Slashdot Mirror
BitTorrent May Prove Too Good to Quash
gollum123 writes "There is an article on washignton post on bittorrent where the author discusses why BitTorrent is here to stay. According to the author it is being increasingly used to distribute software and entertainment legally. It also mentions that in BitTorrent, unlike many other file-sharing programs, legitimate use doesn't amount to a token minority. It's central to this program's existence. It concludes by saying that the MPAA may be able to drive BitTorrent movie downloads into what Green called "the dark corners of the Internet," but this program isn't going to go away. It might, however, be just what movie studios and record labels need to market and distribute their own content efficiently on the Web."
It might, however, be just what movie studios and record labels need to market and distribute their own content efficiently on the Web.
Well, at least someone realizes this, instead of tacitly - or overtly - arguing that it's okay for them to be unabashedly ripped off, coupled with myriad ridiculous justifications and semantic acrobatics about how it's not really "stealing".
Frankly, the content industry convincing major ISPs to enable multicast on their networks may go a lot further toward efficiently distributing non-"on demand" content than something like BitTorrent.
But backing up a bit:
One reason for this change of heart may be that in BitTorrent, unlike many other file-sharing programs, legitimate use doesn't amount to a token minority. It's central to this program's existence.
Not that I don't recognize that BitTorrent is currently used for many legitimate applications (whereas that was extremely difficult to argue with a straight face with P2P), but I think this statement is a little overboard. I'd say that, currently, "legitimate" use of BitTorrent is a "token minority" of its use. The vast, vast majority is pirated software, pirated movies, and pirated TV shows (and, to a lesser extent, music, just because of the nature of BitTorrent being more conveniently applicable to small amounts of large files, rather than large amounts of small files).
Anyone not admitting that at this particular point in time is lying to themselves.
Note that I agree wholeheartedly that BitTorrent isn't going to go away. Neither did P2P. But the content owners will continue to rightfully go after people and sites who distribute copyrighted content unlawfully, no matter the mechanism (please, no fringe examples of 83 year old grandmothers and dead people). But yes, I get the point - and agree with it - that BitTorrent could potentially have much more legitimate use than traditional P2P.
The point is valid: the fundamental distribution mechanism of BitTorrent is a novel and good one; there is no reason that BitTorrent couldn't, for example, be made even more robust and further "protocolized", and integrated into browsers and other download clients, allowing content distributors of any stripe to take advantage of its clear benefits. And in order for it to be a compelling solution for real content providers, that's exactly what will have to happen.
The MPAA will still want to charge about the same price for a download as a store-bought movie.
Unless they prove me wrong, their torrent distribution model is not viable.
One of the things for which I love BitTorrent is the ability to get movies and television programs not available in the 'States. I'm studying Japanese, and don't like most of the Japanese media that is available in the US, as it is marketed, by and large, for the otaku crowd. I mean, yeah, there's some good stuff in there, but most of it is crap.
Having access to BitTorrent means that I can download regular TV shows, dramas, historical programs, and recorded news broadcasts, all of which would be completely unavailable in the U.S. I can download anime that I like, but which isn't popular enough to make it into the U.S. market. These are all very effective study tools, and have helped me improve my listening comprehension markedly.
--
I Hit the Karma Cap, and All I Got Was This Lousy
It's increasingly likely that in the years to come it will be possible to rent videos by having a set-top box coupled to a DSL or Cable broadband pipe, which downloads DRM-enabled video files from a central server.
What better way to save bandwidth - the single killer cost when each film might sum a gigabyte - than by having the box download the film using a restricted version of bittorrent, and use a proportion of the available upstream bandwidth on the local connection to supply other people renting the same film? As the file's encrypted piracy wouldn't be a concern as the key to play it would only be issued by the central server, over an encrypted channel.
This would have the effect - exactly opposite to a DVD-rental shop - that popular videos would be available more quickly than rarely demanded ones. The system has the same priorities as the company behind it.
I have been a user for about 10 years. This ends Feb 2014. The site's been ruined. I'm off. Dice, FU
BitTorrent is a really big change, because with it we can finally upload data directly to "the network". The physical location of the data is immaterial. It's a really distributed database, where the schema is determined by the content, unlike the previous top-down schema designs. And it works - especially well on large media objects.
It's just getting started. A few changes will make it the global distributed computing system we've each been coming at like blind men at a seeming menagerie that's really just one elephant. Distributing the catalog, so any centralization is redundant. Ensuring that any bit is always replicated at least once. Implicit hyperlinks among data chunks for content-specified traversal of the infospace (like HTTP/HTML/URLs). Search engines full of metadata. Asynchronous, realtime streaming protocols layered atop the application - including multicasting.
Maybe it won't be "BitTorrent" that gets these revs - after them, it would hardly be recognizable as BT. But BT has gotten us across a major watershed, the way the CERN HTTPd v1.0 did in 1990. Like anything else that hundreds of millions of people are doing simultaneously, throughout the day and night, it's too late to stop.
--
make install -not war
if the bittorrent protocol would be updated to look like HTTP or FTP or something else, to make it impossible for ISP's to filter it.
If the price is lowered becuase it's distributed by bittorrent, of course I don't mind using it.
However, I can almost guarantee you that they wont lower prices because they're using bittorrent. That's what I have a problem with. This is a plus point for greedy distribution companies (talking about the big boys here), there is no plus point for us consumers, as the savings will probably not be passed down to us.
Have you actually tried downloading a DVD with bittorrent? As it now stands you could be spending days downloading something you paid for, and it will be eating your upstream. While bittorrent is great in that it empowers people to release whatever they want no matter how big, it's not so great when I'm paying to download something.
And yes, someone could have cracked the encryption, but they've cracked dvd anyway. Point is that this could be used by the common person who just wants to download a movie, and doesn't want to have to get out of his chair. Because heaven forbid we actually get out of our houses, or have to actually go *get* something. People pull back muscles doing that sort of stuff, right?
What is leeching in a BT network? (Yes, that is a real and sincere question.) I thought that with BT you had to upload the file you were downloading, so how can you leech? I'm not saying you can't, but I would love to hear what it means.
San Francisco Photographers
Do you think that you're not paying for the distribution costs of every CD/DVD/etc that you buy right now?
The idea is that if they can cut costs, the price of the product will drop (as evidenced by the $0.99 song sales -- $8-12 on an album is still cheaper than most CD's). If they could further cut costs, I'd be game...especially if I don't have to go to Best Buy to purchase a DVD. To take it a step further, most folks with broadband don't use their upstream bandwidth for anything but page requests and ACK's. Those people will likely never care -- it's like selling a commodity that is effortless to produce, and rarely (if ever) missed.
If you're a cynical anti-corporate kinda person, well...I'm not going to sell you on anything. The rich are eating your lunch and you'll probably never be convinced otherwise. If this is the case, go nuts & have fun. If not, think about what an album costs when you cut out the physical distribution media. Maybe BitTorrent isn't an exact match for this, but why can't the costs of online distribution be cut down even further? If it's not worth it to you, I'm sure that there will always be a distribution of some kind of physical media out there that you can buy. In the meantime, I'll gladly share a portion of my upstream bandwidth and save some cash.
-Turkey
That's the fundamental problem of BitTorrent: no incentive for seeds to stay. This isn't really a *problem* at all. Sure, it means that your BT download might not go as fast as if they seeds stuck around. However, in the worst case, BitTorrent speeds simply break down to FTP levels (i.e. everyone is getting their chunks directly from one central copy.) Anyway, it's their bandwidth, they don't have to seed if they don't want to. To review, worst case: as good as FTP best case: WAY better than FTP
Your upstream has no value to you when it idle.
Putting it to use to distribute content you like means the content you like can be distributed to you without the vendor having to bundle in the cost to you of building a distribution infrastructure that duplicates resources you are already paying for in the form of your idle upstream capacity.
The article mentions that media folks (riaa/mpaa) take down sites hosting the torrents. Assuming the seeds themselves are hosted in countries disinterested in riaa and mpaa takedown notices or spread out across many individuals, what stops people from using p2p networks like kazaa to distribute the torrent files. It seems that this would be the next logical step. I'm not condoning this, I'm just curious why it isn't happening... yet.
-- john
Right! That's exactly how it should work. Make it a sliding scale, even. Work it like this: Downloading the bits of two hours of content costs you, say, $6. Add that to my monthly bill. And then for every, say, hour of content I upload, credit my account $1. This credit will never be paid to me in cash, but can be applied to future downloads.
Then, in the software or set-top box or whatever, give me some options:
People who want to leech can do so freely, at $6/movie. If I rent a movie a month, I can set mine to trickle up to slowly pay for my movie without me noticing it. Or maybe I watch movies in the evenings, so I can crank up the upload and it'll be 'done' overnight. Or maybe I 'rent' a lot, so I give it a decent upload all the time, so I always have a healthy credit to rent movies at no effective charge to me.
One simple mechanism, many ways for customers to use it to fit their style. Am I the only one that thinks this would work awesome??
Doug
How are you going to bulletproof the 1s and 0s
There are many ways; it doesn't even need to be encrypted securely, just mangled reversibly in a way that the firewalls aren't smart enough to deal with.
A bandwidth-eating encryption scheme
I'm not sure where you get the idea that encryption is bandwidth-heavy. It is CPU heavy. Your secret-key ciphers have almost no bandwidth overhead except for having to round up to the nearest block (which are usually small - Blowfish, for example, is only 8 bytes), storage of the data length, and the initial transmission of the key. Public key encryption systems have a lot more overhead - often a dozen or two bytes per 256 byte block, or something to that effect, plus the block roundoff and data length - but you generally only use them to establish a secret key session.
that will just be broken
Few encryption schemes are ever "broken". Nowadays, they're generally only "weakened" - the search space becomes a lot smaller. If you use excess bits, even a weakening will generally not put your data at risk. And most algorithms have never been dented, despite a strong mathematical interest in breaking them.
"Here's a fun fact: the moon has turned to blood!" -- Newscaster, "Jesus Christ Supercop"
Are you sure about this? My problem with bittorrent is for every file im downloading I end up uploading two or three times as much. A few weeks ago I missed a certain race on tv, it ended up taking me 2 days to d/l the 700 MB file, in that time I had uploaded 4 GB. Because of this I think there is alot of real leaching going on where client download but don't upload.
Yeah... most net protocols don't do much of anything to grant you anonymity. That's why I've been working on Uso (site still under development, just like the project). While I hate to put a date on when the first version will be out to the public (because I was so wrong the last time I estimated), I'm especting to start debugging my last mess of changes later this week, leading to a release around the weekend after next. Check the News/Updates section on the site for updates.
There's a moderately thorough whitepaper on one of the protocols implemented at the site. The other hasn't had its internals completely fleshed out yet, so there's no whitepaper yet.
To sum up, the first method is for bidirectional traffic; both machines send with source addresses using the maximal degree of lying about their sources that their routers will allow. For destination addresses, they have as much lying about where on the target subnet the target machine is as possible. Recipients of packets ignore the source and destination addresses, and instead use user-unique codes in the payload to look up where to send responses to. If the machine is behind a switch instead of a hub, it can attempt to use arp flooding or other techniques to get packets sent to other machines. Regardless of the configuration, the sender doesn't know how much lying the recipient is doing about where it is located.
The second protocol, which doesn't have a whitepaper yet, is unidirectional. Transfer requests and acks are proxied in bulk via an uninterested third party. The uninterested party never knows what the packets that it is sending are for, and while the transfer-requesting machine has to give out an address on its local subnet to receive what it asks for, the sender can fib about where it is to the extent that the routers that it passes through will allow. Since the proxying involved is only acks in bulk and transfer requests, proxy loads are kept to a minimum.
Both protocols use encryption, with a SSL-style three way handshake to establish a session key (public key=RSA, secret key=Blowfish). The unencrypted portions of the packets are designed to have no general distinguishing characteristics, to help prevent filtering; instead, each client scans for its unique code in the destination code field. Both protocols are designed to allow communication through NAT by determining their information through a postconfiguration stage with an outside "friend" client.
The project is in C++, uses libnet, libpcap, and openssl to do the dirty work, and is designed to be as cross-platform as possible (although I don't plan to port to Windows until I've at least got an alpha version released for Linux).
"Here's a fun fact: the moon has turned to blood!" -- Newscaster, "Jesus Christ Supercop"
I'll break protocol and tell you this. (Since it is automatic these days, and the requirements have changed)
.6, I would let you slide. .6 to .4 was a warning, and flagged for review in 5 days. .4 and below... it depended on your usage. If you had over 10GB down, you should know better, and I would ban your ass in a heartbeat.
Back when it was manual, I would go through accounts that others had described as being leechers. Sometimes 200 to 500 a night.
Now, different admins have different scales they use, and I liked it that way. I tended just to tackle the ones that were obvious offenders, and I would usually ban 100 or 200 leeching asshats per night.
If you were above
The excuses are all the same...
But.. but... I've got shitty upload speeds...
But.. but... I didn't know I had to leave it open that long...
But.. but... My roommates were using all of our connection...
But.. but... I *was* uploading, but no one would connect...
But.. but... My ISP restricts my upload speed...
But.. but... I could not leave torrents open because of my gf...
But.. but... I have dsl and my IP address keeps changing...
But.. but... I was sharing on other sites too...
But.. but... I didn't know my client would automatically stop my upload when my d/l was done...
Screw em!
I have a shitty asynch connection, and I never dropped below a 4.* because I took the time to learn about the fundamentals before I started using torrents at all. I got Azureus, (I like BitLord better now) and set it up to run up to 2.0 on every torrent, and continue seeding if there were less than 3 other seeds. If not, drop it there.
I also capped my upload speed at 60% of what my connection could handle, (after trial and error) so that it did not make my connection almost unusable.
I think that is the biggest reason people do not seed after the dl is complete. They have their upload set *too* high, and suffer because of it, therefore they do not want to seed. If they took 10 minutes to find out what speed they could continue to comfortable share at, it would not affect them in that way.
Too many people came in with the Napster and Kazaa attitude of "get what I want, and go" not realizing that torrent was a completely different system. They either learn fast, (and I will teach them how to use it properly, if they want to learn) or they get banned.
I'm not all that active on torrent sites anymore, Big E included, because life has been really hectic, I'm learning new skills (goldsmithing and vet tech) and with the new system, not so much maintinence is required.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order- Ed Howdershelt Via Tass