I finally realised that an engineering degree such as sofware engineering or even computer science just isn't for me. Instead, I'll focus on getting a business degree. Most of the technial stuff is pretty easy, anyways.
Besides, I can learn all my engineering skills from Slashdot!
The future of video distribution will be based on decentralized, P2P distribution technologies. Due to tit-for-tat, these tend to be as fast as the average UL speed of those in a swarm. Thus, having faster UL speeds is the real concern here. It doesn't hurt to have a 1024Mb/s DL connection, but if UL is still 256Kb/s, then it's all still moot. On the other hand, if the 1024Mb/s can be used as UL bandwidth, then we're all set for 512Mb/s each way.
Also, if we're trying to do video in real time, 512Mb/s can cram a pretty decent AV stream in such a bitrate.
By utilizing BitTorrent or other decentralized internet protocols, media holders can distribute their works at lest cost, meaning more profit for them and less cost for us.
Issues with copyright infringment don't go away, but they may be negligable by legitamite alternative means, such as advertising through the site providing the core infrastructure for distribution or even directly in the media itself.
For premium content, pay-per-torrent solutions are also possible, and even DRM systems as a last resort.
Yes, but one would think a 160-bit key would be good enough for most things. I doubt data corruption wouldn't get caught using this.
As for the popularity issue, that is a problem that will need to be addressed. Structured overlays may be the trick here, but time will ultamitely tell.
What about Decentralized (P2P) Replication of data? Perhaps the sheer number of nodes on the internet today combined can replicate all data to the point that a file's SHA-1 or whatever can be used to retrieve that file at any point in the future. And, 160 bits isn't hard to physically write out to another medium, either.
The answer to every question is Decentralized P2P. Or at least I think it is.
Not only is P2P technology capable of substancial non infringing use, it is also commercially beneficial.
The same 'viral' distribution that the industries loathe so much can be turned around to form a zero-cost distribution medium, meaning more profits for the holders and cheaper prices for the subscribers.
There are two things here: P2P and copyright infringement. The problem is, they are not the same thing. By saying P2P is Legal because someone was declared not liable for copyright infringement is irrelevant.
The true test of P2P legality comes from issues such as MGM v. Grokster and the various INDUCE bills floating around the US and particularly California.
The issue here isn't the current state of affairs with P2P being a massive utility for copyright infringement, it is the potential for a zero-cost infinite distribution channel for publishers. That potential is what P2P is really all about.
First, I never had to do that for Ogg Vorbis audio streams.
Second, I was not referring to Ogg Vorbis audio-only streams, but streams with both Vorbis audio and Theora video.
Unfortunantly, at this time I am unable to determine whether PeerCast is using a standard stream format for these, because they are labeled OGM everywhere I can find it. With a little bit of Googling, I found OGM Container, which seems to say that this format is a derivitave.
So now I'm confused. Can OGG streams be used to stream both a/v? Then why does PeerCast call Ogg Vorbis streams OGG streams, and Ogg Theora+Vorbis streams OGM streams?
I'm sorry, I just made things more confusing than I wanted to. It looks like the majority of PeerCast users are Windoze idiots who like their Windows Media and Real Players.
Yeah, but PingBack is somewhat better, in my opinion.
First, Pingback uses an extra header on the document, versus the extra XML embedded into the document for Trackbacks. Second, Pingbacks have no discription provided by the pinger. The trick here is that the pingee checks the URL to ensure that there indeed is a link to its page, and while its at it, it grabs some text around the link.
Magnet support? Then it's just hash-based, then. eDonkey and Gnutella have been doing this for years.
eDonkey uses a splitfile with MD4 Gnutella uses SHA1
The problem with Supernova was twofold. First, the bandwidth needed for trackers and torrents was high. Second, the linking to the torrents or the torrents themselves was a potential liability. Gnutella and eDonkey solve the former problem, but the linking liability problem will always remain. There has to be some point of entry, and there is where the technophobes will try to attack.
There are no legitamite uses of P2P file sharing technology.
Why would any distributor want to distribute their data with no cost to them? That would be stupid; distributors make their money by putting profits on the tail end of the distribution price itself! Therefore, unless the data has no IP value to the distributor, like silly communistic "open source" software, all file sharing would do is eliminate the possability to pad a distribution fee to make their buck.
And, duh, all open source software is communistic. It's against the capitalistic and corporate philosophy of this great conservative country!
Get an HTTP client that can resume where it left off, silly!
I never understood why a failed HTTP or FTP transfer had to start over. First, the clients didn't support it. Then, and still, some servers out there don't support it.
It doesn't matter, really. Hash-based P2P will change the world of file transfer, anyways.
PAR2 error correction, RAID collocated, and Subversion. Simple.
P2P filesharing is also becoming a great way to backup popular files. Using networks such as Gnutella, you can even retrieve the file using a SHA1 hash. Magnet URIs, anyone?
Slashdot Mirror
I finally realised that an engineering degree such as sofware engineering or even computer science just isn't for me. Instead, I'll focus on getting a business degree. Most of the technial stuff is pretty easy, anyways.
Besides, I can learn all my engineering skills from Slashdot!
The future of video distribution will be based on decentralized, P2P distribution technologies. Due to tit-for-tat, these tend to be as fast as the average UL speed of those in a swarm. Thus, having faster UL speeds is the real concern here. It doesn't hurt to have a 1024Mb/s DL connection, but if UL is still 256Kb/s, then it's all still moot. On the other hand, if the 1024Mb/s can be used as UL bandwidth, then we're all set for 512Mb/s each way.
Also, if we're trying to do video in real time, 512Mb/s can cram a pretty decent AV stream in such a bitrate.
Don't use passwords. Use physical keys (USB Stick, Keycard) and biometrics instead.
Just say no to passwords.
By utilizing BitTorrent or other decentralized internet protocols, media holders can distribute their works at lest cost, meaning more profit for them and less cost for us.
Issues with copyright infringment don't go away, but they may be negligable by legitamite alternative means, such as advertising through the site providing the core infrastructure for distribution or even directly in the media itself.
For premium content, pay-per-torrent solutions are also possible, and even DRM systems as a last resort.
Yes, but one would think a 160-bit key would be good enough for most things. I doubt data corruption wouldn't get caught using this.
As for the popularity issue, that is a problem that will need to be addressed. Structured overlays may be the trick here, but time will ultamitely tell.
What about Decentralized (P2P) Replication of data? Perhaps the sheer number of nodes on the internet today combined can replicate all data to the point that a file's SHA-1 or whatever can be used to retrieve that file at any point in the future. And, 160 bits isn't hard to physically write out to another medium, either.
The answer to every question is Decentralized P2P. Or at least I think it is.
Not only is P2P technology capable of substancial non infringing use, it is also commercially beneficial.
The same 'viral' distribution that the industries loathe so much can be turned around to form a zero-cost distribution medium, meaning more profits for the holders and cheaper prices for the subscribers.
Pay-Per-Torrent, anyone?
There's also Ruby.
There are two things here: P2P and copyright infringement. The problem is, they are not the same thing. By saying P2P is Legal because someone was declared not liable for copyright infringement is irrelevant.
The true test of P2P legality comes from issues such as MGM v. Grokster and the various INDUCE bills floating around the US and particularly California.
The issue here isn't the current state of affairs with P2P being a massive utility for copyright infringement, it is the potential for a zero-cost infinite distribution channel for publishers. That potential is what P2P is really all about.
I guess donuts will become very expensive with this.
Why not leave the GPS device in your garage?
lynx http://www.dec.org.uk/
Now for a [insert word that means self advertising]: Project nuWeb is aimed at making IM+P2P come true.
First, I never had to do that for Ogg Vorbis audio streams.
Second, I was not referring to Ogg Vorbis audio-only streams, but streams with both Vorbis audio and Theora video.
Unfortunantly, at this time I am unable to determine whether PeerCast is using a standard stream format for these, because they are labeled OGM everywhere I can find it. With a little bit of Googling, I found OGM Container, which seems to say that this format is a derivitave.
So now I'm confused. Can OGG streams be used to stream both a/v? Then why does PeerCast call Ogg Vorbis streams OGG streams, and Ogg Theora+Vorbis streams OGM streams?
I'm sorry, I just made things more confusing than I wanted to. It looks like the majority of PeerCast users are Windoze idiots who like their Windows Media and Real Players.
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.
There are no lines - only line break characters.
Seriously, folks, count bytes, not lines.
Yeah, but PingBack is somewhat better, in my opinion.
First, Pingback uses an extra header on the document, versus the extra XML embedded into the document for Trackbacks. Second, Pingbacks have no discription provided by the pinger. The trick here is that the pingee checks the URL to ensure that there indeed is a link to its page, and while its at it, it grabs some text around the link.
Magnet support? Then it's just hash-based, then. eDonkey and Gnutella have been doing this for years.
eDonkey uses a splitfile with MD4
Gnutella uses SHA1
The problem with Supernova was twofold. First, the bandwidth needed for trackers and torrents was high. Second, the linking to the torrents or the torrents themselves was a potential liability. Gnutella and eDonkey solve the former problem, but the linking liability problem will always remain. There has to be some point of entry, and there is where the technophobes will try to attack.
Nothing to see here, move along.
P2P technology itself is not evil. I hypothesize most legal data distribution online will take advantage of P2P technology within the next few years.
There are no legitamite uses of P2P file sharing technology.
Why would any distributor want to distribute their data with no cost to them? That would be stupid; distributors make their money by putting profits on the tail end of the distribution price itself! Therefore, unless the data has no IP value to the distributor, like silly communistic "open source" software, all file sharing would do is eliminate the possability to pad a distribution fee to make their buck.
And, duh, all open source software is communistic. It's against the capitalistic and corporate philosophy of this great conservative country!
(Hint: Today's word is sarcasm.)
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.
Get an HTTP client that can resume where it left off, silly!
I never understood why a failed HTTP or FTP transfer had to start over. First, the clients didn't support it. Then, and still, some servers out there don't support it.
It doesn't matter, really. Hash-based P2P will change the world of file transfer, anyways.
Congradulations! You've just reinvented a hash-based filesharing network. You're not the first, though:
.torrent block hashing thing. Most of these networks deploy swarming, too.
* Gnutella (BASE32 SHA1)
* eDonkey/Overnet (Tiger Tree Hash)
* KaZaA (KZHash)
* Freenet (CHK)
* Mnet (?)
Mnet even does the full
The coolest thing is magnet-uri's. I've even written a redirector for SHA1 links here.
PAR2 error correction, RAID collocated, and Subversion. Simple.
P2P filesharing is also becoming a great way to backup popular files. Using networks such as Gnutella, you can even retrieve the file using a SHA1 hash. Magnet URIs, anyone?
Just tax internet usage. Wait, that's the tax on the payments to the ISP. Nevermind.
Why do I not have mod points when I need them, and have them when I don't?
The mod point system from Slash should be redesigned, In My Not So Humble Opinion.