Slashdot Mirror

Very interesting concept, and I'm surprised nobody thought of it sooner.

In fact, they did.
http://www.peercast.org/
http://p2p-radio.sourceforge.net/
http://www.streamerp2p.com/

The only difference here is the budget. Not to be a prick, but I don't see anything inovative here. Except maybe the bittorent roots (22m for a modded BT client with an embbeded media player ? who's to say that a bittorent type algo is better than a p2p algo specifically designed for the task of streaming ?)
This development will not change much. People prefer to have the files on their computer and build collections, not stream them. They want to move them arround to other devices not connected to the net.
In very a distant future (*), when a huge library of pirated/cheap material becomes available, and most mobile devices have broadband internet connections, and the streaming is so damn perfect and flawless that it's indistinguishable whether you play a local file or a stream, than maybe something like this becomes relevant.
For commercial online TV and the like, this technology it's still unproven, and I'm not referring to SwarmPlayer specifically, but to alternatives that have been available for years. As it turns out, the cost of the bandwidth is not that large. It remains to be seen if a p2p method comes close in reliability to a well provisioned CDN, until now it has not. Digital online TV has other, much larger problems, for example the fact that it's a nightmare for most ISPs, who have designed their networks so that each user is able to browse for an average of 100MB/day, and now for the same user to view 5Mb/s digital TV 10 hours/day they need to increase the capacity 200x.

(*) 2 years in internet time

Is it actually a standard torrent? How is this different from something like Peercast?

Peercast May be the solution. It allows P2P music streaming, and it's hard to find the source. More importantly, the source only has "one" listener, and it can be hard to tell how many total listeners there are.

Unless your station is talk-only, 64k stereo/22k mono mp3 is totally unlistenable. If you don't have the bandwidth for a "full" 128k ogg or aac stream, at least provide a ~80k stereo ogg/vorbis stream. Not much more bandwidth but it will be listenable without giving headaches. Or send a single "full quality" 128k ogg stream to the PeerCast network while providing the modem/isdn-streams like you do now.

http://www.peercast.org/features.php
Anonymous broadcasting - clients do not tell each other if they are the source or just listening

"Icecast has had Theora support for a while now."

It has been a while since I messed with it, but I think I had Theora streaming with peercast as well.

For those that don't know, peercast does peer to peer streaming.

http://www.peercast.org/

all the best,

drew

Anybody remembers PeerCast? I think it was made for situations like this :) Some people in Japan already set up some stations, but not enough...

Can somebody set up one station closer to the US, to get my fix at lunch/work? (I don't mind the language).

Take a look at peercast as well.

And while you are at it, you may as well check freecast.

http://www.peercast.org/

http://www.freecast.org/

I have used peercast for years. I am still meaning to try freecast. One of the people involved in freecast hangs out in the #rivendell channel on freenode if you have any questions for him.

all the best,

drew

http://www.ourmedia.org/node/111123
A link to "Tings" the first draft of a novel I wrote in November 05 and put under a CC BY-SA licence. (think copyleft)

It would be fairly simple to build a p2p broadcast client that operated like bittorrent (so as to eliminate the need for a single server to serve bandwidth to everyone).

Simpler than building something like that would be to just use peercast. A peer to peer audio streaming program designed to eliminate the need for a single server to serve bandwidth to everyone.

(Amusingly another reply to the parent post is some-one pointing out how this can never work. What I'd call 'management material' there..)

Actually, there is such a program. Its called Peercast, and you can read about it here.

Now the personal experience... When these guys first came out, they 'promised' to make the source code available. After many, many, MANY months I think they finally may have, but not before missing target dates. And not before pissing me off enough to make me think that the 'promise' was just bait to get more users. Eventually, I left, and stopped running the software. It may have grown since then, but something tells me the egos of the people who run that site have gotten in the way more than once.

Let's have a look to :

http://www.peercast.org/
and
http://www.freecast.org/ (java based)
;)
yomguy

What you're describing sounds like PeerCast. http://www.peercast.org/ I've tried it with some radio stations.. connectivity rate seems to suffer, but otherwise, there aren't any issues with quality (but I only used for radio, not video)

I`ve been looking for an open source video server for Linux that can send TV/video/audio to my laptop etc.. does such a beast exist?
Peercast or Videolan would probably be a good start.

Peercast does have some live video feeds being p2p-broadcasted. On the peercast yellow pages, you can find live anime feeds running for a few hours a day.

Peercast does have some live video feeds being p2p-broadcasted. On the peercast yellow pages, you can find live anime feeds running for a few hours a day.

http://www.peercast.org/

This does the whole live-stream multicast over P2P thing. Suprisingly, it sometimes works.

I think he is referring to the problem that, what is it, the affero license is trying to correct?

I could be mis-reading that though.

I don't buy that I, as a GPL coder, am a big loser though. I may not be as big a winner as some, but I cam certqainly no loser in this game.

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

"This guy is following the license and spirit of the GPL, and making money doing it. People should be patting him on the back, not giving him a hard time."

If you read the posts, you will see that some people do not think he is following the spirit of Free Software. Making money from it is not the problem for them though.

Isn't this software based on java? Care to comment on his devotion to the spirit of Free Software?

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

"Now, the trick they use is to purposely not document their work, it is free indeed, but just try to use it. Oh, you want support? Write a check to....."

"One thing with people who only do it for the money is that I tend not to trust them not to make things unnecessarily complex in order to earn the service/consulting money."

My point exactly. I think the tactic is so unnecessay and counterproductive. There are always going to be real problems and opportunities and people will always pay to make their lives better. We don't need to create fake problems to get them to pay us to fix.

When I find people who do this, I try to stay well clear of them.

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

Here is a link to a relevant page on the jboss site:

http://www.jboss.org/company/pos

So, he obviously seems to think you can make money writing "open source" software.

Lately, I am thinking that a more interesting question is one that should be on the minds of Free Software users...

What are some important factors that I should look at in assessing "the total development environment" of the software I am depending on?

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

"It's there because he bought a piece of equipment that effectively installed its own "Open House" sign."

A point I made earlier (in me reading.) So bring the device maker up on charges for wrongfully installing the open house sign, not the person so saw the sign and walked in to the open house innocently.

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

"It's even more than that. The wireless router received a standard, "can I have legitimate credentials on this network?" request in the form of a DHCP lease request."

Yes, and the real culprit here is the device manufacturers who ship these devices with default open policies to make it easier to just plug and go without knowing anything and without having to configure them on initial setup.

all the best,

drew

http://yp.peercast.org/?find=bysa&Submit=Search

Hasn`t peercast being doing this for about 3 years so far?

theres alot more interesting content on their Yellow pages too. 200+ channels etc.

Hasn`t peercast being doing this for about 3 years so far?

theres alot more interesting content on their Yellow pages too. 200+ channels etc.

P2P broadcasting isn't new. Pearcast pearcast.(www.peercast.org/) has been doing it for quite a while. The windows client also has OGGCap Video Broadcaster included. It allows you to broadcast in ogg or ogm, the input from a capture card to P2P on the fly or saved to disk, in various quality levels. The yellow page directory is at http://yp.peercast.org/. We've tested peercasting a local public access chanel. Pretty simple actually.

Also, There is a free 24 hour Adult XXX program P2P TV network supported by tv commercial type ads (free registration). http://www.adultinternet.tv/ It uses the same peercast client as the generic peercast above but it's branded. The the generic client works the same except it doesn't direct you the the right yellow pages http://www.adultinternet.tv/yp/. If you use the generic client and go to the yellow pages you don't need to register (which is a stupid requirement).

In order for these to work, you must open port 7144-7145 (which is different from bittorrent) on your router's firewall. Using trigger ports works well as opposed to port forwarding if you have more than one computer on your home network.

We are working on implementing a model that will encourage greater participation (not affiliated with the website that is the subject of this article).

done. http://www.peercast.org/

But skype is p2p, so that instead of you streaming directly to your audience, listeners may stream from you AND some other listeners, obviously minimizing the bandwidth required of the originator.

PeerCast does try to do what you describe, but last time I checked it didn't do a very good job of it.

Could this be fixed by ISPs putting in their TOS that you are not allowed to make any content available on the net unless you are willing to have it freely copied with a minimum of a copyleft license?

Would this work? If not, why not? If not, could it be made to work after a simple tweak?

all the best,

drew

http://www.peercast.org/

An ISP can better control voice quality by avoiding using the public internet.

I think giving their own customers priority wouldn't be effective, except for local calls in an area where the ISP has a monopoly. From what I gather about VOIP, it is supposed to also get even cheaper if the recipient is also using VOIP. Two parties on different ISPs using VOIP to call each other wouldn't benefit if both ISPs are crippling each other's packets. That could be the case in areas where ISP coverage overlaps. That would also be the case for international phone calls, like one to Bulgaria mentioned in the article.

On a side note, I'm curious about how networks will be able to distinguish VOIP packets from others. Perhaps it would be easy to do that with hardware-based VOIP connections, but with software it could be trickier, like with theswitchboard.ca example in the article. Calling someone through a computer may end up being an integration of Instant Messaging audioconferencing and VOIP. The call could be routed different ways...

1. If the person being called is online using the same Instant Messaging software, the audio is streamed between computers.
2. If the person being called is not online but has a VOIP hardware setup, then the call is routed to his/her phone that way.
3. If the person being called isn't online and doesn't have VOIP, then the call is handled as a regular call as close to the recipient as possible.

I also see there is a major parallel between the functions of VOIP, Instant Messaging, and P2P software. Maybe that was the whole reason for the KaZaA/Skype bundle. Conference calls and filesharing with multiple users is technically very similar from a software perspective and could use the same code. In fact streaming an audio call to several recipients is just like uploading an audio file to several users simultaneously, except that is is being recorded on-the-fly. One application could handle all the following, using much of the same code, blurring the distinctions between them so that it's all pretty much the same thing...

• Address Book (automatically determines if call recipient can be reached by email, IM, or phone number)
• Email
• Instant Messaging
• Audio/Video Conferencing
• Filesharing
• P2P Audio/Video streaming (like PeerCast)
• VOIP

Yes. In fact P2P has already been applied to streaming. For example there's PeerCast P2P Radio.

http://www.peercast.org/ ?

PeerCast.org is one solution. You can set it up on one PC letting only clients on the local network connect.

You'd be better off with an open-source program like Peercast ;)

Thanks, what I saw at the download page was just a binary for my OS and some documentation. I must have missed the link to download source code under a free software license.

http://peercast.org/code/

Use PeerCast to stream an Ogg Media format stream. Hook up VLC to it and you're all set.

Yes, this all works on Mac OS X.

P2P Streaming will revolutionize high quality broadcasting on the internet.

"Also, Shoutcast swarms could be especially nice."

Peercast does this.

While the user can't skip ahead, they can efficiantly broadcast to an infinite number of other users.

PeerCast

There are other distributed streaming solutions out there, but this one seems to work the best for me.

Peercast already allows for P2P video streams in most popular formats.

I've had a go with it and its not too shabby.

With clients for Mac, Linux and Windows, availability is good. Unfortunately, Peercast doesn't advertise themselves too well which means there aren't so many video streams available yet (typically 5-15 video streams and 100 or so Audio streams.)

Mods: this is ON topic, this is the sort of project projects like Theora and Dirac, and even the Xiph project as a whole, seek to enable.

Theora is usable now. As for real-time P2P broadcast, have a look at Peercast and develop from there. You may want to keep the bitrate low. Real, real low.

(Note: The following is publication, and may act as prior art to any future patent on this issue.)

I warn you, what you REALLY need is a peelable codec - one in which you can not only reduce the bitstream trivially to an optimal lower bitrate form, but reconstruct the higher bitrate form trivially (ideally, by just tacking the extra bits onto the end of each chunk).

Vorbis has a structure like this, but current encoders do not produce bitstreams that are optimally ordered for peeling, so quality will not be as good; at the moment, no video codec has a structure like this.

It's possible, once that is sorted out, to organise a network based on latency and the knowledge (dynamic measurement) of latency and bandwidth between links so that the network organises a pretty good distribution network by itself, so that peers automatically get the bitrate they need, and redistribute what they can; and so that the parts can be reconstructued so that low-bitrate stream bits are easy to get and plentiful, and peers with higher bandwidth can get higher bits from the stream from higher bandwidth peers, and redistribute just those (which are rarer).

Bearing in mind that you want to keep a T-latency system, dynamically figure out how you need to buffer ahead and reduce bitrate as required (and in particular, buffer low bitrates ahead, getting higher layers as the estimated curve is satisfied)... and also bearing in mind that many ISP connections are asymmetric, severely harming P2P networks, but this is a situation in which multicast, especially if used dynamically and cleverly, would be stupendously useful... and ALSO bearing in mind that typical ISP connections have a modellable, annoying behaviour where the latency increases as the data throughput increases, it's a challenging and exciting NGP2P network design.

This is, I might note, an ongoing research project. (It'd be spiffy if someone would fund its development, but thanks to the new emerging dumb-laws-to-be in the US like the INDUCE Act, I can think of several reasons that'd be a really dumb idea, no matter the intent or capability of the network.)

And don't have an alternative system of their own. Which in turn makes boycotts much easier.

On a side note, the technology is impractical - buffering doesn't matter - unless the stream is multicasted, a large enough group will either bring down the server/s, or impose a huge bill for bandwidth.

PeerCast. Somebody suggested it here earlier too. Or capture the stream to a file and Bittorrent (or Freenet) it out - not exactly realtime, but close to that.

PeerCast is an Open Source (not sure about the license but the sourcecode is available using Subversion) P2P broadcasting system which works great!

Mr. Stallman would approve: not only does it run on GNU/Linux, but it's under the GNU General Public License.

Maybe this is a good way to promote P2P broadcasting?
PeerCast is an Open Source (not sure about the license but the sourcecode is available using Subversion) P2P broadcasting system which works great! I've not tried broadcasting/viewing videostreams, just listened to radiostations, but it has support for MP3, OGG Vorbis, Theora, WMA, WMV and NSV streams.
Very easy to install and use, it's just a single executable!

You just point it to a streaming source (for example your own IceCast server, a WMV stream which you have access to or your favourite internet radiostation) and the stream is available on the PeerCast network for everybody to listen to or watch, just pointing your favourite player to a http://localhost-URL.

Does anyone still remember Massinova? They offered near CD quality streams, a great request system, etc etc...

And to thank them for their efforts, the RIAA sued and screwed em, and now that great Trance stream is no more.

Ah, but it lives on at Massinova: Reborn -- http://massinova.db140.com/. The stream itself is at http://66.135.33.226:8000/listen.pls. The request system is gone, I'm afraid, but the tunes that made up the soundtrack of many late night coding sessions are still there. Enjoy!

Also, you might want to think about helping to keep Massinova and other Webcasts up -- check out PeerCast.

The maker of P2P Client, Shareaza, is working on a new Radio P2P project called Mercora. This network gives users the ability legally webcast music to other users on the network.

How is this different from PeerCast? I glanced at the Web site, and didn't see anything that was revolutionary -- looks like PeerCast combined with IRC to me.

Though, perhaps they have fixed the problems PeerCast seems to have with bandwidth -- I've used it off and on, but it seems to always suffer from lag. Perhaps that would go away if there were more users, or perhaps it's just inherent in the design of the network -- I've never bothered to look at the technical details.

Anyway, I think the more exposure Webcasting has, the better. More variety, smaller players that can appeal to niche audiences, and lack of corporate interests playing to the lowest common denominator for the highest advertising profits are all advantages Webcasting has over traditional, ClearChannel-dominated radio. At the moment, at least. :-)

CDex
Ogg Vorbis Audacity

CD-DA X-Tractor AudioCoding Mp3splt Mp3Wrap

Alba Extractor PeerCast GNUMP3d Mp3 Tag Tools

GramoFile FFmpeg

JAZZ++ Open Sound World

Wow. Slashdot really sucks. If the lameness filter actually prevented ascii art, they might have an excuse.

I stream most of my music like radio now - couldn't be bothered to actually download mp3s (free or not) - and I can't wait to see something like peercast p2p radio combined with a "people-who-like-this-also-like-this" collaborative filter.

--

Peercast is a pretty good solution. The only problem is that it has very few stations. I've uninstalled it and I only listen to shoutcast now.

Peercast is a pretty good solution. The only problem is that it has very few stations. I've uninstalled it and I only listen to shoutcast now.

It's called Peercast and you can find it here.

I just wish more people would use it....