Roll Your Own Television Network Using Bittorrent

Cryofan writes "Mark Pesce, lecturer at the Australian Film Television and Radio School (AFTRS) writes here and here about using p2p networks, specifically bittorrent, to create a grassroots television network. He cites as an example the BBC's "Flexible TV" internet broadcasting model using that as the core of a "new sort of television network, one which could harness the power of P2P distribution to create a global television network." Producers of video entertainment and news would provide a single copy of a program into the network of P2P clients, and the p2p network peers distribute the content themselves. Thus, a virtual 'newswiki' where the content is distributed bittorrent using some sort of 'trusted peer' or moderator mechanisms as a filtering/evaluation mechanism. So what is stopping anyone from doing this now? Awareness of the concept, perhaps? Lack of broadband connections? Lack of business models for content producers?"

Comments

Content

[sport_160]

I think what is stopping people now is a lack of legal content that they can share. You can bet that nobody wants to watch my home videos.

Re:Content

[d34thm0nk3y]

I think what is stopping people now is a lack of legal content that they can share

I agree, and frankly, what is availible usually isn't very good so it requires a lot of "filtering" to find much you like. I think that this will change, though, once artists realize they can make money more directly.

upstream quota

[discord5]

How about the average broadband connection having an upstream quota cap. 1.5GB of upstream traffic a month for me, and not a byte more unless I "contribute" a generous amount to my ISP.

This is still one of the major issues for me when it comes to ISPs. If I would download something popular from bittorrent or edonkey, 1.5GB is absolutely nothing. So the only solution would be if I were to firewall incoming connection and be a leech, or put QOS on all traffic going out, limiting it to 0.5K/s.

This all is of course hypothetically speaking... ;)

How do you advertise?

[cranos]

I think one big hurdle to this sort of thing would be how do you cover you're costs.

Producing even a basic news show still costs money, even if all the people running it are volunteers.

Re:How do you advertise?

[Cryofan]

I was thinking maybe product placement commercials, or banners occasionally running across the screen.

Still too hard for the average user

[papasui]

While I think Bittorent is pretty easy to use when I tried to explain it to my sister she had no idea what I was talking about and wanted to know why it was better than Kaazaa. In order for this to take off beyond the geek community to average users it needs to be somehow streamed to a easy to use media player or embeded in a webpage. There is a lot of potential with this type of technology, but it really needs to be super-easy to make any kind of splash. And I can also see this type of network abusing the end user who isn't smart enough to exit the program and then can't figure out why their internet connection has been moving at dial-up speed for the last 3 weeks.

mass tv over p2p?

[seramar]

What happens when people start embedding viruses and worms into media files? With the GDI+ vulnerability, it's only a matter of time. And it'd be easy for people on a p2p network to modify the file and start sharing it. Sure, you could have moderators etc, specified distributors, whatever, but that sort of destroys the point of having something like this utilize a p2p network. And if it's very popular, then you know the files would have a high likelihood of being modified and corrupted. Or how about simple work arounds to make the file appear to be of one media type when it's really another? Sure, few people on slashdot would have to worry about getting tricked. But we're not the masses. And isn't that what this sort of thing is aimed at?

The Real Problem

[techsoldaten]

The real problem with this idea is ubiquity of signal. Anyone can post anything they want, even if broadcasters closed off a single p2p service just their programs there would always be competing services. Pr0n, wicked graphic hunting shows, and real-life stuff would dominate the bandwidth, things we may want to keep our kids away from.

M

OMG WHY NOT PPL!

[Anonymous Coward]

seriously, these articles are strange. the author takes a really new concept that no one has really heard of, or at least the implementation is new and then writes something saying:

OMG TEH MAN HE DOES NOT LIKE P2P! THIZ NEW IDEA IZ PROOF! EVERY T>V MUST USE P2P, BITORRENT IZ JEZ"US! I HATH PROVEN THE MAN HATEZ GOOD IDEAZ!!!!!!! HAR HAR! DOWN WITH CORPORATIONS Who WATZE!

when in reality no one has really considered the concept in all actuality. and for some reason the author fails to notice, could make the whole idea worthless anyway. but we get an interesting slashdot read.......

One little problem...

[TheNarrator]

The one little problem is that bittorrent is not a streaming protocol. It cuts up the whole file and sends a different piece in random order to each client. Each client then trades there piece with the other clients. So you can't go linearly through a video segment without having the whole thing. You could make smaller downloadable segments that would download and then auto load sequentially. It wouldn't be live though.

Waiting too long for a show

[prozac79]

I don't know about the rest of you, but when I try to download something from a bit torrent source, it takes several hours over a DSL modem. This even happens on torrents that have a lot of seeds and a lot of downloaders. So how feasible is it to have P2P, on-demand television? Even if you could stream them, the download rates are far from constant so you would have to pause a lot to accumulate a buffer.

Bittorrent is not the right way to do this

[hyc]

Any connection-based protocol suffers from scaling problems, especially on the scope this article implies. If you want to do a media broadcast, you should be using IP multicast in realtime. Then you don't need to worry about upload rates either, you get maximal efficiency and data only has to move in one direction around the network.

All of the P2P networks have this problem because they are connection-based and on-demand. A TV network is not on-demand, it's a fixed message delivered on a published schedule. That's the model that works most efficiently, making the most efficient use of the transport medium. For the internet you can be somewhat flexible and start redundant broadcasts at staggered time intervals, but in general, if you don't start listening/downloading when the stream starts, tough.

For compressed video you need to make sure that there are plenty of I-frames in the stream so that people can come in at any arbitrary point and sync up, but that's no big deal. Also if you take this approach you don't need to broadcast multiple streams of the same content at different resolutions/bitrates, the network itself will provide rate reduction by dropping frames that the receiver can't pick up fast enough. (Tho doing that will make the audio pretty noisy; I guess you can do low bandwidth streams if you really want to. Or just do separate bandwidth streams for the audio. That way if one audio stream needs too much bandwidth and is losing too many packets you can just select a lower bandwidth stream instead.)

Re:Bittorrent is not the right way to do this

[An Onerous Coward]

Why IP multicasting? What real advantages does it give you? Aside from the "Ooh, look, this is happening *right now*" factor, it seems like live streaming just takes all the problems of regular broadcast television and imprints them on a much more flexible medium.

Live feeds have their purpose, but I'm having trouble seeing how they would work well under a bittorrent system. It could be set up under a telephone tree model, where node A feeds nodes B C and D, which each provide feeds to five or six other nodes. There could even be some redundancy built in so that a dropped packet to node D doesn't propagate to all its clients.

But for things where time isn't critical (read: 90% of what we watch on TV), Bittorrent is ideal. Unlike normal Internet broadcasts, supply scales with demand. Even a delay of a few minutes should be adequate for most purposes.

Want everyone to watch your thingy at exactly the same time? Send it out to all the nodes that want it in some encrypted format, then when enough nodes are seeded to meet demand, distribute the key.

Varying quality streams are possible within a single stream. For example, with Ogg Vorbis, you can get a low-quality stream from a high quality stream just by removing portions of the stream (no re-encoding necessary).

I think all your objections can be overcome,

Re:Bittorrent is not the right way to do this

[Anonymous Coward]

Varying quality streams are possible within a single stream. For example, with Ogg Vorbis, you can get a low-quality stream from a high quality stream just by removing portions of the stream (no re-encoding necessary).

That's an argument for multi-casting. BitTorrent doesn't stream, period. The part of the file you're being given at any time is effectively random. You're just as likely to have 5 seconds of high quality video as you are to have 15 seconds of low quality video. With multi-casting, the video is pushed out in a known order. If a few packets get dropped, then you just have a low-quality stream, but you still have a stream. With BitTorrent, you never know what you have until you have the whole file.

It's just a question of what your requirements are. BitTorrent is an all or nothing protocol, but you can get it any time. Multi-casting is, you get it as it comes, but you know what's coming. For video, I think people really do want the video ASAP, and can give up some quality if they need to.

I'd like to see a NPR/PBS style approach

[Dekks]

Have a network of members and affiliates who all shoulder the cost, donations go to the pool and appropiated by a commitee/board to fund different projects and shows. This way you could have a world community, that drills down to a national community, that can still drill down to a local community, mix and match the international shows with the national and local.

The AFTRS

[mathgenius]

I've seen some of the short films that AFTRS students produce, and they are world class productions. Really brilliant.

Simon.