Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Networking The Internet Technology

Continuing the Distributed DNS System 77

bs0d3 writes "Last year, piratebay co-founder Peter Sunde gathered coders to begin a decentralized dns system. This is a direct result of the increasing control which the US government has over ICANN. The project is called P2P-DNS and according to the project's wiki, this is how the project is described: 'P2P-DNS is a community project that will free internet users from imperial control of DNS by ICANN. In order to prevent unjust prosecution or denial of service, P2P-DNS will operate as a distributed and less centralized service hosted by the users of DNS. Today the project continues, barely. A majority of interest shifted to namecoin once the idea was realized, but coder Caleb James DeLisle continues on the first project. So far he has DHT nodes and routers worked out, and awaits help on his IRC channel whenever volunteers are willing to join."
This discussion has been archived. No new comments can be posted.

Continuing the Distributed DNS System

Comments Filter:
  • by MightyYar ( 622222 ) on Tuesday October 18, 2011 @08:07AM (#37748872)

    I know they are the "pirate party", but there needs to be some way to protect trademarks. As much as I like the dark underbelly of the internet, I also like being able to use online banking and such without having to remember something like the_REAL_bank_of_america.com

    • by bfree ( 113420 )
      What do you propose is done about apple.* or any other (generic) words which are trademarked only within specific fields of endevour? Should smeg.* really be out of bounds for all dwarfers because somewhere in the world some company has a trademark on it for some specific use?
      • There's over a hundred years of trademark law to handle these disputes, and in general it works well. Apple.com could have gone to The Apple Growers Co-Op or some such if they got there first, but applecomputer.com is a clear trademark. This is a solved problem, with only a smattering of cases that cause real issues (Nissan being one).

      • This is safe for work [smeg.com]

    • by Hentes ( 2461350 )

      Trademarks only apply to a specific area of the market. Internet is used by all kinds of companies and people. Trademarks should not determine who gets a domain name.

      • Trademarks should not determine who gets a domain name.

        I don't think it is a helpful DNS system that doesn't incorporate trademark law. The only people it helps are domain squatters. Trademark certainly helps companies, but it also helps consumers. It's not desirable to have some random guy selling software with the domain name Microsoft.com just because he got to it first. I mean, we'd all learn to live with it, but I certainly don't see any system without trademark protections gaining ground on traditional DNS. Who the heck would adopt it? Certainly not any c

        • I think free market could handle that. If a name is important enough for a company, they will buy it from its owner. In fact, domain names should be sold on auctions. Also, what I was trying to point out that sometimes several entities from different fields own the smae trademark. But the biggest problem with that is that it gives companies an unfair advantage as they are the ones controlling the majority of trademarks. The current system encourages excessive IP herding, as it is a requirement to get domain

          • If a name is important enough for a company, they will buy it from its owner.

            You are describing the existing system. The only time trademark interferes with this in the current system is when someone tries to use a trademark in commerce. I could set up a site called "microsoft_sucks.com" and be just fine. But if I started selling software from that site, then I would probably lose a trademark case.

            But the biggest problem with that is that it gives companies an unfair advantage as they are the ones controlling the majority of trademarks.

            I don't know that I agree that this is a problem. Trademarks are meant for commerce - it makes sense that companies would hold most of them. Under the current system, an apple-selling coll

            • by Hentes ( 2461350 )

              I don't know that I agree that this is a problem. Trademarks are meant for commerce - it makes sense that companies would hold most of them.

              Yes, trademarks are meant for commerce, and the web is not a commercial thing, but a general communication medium. I don't have a problem with companies holding trademarks, I have problems with domain names seized based on trademark claims.

              My concern is where someone acquires "microsoft.p2p" and then starts selling software. Or "bankofamerica.p2p" and starts offering banking. It's confusing and unnecessary, and only aids fraudsters - not companies or consumers.

              I'm not sure what you meant...you have a problem with TLDs?

              • Yes, trademarks are meant for commerce, and the web is not a commercial thing, but a general communication medium.

                OK, but trademark law applies to other communication mediums such as paper, TV, etc. The internet is unbelievably cool, but it is hardly the first "general communication medium".

                I'm not sure what you meant...you have a problem with TLDs?

                Not at all. I was using "p2p" as an example because that is what the distributed DNS system uses.

  • by Anonymous Coward

    What kind of genius runs his webservers on port 82? Thanks, now a fifth of the readers can't see it due to firewalls.

    • by Anonymous Coward

      What kind of Slashdot reader doesn't have immediate access to an outside proxy?!?

    • by Anonymous Coward

      One of my co-workers does. He started on 80. had to pay extra to have his service re-enabled. Then went to 81; succeeded for three years before his ISP found that. Paid extra to re-enable service. Next, tried port 82. It's been a year and a half or so and he's still runnin' strong. lol

  • by elrous0 ( 869638 ) * on Tuesday October 18, 2011 @08:11AM (#37748898)

    Don't worry, I'm sure the U.S. will just declare this a terrorist movement and start actively blocking all alternative DNS servers by IP. Then we can get into the escalating you-jailbreak-it-we-find-a-way-to-stop-the-jailbreak-you-jailbreak-it-again game that seems to be the inevitable result of a conflict between those who want total freedom and those who want to stifle freedom within their own self-serving carefully defined boundaries.

    • Re: (Score:2, Flamebait)

      by cavreader ( 1903280 )
      While there is a lot to criticize about the US government their administration of the DNS servers has been pretty solid. Maybe you would like to transfer control to the UN? Most of those fuckwits need their aides to show them how to access their e-mail and Twitter setup and do not have the slightest understanding of what the DNS actually does.
      • They have done a good job over the years the primary issue is lately they have been using that control to assume all non country tlds fall under us law and summarily removing domains without any recourse of procedure. The legislature needs to specifically protect the DNS system from us law enforcement. Short term a presidential order would cover it. Require a judges order in the country that the domain is registered in.

        • Require a judges order in the country that the domain is registered in.

          .com, .org, .net, etc domains are registered in the USA and so come under USA jurisdiction. Your nation has a national registry. Use it. Much better to have many national registries than one monolithic central one whether controlled by the USA or the UN (better yet would be a decentralized system, of course).

  • by John Hasler ( 414242 ) on Tuesday October 18, 2011 @08:18AM (#37748940) Homepage

    ...to the UN, which would hand DNS over to something like WIPO. At first glance Namecoin looks great. Let's try to make it work. We need to get it up and running before major governments start outlawing what they will call "rogue" DNS.

    • Re: (Score:1, Insightful)

      I haven't tried namecoin yet, is it responsive enough to host real web applications on it?

      One other thing google didn't answer, in bitcoin, minters are supposed to also get money off of transactions facilitating a means to keep up with minting costs once all block are generated. Isn't that (minting costs + DNS resolution/hosting costs) going to be disproportionate when dns traffic (lookup requests) get very high if namecoin succeeds?

    • by Xest ( 935314 ) on Tuesday October 18, 2011 @08:51AM (#37749270)

      Yawn, I've said it many a time, anti UN whinging about this is just stupid and demonstrates massive ignorance. The UN already objectively handles this sort of thing well with it's management of international telephony (ITU), international maritime standards (IMO), international air standards (ICAO), international post (UPU) and so forth. The fact you apparently don't realise this is a testament to the great job they do, have you never wondered why you can easily send post to different countries with no wonder about htf it's going to get there? have you never wondered why there's no problems with planes flown by people from different countries with different cultures, speaking different languages crossing tens of borders within a single flight? have you never wondered how painlessly you can make a phone call abroad despite the plethora of different national telecommunications laws, concepts, and technologies between countries?

      The UN is perfect to this role, because unlike the US it means no single country can enforce anything, it means consent is required of all member nations to push through things like web blocking. That means no more arbitrary US censorship with ICE, no more arbitrary effective shut downs of other country's companies corporate sites because some Texas patent troll court wins an injunction the victim never knew was filed against them, no more ability of redneck states to shut down the likes of Antigua's online gambling industry domains simply because of their own ass backwards moral standards. That's of course before you go into the drama of the new buy your own TLD plan which destroys the hierarchial structure of DNS, is technical idiocy, but was allowed through because it means far more cash for ICANN's staff and directors to pocket.

      Global consensus in a world with such vast disagreement meaning controversal stuff inevitably finds itself at least one veto, whilst mundane stuff that's pretty essential (like changes relating to DNS Security) passes easily is far more sensible than everything goes P2P, or everyone being forced to adhere to the lowest common denominator state of moral standards in the US.

      As for WIPO, would now be a bad time to point out that WIPO worked well like this originally too, with poorer nations vetoing over the top IP protection for the pharmaceutical industry so that the suffering amongst their populace could afford medicine and the like too, but as a result of this, primarily the US, pushed this sort of thing into a new organisation - the WTO precisely so it could bypass the fairness that WIPO originally offered?

      Yes, that's right, the UN isn't the problem, a minority of countries like the US is, which is precisely why it shouldn't retain control of ICANN. The only parts of the UN that don't work well are the parts that aren't truly representative of the global community - the likes of the security council, the WTO and so forth, though even these are still better than organisations controlled unilaterally like ICANN. This is why ICANN should be moved to a representative UN organisation like WIPO was before America gimped it, like the ITU, ICAO, UPU etc. thankfully still are

      • by Infernal Device ( 865066 ) on Tuesday October 18, 2011 @09:36AM (#37749756)

        Yes, the UN generally does an OK job, except for that bit where they want to censor speech that makes other parties feel bad.

        http://www.nydailynews.com/opinions/2009/01/14/2009-01-14_unacceptable_censorship_the_united_natio.html [nydailynews.com]

        There is not one common standard for "defamation". An atheist might say that it is impossible to "defame" a religion, since they're all made up anyway. A hard-line Christian or Muslim might conclude that any criticism whatsoever was defamation. Additionally, the freedom of speech in member countries is not synchronized in the least. Consider that in the UK, a newspaper can be ordered to not publish certain articles about individuals, which is practically inconceivable in the US. German courts have ruled that the names of criminals cannot be published alongside their crimes, regardless of the fact that they actually committed such crimes (not "may have committed", but "actually did commit").

        Where does it stop? Once we allow the UN the toehold in determining what is acceptable speech, where is the line that cannot be crossed?

        • by Xest ( 935314 )

          Right, but that's precisely my point. You're referring to a section of the UN that relies on simple majority vote, rather than the sections I pointed to which rely on international consensus to operate.

          I quite agreed with you in my original post that the sections of the UN that don't work are those where consensus is not required and where vested interests can succesfully pursue their agenda like the US does with the WTO.

          The problem with your argument is that you're then extrapolating this to imply the whol

        • by sznupi ( 719324 )

          German courts have ruled that the names of criminals cannot be published alongside their crimes, regardless of the fact that they actually committed such crimes (not "may have committed", but "actually did commit").

          The real goal doesn't need to be, typically implied by critics, "protecting the criminal" and such. This can be also easily about protecting random others who will get caught in the debris. Kinda like what Xest points out, "in the article you linked there's a fair argument that condemning religious hate speak has the goal of preventing unnecessary violence in the world"

          Opposition to such anonymity perhaps partly stems, also, from "traditional" outlooks at punishment... but remember, those were formed whe

        • by sznupi ( 719324 )
          Generally, Germany tries to be (with mixed results) quite cautious, vigilant about things which can negatively brand a group of people by some vague association. You see, they had a bad experience with such practices, in the first half of XX century.
          For example, AFAIK, German authorities don't really follow the statistics about ethnicity, skin colour, "race" of their population, they don't really know how it's distributed (does Berlin has, say, 100k or 300k black people? Who knows; at most one can tell, I
      • People also forget that it wasn't ICANN that let ICE seize those domains, it was VeriSign, who administers the .com domain.

    • by ledow ( 319597 ) on Tuesday October 18, 2011 @08:52AM (#37749282) Homepage

      The problems I see with namecoin appear to be:

      1) You have to buy the name-IP links. This is usually by crunching a certain amount of numbers for either bitcoin or namecoin. To me, this says that he who has the most computing power (i.e. government / banks / spammers) wins all the interesting names.

      2) Because they have monetary value (and are created by it), you can potentially run into problems when transferring them to someone, or receiving - unwittingly handling stolen goods, etc.

      3) The above = first come, first served. Yes, DNS has this problem too but first-come, first-served on something that is starting from scratch seems stupid. Why is there no thought yet to integration with existing DNS? The value of the DNS system is NOT in the management of it, but in the data included in it. Without that data, .namecoin is just as bad as any other .obscure TLD.

      4) The name links are purely one name to one IP. No load-balancing, mailservers, SPF, DNS tricks, etc. This I would consider to be a HUGE disadvantage. If you're going to replace DNS, then you need to replace DNS. Not go back in time to pre-DNS days and implement only the single, simplest job that it does and which would be insufficient for huge swathes of the Internet. Hell, even if they'd just set it so that was the nameserver it linked to (which could then answer more questions about the subdomains, other records etc.), it would have been infinitely more useful, but that's NOT what it appears to do.

      5) It's all a bit convoluted. Patches to this and that, here and there.

      It seems to me that a P2P DNS alternative should act as nothing more than a P2P lookup of the already-existing and conventional DNS data. If enough people think that www.example.com resolves to 1.2.3.4, that's what it should resolve to under P2P DNS. There's no reason that a trust / reputation system couldn't do this, and still be secure to those people who trust nobody.

      Sure, you can have other TLD's of your own choosing for purely P2P services, but what people want is a replica of current DNS under a different backend protocol, not a half-assed replacement.

      I still need to look up google.com much more often than I would anything.namecoin and there's no facility to account for that currently. And, what the hell do you do about DNSSEC? If that comes first, you lose.

      • It seems to me that a P2P DNS alternative should act as nothing more than a P2P lookup of the already-existing and conventional DNS data. If enough people think that www.example.com resolves to 1.2.3.4, that's what it should resolve to under P2P DNS. There's no reason that a trust / reputation system couldn't do this, and still be secure to those people who trust nobody.

        The issue is how you define "enough people" such that it's not trivially exploited by having a single guy rent a bunch of proxies. CPU tim

  • i've mentioned this before, and it's worth mentioning again: the p2p-dns project's goals appear to be to create a separate ".p2p" top-level domain rather than to provide a complete distributed DNS replacement. that's just simply not ambitious or useful enough. we need a replacement where you put in the alternative DNS into your system and it takes over and goes seamlessly from there. you then rely on "crowd-sourcing" from the intelligent people just like cloudmark do distributed spam filtering to check t

    • Well, from a technical standpoint if they get .p2p to work, it's fairly trivial to then change it to work for all domains.

      The reason that these projects use their own top level domain is so they can avoid spending time on resolving name space conflicts.

  • Namecoin = stupid (Score:4, Interesting)

    by GameboyRMH ( 1153867 ) <gameboyrmh.gmail@com> on Tuesday October 18, 2011 @08:59AM (#37749356) Journal

    Namecoin has a first-come-first-served system of domain ownership with no oversight. That means it will only ever be popular with Internet libertarians, because nobody else would want to touch such an anarchic hellhole. It would never ever be practical for business. Ever. And .onion is technically better so it's not even useful for uber-geek tinkering.

    • That they are using the same method of "earning" a domain name makes me about as uncomfortable as I am with bitcoin. Additionally, it will become near impossible to purchase a domain name in the future, as the difficulty in mining namecoins rises.

      I feel like their project is missing the point - the issue is censorship and bypassing it, rather than purchasing names.

      • by Zan Lynx ( 87672 )

        The issue that is solved by "buying" names is spamming.

        A DNS replacement that had no cost per name would be drowned by people running scripts to register {/usr/share/dict/words}.com

    • by Hentes ( 2461350 )

      The Internet became what it is today because of its anarchy.

      • What? The Internet is very hierarchical. The only reason the US government doesn't have sole control over DNS is national TLDs. Darknets that run on top of the Internet are another matter. That said, the only part of the Internet that actually needs any hierarchy is DNS.

  • Could convergence be implemented for DNS resolution as well? Could you resolve a domain through the standard DNS hierarchy, through a p2p net, and through another means then take the most likely value from what you get back? Seems it would offer graceful degridation.
  • "We'd like you to work hard to spread free entertainment" isn't a compelling cause. Who would have guessed that.

  • instead of using a distributed dsn service of questionable reliability why not just install bind9 on your computer and it will be its own dsn server. anyone with Linux or a mac could do this and as for windows users they could run say damn small Linux or another tiny Linux variant in a vm with bind9 dsn server installed on it. and route your dsn requests to it?

You are in a maze of little twisting passages, all different.

Working...