Slashdot Mirror
Pirate DNS?
"What I had in mind was a system that was totally open and free, anyone can register any domain name or TLD. Since it was kept separate from NSI, this would include domains already registered there. In order to keep this DNS from leaking into the old DNS, I guess the servers could run on different ports. I also had in mind a modified version of the DNS protocol, where data could be distributed in a Gnutella or FreeNet fashion. This would eliminate the need to have root servers that had to handle a heavy load. However it would be vulnerable to spoofing and faking data.
One question I have though, since this is essentially a private network of computers (even though it is distributed throughout the Internet), can companies bring trademark law to bear on what I choose to name my computers on a private network? Can I call a machine microsoft.com on an intranet or on a network based on a naming scheme different from the standard DNS?
Another alternative would be to only use this system to add new TLD's and use the NSI DNS for .com, etc.. This would allow TLD's such as .god, .sucks or .anything-you-want to be set up without interfering with servers not recognizing this new DNS. This would avoid naming collisions with the old DNS.
My hope was that something built in this fashion would be controlled by the normal everyday users of the Internet, not by corporations. A distributed, cooperative naming system where hopefully less bullying could take place. Is this feasible? Or even desirable by anyone else?"
Look at all the top level heirarchies in USENET to see how well a distributed DNS system would work. Broken distribution, "not carried here", namespaces getting split into non-connected segments. Newgroup/rmgroup wars. Disputed meanings (Is .ca California or Canada? .la Los Angeles or Louisiana?). The only group names that work well are the big 7... which work because they have CENTRALIZED CONTROL.
There's 12 root servers, these servers handle all authoritative information for registered domain names. When you do an looked for a host, here's the process...
You'll typically have your ISP's DNS servers in your TCP/IP configuration. When you hit www.slashdot.org, you send an A query to your ISP's DNS server for "www.slashdot.org"
Your ISP's DNS servers think "well hell, I don't know who the hell www.slashdot.org is, let me contact one of the root servers to see who has authority over that domain."
Your ISP does an NS query on "slashdot.org" to see what servers are authoritative for that domain.. In this case it will return:
slashdot.org nameserver = ns3.andover.net slashdot.org nameserver = ns1.andover.net slashdot.org nameserver = ns2.andover.net
Your ISP's DNS server now knows who has authority for the domain and who can give you the A record for "www.slashdot.org"
Your ISP's DNS server contacts one of the authoritative name servers to do the A record lookup for "www.slashdot.org", basically saying "what's the IP for www.slashdot.org"
Andover's DNS server returns 64.28.67.48
Your web browser hits 64.28.67.48, and your web browser pulls up the site.
Please, please, please.. UNDERSTAND what you're talking about before posting things like this. You can't just "HIJACK" the DNS system, and NSI DOES NOT CONTROL IT.
The root servers are 12 servers placed all around the world, and no 1 corporation owns them. I believe it was the IETF that put them in place, who are the saviors of the internet world.
I just wonder if it would be possible to assert authority - what if nsi wanted to start using .music or something and a small segment of this 'pirate' system was already using it - wouldn't nsi be able to demand relinquishment of it with federally backed authority?
I don't see why. No law says that they are the one and only way to resolve names. NSI has only an old contract from the NSF and consensus to make them an authority. I doubt that music is a trademarkable term.
As much as this sounds a inviting idea, it would not work without some degree of control. Look at what happened to the alt.* groups on Usenet. A lot us usefull stuff, but even more junk.
The junk doesn't matter much. Most people see the junk alt. names, laugh if it's funny, and move on.
It is possable to configure a news server to NEVER accept a newgroup alt.this.is.totally.useless.nose again.
I don't know much about the GNUtella protocol, but it sounds like the wrong tool for the job.
There would need to be a signature and web of trust system to avoid scrip-kiddies and other nonsense in the system.
One way to handle registrations would be to sign and timestamp the message. Messages w/ timestamps too far in the past are dropped silently. Only the signer can change the registration info, registrations drop after an expiration if not renewed.
There are problems, but they could be worked out with careful thought.
There isn't much to be done about forcing propogation of an alternative DNS. It is a chicken and egg problem. Unless it is in use, there is no interest in supporting it, and there won't be a great attraction to use it unless it is widely supported.
The answer IMHO is to start with specialized use TLDs where it is not so important that the whole world be able to resolve it. Ideally, the domains registered will have a big 'cool factor' so that people will WANT to configure for it, and will pester their providers to support it for them. The next wave of domains would be those who already have a domain in a 'standard' TLD but want the 'cool factor' of an alternative TLD.
I agree that it is very important thet the TLDs remain consistant. The best way to do that is to include the 'stabndard' TLDs by reference (forward lookups to the NSI root servers), and to make sure that each alternate TLD has some sort of consistance mechanism in place and that it works. An inconsistant TLD won't stand a chance of becoming popular.
Hopefully, there will be public DNS servers that get configured to support the various alternative TLDs. ORSC is doing a pretty good job of that one so far.
As far as officially setting up new TLDs, that won't solve the problem at all. Those will still fall under NSI and ICAN'T and have all of the problems the current ones do. (Many corperates registering under com,net,org,biz,shop,sux,tm,md,dds,dvm,etc).
1.Court systems (in whichever country you are in) still exert juristiction. Moving to Sealand isn't going to help. While you (the new DNS people) may not have problems, people using your service still have to face liability in their country of origin. As long as the court systems seems to think that Domain Names are trademarkable, well, we're screwed.
There's nothing technical that can be done about that. However, simply not disableing a domain name UNTIL a court order is issued would be an improvement over what we have now. Ideally, the court order should go to the domain holder and not the DNS provider. That can be done if the DNS provider maintains that the registrant owns (or purports to own) the name and the provider simply propogates it as a service to the registrant.
2.You'd have to get the big players (most of the major ISPs) to go along. For legal reasons (see #1), this will never happen. So, if the people that provide 90% of the internet users capacity don't use you, what's the point?
Even windows allows manual selection of DNS. Nothing requires you to use the DNS provided by your ISP (I don't). We don't need a sweeping revolution, it can be one netizen at a time.
3.As to this, there is currently AlterNIC. They don't have many of the problems we associate with the current system, but guess how successful they've been?
You've heard of them, I've heard of them, everyone reading this has heard of them... Their TLDs resolve for me. Again, one at a time.
4.DNS requires a controlling entity. Distributed control isn't really a good idea. (Distributed operations are, though).
Why not? As long as a mechanism exists to prevent cheating it should be no problem. The only authority needed is private agreements between the participants, which could perhaps be informal. I need no legal authority to claim that www.microsoft.com = 207.46.130.149. You are free to take my word for it or not.
A key to avoiding trouble in that area is to stay away from the current TLDs. Let ICANN and the rest have .com etc.
Is thier any way we can cluster this sort of thing? Have no central database, thusly keeping problems like this from happening?
I realy don't know much about this, but it's worth a shot of asking.
http://www.xpurple.com
iDNS is a project to extend the current DNS so that we can use Unicode instead of 7-bits (F)ASCII code. It seems that the main servers are operated from East Asia.
__
__
Men with no respect for life must never be allowed to control the ultimate instruments of death.
GW Bu
IIRC, it was Alternic that tried this once. It wound up breaking a bunch of stuff.
Also, if it COULD be made to work, who'd admin it? How would they get paid? Would their service be any better than NSI?
Chas - The one, the only.
THANK GOD!!!
Chas - The one, the only.
THANK GOD!!!
This post wasn't a troll, this is a valid concern. Any thoughts on how spoofing would be controlled? Perhaps using public-key encryption and signed DNS records?
You could keep public keys on multiple servers as well. To make it all work, you'd need to verify a DNS record from server A with a public key from server B. With proper client-side caching and forwarding a'la Freenet (that's what a distributed network is all about, right?), it'd be pretty hard to spoof all of the elements used to validate a DNS record, wouldn't it?
Because the public keys can have larger scope (eg. covering a set of domains, rather than just a single domain or host) it's more feasible to have fewer public-key servers, with more energy put into those servers to protect security. That also makes it possible to have public-key servers of varying authority, as the number of PK requests should be far fewer than the number of DNS resolution requests. The "varying authority" comes from the fact that keys can be signed by others to verify their authenticity. Basicly, this amounts to building a web of trust with public keys, and then using that to secure the distributed-DNS network in parallel. In a sense, a public key acts as the registrar. Kinda.
Thoughts?
--Joe--
Program Intellivision!
First of all: It's 13 root servers, not 12
$ dig . ns |grep NS |grep -v \; |wc -l13
NSI is really split into two beasts, the registrY and the registRAR.
The registRAR is the people who auction off old domain names.
The registrY is the people who maintain the gTLD servers, and only two of the root servers.
The root servers are maintained by:
A: NSIB: ISI.EDU (California)
C: PSInet
D: UMD
E: NASA
F: ISC/NOMINUM
G: DOD/DISA
H: ARL (ARMY)
I: Univ Stockholm
J: NSI
K: London (LINX)
M: Tokyo University
Currently, the COM/NET/ORG are hosted on only a few root servers, namely, A, E, F, and G. There is currently a transition of moving COM/NET/ORG off of the root servers, onto gTLD servers.
These servers:
$ dig com. ns |grep NS |grep GTLD
(snipped)
K.GTLD-SERVERS.NET.
A.GTLD-SERVERS.NET.
M.GTLD-SERVERS.NET.
H.GTLD-SERVERS.NET.
C.GTLD-SERVERS.NET.
I.GTLD-SERVERS.NET.
F.GTLD-SERVERS.NET.
J.GTLD-SERVERS.NET.
All of the gtld servers are administrated by the NSI RegistrY.
And although the A server was once the master of the Root servers, it is no longer. The root servers use a stealth-primary (that would be one that you don't see) for distributing the root zone.
...And NSI does not control policy regarding the root servers, IANA does. If IANA told NSI to stop hosting all of the gtld servers by a certain date, NSI would cry, scream, kick ans wail (maybe even sue), but NSI would eventually have to give it up.
NSI does not control each of the root servers. If NSI told the root server operators to do a certain thing, like "All root servers must run NT", each individual root server operator could individually refuse to cooperate. (And I certainly hope that they would, NT is evil incarnate).
And while you're criticizing NSI, maybe you should look at the response time for the COM zone on all of the servers NSI administers. Depending on your connectivity, I haven't ever seen anything more than 100 ms -- Way better than the sucky E and G root servers, which regularly respond after 4000 ms. NSI is throwing a lot of money into making the gTLD servers more responsive, and accurate.
-- If you met me, you probably wouldn't remember me. I'm pretty hard to remember.
Part of the problem with the current DNS system is that the TLDs weren't developed with the idea that soon there'd be umpteen billion websites. None of the growth of the Internet was adequately planned for.
.com TLD goes to multinational companies, and can only be registered for their tradename and trademarks. They can only register domains for products (www.slimfast.com), trademarks (www.therealthing.com) and tradenames (www.exxon.com). If they don't own the words, they don't get to register them.
.com domain.
.xxx/.sex domain is needed, without a doubt. It's open for businesses and people, with no rules about names: you can register phrases, words, tradenames and trademarks that you own, what-have-you. No country/state/city codes are mandated. If you want them, you can have them (allowing www.goatsluts.redmond.wa.us to list prices for the locals).
.org domain goes to registered non-profits. The big user groups and so on will have to get a bit more formal. Non-profits that operate as a business (Oxfam & such) will probably also want to register their .com/.cc domain.
.org; national/state/provincial ones will be in the appropriate form of www.goatsluts.city.state.cc.
.coms applies to .orgs. .com/.cc entities. If they want to the TLD entry, they'll have to provide proof of international business registrations!
.personal.cc page, but you're at risk of losing it if you're successful. 'cause if you're successful, you should be a registered business (otherwise you take the far greater risk of having your ass nailed to the wall for tax evasion!)
.cc TLD, however they see fit (could be a government service; could be privatized; could be contracted to NSI, even).
.personal users aren't going to cough up big bucks.
.state.cc postfix).
MY PROPOSAL:
The
The country code TLDs go to big businesses with national registration. Same rules apply as for the
The state/province domains go to small businesses, ones that are not nationally registered. Same rules as above.
City/county-specific domains go to businesses as well. This allows franchises and such to deal with their local community.
You can't register a name that's being used more broadly than your use of it: you can't get a state name if someone has the national registration.
This exactly matches the real-world rules for registering your business with the government. There can be no domain name conflicts, because the government doesn't allow real-world name conflicts. Where there are, the domain solution will exactly match the real-world, government/court solutions.
New TLDs are created for other uses.
A
The
International non-profits get a plain
The naming rules of the
Note that a lot of community groups (ARSTechnica, PlanetNameYourGame) are profitable ventures and would be registered as
Where do personal pages fit in? As sub-domains of a country code. I suggest www.goatsluts.personal.cc. There are no naming rules: you can use tradenames, trademarks, phrases, words, whatever. And, no, companies can't shut you down for using a trademark or tradename: the "personal" subdomain makes it very clear that this is *not* a business page.
I'm not stuck on naming it "personal," but it does have to make it clear that the page isn't a business-authorized one. An internationally-recognized word would be good.
ISPs will be responsible for not allowing business to be conducted on personal pages. No ad banners, no shopping carts, no promoting one's business.
The enforcement rule: if someone wants your domain name and discovers that you're doing business, they'll report you to your ISP, and you'll lose your domain.
Ergo, you *can* sneakily do business on a
And, of course, if you do business using someone's trademark or tradename, then you're going to get into hot legal water when they discover you.
Most registrations will be handled by the country represented by the
The international domains will be handled by NSI.
The key advantage to all this is that it opens up the domain name space.
It sensibly restricts what names businesses can use, while opening up all possibilities for private users.
It eliminates camping: businesses own their trademarks/tradenames, and
It recognizes that non-competing businesses (ie. businesses in different states/countries) may want to register the same name (and differentiates them by the
It recognizes that big businesses own their names/marks, and that little businesses don't get to name themselves after a national/international business.
In short, it seems to work very well, and for that reason alone will probably never come to pass...
--
--
Don't like it? Respond with words, not karma.
I just wonder if it would be possible to assert authority - what if nsi wanted to start using .music or something and a small segment of this 'pirate' system was already using it - wouldn't nsi be able to demand relinquishment of it with federally backed authority?
While a noble idea, there are several major reasons why you would never be able to get it accepted:
In the end, however, you fall into the same trap virtually everyone does when attempting to "Reform" the DNS system. They make the assumption that names have connotation. That is, that there is some meaning to the name www.microsoft.com other than it's easier to remember than 207.46.130.149. The DNS system was designed, and SHOULD REMAIN simply a pneumonic (sp?) that makes life easier for machine identification. What we've loaded onto the DNS system is content location, something it's completely unsuitable for.
Fundamentally, I should NOT be typing in "www.microsoft.com" in IE if I want to look for Windows 98 crap. I should type in "Microsoft Windows 98". There should be no end-user mapping between content and DNS name. Content should be divorced from DNS completely, in the manner that DNS is divorced from IPs. Meta-searching and content discovery/cataloging need to be avanced to the point where honestly, the end-user should NEVER KNOW ABOUT URLs. Does then end-user know about IP addresses nowdays? No. Neither should they need to know about DNS names.
We need to fix cataloging and searching first, then the DNS problem will go away.
-Erik
There are always four sides to every story: your side, their side, the truth, and what really happened.
OK, let's say we take your proposal, and create the World's Best Registrar(tm). We have lots of nice, end-user friendly policies, responsive customer service, and we don't screw people over. In short, the polar opposite of NSI. So, everyone hears about you, decides you're cool, and we all switch over to you.
Heck, you're so nice and cool, and we all trust you so much that you go and create a whole bunch of new TLDs. We like 'em, and use them wonderfully.
Everything is hunky dory up until the point where you decide to let someone other than Mr. Gates' company register microsoft.com. Or even microsoft.xxx. Suddenly, a whole passel of MS lawyers show up at your doorstep, and demand that they get back the microsoft.xxx domain, because, damnit, they own the trademark.
Guess what? You're going to have to give the name back to MS. The reason is the current interpretation of law. You, as the service provider, are responsible for following the law, and the law states that MS has the right to the microsoft.xxx domains. You have to comply. Sorry, no way out. This isn't just in the US, anymore, since WIPO decided to essentially impliment the US trademark-on-domains philosophy into internation trade law. So you can't move elsewhere.
Fundamentally, right now the boundaries are set up for what we can and cannot do on our own - certainly, there is a whole lot of room to improve over NSI (and I'm certainly moving my business from them to someone else), but the current legal atmosphere limits what policies you can put in place for domain registration.
-Erik
There are always four sides to every story: your side, their side, the truth, and what really happened.
Alternic tried this back around 1996. Here's a link to a boardwatch article that discusses their system. It transparently handled regular Internic (now Network Solutions) requests as well as their own names/TLDs. They mentioned that you could get your own TLD for $98/year. How cool is that? :) I never actually tried changing my DNS servers over to theirs, just because it didn't seem to be catching on at all.
/ \
\ / ASCII ribbon campaign for peace
x
/ \
The domain name system is about dead. So many hose beasts have bought up all the good names and aren't even using them and you have all that fighting over new TLD's as if they really even matter. The majority of people I've watched don't even understand domain names and URL's. They go to the search engine default of their browser and click'n'search from there or find sites out of their bookmarks.
Would be an interesting experiment to create a distributed system that used encryption keys to lookup the IP of the machine. This encryption code could be either for the machine itself or for a certain user on that machine. Once you located the machine you wanted to talk to you could send your public key to it encrypted by it's public key and then carry on secure communications from there. By this system even if you knew the machines IP you couldn't talk to it unless you also knew it's key. I'm sure it'd need a lot of ironing out but it might fix some of the problems and prepare for a Net that is much less centrally controlled such as peer-peer wireless internetworking.
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
Not to have problems with trademarks, such a new DNS database should not have the same chaos of TLDs as today. My proposal is, such a system should have a smaller set of TLDS, with subdomains and subsubdomains, for different categories of domains. Like *.linux.free.soft.comp for all linux-software sites, *.rel.phil.cult for religious sites, and so on. A company can not, in most countries, claim any right to their trademarked name when it is used in a totally different usage domain, like Apple can not claim any right to the Macintosh name of fruits.
--The knowledge that you are an idiot, is what distinguishes you from one.
--The knowledge that you are an idiot, is what distinguishes you from one.
Recently I read a comment comparing the DNS to Usenet on the basis of where the renegades end up. It basically said that the creators of Usenet had the foresight to create not only the top level newsgroup for "proper" discussion forums, but also the alt newsgroup where anything goes. To my knowledge no trademark infringement under the alt umbrella has ever been challenged in court, and due to the nature of that part of Usenet, I don't think it ever could be. It is well understood that alt. is no man's land and on that basis, a sufficiently talented attourny could convince a judge that a mark used under the alt newsgroup does not dilute a trademark due to that understanding.
.com, .net, .org, .edu, .int, .gov, .mil, and ISO country code domains and leaves no place for all those registrants who would more appropriately occupy a .alt domain. If the DNS had started out with a .alt domain that then became a well understood DMZ, we probably wouldn't have had the domain dispute problems that we've had without it. Adding a .alt TLD now may or may not alleviate the problem of excessive litigation, depending on whether net users at large can embrace the new domain's intent. In the beginning it would have been easier to get users to understand something like that.
By contrast, the DNS traps everyone into the
As for usurping root service from NSI, it really depends on how widespread pirate DNS usage becomes before that first legal challenge. If the pirate DNS can proclaim loud and clear to all its users from the beginning that its subdomains are merely place names, like those found on a map, and can get a large portion of the Internet using its roots, then appropriate backing from organisations like FSF and EFF and such could get the courts to recognize domains as place names therefore nondiluting.
I like to play children's songs in minor keys.
"We're all sons of bitches now." --J. Robert Oppenheimer
I put together a concept a few years ago where those whoe run (or want to run) their own DNS would simply set it up as a root server by being authoritative for the "." zone. Within that zone, they put in the TLDs they want, and reference the name servers where they want that TLD data to come from.
The idea came to me as a result of TLD wars between different people who wanted to be the authority over the same TLD and had even already accepted registrations in those TLDs. I thought about who really should decide who is the authority for a given TLD (a number of different groups were trying to emerge at the time to do that). My thought was that it should be the open market, the people, the grass roots.
Roots?
So I coined "Grass Roots Servers" to designate the idea that each individual operator of a DNS server can choose for themselves what TLDs and who supplies the data. If you wanted to let NSI be authority for .com then you can (and I did). But you wouldn't have to if you didn't want to. If you wanted to leave .xxx out for a religious oriented ISP, great. If you wanted to hook into an underground source for the .mp3 TLD zone (now what what could that possibly be for? :-) you could.
The big stumbling block I saw was the difficulty in building a complete "." zone and keeping it up to date (servers do change) without destroying your selections in the process. So I gathered up as much data as I could get at that time (it is now out of date a bit) and built a web page CGI that would let you choose from known sources and it would build a "." zone file for you. It would also provide a means to save a page with hidden input fields that would come back with your original selections so you could regenerate the zone file again from your selections and new data.
It's open sourced (you can download source and the old database at the bottom of the page) because I wanted there to be many sources of info about TLD sources, just to make sure no one entity could grab control.
It is still online at http://grs.ipal.net/ . I have not updated it. If there is interest, I can resume it, or you can grab the code and data and go for it yourself.
now we need to go OSS in diesel cars
If you put it in the hands of a known agentcy such as the Free Software Foundation then it would be out of reach of corprate intrests.
In my view RMS and the FSF are IDEAL for running such a system. The pre-existing contempt for all the elements that give rise to the current DNS situation means the FSF would not cave to corprate pressures. Not at all.. they would be stronger as then exist in a state of battle.
It's a whole mindset thing. Basicly they allready hold contempt for corprate mentality so when corprate mentality comes along with "Our way or no way" the reply is swift and automatic "Go To Hell".
Anyway it dosn't need to be the FSF.. as long as it's a known agentcy with similer addatudes. Or form a new agentcy with RMS, ESR, BP and a few other advocates at the head. As long as it's a "known" in that we know it will behave to the benifit of the net and not the benifit of corprate culture.
I don't actually exist.
The point here is not that you can or cannot register domains with different people. The point is that the NSI/ICANN/other-large-corporate-entities have determined that the Internet is a Good Way To Make Money(tm), so they're going to make damn sure it stays that way. Since they also like trademarks, they're gonna want to enforce those rules as well. Since new TLDs would allow others to possibly register their precious trademarks, they either want to not have or severly abridge the ability to register names under this space.
Ergo, in the opinion of a lot here, a totally unbiased third party (who is capable of telling Monied Interests to shove off) is needed to run the central list of names and ccTLD delegations.
--------------------
As much as this sounds a inviting idea, it would not work without some degree of control. Look at what happened to the alt.* groups on Usenet. A lot us usefull stuff, but even more junk.
What's the problem? Why would you want to protect against it? People are free to use *ANY* lookup service they want.
The Internet does not = InterNIC + HTTP. It is what we make of it. DNS is just one protocol of the countless numbers we *could* be using.
Nothing.. *NOTHING* prevents, or dictates in *ANY* way that an ISP must use the global DNS system. THey just DO because it's the only one there, and because it seems to work for them.
The problem, of course, is that any 'pirate' (really bad term.. REALLY bad term.. it's not piracy in ANY sense) DNS system would have to get off the ground. That means ISP nameservers have to *chose* to use it. If they all chose to use it.. no problem. If they don't, it becomes useless.
I say the real answer is get off the whole damn DNS thing. It's just a tool to map IP addresses to names. How we've virtualized it and use it (abuse it) for things it wasn't intended for.
your web site will be *just* as good no matter what hte URL is.
Hacking up a DNS to make things go to where some local bigwig thinks they should go, rather than the registered site, is very bad policy.
.gov site, they wouldn't understand what the article was talking about.
Sure, it seems harmless on the surface, but breaking the basic URL mechanism could have unforeseen effects.
What if someone wrote an article about how wrong it is for porn sites to grab hits (likely from minors as not) from mistyped URLs? What if they used the whitehouse.gov/whitehouse.com thing as an example? Anywhere it was remapped so both go to the
BTW, I doubt you could get sued by whitehouse.com. They don't have any higher trademark claim to the domain than the US government does.
He wouldn't mess with the protocol! He's not that kind of egomaniac!
. not.unix.stallman
It would, of course, be: http://fsf.www.emacs.domainname.lignux.com.gnu.is
What's with the assumption there? I certainly wouldn't touch a DNS server that some random person told me about on IRC! The fact is, that by using the "official" servers I can pretty much guarantee that when people give me a URL or machine name that I will be able to successfully look it up; if I've using your rogue service, I could get anything. And I think that quite a few of the users of your ISP would cotton onto this as well.
I mentioned a "missed opportunity" in the subject. Here's what it was. We had a chance to completely BYPASS the entire DNS crap years ago. Where? At the web browser itself. Now, I'm going to use a word that a number of people won't like, and that's "keywords". What if in Netscape, when someone typed in "glorious shoes", it didn't try to DNS it, or search engine it, but to look it up at a keyword registry site? (Sure, okay, its a little too close to AOL for some.) But the idea is to put the power of name lookups into the browser itself, bypassing the DNS mechanism. If this would have been done, NetSol would have been somewhat moot. (But then again, we could have created a new monster.)
Of course, it isn't TOO LATE for this to happen, but there better be a central keyword registry or Netscape and Microsoft will fight with registries of their own. Fun fun fun.
BTW... if someone can pull off this idea, they can become a VERY VERY rich person. If this becomes the case, please hire me. :)
just occurred to me (probably due to lack of sleep). Imagine that a service similar to internic was established (I could do it, I got a spare 486 around here somewhere) that acts as a root domain server. I point my nameserver at it to feed me addresses. The root nameserver will check its own databases first and if it doesn't find a match, it will then query the internic root domain servers. This would allow me to use microsoft.com or any other domain that has already been taken.
Obviously, this has limited use when there's only one person using it, but nobody is FORCED to use the primary internic root nameservers like everybody does. A nameserver could use ANY root nameserver it liked.
Lets say I run a small isp, something around the size of AOL. I set up my nameservers to use the alternate root nameserver (still using the 486 of course). Then I decide that I don't like microsoft (can't think of any reasons at the moment, but I'll worry about that later). I therefore "register" microsoft.com on my root nameserver and now all the 18 million users using my isp will see the NEW microsoft.com. Microsoft still owns microsoft.com. The original microsoft.com site is very much in the same place it always was. Every other user on the internet is able to access it directly. But not my users. My users see a glimmer of a gentler, kinder microsoft. Perhaps a microsoft that promotes linux.
Microsoft will eventually catch wind of this.... maybe one of their employees prefers my isp over MSN and notices that the microsoft homepage looks funny, and reports it. After several multimillion dollar lawsuits against internic, they'll eventually discover that all my users are not using internic, but a different name service provider. Microsoft is not pleased. They want microsoft.com.
But they already HAVE microsoft.com, and despite the fact that they are now in debt greater than the US government, internic insists they STILL have it. At least the lawyers are happy. Microsoft attempts to get us to fix our problem, but our name service provider doesn't conform to any federal regulation. Nobody is forced to use my service any more than a user is forced to use windows on a PC. Of course, my 18 million users might start leaving in droves if they discover that I've been deceiving them all along, but then again... an internet without Microsoft... worse things could happen.
Lets say that all the microsoft lawyers go on a 6 month vacation and therefore microsoft doesnt' press any legal agenda toward reclaiming their domain from us for a while. I decide to set up another top level domain (.mine) for my personal name service. This way, someone can access microsoft.com.mine to get to the microsoft.com that I provide from outside of my isp service. This means that everyone on the internet now can access my domains even if they don't want to use my name service exclusively. (Its starting to look as if I might have to upgrade that 486).
But why SHOULDN'T they use my name service over internic's? After all, its just as good. So I go out to IRC and spam everyone to change their nameservers to point to the new root name server instead of the internic one. A few might be unwilling, a few might be confused, but the great mass of users will jump at the chance. I can also send out a large number of emails to all the isp's to use my name service instead of internic's. Promote the advanced technology of the 486 over whatever it is that internic is using. Chances are, it won't be too much of a stretch. Once I have a significant portion of the internet using my root domain servers, microsoft will gradually disappear from the web, and nobody would be the wiser. Oh sure, the news will be blasting it out to everyone every hour of every day, but ILOVEYOU and MELISSA were able to spread despite massive news reports, so I can't see how it will make any difference.
Internic was established to provide a single source of name service and was regulated by the government (to some extent) to provide those names. However, I know of no law which forces me to use them as my root nameserver, nor do I know of any law that forces me to provide customers with internic root name server access.
I'm sure I'm missing some key points in trademark law which would bite me in the ass before I got too far with this. But someone with the time and the resources could pull this off on a small scale and get away with it for a while and depending on how the laws work wherever that company is..... it might be viable. But with only a single 486.... very VERY slow.
-Restil
Play with my webcams and lights here
While I agree that the present primacy of NSI over .com, .org and .edu raises several problems, it is important to note from where NSI obtains its authority, and seat the blame appropriately. This solution may work itself out in time, once there is a non-NSI infrastructure capable of picking up the registry slack. Outfits like register.com are proof that sound competition is possible.
Now the trademark issues do not derive from NSI itself (although certainly some of them do), but from the tradmark laws, really stupid special-purpose legislation (the cybersquatting bill) and from limitations of the technology itself. Eliminating NSI from the mix would change none of these things.
The fact of the matter is that it is unlawful to use a mark in violation of trademark laws. Mere registration is not, by itself, an infringement, subsequent use of that registration will be. It doesn't matter who is registrar for the domain names -- the lawsuits will ensue. The anticybersquatter act further ensures this.
The problem is not with the law, either. Trademark law has developed reasonably over the centuries, and can readily cater to plural persons being able to use the identical mark. (ABC for a TV network, an unrelated pizza chain, an unrelated chain of liquor stores and countless garages and other small businesses; AAA for everything, and Acme for everything).
The difficulty derives from the fact that unlike a telephone listing, a domain name is unique. This doesn't have to be, although it is very convenient to have it be so, and technology could evolve, and maybe should evolve to permit plural acme.com's. See, e.g., an old white paper I wrote about five years ago on possible alternatives.
But the bottom line is this: NSI overreaches regularly, only because they can. They can only because NSF lets them, which it does because no one else was ready to step up to bat last time it was time to renew a contract. This is changing as we speak. And competitive registrars will probably intervene to protect their territory and prevent further overreaching, at least to some extent.
I see no reason to believe that another registrar, however well-meaning, will be better or worse than the status quo, absent a uniform set of black-and-white rules enforced by a meaningful authority.
Anarchy will make the problem worse, much worse, and not any better. Further, fear of the consequences of such anarchy will preclude the necessary critical mass to build to make the new top level domains possible.
I think we should rely on, and exploit, existing processes to hold NSI in check. This requires some patience, and perhaps the creation of some new technologies, but it can work. Other solutions proposed thus far seem only to introduce new problems, and probably no real new benefits.
This thing has been done already. Even the .god TLD has been done. There's EVEN been news stories about that (I don't know the URLs since it was a while ago, but I beleive CNN carried it).
Where are those now?
Sorry, but the established DNS network is there for a reason. Guess why? =)
Isn't it what the open root server confederation is about?
When I caught him going to Yahoo first and typing www.amazon.com, it opened my eyes. I've since learned that very many people use portals and indexed catalogues as namespace locators, even using them as URL entry forms.
My mind still boggles at this. I have a web hosting client, an attorney who is a bright guy. A while back he registered several related domain names and I pointed them at his site.
About two weeks later, he calls me and he's pissed, claiming I haven't done the job he's paying me to do! The domains don't come up, he complains. But they do! Just type it in your browser, I say. In frustration, I have him describe EVERY step he is taking, and discover that this is EXACTLY what he does, everytime - he goes to Yahoo! and types into the search field.
I am dumbstruck! This makes as much sense to me as putting on pants to take a shit! And this horrible glimpse of another reality dawns and hits me square in the face - he totally lacks distinctions with regard to technology. In fact, in a very real sense, technology=computer=internet for him, no distinctions. He doesn't distinguish the Yahoo! document from the browser it is displayed in from the OS running the browser from the machine running the OS. He can't. Not because he's stupid. He's never needed to. Or known he could. Now considering all of that, he has still managed to make this web shit useful for himself. Confronted with an undistinguished jumble of computer, monitor, browser, preset home page and a lot of text, he DOES distinguish the button labelled "Search", and it produces useful results for him. And it has worked quite well for many months. Until something he wanted couldn't possibly exist yet in the search engines. And he, in the year 2K, is a veritable technogeek compared to most of the population.
But the most stunning aspect of this was what it said about me! About the assumptions I make everywhere: in performing services for people, in delivering information to people, in my day to day conversations with people around me. About how very, very different the same thing can appear, depending on who is doing the perceiving.
"I will gladly pay you today, sir, and eat up
Sacred cows make the best burgers.
But I can't help but think that this is more a matter of typical micro$oft "we're gonna sue you into oblivion if you don't follow the word of bill" than actual law.
For instance... an example from real life...
When I was still in school I was on the admin team for the UNIX network. Now, at one point while I worked there, some drone from one of the pedantic majors (future marketdroid or mba type I think) went looking for information on the US goverment, and surfed over to... you guessed it: whitehouse.com. Well, little miss anal-retentive promptly threw a fit and complained to everyone who would listen... including a nasty letter to the school president. Eventually the order came down from somewhere above: We were to redirect whitehouse.com to whitehouse.gov in our DNS. We did, leaving a easy to exploit way to get to the real whitehouse.gov if anyone cared... as the OFFICIAL policy was not to censor net access (this was well BEFORE the RIAA began harassing schools to block napster).
So, were we, therefore, in violation of the law, for changing DNS on our own PRIVATE network (at a private school, not a state one)? Could whitehouse.com, had they gotten word of it, sued us over the remapping AND WON????
At home, I have microsoft.com remapped to www.kmfms.com so if any of the two other people who have access to these boxen suddenly become drones, they will be sent to a place they can get help. It is ILLEGAL, to change DNS???? On my own PRIVATE network???
That seeme totally idiotic to me, if it is true.
john
Resistance is NOT futile!!!
Haiku:
I am not a drone.
Remove the collective if
Imagine all the people...
We really can have a free world wide web (the way it used to be). Remember all those little banners you used to see a couple years back? "Corporate Web Sites Kill The Internet Dead" I thought the loonies who posted those at that time were paranoid. But they were 100% right. It's happening. Witness the multi-billion dollar corporations bullying private citizens and shaking them down for domain names. It's sick. But how can we stop the big money folks from destroying the new (let's call it FreeDNS) naming service?
We look to GNU and its GPL. That powerful license has kept free software free up until now. The basis of free software is "you're free to play with us, but if you do, you must play by our rules". The GPL says that if you want to use our GNU software, you may, but then you have to play by the rules in this license.
Imagine a contract for new domain name registrants. It will have them agree to play fair if they're going to play. Here's a sample of what I'm thinking of.
FreeDNS terms of service:
1) By registering a domain name with the FreeDNS service, I am agreeing to adhere to any and all terms in this FreeDNS terms of service contract. Where there is a discrepency between this and any other contract, this, the FreeDNS terms of service contract, shall take precendence.
2) By registering a domain name with FreeDNS, I agree to not challenge at any time the ownership of any FreeDNS domain name, neither existing domain names, nor domain names yet to be registered. I agree that any domain name registered by any person or organization containing any of my trademarks or other intellectual property.
3) I understand that if I ever do challenge, civilly or criminally, any domain name held by any other person, I forfeit my rights to participate in FreeDNS. Although I may legally prevent others from using my trademarks and other intellectual property in FreeDNS domain names, I understand that this shall preclude me from participating in FreeDNS.
4) By registering a domain name with FreeDNS, I am relinquishing any domain names such that I have gained control of through civil or criminal prosecution, or through legal settlements, or by coersion via threats of civil or criminal suits, to their previous owners.
5) By registering a domain name with FreeDNS, I am agreeing never to sell any FreeDNS domain name to another party for any fee or in exchange for any good or service or favor of any sort. I understand that any transaction is invalid, and such a transaction puts the sold domain name into the pool of registerable domain names. If it is discovered at any time that I sold a FreeDNS domain name, I make myself liable for any and all damages arising from the breach of this contract.
6) I agree that failure to adhere absolutely to this contract voids any and all FreeDNS domain name registrations that I hold, and that they shall return to the pool of registerable domains.
This does a few different things (and IANAL, so obviously, this would have to be beefed up by a real attorney). First, it says that if you want to participate in FreeDNS, then you can't try to take anyone else's domain name for any reason. If you choose not to participate, then you can sue to have, for example, microsoft.sux removed from use and removed from the pool of registerable domain names. However, if your company, for example, microsoft decides that it DOES want to participate, it will have to give back any siezed domains and play nice. So if my company, DerMarlboro Enterprises, registers dermarlboro.com, and some yahoo registers dermarlboro.sux or dermarlborosucks.com, I can't challenge that, and I can't even threaten the registrants of those names or else I will lose my right to participate. Plus, sale of domain names voids them, and voids all other domain name registrations by the seller, so the motivation to cybersquat is removed. You can't sell it. If you try, someone else can get it for free.
Personally, I think it should have been done this way to begin with. But when nameservice first appeared, who would have imagined the friendly bearded sysadmins who owned the domains SUEING one another because one owns sysadmin.com, and the other registers sysadmins.com? It turns out to be true. Corporate web sites really do kill the internet dead.
The DNS landscape is littered with aborted attempts to do this: Alternic, EDNS, uDNS and other good ideas that turnd sour. As of today the only efforts that still exist are ORSC (which can trace it's roots back to the original "new domains" mailing list 5 years ago ("newdom", see http://www.newdom.com/archive), name.space, TINC (http://tinc-org.com) and Adam Todds irsc/narsc/aursc stuff.
I'm biased because I am heavily involved with ORSC but I urge people to look at all them and make your own decisions. I did and have found the Todd and name.space do "not play well with others". TINC is an exception; they're cool and have a major clue. Where we disasgree witht them is TINC belives "no more than one TLD to a customer" and while we're not sure what that numebr should be, we know it's not one. So, we go our separate ways but work fairly closely.
There are a couple of errors in the orignal post I'd like to correct. First of all the venom directed at NSI is undeserved. NSI operates under a contract with the US departent of Commerce and has it's hands tied so tightly it's a wonder they can do anything at all. I'd like to point out that NSI has done more to help the alternative domain community than any other company to date.
So, I have to say NSI is not the great Satan here - the coopted US Department of Commerce is. Large three-lettered companies have spent almost a billion dolars to make sure no new tlds ever see the light of day and the DoC has it's strings pulled by these clowns - or hadn't you noticed that in almost two years of ICANN existance the only thing they've done is make big lawyers happy by implementing the UDRP that helps trademark owners ans screws the avrage domain owner. (see http://www.news.com/Perspectives/Column/0,176,459, 00.html) and have done nothing except talk about new tlds.
Worse, ICANN does not have the power to create new tlds! All they can do is make suggestions to the USG Department of Commece who actually control the legacy root zone that IANA used to own. You can verify this by reading the GAO's whitewash of ICNANs illegitimate birth where they state outright that "the DoC has no plans to hand control of the root zone over to ICNAN" at www.gao.gov/new.items/og00033r.pdf)
The problem is not one of gnutella or distributed whatsits, the problem is one of education.
New tlds registries exist, and some have existed for 5 years. Alternative root servers exist and can be used by anybody.
Forget Alternic and eDNS; they're dead Jim. They once enjoyed some resonable support but now exist as names only, haveing been sold to other people for the name value (such as it is).
In conclusion, there is really no need to reinvent the wheel. If you want to play in the new domain area and outside the government controlled root zone you can do that now by pointing your nameservers away from the legacy root zone.
There's more than one way to do this, but my favorite is to secondary the ORSC root zone; in this manner you become your own root server and save one level of lookups as your server now knows where all the tld servers are.
What's (very) important to understand here is by doing this you will still be able to use com/net/org but now will also be able to see new domains such as http://lighting.faq and http://free.tibet - it's not an either/or situation.
For more information look at these urls:
- ORSC Root zone: ftp://a.root-servers.orsc/pub/db.root also available via http://dns.vrx.net/tech/rootzone/
- ORSC website: http://www.open-rsc.org
- How to point to new root servers: http://support.open-rsc.org/How_To/
- ORSC mailing lists: http://www.open-rsc.org/lists/
If enough people do this we can take control of the net back from the lawyers and inept government wonks that control it now.Don't just sit there with your thumb up a penguins butt, DO something!
Need Mercedes parts ?
This hyperlink selection would start gnutella and could be passed through the gnutella plugin and shown again in the browser (Note: no DNS needed for the domain as we're using the other users IP).
Even after all of this, we still come back to the point you raised about bandwidth. Every peered computer serving up content would have to have a 256k+ connection in order to make such a scheme even remotely useable because they need to handle the existing searches of a growing host list and they need to handle new search bots consuming their bandwidth with "deep scans". Perhaps DSLnet would be a better name for this.
This does have it's advantages, as it would be:
1) Built on existing technologies (Browsers, Gnutella packets, indexing bots).
2) It would give a point and click interface to a distributed file sharing network with no central control.
___
What I know then there is already such thing. Some guys in .cz decided that they have problems to get their domain names and decided to create new top level domain '.exe' simply by adding a fake root server for this domain to all their DNS servers and tried to promote the domain to others. I am not sure if it still lives or whether it died out already, but I remember that at some point there were at least 500 subdomains registered.
If programs would be read like poetry, most programmers would be Vogons.
Someone should sit down and think the problem through and come up with proposals. There are already multiple name spaces (e.g. 'RealNames', 'Yahoo') on the net apart from Nic.
The reality is that at the end of the day, you are going to have disputes, and you are going to need some way to resolve those disputes. That is an undeniable fact. Someone else is better placed to answer this one.
-- Matthew - matthew.gream@pobox.com, http://matthewgream.net
This brings up the second problem on the internet - running out of IP numbers. There are 4 billion possible combinations of numbers in 32 bits, but for various reasons, there are a bunch that are invalid and because of the allocation system, which assumed that the internet was small, there is a lot of waste. For instance, MIT has 24 million IP addresses (a "class A" network - Apple also has one). MIT is obviously not using all of them, but nobody else can have them. So, any IP address that starts with "18.x.x.x" belongs to MIT. Carnegie Mellon University (and a bunch of others) has a class B address space with 65,536 addresses (anything in 128.2.x.x) belongs to them. Smaller organizations have "class C" (256 addresses) spaces. There was (originally) nothing in the middle. Then they came up with this thing called CIDR (classless interdomain routing) that allowed for arbitrary numbers of addresses per network (as long as they are of the size of a power of 2). That has saved the internet so far. The end solution is IP version 6 (present version is version 4 - what happened to 5???) which has a few trillion IP's so that should solve the problem nicely :) However, it's taking forever to get implemented because everyone has to change their OS to support it.
Anyway, that is why you can't just add something to the IP address.
Go Badgers! -- #include "std/disclaimer.h"
why a different port?
.dns.org network first, you can all resolve company.com (really company.com.dns.org).
why different resolvers?
why not use the existing dns system (including the root servers)?
say i set up dns.org (i'm sure that's taken, but let's pretend). on top of that, i build com.dns.org, net.dns.org, sex.dns.org, sucks.dns.org, and so fourth.
now, if you all change your dns search order to look at the
if you want the real "company.com" just try to resolve company.com. (trailing dot).
and the best part: this is entirely outside nsi's jurisdiction. they have no authority on subdomains, and neither do the courts.
my 2bits,
pchan
sublimate the masses!
the slashdot method
or: How Bob got his post moderated down
 -By FluXraD
Open Curtain
Bob: Funny Post
Dork: That wasn't funny because, technical technical technical, point out technical innacuracy of humorous post. See, now everyon on slashdot knows that i'm at least smarter than you.
Bob: Well, technical technical technical, see how much funnier that is now that you've reread my post
Dork: Technical technical technial, you see, technical technical.
(Brief pause while the audience realizes that what had started as a humorous post by Bob has now been decimated by the technical idiocy and humorousnessness of Dork. They also begin to ask themselves why Bob would defend his post - they are about to get their answer)
Bob: Post that i think is funny, but really isn't anything more than a poor attempt at recovery.
Dork: Flame
Bob: Flame, technical
Enter moderator: Moderator reads 2/3rds of the thread and decides, based on Dork's joke killing posting methods, that Bob is neither funny nor knows as much as Dork would like to pretend. Moderator, with a look of stoned exhasperation moderates Bob's original comment down as [Flamebait] or [Overrated] - Moderator2 enters quickly to follow suit. In an ironic twist, Dork's original post is moderated (Score 4: Informative). Only 10 years later do we discover that Dork is actually a replicant.
Curtain.
FluX
After 16 years, MTV has finally completed its deevolution into the shiny things network
"It is seldom that liberty of any kind is lost all at once." -David Hume
Try www.name-space.com, an alternate registry with some 500-odd gTLDs already set up. All you have to do is point to their machines as your resolver and you can access the .bicycle, .sex, .shareware, etc. gTLDs. (so if you have that done, they're also www.name.space). Whether that's a good idea or not is something else: they'll let anyone register any gTLD. They're hoping to get enough popular support (people giving out their "new improved" names) that NSI will have to honor their gTLDs; I tend to think that won't happen. But FYI.
You'd still be faced with the same problem that caused the creation of NSI in the first place: who is ultimately in charge of the database? Some entity somewhere has to have the final say in what's in there, otherwise you have chaos. Not to mention trying to persuade overworked network administrators to add your alternative DNS to their name servers, some of whom wouldn't do so simply because they wouldn't like the idea of, for example, a .guns domain.
I think we're stuck with the present system, so our energies ought to be brought to bear against NSI and its increasingly fascist policies. Where's a tobacco lawyer when you need one?
I would propose closing down .com for further registrations as it is now beyond repair (every company in every country fighting for a name in one namespace was never a good idea even when America was the only county you expected to matter to the 'net). All companies would have to register in the .com/.co/whatever sub-domain of individual countries and could only do so if they held a registered trademark in that country.
.net and .org should stay open but vetting should return, perhaps by ICANN or a UN body.
.gov should be closed down and phased out. At the very least any further registrations should be requied to add at least one country-code subdomain (e.g., .uk.gov .us.gov .cx.gov) to recognise the international reality of the modern Internet. The same goes for .mil (oh, that'll happen!).
Add a .xxx TLD to shut up all those twats that want to censor the web. It's pretty obvious that the sites themselves want this TLD and it would act as a one-point-of-control for parental filtering which actually makes sense.
Add .home for individuals and non-registered companies (in the UK this means non-limited companies) but require at least two levels of sub-domain (e.g. www.tww.farnham.surrey.uk). Registered companies may not apply so no confusion can ever occur or be claimed about trademarks etc. Non-registered companies (high street shops etc) would have a two year change-over period if they did register where they would have both a .co.fr domain and their old .honfleur.normandy.home after which they could only retain the .co.fr one.
Setting up a new system seems like a good idea but the best we can expect, given the size of the net now, is to improve what we have.
.int can be closed down (for further registrations) and I doubt that anyone will miss it.
Don't reject every solution because it's not perfect as there is not going to be a perfect solution. We do need, however, a good solution. Soon.
Newsnet has not inspired me to think that the "anarchy" approach would work. As an anarchist I feel that anarchy really only works with groups of under 200 adults.
I think that's all I have to say about that.
TWW
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
I use my Dad as a benchmark. He doesn't care to understand the underpinnings of the protocols or the technical whys and hows...all he wants to know is how to push the buttons and get at the information he seeks. So, he gets familiar enough to understand the tools and the routines. When I caught him going to Yahoo first and typing www.amazon.com, it opened my eyes. I've since learned that very many people use portals and indexed catalogues as namespace locators, even using them as URL entry forms.
The point being: devising clever new URL structures, alternate domain name schemes, different protocols...you're focusing heavily on "McGyvering" the system, and forgetting to consider the basic social engineering that is required to ensure these work (at least among the naive and untrained who will be expected to follow). AOL is a hit for a lot of reasons, none of which are performance or technically oriented: their KEYWORD system being one of them. All I'm saying is as you apply your creative juices to the problem, try to imagine how it will benefit my Dad.
Get Veiled
We already have the perfect system. It's impossible to make a system everyone likes. But it is possible to make a system everyone hates equally, hence what we have now.
The first is that many protocols utilizing the DNS lookup function would not work, and would have to be re-configured. While this may seem a small task, remember that (unfortunately) many people rely on so called "point and click" functions of their servers, and have no knowledge base on re-stringing queries.
The second is the functionality behind the current DNS scheme. Domain names are handed out, and related to IP addresses, which works great. Your normal first timer can type in known addresses to get known companies. (.org and .net fall under this too) This easy familiarity makes surfing the internet a breeze, even if some people have to suffer for it.
Another thing is non, or misuse of a new DNS registry system. I don't beleive someone should be allowed to camp on a site that has an easily identifiable corporate name (Wal-Mart, for example) and use it to either coerce the company for money, or for other things, such as placing 600+ ad banners in an effort to get paid. Such misuse demotes the true intent of the internet, which is the free distribution of information. A good example of non-use is the so called "undernet" and "cobranet" that cloned the IRC a few years back. These clones, while functioning perfectly like the IRC, and perhaps even a little better than the IRC, were hardly ever used, and quite a number of them went defunct in a matter of a few months.
An alternate system I support is the "extension" of the current DNS configuration that was rumored to be in the works back around 1998. It involved pushing the extension barrier out a couple steps to include things like .bus (business) .que (web queries or searches) and so on... Rumor has it that someone even came up with the idea of giving each state/local province their extension, (so North Carolina would have the extension of www.*****.ncs.us, and the venezia regional government of italy could have www.*****.vez.it
I firmly support pulling the dollar power away from NSI. I think the distribution of domain registries should incur a one time type in fee only, and that should basically be the cost of the man-hours required to make the changes to the system. Require domains to confirm that said domain is still in use once every month or quarter (and require it to be filled out in a predermined format so it can be parsed out by a PC updater) and that's that.
krystal_blade
It will be easy to motivate our fellow man; there is hardly anything people treasure more than not being annihilated.
A few weeks ago, after submitting the ask slashdot, I put up a web site at www.piratedns.org if anyone is interested in contributing.
Instead of having a single, context-insensitive name space, allow anyone to create a name space, or many. Bare with me here.. I'll try to illustrate how it would actually work in practice with some examples:
First, I would have my own name space (actually, I would have many; more on that below). This would essentially be a mapping from names as I know them to Poi numbers (Permanent Object Identifiers, a unique numbering space for everything--see below). Generally this would behave like a local cache, fairly transparent to the user. But the key point behind it is: when I use a name, I get what I got last time, period (unless I've gone out of my way to specifically reassign the name to something else). So, for instance, the Slashdot home page would have an associated Poi number (forever!), and in my name space it may simply be called "Slashdot". (Note, in practice I would have a specific name space for web pages, which would be different than my name space for email addresses, and so on, so in theory I could simply enter "Slashdot" in my web browser Goto: line and be there.) If slashdot moves to a new domain, or IP, or even changes its name to something else due to Microsoft inventing a time machine and going back in time to trademark the name, Slashdot would nonetheless keep the same Poi number (forever!) and so I'd still go where I wanted when I typed "Slashdot". Key point: the Poi space is effectively infinite and context independent (truly global) and has *no implicit mapping to reality whatsoever* so there is no risk of losing a Poi number over trademark, geographic, or network topology issues.
Next, say I do a Google search, and click on a link.. What happens? Well, the link just contains a Poi number (and optionally Google's name for it), so I'm brought straight to the web page (I'll discuss mapping Poi to IP's and whatnot below) and if I bookmark it, it's bookmarked by Poi number which means even if the web page is moved to a different host my bookmark still works. Note that by bookmarking it, I am effectively expanding my personal name space. Bookmarks would again be a separate name space from the generic "web page" namespace, though my browser may allow me to assign a name in that space too: E.g., I may have a bookmark called "Slashdot User Profile" which I also call "slashuser" in my "web page" namespace (so I can just type "slashuser" at the Goto: prompt). A simple utility could search all my name spaces to show me aliases of the same Poi in case I forget one.
Now, what if a friend wants to tell me a web page over the phone? In a pinch, he could read me the Poi number (a pain, but doable), but more likely he'd point me at a *common name space* and tell me the name. E.g., he'd say "Go to Yahoo/Companies/Redhat". Here it is assumed that Yahoo is already in my private name space (maybe it came with my browser; maybe I got it from a friend; whatever -- more on this below!), and the "/" is a path separator as in a file system, except that instead of assuming a hierarchy it's just a search through consecutive name spaces (may be arbitrarily complex graph, not a tree): Start with Yahoo (who's Poi I already know), ask Yahoo for Companies (returns a new Poi number -- namespaces themselves have Poi numbers too!), ask Companies for Redhat. Done!
So, in effect, it creates a competitive market for name spaces, which will most likely result in a few dominant players (e.g., Yahoo) at any given time, but affords no inherent monopoly to anyone. Note there is no one root! The "root" of the name space is implicit in the current body of popular name spaces!
Note also that the name spaces are of varying specificity, so for example say there are twenty companies in the country called "ACME Services" -- Yahoo/Companies/ACMEServices may return nothing, may return a list of all twenty, or may return a default one, according to Yahoo's (or the requester's?) choice, while Yahoo/Companies/California/SanDiego/ACMEServices may return just the one you're looking for.
Obviously the line between searching and DNS is being blurred here (see also Erik's post, subject: "This is bound to fail"), but the distinction from pure searching is that there would still be well-known name spaces (though many instead of just one) and the names within those spaces may still be concise and definitive so that, for instance, once I get yahoo/people/brandyn to Poi'nt to my page, I really can tell someone that over the phone and it's not like they have to do a Google search and sift through fifty returns (assuming yahoo was maintaining uniqueness in that particular name space).
Next there's the question of mapping Poi to IPs and whatnot, which could be implemented as a cached, distributed database -- ideally when you Get a Poi (e.g., via a Google link or whatnot), the Poi of the giver (Google) is also remembered with it, so when you look up a Poi, if it isn't already locally cached, you ask (first) the associated giver (Google) for the info. Once you actually get the Poi info, part of the record would be the current preferred giver, which would be the first place you would go later if the cached info proved antiquated. Typically, the preferred giver would be the host actually serving the Poi, which would provide forwarding service (that is, it would tell you who to ask instead) if the Poi were later moved away.
Note that what a Poi maps TO may change with time. For now, for instance, most Poi might map to urls (with hard-coded IP's in place of host names), but they could also map to other protocols/hardware addressing schemes entirely. The good thing about a Poi is it is NOT an IP address, nor a host name, nor an Appletalk node name, nor a filename, nor a phone number... it is just the universal "name" of the object you want to reach. Today it maps to an IP and url (or IP and port number for other services like telnet), but tomorrow who knows.
Lastly, there's the question of the Poi numbering space itself. How do you prevent monopoly concerns here? Obviously we need *some* central control (world-wide!), so I would propose something like this:
Some non-profit org would sell large chunks of Poi space -- let's say they sell chunks for $10 per bit, so I could buy 65K unique Poi from them for $160 (but I wouldn't -- read on). The presumption, then, is that other organizations would buy huge chunks of Poi space, and resell them at a lower cost per bit (but a higher cost per Poi). As an end-user, I ought to be able to get Poi practically free (ISPs could pay trivial fees to provide an endless supply to their users), to use for all sorts of things.... This scheme keeps Poi arbitrarily cheap while still not burdening the top-level Poi service with lots of small requests. Each re-seller would be responsible for verifying a Poi's authenticity, simply by identifying which block it came from and where that block was purchased (thus providing an audit trail straight back to the top level). Note that these services *only* assure uniqueness of the Poi -- they do not store any information about their use!
(I already have this set up on my machine for my personal use. Poi can be any number of bits, so you can't run out, and even if I chewed through them at thousands per second they still wouldn't get very large before the sun went cold.)
-Brandyn (Poi #1000000000000001b;)
"No one ever went broke underestimating the intelligence of the American public" -PT Barnum
"Yeah, what he said!" -Bill Gates