Become a fan of Slashdot on Facebook


Forgot your password?
The Internet

The Tragedy of the Digital Commons 197

Frog writes "The New York Times (reg. req'd) writes about a study by the Xerox PARC Internet Ecologies Area which shows that only a small percentage of Gnutella users actually share files -- the rest just take 'em. The researchers note it's 'hard to generate spontaneous cooperation in large anonymous groups.' As a consequence, the system has degraded performance, and is more vulnerable to censorship or legal action. Maybe the solution is to implement a market system for resource allocation, but how to prevent cheating?" Reminds me of the BBS days of file ratios - 'course then we'd just take an image, resize and upload it, so that idea didn't exactly work as intended.
This discussion has been archived. No new comments can be posted.

The Tragedy of the Digital Commons

Comments Filter:
  • Hmm, wonder what kind of images you're talking about :)
  • by Zurk ( 37028 ) <zurktech AT gmail DOT com> on Monday August 21, 2000 @09:16AM (#839794) Journal
    filesharing shouldnt be forced and anyway, gnutella hosts 20+ TB of files right now...yep thats terabytes. just like the old bbs days more and more people will share files if they cant find what theyre looking for and the system will keep growing. its no biggie - gnutella can be censored simply by injecting garbage into the system which is what flatplanet tried to do. freenet on the other hand...
  • by cs668 ( 89484 ) <> on Monday August 21, 2000 @09:17AM (#839795)
    I think that faster net access would make it easier to share data instead of just downloading it.

    My connection is slow enough that I can not share anything, but I would if I had DSL or Cablemodem access.
  • Is this not just an example of the sort of freedom these type of networks allow? Why should anyone be forced to share? There have always been leeches and lurkers, and I don't see any reason why there shouldn't be. For example, I have my max uploads set to zero on my napster client at work (to avoid potention problems). I share at home because it is a different situation. The people who are sharing are doing so because they want to. Just as with the early days of the web, or even back to BBSs -- Why spend the time doing a website, or the money and time it takes to run a BBS when there is nothing in it for you? Apparently there is something in it for those willing to share, even if it isn't tangible. Bottom line: these types of services are about freedom.


  • by Dman33 ( 110217 ) on Monday August 21, 2000 @09:20AM (#839797)
    C'mon people! Do you really think that this is any shock? I could have told you that people would rather receive than give. Especially when it is anonymous.

    I think a ratio type of thing would be a great idea, but how in the world can this be done? Obviously this is not too practicle in an anonymous situation; so is this the great paradox of anonymous vs quality?

    In other words, if it is purely anonymous as many would insist it to be, then it will lack quality and usability by nature. So in order to have good quality, you have to sacrifice security?

    Can anyone think of a better way?
  • Even with Napster this can happen. I should know, I do it too. With my connection I can't have a lot of people accessing my computer. It slows the bandwidth way down. Most people just don't know how to turn off sharing on both Gnutella and Napster. Most just don't care, except for those who actually read Slashdot.

    Even the samurai
    have teddy bears,
    and even the teddy bears

  • One of the reasons at least I don't share my stuff is because I am scared the feds will come after me ! You know downloading is less illegal than actually distributing - difference between buying drugs and selling them. Also, some people use computers at office to do all this stuff - definitely don't want to get into legal trouble there !
  • Make the download and available upload path one in the same, so that anything downloaded is available to others.

    It would help to only show files that are _completely_ downloaded (I do not know how this would be done efficiently).

    Some people (especially those with thin bandwidth) tend to become selfish but this would force some level of sharing without too much restriction (although it is not a perfect solution).
  • ...Is that, by default, whenever you download something, it's shared (at least, as far as I recall). So that if users just don't move it out of the download directory, they are sharing it... This is how files spread through the network. If gnutella doesn't do this (and I don't think it does), then that will of course lower the amount of users sharing, because the casual user will not go out of their way to enable it.

    Josh Sisk
  • by don_carnage ( 145494 ) on Monday August 21, 2000 @09:22AM (#839802) Homepage
    "As a consequence, the system has degraded performance, and is more vulnerable to censorship or legal action."

    The article notes that if someone (such as the RIAA) were to sue, they wouldn't go after the individual users, but rather the big fishes.

    However, if Napster were to be shut down for good, it would inject a surge of "survival instinct" into the whole operation and everyone would start sharing files -- If there's something to rebel against, people will rebel against it.


  • it isn't buying vs. using it is how much you have on you. Your analogy is sooo wrong. -davidu
  • It depends upon what's being "shared". When I download the latest Linux kernel, there are a lot of files in there. I may only contribute changes to a few of those, which makes my download/upload ratio pretty awful. Particularly if you count only my revisions which are accepted by Linus.

    I'd expect there to be more downloads than uploads, unless the type of content is something which everyone can produce easily -- and how useful would such content tend to be?

  • by ibot ( 219510 ) on Monday August 21, 2000 @09:23AM (#839805) Homepage
    Because of the fact that more people have lower speed connections (mostly dial up), they are likely to stay online only as long as it takes to download a large file. Anyway who'd be interested in downloading from a source that can channel only a few kilobytes a minute.

    Founder's Camp []

  • by Apuleius ( 6901 ) on Monday August 21, 2000 @09:23AM (#839806) Journal
    Mojonation [] seems to address this problem. From their site:
    Micropayments and Scalability Mojo Nation compensates users who provide the resources, content, and indexing services. Effectively preventing cheating, denial of service, and freeloading, Mojo Nation fosters an information market for all types of content. This is accomplished through a micropayment system which denominates the internal tokens, called Mojo, in the same resources needed to provide the services: disk space, bandwidth, and CPU cycles. In time you will be able to buy and sell these tokens, turning Mojo your earn into real dollars.
    Of course, I've yet to use Napster or Gnutella, so who knows. But this thing gives you incentive to give as well as take.
  • You can achieve 100% proof of identity without needing to know who anyone is.

    Of course I'd need to sit down and think for a while to work out how to apply a cryptographic system of checks and balances to something like Gnutilla. It's an interesting problem.

  • Exactly.. There isn't enough bandwidth on a 33.6 modem to make just leeching of Gnutella terribly appealing, let alone when half of your bandwidth is sucked upstream.

    If we all had T's, we'd all share. Until then...
  • First Karma inexplicably disappears. Then it starts moving again--but only down. Then we have this same article posted in the morning but marked "(- duplicate)" and removed. Then it's posted again. WTF is going on in there? Why don't we get updates on Slashstatus anymore? Anyway, here's what I already posted on this topic:

    I agree somewhat with their conclusion, in general terms. It IS hard to get a large anonymous group to share.

    BUT, their experiment is highly flawed:
    1. not everyone has bandwidth enough to allow a million d/ls
    2. not everyone has the CD collection to share out
    3. not everyone has the storage capacity to host all the files
    4. 24 hours is too short a time to test--there are no "major events" in there. What if a new Eminem CD had been released that day, I bet we'd see a lot more people offering that one item.
    5. All it tracks is sharing via those two channels. What about other software, including ftp, http and sneakernet?
    6. (added just now) This is not a "tragedy of the commons" situation. A TotC happens when there is some finite resource that everyone exploits "ruining it for everyone". Gnutella and Napster need not ever reach that point since the resource can grow arbitrarily.
  • by Megane ( 129182 ) on Monday August 21, 2000 @09:23AM (#839810) Homepage
    Reminds me of the BBS days of file ratios

    The reason it reminds you of this is because it is exactly the same thing. The difference is that in the bad old days, the average BBS would only have one phone line, and the leeches could tie it up 30-60 minutes each. Now there are bandwidth leeches, sucking up the limited bandwidth of cable TV modems, and are the reason for upload bandwidth caps. The bandwidth leeches are a large fraction of the top 20%, and the traditional leeches are the bottom 70%.

    'course then we'd just take an image, resize and upload it, so that idea didn't exactly work as intended.

    The really high tech leeches of the day came up with a slick trick: "Leech Zmodem". When it received the last block of a file, it would NAK it, asking to restart the transmission near the beginning of the file, then cancel the download. Most BBS Zmodem implementations would only debit the leech with the point at which the download was canceled, if it debited their download quota at all!
  • The somewhat mandatory sharing of files with Napster is a good thing IMO. While you can hide them, or be a jerk and put your bandwidth down at 28.8 to discourage sharing, eventually you learn to love sharing. And isn't that what sharing is really about? Love? Hmm, I really wish I could some sort of love-sharing device IRL. Would really enlarge my self-esteem (among other things).

    You are more than the sum of what you consume.
  • What CmdrTaco said about file ratios is pretty accurate. I ran a BBS for several years and found this to be true.

    It was also true for Post/Call ratios when implemented on some boards (ie, not letting the user play LORD until he has reached a Post/call ratio of 1, or whatever). This basically created a message base full of crappy 1-line posts (also true on some web-based forums where users with X posts are rewarded with a special 'ranking').
  • Actually there is something like that out there that I initially dismissed as it could/will involve some money but it looks rather intresting.

    Its called mojonation ( and it works on a ratio system with credits (reffered to as mojo) and you spend a little mojo to do a search and a little more to download the file... you can buy mojo or make mojo by hosting files yourself.
  • by Megane ( 129182 ) on Monday August 21, 2000 @09:26AM (#839814) Homepage
    This is accomplished through a micropayment system which denominates the internal tokens, called Mojo, in the same resources needed to provide the services

    Sounds a lot like the Slashdot karma system. Do they have Mojo whores there?
  • by don_carnage ( 145494 ) on Monday August 21, 2000 @09:26AM (#839815) Homepage
    Can anyone think of a better way?

    I think CS668 (further down in the post) has a pretty good sense of what's happening: There are still a whole lot of people out there that are still on 56K connections. Cause and effect: people don't want to waste their time with a 4k/second host when they can download 4 files from a DSL or Cable host at 20k/sec each instead.


  • When you install Napster, it automatically creates an index of every MP3 on your system, and starts sharing from the second you connect to the server. It also doesn't exit if you merely click the Windows "X" button; that only minimizes it to the tray (!). This keeps Napster going on high-speed persistent connections, like those at universities where it's very popular.

    Gnutella is too nice. It doesn't automatically share every media file on your system unless you tell it to specifically, and it exits when you say so. Most people don't mess around with configuration settings. Napster's defaults benefit Napster. Gnutella's defaults benefit the user who wants control. Guess who wins.
  • because they're afraid.

    Getting busted for possession isn't nearly as bad as it is for distribution.
  • Gnutella and Gnotella (even with the skins) still have kind of crabby user interfaces, whereas Napster beta 7 has a funky tree thing to say what you're sharing. I'm sure that's partly to blame.

    Another problem is that ''56K'' modems have a maximum 33.6kbps user-to-ISP bandwidth, it does somewhat hurt your downloading prowess to have people sucking up all your ACK-ing bandwidth.

    People with the broadband connections don't seem to mind sharing, so I guess this will fix itself.

    Also, around the time of the hearing for Napster when things looked bleak, Gnutella fell to its knees; it was very hard to get a reasonable connection without listing a dozen peers. There's something wrong with that there scaleability!

  • Napster does share anything you have downloaded, but also shares stuff you are in the process of downloading, which isn't very useful...
  • by sandler ( 9145 ) on Monday August 21, 2000 @09:28AM (#839820) Homepage
    I noticed that sharing seems to be a late "feature" in many clients. For example, the last version of Knapster I used and (I believe) the last version of gtk-gnutella I used both did not yet support uploading. In fact, the clients may never support it, and people may not want to upgrade to a new version that does. This is always going to be the least demanded feature, and since these types of services lend themselves to a plethora of Yet Another (tm) clients, this could be a significant problem.
  • I'm guilty of sequestering my files to keep them away from others when I'm using Gnutella, but there's a reason. As soon as some idiot finds, for example, my complete collection of Barry Manilow songs (not really), he starts downloading the entire collection all at once. My cable modem uplink is capped at 256k, and he's just saturated it. If people weren't such hogs, I'd be more willing to make stuff available. Maybe I'll relent when I see a Gnutella client that can be configured to limit the number of uploads a single user can have running at one time. One can do this with Napster, so I have no problem there.
  • When I see someone on Napster with a LOT of files, and claiming a slow connection, I try it out. See what kind of speed I get. Then, if it turns out that their claimed connection speed COULD NOT be for real, I suck them dry. Punish the liars.

    Regardless, I can understand someone with a slow connection not sharing at all, or severely limiting it. But if you've got a fat pipe, share it! (since you're probably leeching more than everyone else, too)
  • The third party software is to blame. Sadly enough many clients don't even support file sharing, and almost all don't default to filesharing. This is because third party developers aren't regulated and encouraged to share. If I recall correctly the napster corporation (company?) has certain restrictions on napster clones, that they try to enforce.

    Joseph Elwell.
  • by tylerh ( 137246 ) on Monday August 21, 2000 @09:30AM (#839824)
    Programmers know this as the "90/10" rule: 90% of the useful work will be done by 10% of the code

    This applies to people as well. I helped organize keggers in college. A small number of us did the organizing/financing/clean up -- everyone else just showed up and partied. That was kind of the whole point. So this "ecology" result is NOT a tragedy of the commons, it' just a another keg party - and you know how hard those are to stop 8)

  • The tragedy of the commons was inevitable in these cases, and solving the problem is what Mojo Nation was designed for. It uses a micropayments system which is denominated in digital resources (disk space, bandwidth, CPU time) and exchanged for services. This is just a micropayment system that is backed by the old upload/download credits of the BBS days.

    Check it out at
  • Don't dismiss file ratios as a legitimate means of submission encouragement. Yes, the system can be abused, but after years of BBS hopping I've seen excellent methods to prevent it.

    My favorite method involved an account system, whereby uploads that other members voted as 'spam' or just worthless space were counted against that user. Upload enough crap and you're booted; add a two-day account activation delay and even the most 'disciplined' media hog is curtailed.

    Of course, a ratio of both file size and number of files is necessary, else you'll get people uploading one worthless 1GB file every few days, and then going to town before they're booted.

    Now, these systems worked fine with one server, but systems like Napster and Gnutella present a new paradigm. How to enforce ratios when there is really no such thing as an 'upload'? A few ideas, based more on the general concept of user-controlled servers than any specific service's setup:

    * For every file downloaded off a user's server, give that user a point. Servers can set a score limit on their machine for downloads (i.e. 'only people with a score of 100 can download from me'). Drop by a point or two for each download, but give 100 points or so when you sign up.
    * Same system as above, but base on file size instead of number of downloads. Or perhaps a combination of the two?
    * Keep track of everything on some centralized computer, and enforce the ratio much like it would be in days of lore.
    * Enforce a per-server ratio, the setup of which is determined by the owner. Not a great system on this scale, but worth a look.
    * Rather than 'earning' downloads, penalize users who download too much too fast with a negative score. Doesn't take into account whether or not they serve, though...

    And, of course, what will probably happen in the end anyways:
    * Keep the system as it is now. As the bandwith dissappears, all the hosers who don't know how to share will leave, and all the savvy slashdot readers and linux lovers will get their connection speeds back.

    Any other ideas for control systems, or arguments against them?
  • A TotC happens when there is some finite resource that everyone exploits

    Actually, it's just a strawman used to attack the ideals of Libertarianism by people with no understanding of it. In the real world, such a situation almost never arises.

    -- Floyd
  • As if faster access would change your ways.

    Do you really expect anyone to believe that besides yourself.

    Like those people who say "When I make more money, I'll give to charity." They don't. Come raise time, they just spend more. Giving comes from the heart, not the size of paycheck, or modem speed.
  • How many people leave their client/server running when they aren't using it? How about a new option to automatically serve files when they're not downloading/searching for something? If someone is already downloading something from you, their connection gets slowed down by X until your done, and no new connections to your server can be made.
  • In almost every community, there are a minority of 'producers' and a majority of 'consumers'. A handful of people write novels;millions read them. A handful of people compose and play music;millions listen. Even when economics is no object, this ratio holds. How many people surf the web, versus how many people have their own websites with any content on them? How many people take from charities versus give to charities? Hell, how many people bother to send 20 bucks to PBS, versus how many use Sesame Street as a babysitter?

    The perfect world of everyone sharing equally will never occur. If your own ability to contribute is hampered by "He's not pitching in!", then, that's your problem. Deal with it.

  • And this is what I had posted as a reply earlier...
    Not everyone has bandwidth enough to allow a million d/ls
    That and the fact you can't throttle the number of downloads or bandwidth being used. If you're on dialup it's rather frustrating to find 10 people trying to download over your 56K connection. If I could limit it to 1 person at a time, it might be less frustrating. As for sharing, it might be a more valid experiment to check out IRC, which is where Napster seems to have it's roots.
  • I noticed that the study did not mention anything about the connection speed of the various peers. I wonder if those with narrowband (i.e. dialup) access might be less likely to share files than those with broadband access.

    After all, one has to wonder: would Napster have gained so much popularity were it not for the high-speed connections enjoyed by many college students these days? Without so much broadband in universities, who would want their uplink saturated by strangers? On the other hand, when you have a T1 in your dorm room, you might not even notice a few people grabbing files from you at any given time.

    So, I would be interested to see how much of a correlation there is between one's willingness to share files and one's bandwidth to do so.

    Dave Bailey

  • Actually, you can disable file-sharing by default, e.g. if you are behind a proxy. Or if you just don't want people on your harddrive. You are still browsable though, as I get people trying to download stuff, but they get connect errors (I am behind a firewall).


  • I think that many Napster and Gnutella users are afraid of sharing large numbers of files because of the MPAA's recent lawsuit against Napster. Should the RIAA actually lose (an event that is not very likely), they will probably search for "wholesale pirates," those with large numbers of shares. Consequently, people are afraid to share their files.

    Even if the RIAA is unsuccessful in court, they will have been successful in turning potential members of the file-sharing community into leeches or non-users.
  • by finkployd ( 12902 ) on Monday August 21, 2000 @09:32AM (#839835) Homepage
    I guess the obvious conclusion is that people will not give away, only take from free sources. But then how do you explain napster? Even though I routinly hear that the quality of music is garbage ("it's all brittany spears and crap like that") I've found a wider range of music there than any other place I've ever been. From The Why Store to Bach and plenty of it.

    I think it's a little early to declare gnutella in trouble, just wait till napster gets shut down, then watch the flood migrate over there.

  • It is by default in Napster. Most users therefore already have that setup. My fserve is set up the same way.
  • Perhaps an important factor is a Gnutella user's view of copyright infringement, since quite a bit of the material on Gnutella is copyrighted.

    So some questions might be: Is it a greater "sin" in the eyes of most people to download copyrighted material, or to make your own such material available? Who would the RIAA or MPAA choose to go after? When you click on a file, don't you find out the IP address of the source so you can connect there?

    Just a few things to think about...

  • Doesn't a slow net connection make for hard downloading also? I have a very slow net connection, and I beg and plead with friends to burn me CDs of gnome, linux, etc because I can't download the stuff in a reasonable amount of time.

    If your connection is so slow you can't upload, why can you download?
    then it comes to be that the soothing light at the end of your tunnel is just a freight train coming your way
  • I use gtk_gnutella to fufill my trance mp3 cravings. But I can't share anything because it isn't supported! Maybe that's why some people don't share - they can't.
  • I guess you share files on a slow connection? Or you have a fast connection?

    If you are not stuck in modem land you should not be so quick to judge.
  • On Napster [sorry, my only filesharing experience really... GNUtella sounds nice, but I seek mp3's exclusively just now{if it's more than 10 megs, I am NOT trying to suck it down over a 56k}] I share freely my collection... however, I make no promises of sticking around. I have one phone line [can't afford more], I can't tie it up forever. If someone can leech from me during the time I am seking my 4 or 5 songs, fine. If not, well, they should have looked for a cable or DSL user [after all, Napster plainly tells you what speed line a person has]. When I have DSL[i.e. when I can pay for it] I'll leave napster [by then most likely an OpenNap version] open all the time... since there will be no need to free the line for voice calls.

  • Exactly... most files are very hard to get at in gnutella now. I share files, and I've had to set upload amounts to one file at a time to save a little bandwidth for myself... and that slot is always busy. The net is full with leeches.

    As for your note on leeches, it's funny how the gnutella-advocating-slashdot editors (in this case Hemos) openly agrees that he goes to some lengths to defeat the sharing idea.


  • by Kiwi ( 5214 ) on Monday August 21, 2000 @09:38AM (#839843) Homepage Journal

    Anonimity definitely has a dark side.

    A few years ago, Time magazine did an excellent piece on the problems to today's society. One of the things they pointed out is that the privacy of a modern household has greatly increased the incidents of child abuse. In the society that we evolved in, one large factor that stopped people from abusing their child was the fact that there was no privacy--if you abuse your child, the whole village knew about it.

    The anonimity of the internet causes similar problems.

    Any system administrator knows that if they put any pornographic images on their web server, their machine or their machine's connection will quickly get overloaded. For example, one of my users put up pictures of attractive women. The women were not even naked, yet the server's connection was still overloaded.

    I have heard it said that the most common term asked for in the leading search engines is "pornography". People who would normally be too embarassed to go in to a liquor store or a peep show have no problem getting porno on the net. The internet makes people do what they would not normally do.

    While pornography is somewhat harmless, other activity on the internet isn't. The actions of the anonymous person who brought down Kiro5hin [] come to mind. As does the random bannings on many IRC channels (where the operators as often as not broke in to accounts or engaged in credit card fraud to get a system they could run a bot on to control the channel), the efforts people go to to cheat in online games, countless breakin attempts any experienced system administrator sees in their logs, the nonstop tide of spam, and so on. All of these are things that poeple do when they do not get a chance to look in the eyes of the person who they are harming with their selfish actions.

    It does not surprise me that the internet is full of people who take but do not give back. Human nature has always had the takers who complain when the stuff they are not taking is not good enough for their selfish purposes, and the givers who get little in return for their giving except complaints from the takers. The anonimity of the internet makes this problem worse.

    Anyway, that is my rant of the day. Time to go back to coding my current open-source project [].

    - Sam

  • by Anonymous Coward
    It's rather difficult to host files for Gnutella (or anything else) when so many ISPs prohibit users from operating servers of any sort whatsoever. I'd love to have an ISP with reasonable speeds, reasonable cost, and which doesn't try to cover up it's own idiocy and lack of knowledge by prohibiting anything it doesn't understand or doesn't wish to be bothered with...
  • When you install Napster, it automatically creates an index of every MP3 on your system

    at least the version i installed, asked me if i wanted it to search and find my MP3 catalog. it was not automatic in the sense that i did not have a choice.

  • by Jim McCoy ( 3961 ) on Monday August 21, 2000 @09:41AM (#839846) Homepage
    > I think a ratio type of thing would be a great
    > idea, but how in the world can this be done?

    It has already been done. The Mojo Nation system was designed as a way for people to exchange services using a micropayment system. This system is different from other micropayment systems because the "coins" are backed in digital resources. It is like the old upload and download ratios of BBS days. You contribute services to the system by "selling" to others and when you need services you "buy" them from other agents. Toss in a distributed, de-centralized data sharing services and you have a pretty cool little item. The coins are like tokens at an arcade, except those who contribute more than they consume end up with a surplus they might be able to sell later; greed is a powerful motivation to get people to

    Cheating is controlled (or at least minimized) by using market-based mechanisms like reputations. By basing the service on something like a market it is possible for distrustful parties to conduct transactions and exchange services. Look around at any stable social structure and you will see a lot of the same techniques employed to fairly allocate resources and control parasites and cheaters.

    jim mccoy

  • I don't really see what the problem is.. The internet, by its very nature is parasitic. That is, its comprised largely from takers, rather than givers. Its an environement set up for people to inhale information without any sort of group correspondance. Gnutella (and Napster, i'd bet) is no different than the rest of the internet. The number of leeches outweighs the number of contributors by at least 100 to 1.

    Thats the nature of things. Doesn't make it evil, or bad, or wrong to leech. Thats just the shape of things. The only thing bad about it, is that some of us get tired of looking at the same pablum every day, and try and improve things by lending a hand.

    The whole Linux community is an example of this..Look where that got us..We've all been had, basically....Tens of thousands of people use yours and my work freely, and physically contribute nothing in return. Its a parasitic process by design. People are building their careers based off the work you've done for free.

    Just stating the obvious, imho.

    My $0.02,
    Bowie J. Poag
  • I would not know because I do not use Napster. I think Napster has the capitalistic ideals of a corporation rather than the cooperative ideals of a society (but if I rant any more then I will be moderated down).
  • PCR worked fine if the sysop had the time to check the message boards. Most of the ones I frequented would dish out harsh penalties for lowsy posters [one line posts eh? not only do you not get credit for those, you now have a PCR of .0000001]. The trouble is on a totally anonymous system, it's hard to go in and smack a user by hand. So the BBS analogy doesn't quite fit.

  • What's the point of getting access to a network unless you get to have the data at all? I mean it really just sucks that you can't get something.

    When I am on the net I am out to collect data and such. Then I can share. If I have nothing to share what am I supposed to do?
  • Good -- Napster is preventing a lawsuit from users who feel that the software takes too many liberties with the computer it's installed on. I believe earlier betas of the software didn't bother to ask. My bad.

    Still, though, most users probably say "yes" anyway just because it sounds safe enough. They can then also use the built-in MP3 player (!) to play their existing and newly downloaded files. (Windows Media Player 7 offers to build a "library" of your songs, as well. It can also rip your CD's into "secure" WMA files, which go right into your library. Man, was I ever glad that the uninstaller worked.)
  • Reminds me of the BBS days of file ratios - 'course then we'd just take an image, resize and upload it, so that idea didn't exactly work as intended.

    Ahh, file ratios.

    An anecdote:
    I was too honest to rename files, but didn't have a worthy collection to upload to keep my ratios up. Finally, in what I thought was surely a nasty trick, I uploaded a collection of short stories I had written, one by one (all dreck - because everything written when you're fifteen is dreck).
    Imagine my surprise at returning to the board weeks later to find a story I had deleted from my archive for being exceptionally lame marked as a sysop-preferred download.

    The moral:
    One man's trash is another's treasure
    Fifteen year old boys are as bad judges of writing as fifteen year old girls are writers
    At least file ratios get things out there that might otherwise stayed buried (even if maybe they should have).

  • When I ran my board I had to approve every file before a user got credit.

    When I ran my Hotline server though, I left it up and running with leech access enabled and I had something like 500 d/ls with 0 u/ls over a 4 hour period. I then had to switch to an account based system. So few people uploaded anything that after a few weeks I took the server down. I was using an ISDN line and it wasn't worth wasting my monthly hours for people to leech off of me.

  • Having been on a 2400 baud connection (lousy cheap backstabbing murderous pack of cigar smoking dogs took my connection, grumble, complain, hate everyone revenge soon) before I have problems even accessing even two simultaneous socket connections or even contacting a name server to get a domain resolved when just doing one task. At best I could only share one file at a time for download and do absolutely nothing. It may be better for other folks but not me.
  • This applies to people as well. I helped organize keggers in college. A small number of us did the organizing/financing/clean up -- everyone else just showed up and partied.

    Dang. Around here the party organizer charges a $5 cover, makes a profit, and uses the money to pay his rent


  • by raph ( 3148 ) on Monday August 21, 2000 @09:47AM (#839856) Homepage
    There are two solutions, I think, to the tragedy of the commons. One is to pay people for their disk space and bandwidth. As several other comments have pointed out, this is exactly what Mojo Nation does, using "mojo" micropayment tokens as the currency. I've been playing with it, and though it's been a bumpy ride, it looks very promising. Check it out.

    There is another solution, I think, which is using trust to define a community. The set of "Gnutella users" is too large and diffuse to actually define a community. Why should I donate my bandwidth for other people who I don't know and don't really care about?

    If, on the other hand, I were sharing files with a much smaller group of people, many of whom I know personally, then it starts feeling more like a community. Of course I want my friends to be able to listen to the music I like.

    I propose that the trust system as deployed on Advogato [] might be a good way to define these communities. Of course, I might be totally wrong about this as well. Only one way to find out :)

    Incidentally, the way Mojo Nation is set up right now, it still has Tragedy of the Commons problems. Currently, you don't get mojo for uploading tasty content. In fact, you actually have to pay for the privilege. However, when you share a file, it's not a continual drain on your bandwidth (or diskspace, fwiw). The actual distribution is handled by "block servers", who do charge for their services.

    Of course, the Mojo economy is still in its formative stages. I hope, and expect, that actual markets will develop for providing and identifying tasty content.

    In any case, file sharing sure promises to be an interesting ride.
  • Actually, any asynchronous net connection puts a strain on the system. Cablemodems are the worst. Some people can download at 300-400K/sec but less than 20K/sec upload. Multiply this buy all those @home and rr people and you suddenly have a large headache.
  • Sounds a lot like the Slashdot karma system.

    Very much so. The /. karma system has been broken for months now, and mojo is still ideaware. Neither work very well.

    Do they have Mojo whores there?
    Cool. Now I can be a MojoWhore as well as a KarmaWhore :-) I've actually got a few boxes sitting on some major backbones just being wasted as honeypots for some security studies. I'd love to turn one of them into a Mojo box, just for the egoboo.

    the AC
  • by angst_ridden_hipster ( 23104 ) on Monday August 21, 2000 @10:01AM (#839870) Homepage Journal
    We used to paraphrase Marx's

    "From each according to his ability,
    to each according to his need."


    "From each according to their assets,
    to each according to their greed."

    Of course, this was in our godless commie Warez swappin' Hotline-usin' phase...

    In the long run, it's OK. There are 90% leeches. But the 10% who make up the providers is not always the same 10% of the people. Today's leech is tomorrow's provider, and vice versa. Sometimes.
    It all tends to work out eventually.
    bukra fil mish mish
    Monitor the Web, or Track your site!
  • filesharing shouldnt be forced and anyway, gnutella hosts 20+ TB of files right now

    If you look at some of the host data that gets passed around, you will note that a few hosts are reporting back share sizes of a couple of terabytes. You can see this effect when connecting to the network as the total share size may instantly jump several terabytes. As a result, you simply cannot trust this number. I would put the total share size between four and seven terabytes - no small number, but nowhere near 20TB.


  • That just shows you that the entire thing is a fad, and when it dies out, the whole Napster lawsuit will just have been a small bump in everyone's collective memory. These things come and go... we've had BBS's, FTP sites, IRC, Hotline servers, and now Napster, Gnutella, Scour, et all.

    They come and go, as the fad takes em. And the whole reason why they go stale is that users take more than they give. DUH. So eventually the people with the fat pipelines that were giving so much get sick of giving and want stuff in return. And as soon as they go to the "membership" model or the "ratio" model, it quickly gets abused or the majority of people leave, giving you a select membership only service (which is what most hotline servers are now). Then again, you still have great places like #macfilez... but those places are becoming fewer and further between.

    Napster will die, same with Scour and Gnutella... lets just see how long they stay around. And again, everyone that's really into pirating will stay with their favorite IRC site, FTP site, or BBS. All of the other's that were recently introduced to the fine art of pirating will get sick of Napster and attempt to learn IRC but will find it too difficult to use so they'll give up and start going back to buying CDs and nothing will change.
  • by TrumpetPower! ( 190615 ) <> on Monday August 21, 2000 @10:07AM (#839878) Homepage
    And, in a related and equally shocking development, other researchers discovered that 99% of people using the World Wide Web don't themselves create web pages but only view them. There can be only one result of this vast and tragically unballanced sucking of vital computer resources: the Imminent Death of the Internet [].


  • I remember back in my college days. we had set up a "kiosk" of 3 pc's in the processing department that along with access to UUnet there was a huge (500 megabyte) file sharing database. users could walk in and use floppies to get what they needed and upload that which was useable by all. when we went online we loaded the thing up with gobs and gobs of shareware, tons of free documents and along with the UUnet connection lots of cool things.. We also kept close track of the users activities (No username attached it was all anon access) for the Phyc grad-students that were researching basic human though patterns when exposed to the digital information age. Well there was a ratio of 900 to 1. of every 900 items downloaded there was 1 item uploaded (and usually by one of us maintainers via modem)

    people are greedy, they take and take and take. and WILL NOT give. need proof of that statement? look at the fact that there are starving people in the USA. There is absolutely NO reason for anyone in the USA to go hungry, without shelter.. yet it happenes because of GREED.

    Every pattern we found on that Kiosk matched the Phyc patterns they had with people giving cheritably..

    Face it, Homo-sapiens are self centered greedy weenies. And that is one of the first things that will be seen by any sentient visitor.
  • There are plenty. Visit [] and check out the third party clones. The gnutella 0.56 isn't even supported anymore, so if you are running that, stop, and get something new. Toadnode [] (win) is a quick up-and-comer.
  • by Raetsel ( 34442 ) on Monday August 21, 2000 @10:11AM (#839883)
    First, a warning: This is for Win32.

    The program is called Gnucleus []. It offers the option to allow or deny a result based on:

    • IP address (frustrated by searches that return reserved IPs? No longer!!)

    • File type (don't want those damn .HTM files? No problem!)
    You can sub-search or filter your search results. You can run multiple searches at once. It even lets you throttle things, although I haven't tried that feature myself. @Home throttles things just fine all by itself. ;-)

    So, the source is out there on SourceForge. If these features aren't in the Linux realm yet, porting should be a simple matter (I'd hope.)

  • because they're *lazy*

    Personally, I'm sticking to USENET, where I've run into a core group of regulars who have similarly eclectic tastes, and I don't have to worry about seeing slow servers or MP3s that are cut off at the end.


    Freedom is Slavery! Ignorance is Strength! Monopolies offer Choice!
  • by Kaa ( 21510 ) on Monday August 21, 2000 @10:17AM (#839888) Homepage
    the privacy of a modern household has greatly increased the incidents of child abuse. In the society that we evolved in, one large factor that stopped people from abusing their child was the fact that there was no privacy--if you abuse your child, the whole village knew about it.

    I am highly suspicious of such claims. Let me point out the two most obvious problems with it. First of all, reliable child abuse statistics are very hard to come by. I suspect that it is impossible to find meaningful (not plus/minus orders of magnitude) child abuse statistics earlier than the middle of century, certainly earlier than the beginning of the century. The fact that, say, prosecution records, show that from 1890 to 1900 there were X child abuse cases and from 1990 to 2000 there were Y such cases does not mean much. It's fairly obvious that at the end of the XIX century most child abuse went unreported and unprosecuted. The rates of actual child abuse at that time are open to wild guessing.

    The second problem is that definition of child abuse changed considerably. Right now in the US leaving your, say, 10-year old kid alone in the house for a couple of hours is, technically, child abuse (that depends on the state you live in). Beating your kid regularly is definitely child abuse now, but was totally socially acceptable a hundred years ago.

    So, sorry, I don't buy that argument about anonymity breeding child abuse. I think it's completely bogus.

    I have heard it said that the most common term asked for in the leading search engines is "pornography".

    People are simpler. The most common search term (at least according the current urban legend) is "sex". Pornography is hard to spell ;-)

    And in any case, what's wrong with that? Evolution is very efficient at weeding out people who are not interested in sex.

    While pornography is somewhat harmless, other activity on the internet isn't

    Okay, it's "mostly harmless"...

    But you are really arguing for a police state: with a cop at every corner and with all you do compiled into your record that stays with you all your life. I do not want to live in such a world. I don't think many people on Slashdot do. Of course there are always those who like such worlds (after all, it's safe -- unless the government takes a dislike to you) and unfortunately they are not too rare. Oh well.

  • Even on a T3 at school, when I tried to set gnutella for sharing, a few hundred people would immediately begin leeching from me. In no time at all, a few hundred outgoing connections would all be crawling along at less than 1 k/s.

    Granted, I was on a 10 mbit line, but the point remains. If there was a way to limit the number of outgoing files by bandwidth, or number of connections, or something, I would share files.

  • Files that you download are not shared by default.

    Let me say that again : Files that you download, with Gnutella, are not shared by default.

    Which means that, regardless of what people might say, Gnutella is not, by default, from a practical standpoint, a peer-to-peer network. It's just like any other client-server network - the clients take and the servers give. It takes work to be a server, even if it's 20 seconds of work in copying files and clicking rescan, and most people are short-sighted nitwits who don't realize they need to do that and wouldn't care if they did.

    At the very least, files you download on Gnutella network should be shared by default. I think, if it's supposed to be a peer-to-peer network, then sharing should be the default behavior, and not sharing an option. If you download a file and the file you have is the same length as the source (it's finished) then the file should be shared.

    I'm feeling like enough of a jerk right now to say that it should be shared regardless of where it is put on the hard drive. Gnutella should keep track of the stuff it's downloaded (Stolen Album-Stolen Song 1.mp3, Stolen Album-Stolen Song 2.mp3) and search your hard drive for those files, and offer them to the net if they are found. What part of "peer-to-peer" do you not understand?

    I don't use Gnutella anymore. It's not what it was said to be. I think of a hundred thousand users who can't even be bothered to put thier incomings in thier outgoing directory, and decide not to bother with it.

  • by puppet10 ( 84610 ) on Monday August 21, 2000 @10:26AM (#839895)
    It does not just allow people to engage in socially unaceptable acts, it allows people to engage in acts which are not locally acceptable.

    The general attitude of people towards certain actions in New York is very different from for example a small southern community. The net allows people of like interests to connect with each other, even if the group is small and scattered, and find people who have similar views toward their actions.

    This is a good thing for people who are not considered normal in their local communities, however those local communities may have a problem with this.

  • Instead of trying to send a 3Mb file up a puny 33.6 connection to the net Mojo Nation is designed to break the file up into lots of small pieces so that you ask 300 servers for for a little piece of data; the design was created to take advantage of the assymetric nature of current network connections. Users have less upstream bandwidth than downstream bandwidth, by splitting the load for sending data across lots of hosts Mojo Nation can minimize the impact on any single peer and still be able to deliver high bitrate downloads through the larger downstream side of the connection.

  • anybody remember good old HSLINK from the "old days?" I never learned how it worked, but it somehow managed to magically allow you to upload and download from a BBS simultaneously with no apparent loss of bandwidth. It was really great for ratios :)

    Anyone know how it worked? If only some kind of equivalent existed for the 'net.. hehehe
  • The software defaults make so much difference. Napster chose sharing over security; Gnutella chose security over sharing. The probable effect of this will be that some third alternative, maybe a repackaged Gnutella, wins, but opens up security holes.
  • And in any case, what's wrong with that? Evolution is very efficient at weeding out people who are not interested in sex.

    I suspect it will also be efficient at weeding out people who are excessively interested in streams of numbers as substitutes for mates.

    Can you honestly say that you think that people who search for "sex" on the internet actually have more sex in real life?

    Of course, one might argue that driving down birth rates is a good thing...

    Despite rumors to the contrary, I am not a turnip.
  • I use gnut and set it to allow cacheing of frequently searched for files on my system. Every now and again I clear out some of the more objectionable stuff (surprisingly little of this actually), move some items into my permanent share directory and let it the cache repopulate.

    It's fascinating to see what arrives, to the extent that I rarely search for anything, just check the cache and see if anything interesting has turned up overnight.

  • by Animats ( 122034 ) on Monday August 21, 2000 @10:49AM (#839911) Homepage
    This article is more hype from Bernardo Huberman [] at PARC. Huberman is an economist of the libertarian persuasion. His answer to all problems is a market, as becomes obvious when you read his papers. His libertarian bias is so strong that he's written some very bad papers. Worse, some of them have been distributed widely because they said some things some major players wanted to hear, such as his claim that per-packet charging on the Internet was inevitable.

    (This happens to bug me personally because he claims to have been the first to observe that the tragedy of the commons problem applies to Internet congestion. He wasn't; I was, in 1985. See RFC970 [])

    As a previous poster noted, the default with Gnutella is not to share anything. That's why so few share. This isn't rocket science.

  • For example, in New York state, I believe, women are legally allowed to go topless. Yet, if some kid in a public library without filters, sees some boobies, OH NO HE HAS BEEN SCARRED FOR LIFE!
  • Of course not. But that's not the point: the original poster I was replying to implied that looking for sex on the 'net is a dirty, shameful, bad thing to do. I was actually taking a swipe at his puritanical leanings.

    Morality is founded in practical concerns for the larger community.

    Everyone using pornography instead of getting married and having real sex resulting in children (or, for that matter, doing anything sex-related except seeking and getting a child-producing marriage) would mean the downfall of a society. That's why it's considered immoral by many (not that they thought it out that way, but through darwinian survival of societies the ones with such "puritanical" morals are more likely to survive, and are thus quite common).

    So, depending on how you look at the world, it can be considered a dirty, shameful, bad thing to do. It's a waste of effort and a misdirection of the reproductive drive that is harmful to one's self and to society.

    That is, unless you consider misdirection of the reproductive drive and lowering of birth rates as good for society, and therefore moral.

    Despite rumors to the contrary, I am not a turnip.
  • by d3a350 ( 203361 ) on Monday August 21, 2000 @11:14AM (#839924)
    I think it's a bit more complex than just "people are greedy and don't want to share". Such as:

    Many user may be at work or some other place where may makethere may be a policy against Napster and Gnutella, or at least the fear that they could get in trouble if caught by a sysadmin. Therefore, they get in, grab what they want, and get out.

    Limits of the average PC. Including low bandwidth and limited storage space. Its harder downloading stuff if people are uploading at the same time. Combine this with the fact that not only a small percent of people on the 'net have always-on connections and fewer of those probably leave their computers on all the time, and its not difficult to understand why there are only 10% of the people who share but don't query. And I don't know what the average user does after downloading a song, but I either toss it away after listening to it a few times or burn it to CD, primarily to save space on my computer. In either case, I couldn't share it easily again if I wanted to. Granted, hard drive space is dirt cheap, but how many older computers are still out there, and how much space does the average computer (not the average Slashdot reader) have? (I have a 2 year old PII with about 3 GB of space. And my MP3s have to compete with Unreal Tournament and those other space-hogging games and apps.)

    Guilt. Its one thing to quickly grab a couple of songs you want, but to share out a bunch of stuff you feel more like a pirate. Granted, I could care less if Brittany Spears gets another nickel from her music, but I've never downloaded her stuff or bought a CD. And the record companies can go to heck. But my favorite artists aren't millionaires or superstars, and I hate to think of them getting ripped off because of something I'm doing. I can justify my own downloads by making sure that I support them in any other way I can, but the guy downloading their song from my machine might just be some jerk who could care less. . . .

    And finally, in an slightly off topic rant, my pet gripe about Napster (other than it's actually a pretty cruddy piece of software) and Gnutella (which I haven't used as much), is that there's no opportunity for discovery. When I share music with my friends in the real world (aka off-line), its more on the lines of "hey, I found this really cool new group", than "here, have a copy of this one track by this artist you already know". Napster is definitely designed for finding something very very specific and little else. To me, sharing is passing on something new to someone, that they may not have found on their own, which means that Napster isn't really my idea of sharing. . . .

  • Of course, you never passed the beer around recursively either. If you had, you might be just a wee bit more skeptical.
  • by FallLine ( 12211 ) on Monday August 21, 2000 @12:07PM (#839933)
    I guess the obvious conclusion is that people will not give away, only take from free sources. But then how do you explain napster?
    GNUtella has failings that are far more critical then lack of sharing users. However, there are a few key differences with sharing. First, napster has enjoyed a wealth of college students and other high bandwidth users who otherwise do not use the internet. Because they hardly use the internet, they do not notice their bandwidth decaying as much as GNUtella users--who are mostly frequent internet users. Second, GNUtella, unlike napster, closes when the user thinks he closes it. These two factor combined have created an environment on GNUtella where THOSE few users who do opt to share, get hit with more traffic than they would on napster. Thus they are further discouraged from sharing.

    The main flaw with GNUtella, as I see it, is its recursive design. Though few people seem willing to bring this point up, it simply CANNOT scale reasonably. It simply could never support napster's load, or even a fraction thereof.

    In short, GNUtella is reasonably acceptable for little splinters of "networks". I could see loose knit warez groups/associatons of, say, 50-200 users forming around it. However, this kind of instability and decentralization does not lend itself to use by the vast majority of people who are technically inept.

    Even though I routinly hear that the quality of music is garbage ("it's all brittany spears and crap like that") I've found a wider range of music there than any other place I've ever been. From The Why Store to Bach and plenty of it.
    Compared to other sites on the internet, this may be true, but I've found tons of stuff impossible to find on napster--even stuff that I could buy at any major CD chain.

  • content of any sort (audio, video, text, sculpture, etc.) has always existed for the sole purpose of being consumed. It's not a surprise that digital content is being slurped at a rate far exceeding its creation - the digital medium is what facilitates that content being distributed more effeciently than ever before possible.

    The analogy to a "commons" is a false analogy. A commons is a resource that is shared by a community. For example, a grassy area where you graze your cattle, or the air we breathe, or Social Security. If a system is not used to regulate how users donate and withdraw from the commons, the commons WILL collapse because short term private advantage (eat all the grass) is always more attractive than long-term common gain (ensure grass is there for your cattle for the next hundred years, for the whole village).

    Napster, Gnutella, etc are NOT a commons at all. They are a content distribution system, where a elite set of users (Britney Spears, people with large hard drives or great bandwidth, etc) provide content to a system of users. A better analogy is a museum - the public comes in to access the content (in this case, sculpture, archaeological ruins, etc.) which was provided by a elite (the original cultures who built the structures, Indiana Jones who saved it from Belloch, the city of New York and the MoMA...)

    given that content is very difficult to produce, and then also difficult to distribute (and don't forget that the sole reason content exists is for distribution. NBC doesn't seal filmed episodes of Survivor in a cave, it broadcasts them, for example).

    its important not to gloss over what a "commons" really is because there really are a LOT of commons' out there which we desperately need to regulate better. Lumping in Napster and Gnutella is a mistake because it dilutes the idea of a commons, and also sets unreasonable metrics of success for these new systems to be measured against. The exaggerated upload/download ratio is not a sign of failure but rather one of triumph.

  • Has nothing to do with being a leech or a jerk or anything, but rather with the way Gnutella (or at least the version I have) works/is programmed. You see, the beauty and the curse of Gnutella is that, unlike Napster, it turns everyone's computer into their own server. Which is great for avoiding the fate that may befall Napster and other centralized databases, except that it means that each computer must do a string search through its own database of shared files with each search request made to the system. The way I set up Gnutella originally, that meant searching through a couple thousand files a couple hundred times a second. Furthermore, unlike Napster, that version of Gnutella didn't have the option to limit the number of simultaneous downloads from your system--thus, I would have 20 or more people downloading from me at a time, all whilst I may or may not have been downloading.

    The upshot of all this is, Gnutella makes your computer act like a real file server, and most people's computers aren't cut out for that. I don't know if it was a bug in that particular version of Gnutella (Win32, 0.52) or what, but it was causing my system to seriously chug. As in, CPU utilization at 100%. I couldn't play MP3s without serious skipping, couldn't browse the web without noticable slowdowns. Even my system clock started running slow!

    Now, it could have just been my system, and it could have been a bug which has since been fixed, but it was seriously unacceptable. Since then I've used Gnutella much much less, and often decide not to share when I do (or I share a smaller folder).

    It does not have anything to do with my not being willing to share, or being a freeloader. If this is an inherent limitation of the way Gnutella distributes its file serving, then the problem will go away soon with more powerful computers; if (more likely) it was just an implementation bug, it is probably fixed by now. In fact, I think I'll go check...
  • Of course, you never passed the beer around recursively
    Pleasae expand - I'm not sure I grasp your point. For certain issues, recusrions would seem to help....What I wouldn't have given for an autonomous recursive bathroom cleaning algorithm.
  • My point is that GNUtella has about as chance of succeeding as a massive recursive kegger. GNUtella, if you were not aware, handles all queries and query replies recursively.
  • Having the same directory for downloads and sharing typically leads to a lot of unfinished, broken files to be shared and redistributed. This leads to quite a bit of frustration... So this is not a good idea. Unfortunately, Gnutella (the Win32 client) allows for the download directory to be shared. There are even some of the typical zero-length message files distributed in Gnutella that ask everyone not to share their download directory. FURI, one of the good Gnutella clients, appends .dl to files while they are being downloaded so that everybody will see something is wrong with the file.
  • by RomulusNR ( 29439 ) on Monday August 21, 2000 @01:40PM (#839954) Homepage
    I'm sure that a good portion of the MP3 and other trading communities will lart me for this, especially #mp3files and other like-minded groups, but I'd like to stand on the wooden box and pronounce:

    Ratios are good.

    They're not great; as with any system, they are bound to be abused, and no system (no, not even Slashdot's hairy moderation system) will be free of abuse.

    Sharing thrives because people always have something to share. And sites grow because their collections grow. And their collections will grow as contributions are made. It follows that a site which gets contributions from its users will grow faster, and ostensibly, get better.

    I've run FTP sites in the past that were essentially anonymous. I almost got run out of certain IRC channels and newsgroups on a rail because I decided, on and off, to make that server a ratio server.

    For me, I liked ratio sites, because they always seemed to have better selections of items. On some sites that weren't ratio, I still uploaded. Plus, I had plenty of obscure items that I wanted to spread, so not only was I contributing to the quality of the site, I was also exposing site admins to new things, and increasing the availability of those items.

    In the FTP arena, a lot of sharing problems just came from a simple lack of respect. Many 33.6 modem pups would initiate a dozen or so connections at once, all multitasking among each other, and let them sit, say overnight or while they were out. Then I would have this chode sitting on my site, eating up a login slot, crawling away at 0.4 kB/s or less. I would normally kill these connections on sight once they got that low. It's simply a show of disrespect. There's no reason why you can't (with the right clients) arrange these transfers so that they go sequentially instead of simultaneously. At 33.6 you can download a meg in about 3 mins. A 4-5 MB file then takes 12-15 minutes. That's acceptable. But 1-2 hours or more is not. I had much more respect for those visitors who were giving my site the respect of not leeching at ignorantly slow speeds.

    Ratio sites are also ALWAYS available -- it keeps the disrespectful leeches away. I never had a problem finding a login slot on a ratio server. On a leech server, forget it. You could try to hit it all day, and basically you were in a massive race condition with goodness knows how many other #mp3files lurkers. This is assuredly why so many noteworthy leech FTP sites then died a horrible death -- to the dismay of all those leech-dependent trading pups.

    Once I went to ratio, of course, I would start to get a fair amount of total crap. Not just dupes, which were annoying, but out-and-out crap -- like 15 minute long news reports taped off radio encoded at 44.1 / 128 or more. Not only did the disrespect of uploading these files get my goat, but the sheer braindeadness of encoding such large, worthless files just to use as ratio cloggers when it would be just as easy to encode and upload worthwhile stuff.

    But this didn't encourage me to turn off ratios, because I would always ban those IPs (zones if necessary) after looking through the logs, and because for every ratio revolutionary, there was at least one visitor who uploaded something worthwhile. And the fact that there people out there who respected my site, and were honestly interested in trading, made me keep my site up as long as I could. (Eventually, I had two logins; a ratio login, and a leech login with limited downloadable selection.)

    I started a server not only so I could spit stuff out into the trading (i.e. leeching) community. I started it so I could also get back from that community. It was give and take, not just take, take, take. Leech sites only work in theory -- they have crap, they're overloaded, and they almost never grow. Their admins eventually get discouraged, and turn away from the trading community altogether.

    It burns me to see people out there on Napster who don't share anything they have. They should at least have the common decency to set their sharing directory to an empty dir, if they are going to drop max DLs to 0. That's just disrespectful. And they don't deserve any respect back.

  • Guilty as charged.

    It's just that when I was a godless commie Warez swappin' Hotline-usin' no-good SOB, I was also a otherly-grammared, politically correct phraseologist.

    Now that I'm a godless Corporatist profiteering flak, I am once again a grammar pedant. I use apostrophes correctly (even with the word "its" versus "it's"), I attempt to preserve voice whenever possible, and I never say "quote" when I mean "quotation."

    Grammar is just another natural transition engendered by such a political shift; others are exemplified by driving a Porsche Boxster instead of a '69 Volvo Stationwagon, dating a woman named Elizabeth instead of one named Sunshine, and drinking dry martinis instead of Ernest & Julio jug wine.

    OK. I admit it. Even with all those changes, I'm still a godless commie Warez swappin' Hotline-usin' no-good SOB. So sue me.
    bukra fil mish mish
    Monitor the Web, or Track your site!
  • Gnutella is like the Beggars in Spain, no limits... put some files up, and every byte of bandwidth will be sucked out of your line.

    I tried it, and whenever I put something up (like some Tangerine Dream concert bootlegs in MP3 format), a few maniacs with unlimited bandwidth start multiple downloads till my internet connection is completely overloaded.

    Until there are options in Gnutella to limit the number of connections per user, and total bandwidth used, I'll opt out (as will most others who want to actually use their internet connection for other things besides Gnutella).

  • by EricEldred ( 175470 ) on Monday August 21, 2000 @07:45PM (#839971) Homepage

    Garbage in, garbage out. Make a model and then dig up the evidence the model fits the evidence. How is it that Slashdot escapes this paradox--do 99% of Slashdot users read instead of posting? I don't think so.

    Commonsense ought to tell anybody that new popular sites get more links than older, static sites. But Huberman has to publish a paper in Nature to prove that. He also contorts his brain to try to prove the power law distribution and page count of sites.

    There's nothing new to these ideas--they are just trotting out the old "Pareto principle" that hardly anybody tries to explain. You find self-help books now in the Business section on how to "apply" the 80/20 rule. But the "law" that Pareto discovered is too universal in nature to provide an answer to inequality in the social realm, and too abstract for anybody to say that it is greed that motivates Gnutella sharing.

    Even if one admits that music sharing is a market (not clear what the costs are) or that some market principles can help in distribution (it's too early to tell, the Net is an experiment, bandwidth increases to meet demand, it is still almost too cheap to meter, it's becoming easier for everybody to publish their own works, people have other motivations to publish than to make money, and so on) Huberman still don't have the right answers--and they can't derive them simply from the data they produce.

    Instead, look at the work of Jean-Philippe Bouchaud and Marc Mezard []. As explained ably by Mark Buchanan in New Scientist 19 August 2000, pp22-26, the natural laws of physics can explain the inequality in the power law distribution much better than economists in the past have attempted to do.

    If one needs to remedy inequality then there are several programs one can attempt. One is taxes--tax the rich and give the money to the poor. But taxes often are captured back by the rich and the poor stay poor. With music sharing, if the receivers paid a small tax to the system, with the money going to the producers, one might or might not see the receivers producing more files on their own. Maybe they are motivated by fame instead of money. Or the big collectors might not increase their collections if they were given a small amount of money.

    Another remedy these physicists suggest trying is simply raising the "temperature" of the system. By increasing the transactions in the system, through free trade, fair rules, more exchange, and some competition, then there is less volatility for individuals--less chance for somebody to go down and stay down.

    Libertarians such as Huberman might be interested in this idea. Certainly they worship free trade. But they ought to be careful of bringing stale models to the new Internet. It might not work the same way they predict in other systems.

    I for one don't really see the Gnutella "discrepancies" as a problem. There is certainly not yet any "tragedy of the commons." Because "tragedy" implies that there are limited resources. For example, on my cable modem segment I am no doubt considered a "hog" because my web server gives out big text files to people all over the world. But even if I increased my bandwidth usage tremendously, I would still be far from saturating the system. All that would happen is that some people might be delayed a few seconds by collisions. If the system ever got saturated, I would have to take my server off anyway, and then the system can work toward a new stability.

    Since I don't think Huberman's economic, market model is very good at the moment, I think that we ought to explore other alternatives. For example, Frances Cairncross, editor of The Economist, in her excellent book The End of Distance, refers to at least three other ways markets could succeed in music publishing: the advertising model (Cairncross attaches Ester Dyson's name to that), the performance model (John Perry Barlow), or the sort of subscription model (Wall Street Journal formally, Stephen King less formally, ideas that date back to de Sola Pool's idea of the Internet as a model for freedom).

  • The scaling problem will never be fixed in Gnutella without having some sort of central server. Its just how the numbers work out.

Life in the state of nature is solitary, poor, nasty, brutish, and short. - Thomas Hobbes, Leviathan