iTV) article a few articles lower. The headline is just linkbait.

Sports fans have been using live P2P streaming for many years, the most obvious being Sopcast (sopcast.org - even the website is copyright 2007). The quality is pretty good, and the 'danger' posed was made clear with some of the first domain seizures being the stream aggregators - myp2p, rojadirecta etc.

The difficulty with this technology 'replacing TV' is that realtime is only truly relevant for a few things - sports, maybe concerts and awards ceremonies - beyond these the on-demand paradigm is much more attractive.

That's not to say that at scale a certain amount of bandwidth offloading couldn't happen to equivalent-timecode peers, but I suspect the biggest challenges to the complete overthrow of the cable networks model right now involve the notorious impossibility of negotiating agreements with the content owners, not the bandwidth/hardware cost of server-client streaming.

Meh, it's not solving the right problem to "kill TV". Content distribution isn't an issue. Netflix, Hulu, iTunes, Amazon, HBO Go and more are all effective distribution mechanisms. At best this lowers the barrier of entry which is good, but that's not the real problem.

The real issue is coming up with a sustainable business model that users are willing to pay for. Thus far monthly then monthly cable bill plus ads is ahead by a huge margin in terms of revenue. Content owners cutting out middle men is a good step, but the harder one is getting users to pay for it.

Of course it could kill TV with copyright infringing streams, but that would be illegal and send people to jail ala MegaUpload.

Content owners cutting out middle men is a good step, but the harder one is getting users to pay for it.

Interesting. By definition, that means middlemen are currently adding value.

For a good chunk of people it is still the cable companies that provide the last-mile pipe that something like this requires. And because they control it they can skew the bandwidth allocation towards their own digital TV side. You never hit a Comcast bandwidth cap by watching too much cable TV. But you can quite easily hit your "unlimited" bandwidth cap on the same pipe if you stream too much video over that very same cable prompting nasty email warnings and eventually termination of service. I don't see how BitTorrent Live addresses this in any way.

As a pedantic note, I wonder if the cap is measured from your cable modem or access to and from the Internet. Because if it's the latter (ever how unlikely) then a P2P protocol could make a difference.

I suppose that scenario could exist somewhere. It definitely doesn't on Comcast in the US.

I mean this in all honesty: how do you know that?

Try a traceroute to your neighbour who is likely on the same physical Comcast segment as you. It's not a simple 2-hop route. It goes way out and comes back.

I'd be interested to know how this compares to Flash multicasting... especially since Flash is already installed in most browsers.

http://www.flashrealtime.com/multicast-explained-flash-101-p...

Joost tried this with their P2P desktop client. I was really impressed with the performance and latency.

Technology was one side of the story but in reality Joost failed to get traction, I hope BitTorrent Live is able to make use of the technology may be integrate with existing bit torrent clients.

It will end like Joost broadcasting second- and third-rate shows.

Technologies like this can indeed reduce infrastructure cost not having to rely on expensive CDNs, but startups don't have the financial muscle to acquire interesting contents.

spotify does this at music-scale:

"Spotify – Large Scale, Low Latency, P2PMusic-on-Demand Streaming" [1]

TLDR: getting low latency, low stutter playback is a lot more complicated than bittorrent, but clearly possible, and already solved.

as dotBen points out, it doesn't prevent Spotify's business from getting re-negotiated into the ground every 3 years by the ip rights-holders.

i would like to see the technology sponsored by a business or industry who cares about freedom of information - like the press - otherwise it's just a few nerds working on a startup. I mean, app, I mean, toy project.

[1] (googlecache) http://webcache.googleusercontent.com/search?q=cache:http://...

wow. All this negativity and dismissiveness about what is potentially a good & impactful technology.

if it works well, it could mean netflix that is free (the difference between free & cheap is big). It might change what people watch (like youtube). It might change how/if people pay for it.

yes. I agree tomorrow morning tv as we know it will be unchanged. I can see lots of paths from here to someplace different that could involve such a service.

How does it make Netflix free when Netflix has to spend half a billion dollars buying the content?

Netflix's content bill is far bigger than their bandwidth bill.

This kind of technology should be integrated in the browsers, and obviously be open source. Google should've been working on something like this for Youtube. It would save them a lot of money. They wouldn't even need to use it 100% of the time, just for the more popular videos who are constantly watched.

I'm not sure why sensationalist headlines like this are necessary. As has been stated, the real problem is getting content. Lots of technologies exist that could have "killed" TV but haven't.

Veetle (http://veetle.com/) has had a functioning demo of such a technology live for at least a year now.

What is this?! The Twilight Zone?

Vetle is a fairly uncommon Norwegian name, and happens to be my first name. Very weird.

Reply on news.ycombinator.com