Slashdot Mirror
Is BitTorrent Search Harmful?
protee writes "p2pnet published a report arguing that the robustness of BitTorrent to free-riding might have been more related to the lack of meta-data search rather than to its tit-for-tat-like strategy. The question now is: how the release of such search engines is going to impact the BitTorrent network?"
Such networks thrive because individuals can find the content they want. Searches will help improve that much as has happened with the World Wide Web. Remember, it didn't become explosively popular until the early search engines like Yahoo!, Altavista and Magellan came about.
Cyric Zndovzny at your service.
That Braham Cohen is so dumb, he probably never considered anything like this when he put together his own official bit torrent search engine. What does he know, sitting there coding in his mommy's basement when all the real geniuses are on Slashdot and p2pnet!
I mean, c'mon... *eyeroll*
...when they can't be traced. Up the encryption and IPsec and you'll find that people will start to share.
The question now is: how the release of such search engines is going to impact the BitTorrent network?
The answer: not at all. There isn't a BitTorrent network, just an application that has caused many thousands of disjoint, single purpose networks to come into existance.
And that disjointness will help protect them, I feel.
If the number free loaders gets too great, nobody will be able to get fast downloads off of BT due to lack of seeds (or whatever they're called). Once that happens, popularity amongst freeloaders declines, service returns to normal. A file sharing system without anybody seeding any files is a waste of time.
Difference: The early Web flourished in .edu circles, where there are likely to be a lot of people dedicated to providing educational works of authorship on fat pipes. BitTorrent, on the other hand, is often blocked by .edu ISPs, and residential customers of commercial ISPs don't have nearly the fat pipes to supply everyone who wants to download a given file.
The main strength of BitTorrent is that it works on individual files. It is not a network, rather a protocol like ftp or http. Ftp sites that offer copyrighted content can be taken down, but the ftp protocol is alive and well.
See charts for twitter trends on Trendistic
The thesis is basically that by causing your client to change identity frequently, you can take advantage of the leniency that BT allows newcomers to the network, and thus "leech" without punishment. This isn't done because you'll get kicked out of the communities that publish BT metadata if you do it.
I don't see it. If you're going to leech, that's the way to do it, but cooperating overall results in even better upload rates; you're not fighting for the few slots afforded newcomers, you will be given as many packets as you can eat as fast as you can eat them so long as you reciprocate. And I'm sure those communities will survive - I suspect that Bram will have thought of how to integrate search with community.
Xenu loves you!
Seeing as how the **IA and its international counterparts have been successful in shutting down the tracker sites and this will help them locate these sites, don't you think the impact will be a move to only legal files being indexed for the search. This could actually lead to a vindication of p2p as a useful piece of software and a decline in the number of sites specializing in illegal copyrighted downloads.
It could be compared to bootlegs being move from inside the music/video/etc. store to the street merchants that have to pick up and move everytime the cop walks near them.
I've never seen a P2P network without leechers. Even those which include an economics system like edonkey still have their share. I don't think there's anything fundamentally different about bittorrent. Now it's pretty much an ordinary P2P net leechers will appear. The economics will help limit their impact though.
I am trolling
<MPAA_Exec> Does a bear shit in the forest?!?!
She's built like a steak house, but she handles like a bistro....
(Note: I'm Swedish, just like the ones running and hosting Piratebay)
So what happens when pirate bay gets busted by the RIAA-imperial navy?
It's not clear whether they are breaking any Swedish laws - that's why they're so smug and play around with all the takedown notices. The only law they _might_ break would be something like "large scale contributing to copyright infringement" but even that's a stretch. There's a reason why they haven't been charged with anything yet, even though the Swedish Anti Piracy Beaureu are all over the piracy sites they know they can bring down in court.
it's in my head
As piratebay has pointed out before, sweden is not a state in the USA.
Sweden is a small country in the north of europe.
RIAA-imperial navy can stay the fsck out of sweden, thank you very much.
More leachers != merrier
More peers = merrier
I think the main problem with some bt clients is that they flood your upload bandwidth... thus killing your DL speed.
A client that intelligently detects/limits/manages ULs is probably the best thing that can happen to bittorrent
As long as uploading is a transparent process that doesn't interfere with n00bs general internet usage, they won't bother to become leechers.
[Fuck Beta]
o0t!
Search = More leechers = More seeders = More health. That means less dead torrents. It's that simple.
~Ilyanep
To get message, take amount of carrier pigeons at each stage mod 2. Then decode binary.
Bittorrent Search could be slower than normal bittorrent usage, if these techniques are used (though I personally find that my download speeds are abysmally slow until I have enough segments to upload too, this "new user window" the report talks about could be a figment of the author's imagination) .torrents for from a website that's trying to provide files for people.
But this will not effect Bittorrent Itself. Bittorrent remains useful for legitimate downloads- of the type that people will be downloading the
Bittorrent may not become more useful because of searching, but it wont become less useful.
-- 'The' Lord and Master Bitman On High, Master Of All
Yes, but the real question is what causes that lack of "altruism" in the first place. My assumption (which could be wrong) is fear of retribution by some media conglomerate. Those with transfer caps would also be candidates for leechhood too, I suppose. However, when nearly untraceable P2P technology becomes the rule, when the Fear of God(tm) and/or somebody's lawyers is no longer a significant issue, I would expect altruism levels to shoot up. Remember the original Napster: everybody pretty much shared their entire collections. It wasn't until the RIAA started slinging lawsuits around that people even thought about anonymity (I think most people assumed they were anonymous.) Well, now they are thinking about it, but so are a whole lot of developers. From the RIAA perspective, I think this is going to backfire bigtime. I mean, they had to know this was going to happen.
The higher the technology, the sharper that two-edged sword.
Afraid not. Bit Torrent depends upon a central "tracker" computer to keep track of who has what pieces of the torrent so that clients know where to go to get said pieces. Remember that Bit Torrent was not designed as an anonymous protocol and was only intended to efficiently distribute large files, such as Linux distros. Consequently, it's almost as easy to monitor who is downloading what as it was for the original Napster service. If you download using Bit Torrent, your IP address is wide open to any media company investigator that wants to query the tracker. And they do, so do it at your own risk. Some clients, like Azureus, offer extensions to the Bit Torrent protocol to support anonymous tracking but I don't know how widespread it is (or how well it works in practice.) But I'd say there's not much question that this is a technological issue that will be effectively addressed in the not-to-distant future, either by Cohen himself or some other equally bright developer.
The higher the technology, the sharper that two-edged sword.
The thesis of the research appears to be that, (1) if they can get away with it, some programmers will write implementations of the bittorrent protocal that are designed to "cheat" in such a way that they can have a higher ratio of downloading to uploading than they can currently get away with, and (2) it is the multiple swarms created by a lack of a central search engine that stops this.
The research is very unsatisfying to me for several reasons. First, its not even necessary to "cheat". On every bittorrent I've ever downloaded, my download has completed *way* before my ratio has reached 1:1, and it is only because I choose not to end the session that I continue seeding (or, more often than not, because I'm asleep, so the choice to continue seeding is made for me).
Second, the example they give of a strategy that beats tit-for-tat is one in which several cooperating strategies are used at the same time, with some taking on a "master" roll and some taking on a "slave" roll. This may make their point on some academic level, but as a realistic example is fails utterly. Who in their right mind would start ten different bittorrent sessions, with some acting as slaves and some acting as masters? The overall download speed would be awful from having multiple sessiosn running over the same wire. Its just stupid. At least come up with a better example of a strategy that can best tit-for-tat.
Third, I don't see evidence that people would use a bittorrent program that was designed to cheat. Maybe they would, maybe they wouldn't... the article assumes people would. My bet is that not enough people would use such a program that it would make a difference. Its not like this is evolution, where the successful cheaters "pass on their genes" to create more cheaters.
Overall, I think the research is a lot of academic mumbo-jumbo that may sound good on paper, but has very little, if any, connection to reality.
My own simpler thesis would be this: bittorrent works so well because a lot of the downloaders fall asleep and end up seeding longer than they otherwise might.
done. http://www.peercast.org/
Brain(s): 0.0% user, 1.3% system, 0.1% nice, 98.6% idle
I've never been in a torrent where there was'nt at least SOME aggregate downstream bandwidth in and above what the uploading clients were taking. But, when you first join a torrent the protocol is written so that seeders that are done downloading are three times as likely to choose new torrent clients with little or no data over clients that already have some data and are uploading. I should say at this point I really don't know EXACTLY how this works. Except that it seems that there is a percentage involved of how much a seeder gives a leecher before it stopps giving anymore (unless th e leecher starts uploading)
---from the article--
Once a client has obtained a list of other peers, it will contact them to try to fetch the data it is looking for. In BitTorrent, file contents is split into small-sized pieces and each client maintains the list of the pieces it holds. After a handshake, peers exchange their piece lists so that each of them may determine whether the other has some pieces they are interested in obtaining.
The bandwidth being a limited resource, a single client cannot serve every peer interested in pieces it holds at the same time. The maximum number of peers served concurrently (i.e. the number of available slots) is configurable by the user. All other peers connected to a client (whether they are interested or not) which are not being served are said to choked. In consequence, each client implements an algorithm to choose which peers to choke and un-choke among those connected to him over time. The strategy proposed by BitTorrent is named "tit-for-tat", meaning that a client will preferably cooperate with the peers cooperating with him. Practically, this means that each client measures how fast it can download from each peer and, in turn, will serve those from whom it has the better download rates. This strategy is implemented for all but one slot which is attributed to an interested client, regardless of its upload rate. This so-called "optimistic unchoking" allows for the discovery of better peers than those currently selected (i.e. those with higher upload rates). This strategy, however, if implemented strictly, would considerably slow down the insertion of newcomers into a running swarm as, they obviously do not have anything to share at the beginning. Thus, clients that have nothing to share are given three time more chances to be selected by the optimistic unchoke. When a client has finished downloading a file it no longer has a download rate from other peers but it can still share (upload) pieces of the file. In this case the choking algorithm is applied by considering upload rate instead. Peers are selected based on how fast they can be uploaded to. This spreads the file faster. Such "seeder" peers that store the whole file are very important to the functioning of a swarm. If a swarm contains no seeders it may lead to a situation in which pieces of the file are missing from the swarm as a whole. In this sense the system requires at least some level of altruistic behaviour from "seeders".
Worse case senario for unintelligent clients of any type on windows - use NetLimiter (I'm using 1.3, it's great). You can set upload/download limits per application, and schedule changes (say you want to upload max when you're asleep or something).
Opera, Proxomitron-Grypen,GPG 0x0A1C6EE3
a problem fixed by the very behaviour of each serious user who downloads then lets the file on his disk (seeding it) 'till it reaches at last a few days there or a good (> 1) share ratio.
True, but as a file transfer system becomes easier for novices to use, it is likely to draw users who aren't "serious", who cancel the upload as soon as the download completes. And if you try to enforce share ratios on a registered tracker, remember that the mean share ratio across all users is exactly 1.0; therefore not everybody can have a cumulative ratio >= 1.0. What happens when demand falls off for a file, and though you leave the upload going, nobody downloads more than a couple megabytes for days?
If it matters, my personal rule when downloading something on BitTorrent or eMule is to let the upload go 24 hours after the download completes, then let it get up to at least ratio>=1.5 or one week, whichever comes first.