Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Programming Patents The Internet IT Technology

Making Freenet Find Stuff Faster 283

Posted by timothy from the chugging-through-the-data dept.
Sanity writes "Many probably saw the recent announcement of Freenet 0.5.2. This release represented a vast amount of work - primarily in reducing Freenet's CPU and memory requirements. However, streamlining Freenet's current functionality isn't all we've been working on. I just finished an article that describes the most fundamental improvement to Freenet's core algorithm since its original design over three years ago, it is called "Next Generation Routing" and has the potential to dramatically increase the speed with which Freenet retrieves information. It could even make Freenet faster than the World Wide Web in many circumstances, all without compromizing anonymity and while remaining immune to the /. effect."
This discussion has been archived. No new comments can be posted.

Making Freenet Find Stuff Faster More

Making Freenet Find Stuff Faster

Comments Filter:

  • Good. (Score:5, Interesting)

    by Squidgee ( 565373 ) on Sunday July 20, 2003 @05:39PM (#6486949)
    I'm glad to see Freenet progressing so well; especially it being resiliant to the /. effect (read: DOS attacks), and it being faster (!) han the WWW.

    Freenet is an awesome idea, and very rapidly becoming one that is neccesary to ensure your protection. Although it is a double edged sword (It can help both good, and bad people), I think it's one that is neccesary. And, if it becomes speedier than the web at large, it'd be just freaking awesome. Now, no one needs to fear censorship, nor do they need to fear the government shoving them into a database.

    Now if only I could get it running on my Mac OS X box...

    • Re:Good. (Score:5, Informative)

      by Ryan_Singer ( 114640 ) <Ryan.SingerNO@SPAMgmail.com> on Sunday July 20, 2003 @05:43PM (#6486963) Homepage
      it's running on my unmodified osx box. just use the unix version.-Ryan
      • Yea, but does it do anything after you run the .sh?

        It never does anything for me..

        • Re:Good. (Score:4, Informative)

          by freedom_leffo ( 605662 ) on Sunday July 20, 2003 @05:59PM (#6487059)
          Well, just wait half a minute or so and then point your favourite browser towards http://127.0.0.1:8888 - and off you go! The Freenet-thingie is running in the background.

          • Re:Good. (Score:3, Funny)

            by Squidgee ( 565373 )
            Ooooh, that's how it works.

            Dammit, I hate it when I miss things.

          • Unless your favorite browser is IE. Then use your secong favorite Mozil--er, browser.

            It's possible to lose your anonymity with IE since it ignores MIME types.
          • The potential is amazing! Think BitTorrent.

            As the number of people on /. that use Freenet increases, stories can start linking to the FreeSite of an article--for example The Freedom Engine [127.0.0.1]--along w/ the 'old' web site.

            Instant distributed mirroring.

            Bloody cool.

            • Mirroring websites on Freenet. (Score:4, Informative)

              by Myself ( 57572 ) on Monday July 21, 2003 @05:03AM (#6489267) Journal
              I've said [slashdot.org] this [slashdot.org] before [slashdot.org]!

              The only problem is that there's no one-click tool to mirror a website into Freenet, yet. Freenet's gateway has an anonymity filter which prohibits out-of-freenet links, and it also disallows a lot of things. If someone wanted to write a simple tool to clean up a site and hack the links to work in Freenet, it would make this a lot easier.

              By the way, using the http://127.0.0.1:8888/KEY@whatever style links is discouraged, because not everyone's freenet node is localhost, and not everyone runs it on port 8888! The preferred format is freenet:KEY@whatever which can then be handled appropriately by your browser.

    • Re:Good. (Score:5, Interesting)

      by Ralanti1 ( 649287 ) on Sunday July 20, 2003 @05:47PM (#6486995) Homepage
      Isn't the www good for both good and bad people though too? any technology that comes out will have a way to exploit it. The fact that it's faster then the WWW is an achievement in itself but would the RIAA/etc try and go after it claiming it's anonmity is the problem? I'm really curious to see how this plays out.

  • Challange? (Score:5, Funny)

    by traskjd ( 580657 ) on Sunday July 20, 2003 @05:40PM (#6486953) Homepage
    "immune to the /. effect."

    If this isn't a challange I don't know what is :-)

    -traskjd

  • A dare? (Score:4, Funny)

    by ATAMAH ( 578546 ) on Sunday July 20, 2003 @05:44PM (#6486969)
    > ... and while remaining immune to the /. effect
    Said the author of the slashdotted article.

  • Hmm, sadly (Score:2, Flamebait)

    by Anonymous Coward
    Those using browsers that support the "mng" animation format (such as Mozilla) can see an animation of a node's datastore specializing over time here.

    That's not true anymore, communists Mozilla maintainers removed mng support to save a 'whopping' 100k download.

  • Will Oppenheim Eat His Words? (Score:5, Interesting)

    by Schlemphfer ( 556732 ) on Sunday July 20, 2003 @05:49PM (#6487012) Homepage
    In a widely publicized interview from earlier this month, RIAA Senior Vice President Matt Oppenheim said:

    Other than the fact that most infringers do not like to use Freenet because it is too clunky for them to get their quick hit of free music, it is no more of a threat than any of the popular P2P services.

    Translation: "Oh Lord, I hope Freenet is inherently unable to have robust search functions, because if it ever develops these, we're hosed. But in the meantime, we can dismiss this software as being a big POS."

    Now, less than two weeks after the interview, it seems the one aspect of Freenet that Oppenheim wanted to write off at is on the brink of being fixed.

    • It isn't search... (Score:5, Informative)

      by Anonymous Coward on Sunday July 20, 2003 @06:03PM (#6487075)
      ...at least not keyword searching as you find in Google and Kazaa. When they refer to searching they mean given a key (a very large number), finding the corresponding data.
      • Agree. The next (distant?) step though will be to turn freenet into a GRID computing engine. One big global (anonymous) computer, on which the mother of all keyword search engines could be run.
    • freenet is a *protocol*, not a client (though they do ship the http proxy client with the main distro). just like the http protocol doesnt have any search functionality built into it, neither does freenet. you can, and people do in fact use regular old web spiders to create searchable indexes of freenet.

  • Easy update for existing freenet users. (Score:5, Informative)

    by anonymous coword ( 615639 ) on Sunday July 20, 2003 @05:50PM (#6487016) Homepage Journal
    Instructions for windows and linux and linux compatables.

    Windows : Right click the rabbit icon in your system tray, then click upate to latest snapshot build.

    Linux : run update.sh in the freenet directory.

  • Beware the Federation (Score:5, Funny)

    by Vagary ( 21383 ) <jawarren AT gmail DOT com> on Sunday July 20, 2003 @05:51PM (#6487018) Journal
    You know what'd be really impressive? Finding a way to make FreeNet slower. It'd be so slow you could make a Beowulf cluster of FreeNet nodes and use it as a time machine. Personally, I'd use it to go back to Ian Clarke's dorm room and convince him to get drunk and high rather than wasting his life making a P2P system that will be useful around the same time we have to start worrying about being censored by the United Federation of Planets. But that's just me.
    • Personally, I'd use it to go back to Ian Clarke's dorm room and convince him to get drunk and high rather than wasting his life making a P2P system
      Been there - done that - didn't help.

  • Distributed algorithm benefits Freenet again (Score:5, Interesting)

    by andyo ( 109338 ) on Sunday July 20, 2003 @05:54PM (#6487037) Homepage Journal
    What I find interesting about this algorithm is that it is applied individually by each node; there seems to be no need for nodes to share data over some complicated protocol as in many distributed systems. Yet (I think we can believe Clarke) this change improves response time through the system as a whole. It's a validation of the basic Freenet model of systems acting alone but providing a service greater than the sum of its parts.

  • peekabooty anyone? (Score:4, Interesting)

    by Snooweatinganima ( 168199 ) on Sunday July 20, 2003 @05:55PM (#6487040) Homepage
    has anyone ever tried peekabooty [peek-a-booty.org], esp. under wine? The reflections on open source development [peek-a-booty.org] the developer(s) feature on their website sound kinda depressed..but then again, the honesty factor speaks for them. Are there any deep flaws in the idea? I personally like the simplicity of their design, but since I'm not a design guru, I may be utterly wrong.

    • Publicibooty (Score:3, Interesting)

      by Anonymous Coward
      The CdC people (and the organizations that spun out of them like Peekabooty) have always been much better at self-publicity than solving real problems.

      They thought it would be cool to design a censorproof network. They weren't interested in supporting what was already in development, namely Freenet, after all - where is the publicity in being part of someone else's project?

      The only problem was that they dramatically underestimated the difficulty of pulling it off - the result? Peekabooty was, is, and



      • They can both become a proxy for someone else as well as hide behind someone else in a ring.

        You connect to me, I connect to joe.

        Joe wants to talk to you, he connects to me, and I connect to you.

        This could be random, every time they log on new proxy rings.

  • Immune to /., perhaps (Score:5, Informative)

    by Anonymous Coward on Sunday July 20, 2003 @06:04PM (#6487084)
    Immunity to ignorant masses of /. users it is not.

    I was in the first /. crowd of joining, and here is the etiquet/advice I have.

    Things to do if you plan on playing with freenet:

    1. Set it up properly.
    1a Set your IP in the config file, read the site for details, but it's freenet.ini
    1b Try to use DynDNS if you have a dynamic IP
    2c Leave it up 24/7 for a few days before you judge speed. You need to let the blood circulate :)
    2. Install a proper version of Java. I recommend the 1.4.2 beta. IBM may work better, I haven't tried.
    3. Fix your browser.
    3a Your browser will crash on some sites (even Mozilla not Opera) because of a GIF bug.. patch it.
    3b Set your number of simultaneous connections up a lot. You request a file from your local store, then it downloads it. You need to request as many in parallel as possible.

    Now, on to advice.
    Get Frost! Frost is like the news groups of the freenet. It's a great place to read interesting ideas.

    If you want to make a site, check out Fish tools, Fuqid and FIW.

    Be aware that there are 3 different kinds of sites, and two modes of getting information
    3 types include interval based, revision, and static. Static sites are one time shots. Revisions you create directories like /1/ /2/ /3/ and link to images from the future. If the image loads, you know there is a more recent revision. date based must be activated every time interval, or they die. Be very careful with these.

    There are SSK and CHK linking methods, which I still don't know a whole lot about, but maybe someone will reply and explain them.

    By /. effect immunity, they mean linking to a site will only make it stronger. Everyone on /. joining freenet is just going to slow it down, because basically, you are creating a great suction on the net without any data to give back. Even worse, when you quit off of freenet, everyone will be looking for you from their cache and not finding you. This is going to cause the most problems, but surely not everyone on /. are going to quit on the same day. ;)

    Get IIP, so you can realtime chat with people that run some sites on freenet. #freenet is dedicated to freenet chat and issues.

    Have fun!
    (Posting anonymously in respect of the freenet principals.)
    • Ok, so I did all of that when I set it up...

      Now the question all the new freenetters really want answered, is - after installing, configuring and letting run for a while.... How do I get some porn off the nextwork? Is there a cache of keys on the netsomewhere that I need to be able to find or what? Is there a crawler app that just keeps track of what it knows it's run across and builds it's own little directory??????

      • OK, not a bad question. not hard to answer, either ;)

        Once you have it up, open the browser proxy page. there are some default bookmarks there. Go to The Freedom Engine.

        When that loads (it'll take a while... it's big...) go looking for porn links. There's lots of non-porn stuff, but just do Find in Page 'porn' or some such and you should find a few. ALternately, YoYo and some of the other default bookmarks have categorised stuff (including porn) but they might be harder to get to load / more out of da

      • Do not download porn from Freenet. (Score:4, Funny)

        by HanzoSan ( 251665 ) * on Monday July 21, 2003 @01:02AM (#6488791) Homepage Journal


        If you download porn, the spyware programmed into Freenet will foward your IP to the RIAA, FBI, NSA, and then post it to a few hacking/warez newsgroups and forums.

        Freenet is NOT a pornster program.
      • One aspect of Freenet is that the content reflects what the community puts out there. If you want to see stuff other than porn, put in up on the network. In fact, it will help to put even more stuff as long as it is of value to other people.

        "If you make it, they will come" is all to important with Freenet.

        Another point to make: If you view the porn and try to download it, you are also spreading this content to other nodes. If you don't want it on the network, don't view it or use it. Indeed, Freenet

  • Hmm.. (Score:4, Interesting)

    by Idealius ( 688975 ) on Sunday July 20, 2003 @06:05PM (#6487088) Journal
    Makes you wonder if Freenet gained popularity over the web whether all "official" transactions would be web-based, leaving Freenet to misc. web sites that are completely information/communication based. The reason I wonder is because if someone gets their login/password stolen from some random service on Freenet which they invested mucho time in, how will anyone else know the difference? That would really irk me.. (Yes, I know the web is vulnerable to this as well, but at least it requires a user have an IP address -- whether or not it's actually legit.)

    • Re:Hmm.. (Score:3, Informative)

      by MyHair ( 589485 )
      There's no login or password to publish data on Freenet. Sites are inserted with private/public key combinations. As long as you never let your private key out in the open no one should be able to impersonate you.

      It is possible to publish data without strong crypto (KSK keys, I think), and those are vulnerable to spoofing, but it also makes for a convenient anonymous feedback system.

      (IANACryptorapher)
    • Since Freenet is really about anonymity, I don't think there's going to ever be any authentication happening. That would kind of kill the anonymity model.

  • It seriously needs it.. (Score:2, Interesting)

    by iamsure ( 66666 )
    I just installed .52 and boy, is it unusably slow.

    Two minutes to load the WARNING page in front of the main 'search engine' of sorts that it has.

    Its worse than being on dialup. I'm all for the anonymity, but I'm on broadband, and it CRAWLS.

  • Make Freenet Free! (Score:3, Interesting)

    by Carl ( 12719 ) on Sunday July 20, 2003 @06:15PM (#6487143) Homepage
    Seriously. It is a bit ironic that the Freenet project doesn't run on a free system like Debian GNU/Linux. So there is an effort underway to Free Freenet! See the developer mailinglist archive [freenetproject.org]. Please donate (Matthew Toseland - Toad - is the "Official Codemonkey" of the Freenet Project).

    • Blame Kaffe - not Freenet (Score:2, Informative)

      by Anonymous Coward
      From what I can see the project does try to stick to Java APIs for which free implementations are available, but it does not hold itself hostage to their bug-fixing schedules. If Kaffe can't run Freenet because it isn't meeting its API obligations, then blame the Kaffe team and pester them to fix the problems, don't blame the Freenet developers, or expect those that donated money to Freenet to have it spent debugging Kaffe.

      Freenet is about Freedom of Communication, not Free Software. Just because there

      • ...does not mean that Freenet should spend its resources advancing the Free Software/Open Source agenda at the expense of its own.

        Well, the problem could have been avoided in the first place by choosing not to develop Freenet using Java. There's nothing so special about java that warrants abrogating your freedom. Seems like an easily avoided goof to me. Probably someone knew java, wanted to learn java, or some silly thing, and now we're stuck w/ a dependancy on non-free software. Too bad, because now

    • Re:Make Freenet Free! (Score:4, Informative)

      by John Hasler ( 414242 ) on Sunday July 20, 2003 @07:24PM (#6487461) Homepage
      > It is a bit ironic that the Freenet project
      > doesn't run on a free system like Debian
      > GNU/Linux.

      Package: freenet-unstable
      Priority: extra
      Section: contrib/net
      Installed-Size: 1532
      Maintainer: Robert Bihlmeyer
      Architecture: all
      Version: 0.6+20021221-1
      Depends: kaffe (>= 1:1.0.6-4) | java-virtual-machine, adduser, debianutils (>= 1.6), net-tools, debconf (>= 1.2.9)
      Conflicts: freenet
      Filename: pool/contrib/f/freenet-unstable/freenet-unstable_0 .6+20021221-1_all.deb
      Size: 1273386
      MD5sum: f1e9f4ae9949f77f618bd1ff6d7a5220
      Description: A peer-to-peer network for anonymous publishing (unstable branch)
      Freenet is a decentralised network of nodes designed to allow for efficient
      distribution of information over the Internet. Freenet's goals are resilience
      to censorship, and anonymity for producers and consumers of information
      through plausible denyability.
      .
      This package provides the software necessary to run a Freenet node able to
      take part in the network used by versions 0.4 to 0.6. Content can be inserted
      and retrieved with a commandline tool, or via the HTTP gateway with any
      browser.
      .
      This is a snapshot from the development branch.

  • plenty of room for future research/tuning (Score:4, Interesting)

    by BassZlat ( 17788 ) on Sunday July 20, 2003 @06:17PM (#6487155) Journal
    the nice thing about the current ng routing scheme is that there's plenty of room for research on how to tune it even further.

    Note: if you haven't read the article, this won't make much sense to you.

    For one, the number of reference points doesn't have to be fixed; if/when memory and cpu power allows us, we could have variable number of reference points per node. This opens the door to other decisions, such as whether we encourage clustering reference points. If yes, we add new ref points closer to others. If not, we remove a ref point the density within some keyspace interval gets too big. Another option is to add a new ref point whenever the n previous estimates turn out to be more than x% correct, and remove one if otherwise.

    Another direction to go into is curve fitting. If cpu power allows us, we could use various techniques of polynomial or Fourrier interpolation within the existing reference points to draw more accurate curve of time vs. keyspace. /me wanders if embedding fortran in java makes sense ;))
    • I find this project fascinating in that it would seem to be a solution on many levels, not just an app that runs over the internet. The way information is stored in redundant fashion, that growth ofd the network makes it more efficient AND more robust, that certain pathways become more specialized over time - it all strikes me as functioning very much like some grand brain. Of course it's "not there yet" but, unlike those who love to troll about how doggedly slow and unusable it is, I see research on this p

  • "remaining immune to the /. effect."? (Score:4, Interesting)

    by RPoet ( 20693 ) on Sunday July 20, 2003 @06:48PM (#6487281) Journal
    Freenet is NOT immune to the /. effect currently. Every time /. runs a Freenet related story, loads of new users seem to get on the Freenet and it just collapses, meaning response times go way up and many freesites become unreachable. I'm sure NGRouting will take care of this, but it's not honest to say it will help Freenet "remain immune" to the /. effect, because it's not immune.
    • When I first tried Freenet a year or so ago it defaulted to be a transient node.

      I noticed the lastest versions default to permanent node and the Windows version also puts itself in your startup folder.

      I don't think a few hundred or thousand transient nodes coming onto and off of Freenet would hurt it, but I think permanent nodes frequently hopping on and off will slow it down. I wonder why they changed the default to permanent?

      If I understand correctly, a transient node doesn't store data, respond to dat
      • As I understand it, there are several reasons for making nodes default to being permanent. For once, transient nodes don't help the network at all; they just leech. But more importantly, if you're transient, you lose an important attribute of Freenet: your plausible deniability. Everything in your data store has provably landed there on your request, and not (as for permanent nodes) perhaps because they were only routed through you.

        So let's just wait and see if all these new non-permanent permanent no

  • The next level (Score:4, Insightful)

    by mcrbids ( 148650 ) on Sunday July 20, 2003 @07:08PM (#6487374) Journal
    FreeNet will have problems for the forseeable future because the average joe can't easily install it and make it work.

    Who will take FreeNet to the masses?

    In other words, who will make a simple, usable client/server program that works on FreeNet? (Think Napster/KaZaA/Gnucleus)

    Will it be KaZaA? BearShare? Will it be some Open Source project?

    How long until somebody with the right skill set takes this to the "next level" so that it's actually usable to people other than geeks?

    • Re:The next level (Score:4, Insightful)

      by Fred Ferrigno ( 122319 ) on Monday July 21, 2003 @01:24AM (#6488841)
      Meh. The actual problem with Freenet is that there just isn't enough worthwhile content on it. I've shown my friends Freenet and babbled on about how cool it is that it's totally anonymous and all that jazz, but the first thing they ask is "how fast can I download stuff?" Of course, by "stuff", they typically mean all the sorts of things the MPAA and RIAA don't want you to have.

      So I have to explain that, well, there isn't really any "stuff" on Freenet at this point, and frankly if there were, it'd take forever and a day to complete, if you managed to find a node with all the data. But, like, there are all these sites that basically just link to each other, though occaisionally there's a site with some Dilbert cartoons that don't load. Oh, and did I mention browsing Freenet sites makes your $50/month broadband feel slower than a 14.4 modem?

      OTOH, I'm all for the concept of Freenet. Every major release I set up a node and run it for a few days to see if it's gotten any better, but I end up shutting it off.

      • Well no wonder it sucks! (Score:5, Interesting)

        by Myself ( 57572 ) on Monday July 21, 2003 @05:41AM (#6489345) Journal
        Every major release I set up a node and run it for a few days to see if it's gotten any better, but I end up shutting it off.


        You're part of the problem! The reason Freenet sucks for a little while after each release is that there's a huge influx of empty datastores joining the network. The network bounces back pretty quickly, as data gets passed around and as routing tables hone themselves, the network gets a lot better.

        Then a day or two later, you and 90% of the other slashdotters drop off, and leave holes in everyone's routing tables. All the contribution that your nodes were just starting to make, gets undone. All the copies of content that got replicated into your datastores vanish. All the routing optimizations that were just sorting themselves out get broken again.

        Tourists hurt the network. If you're judging Freenet based on it's performance the day after a slashdotting, you're not getting a full or fair picture. Come back and stay a while! Let your node run for a week and I think you'll be impressed.

        When they say Freenet is slashdot-resistant, they refer to content within the network. Any piece of data, be it a single file or a whole freesite, will simply propagate more as more people request it. The network itself definitely labors a bit as empty datastores dillute it. The best way to improve Freenet's performance is to encourage those tourists to stick around, so they and the network will benefit the most.

  • Freenet not a panacea (Score:4, Informative)

    by acceleriter ( 231439 ) on Sunday July 20, 2003 @07:08PM (#6487375)
    A "rights" holder knows the freenet key of certain material. Can the holder not hust write a script hop onto Freenet, request that key (and only that key), and fire off C&Ds to all the ISPs whose allocations include the addresses that respond? Seems simple enough--even with blinding of requests, the intellectual "property" holder can still point to the nodes that respond as having distributed the material--just as the exit server from Mixmaster (or freedom.net, before it became a casualty of 9/11 hysteria), etc. is vulnerable to legal attacks.

    This might be able to be foiled with some kind of chaffing in which nodes respond even if they don't have a piece of the data in question, but that would introduce more inefficiency.

    In particular, those who are "willfully blind" to infringement losing safe harbor provisions, I don't see how Freenet will survive as a means of propagating "questionable" material. And since that's it's raison d'être, then it probably won't survive at all in the U.S.

    • Re:Freenet not a panacea (Score:2, Informative)

      by Anonymous Coward
      ..can still point to the nodes that respond as having distributed the material..

      You can't prove whether those nodes were sending you the material (thus hosting it) or simply forwarding it from another node.

    • Re:Freenet not a panacea (Score:2, Interesting)

      by edheil ( 38857 )
      The trick is that by requesting the key, the person is actually propagating the material.

      If you request a key and my node hands you that file, there is no way for you to tell whether I had that file on my machine already and just sent it to you in response to your request, or whether my node went out and got that file from ANOTHER NODE in response to your request, and then passed it on to you, caching it on my node in case of further requests.

      In other words, by trying to 'police' freenet in this fashion,

    • Re:Freenet not a panacea (Score:4, Informative)

      by SWroclawski ( 95770 ) <serge@@@wroclawski...org> on Monday July 21, 2003 @05:29AM (#6489322) Homepage
      Your argument, if I understand it, is that given key A, then you find find the nodes that have it and shut them down.

      On Freenet this becomes a non-trivial task.

      First- all communication between nodes is encrypted. You'd need to do a real time decryption of the communication in order to spy.

      Secondly, nodes will often respond even if they don't have the data- that's the point. Even with NG routing- it's still onion routing. A node that responds that it has a peice of data may just be lieing. And by requesting the data in the first place, due to agressive caching- you're spreading the data across the network.

      As to then shutting down the nodes- you'd have to shut down nodes in places all over the world.

      Lastly, you could just make a second copy of a given data, new key and then then your plan is foiled.

      You should really read more of the Freenet docs- they explain all this.
  • I see contradictory claims. Help me out here...

    From the current announcement: It could even make Freenet faster than the World Wide Web in many circumstances.

    From the Freenet FAQ [sourceforge.net]: While it is unlikely that freenet sites will ever load faster than regular websites, it does adapt to sudden surges of visitors (which will often occur when relatively unknown sites get linked to from a big site) better, and high download speeds for big files are feasible too. Just don't expect very low latency.

    I'm about to

    • Re:Can it really be faster than WWW or not? (Score:4, Interesting)

      by QuMa ( 19440 ) on Sunday July 20, 2003 @09:09PM (#6487941)
      The text in the FAQ is mine, I don't think Ian is claiming freenet will get better throughput/latency for browsing random websites, however freenet can be faster for downloads from websites, websites with flash animations or big applets, those kinds of things. (At current the anonymity filter (a piece of code that filters potentially anonymity-compromising parts from freenet websites) will remove all plugins and applets, but we're talking middle to long-term future here).

      Basicly, freenet latency is bad, freenet throughput is good. (and freenet reliability is different :-))

  • meta data? (Score:3, Insightful)

    by DrEasy ( 559739 ) on Sunday July 20, 2003 @09:57PM (#6488135) Journal
    I've been reading about Freenet, and I'm trying to imagine how a potent search engine could be implemented on top of Freenet. Ideally it'd be great to use meta tags and such to index pages, but then how do you find the files if you do not know their keys in the first place?

    Yes, I have heard about Frost. As far as I understand, it's some sort of anonymous newsgroup. I guess a search engine could harvest the keys posted on Frost, and index them after retrieving and analysing the content and possibly the meta tags. But then the question becomes: how do you host such a search engine anonymously? Aren't you liable/vulnerable if your search engine is known to help you retrieve questionable content? Can't Frost be attacked ultimately for that same reason? Or is it distributed/anonymous? Am I missing something? Should I RTFA?


  • $10 a month to freenet and get all the music and movies you want,

    Or pay the RIAA $100,000 dollars per song.

    I think we dont have a choice but to make the logical business decision just like the RIAA made the logical business decision to sue 60 million people.

    Here you go, Subscribe now FreenetSubscription

  • Freenet: just a few notes (Score:2, Informative)

    by Anonymous Coward
    1. Freenet does work, its slow, but it works, I run it on dialup, all you people with bband stop moaning.
    2. Whatever connection you use give it time to integrate into the network.
    3.Stuff you may not agree with can and probably will be stored on your node.
    4. You cant be done for 3. Unless certain western goverments get really upset with freenet users.
    5.Download it. Run it. Leave it as long as you can. Repeat. Eventually it will work ok.
    6.Remember its worth it. Support this project you might need it.

  • This routing has its problems. (Score:3, Interesting)

    by leuk_he ( 194174 ) on Monday July 21, 2003 @03:44AM (#6489108) Homepage Journal
    1. Its theorethic. The original freenet concept contained simulation to show that it should work. I am missing this here.
    2. It does not fit really well in the freenet sources. In the current freenet implementation the network layer and routing layer are split. Unless you develop it yourself this will not be implemented in freenet (soon).

    DNF: estimate if they are legimate by estimating their time. This does not work on a saturated network. And freenet is always (by design i think ) full.
    There are some asumptions here that do not work. Also there will be things in freenet that will try to hide the location /no hops it took because it leads to security problems.

    Inherited Knowledge:
    Make nodes learn faste by assuming some kind (vague!) of trust between nodes. read: create trust by an estabished node and new (unreliable?) node. This is against the freenet paradigma and creates all kinds of security problems. This kind of thing should not be implemented in freenet where the 1st priority is security.

    The only positive thing this article is suggesting is to time the data and so optimize the flow of messsages according to the internet structure. In freenet this is an implementation problem.

    There were more of these kind of suggestions on the freenet tech mailing list. I unsubscribed it (too much spam, too much interesting ideas from people who had no clue)

    If you write such articles please investigate other p2p solution as well! (gnet/gnunet india network and many others.)

  • Lets see if we can use them all...

    Freenet is now being 'fixed' like a leaky faucet is fixed.

    The RIAA wants the digitial audio/video market 'fixed' like a crooked horse race is fixed.

    With the new Freenet the RIAA is about to be 'fixed' like your dog at the vet's is fixed.

    I think that about covers it.

Slashdot Top Deals

Every cloud has a silver lining; you should have sold it, and bought titanium.

Close