If ICANN Can't, Who Can?
alanjstr asks: "After reading this article at The Register, I no longer understand how domain registration really works. Quite a few posts have come across Slashdot about ICANN elections and rights to domain names. It sounds to me like it started off as a good thing but is struggling to move to be autonomous. ICANN was created in an attempt at
who should run it and How should it be paid for. Clearly the Who has become a problem with many complaining about not being represented. The How is a problem that is still unresolved. The more I think about it, the more it seems like we're setting up a new government to rule the land of Domains. How should be go about fixing this dilemma? The first thing that comes to mind is to write a Constitution to lay the groundwork. How would you complete the following: We the People of the Digital Planet Earth...." It all boils down to ICANN asking most of the ccTLDs to pay a third of it's operating costs without allowing them representation in ICANN itself. Now that doesn't sound very fair, does it?
Fromm Nominet's 2000 AGM notes.
Matt Thompson - Actuality - Insert product here.
Now, I am a relative newbie to Unix in general. Don't even think of asking me to say anything intelligent about how DNS works or about how BIND works or about exactly what happens when I type a domain name into my web browser past the fact that it goes out to a DNS server and fetches an IP address.
/home/inquis. All the user directories are subordinate to the /home directory. The /home directory and my home directory within that /home directory are logically linked.
/home directory, looking at the contents of that directory, and finding it filled with /sbin crap.
y /lowerLevelCategory
My feeling for a while now is that while Uniform Resource Locators make sense, domain names don't. Think about it from this perspective, and see if it makes any sense.
In the context of your computer, there is a string that you can specify that can point to any particular file, or resource, on your entire machine. For example, my directory would be
Jump over to a win9x box. The contents of the Windows directory are logically linked to the identity of the Windows directory itself. Everything in the windows directory belongs in that directory, because everything in there is a part of windows.
Now look at our idiotic system of using domain names to access resources over the web. First of all, nothing requires that the domain name itself have anything to do with the content that can be accessed by using that domain name. This would be akin to sitting down at your linux box, moving to your
Another problem I have with DNS is that related content is not grouped together by default. This harks back to the previous problem (you can't tell the content from the domain name). And I'm not simply talking about going to a portal that indexes web content and drilling down through the links, I'm talking about a fundamental archetecture change.
Look at it this way. Say you want to look up newbie Linux sites, but you don't know where to begin looking. As it stands now, you can go to Google and hope that their spider has picked all of them up; you can go to Yahoo and hope that they have manually indexed them all; either way, you miss out on content.
now, check this out... wouldn't it be easier on you and everyone else if you could just do this?
http://xml/linux/newbie
transferProtocol://contentType/highLevelCategor
As the web stands now, it is analogous to a linux box with every single file on the entire machine crammed in the root directory. You have to know what exactly you are looking for and how to find it before you can actually find it. A more efficient system would allow even the most braindead user to shoot in the dark and still manage to find somehting useful quite quickly.
(Response to one obvious counterpoint: you can grep a directory to find what you are looking for quickly even if you don't know its name. However, grepping the web is not trivial. The closest tool we have for doing that is Google, and we all know that while it is pretty good, it is not perfect.)
Allright, this now ends my directionless rant. Mods, respond to this if you disagree instead of modding it down.
Thanks, and everyone have a good day. I just pulled an allnighter writing polysci paper, so I needed a good rant.
-inq
Until then, I think the best way is to open it up to everyone and get the registries to allow mass voting on what new TLDs are added. WIPO can handle the problems with trademarks.
Another point I feel worth mentioning, is that the problem also lies with the fact that WE are not taking advantage of the alternative DNS systems out there. If we make a mass migration over to alternative DNS systems, ICANN will lose its clout. This should send a message to any future organization or government that would like to manage DNS. Check out and support:
Open DNS Technologies
AlterNIC
ADNS
Open Root Server Confederation
Name.Space
There's more too.
Chris
Open DNS Technologies, Inc.
Everyone's b*tchin' about ICANN but not many people are really doing anything about it - but in the Register article there's a link to The SuperRoot Consortium that has a proposal and is actually doing something about the TLD problems.
I know that I'm switching over, if everybody starts using their rootservers ICANN will loose it's power and all of us will be happier. Think about it, ICANN depends on their rootservers to stay in power, use other rootservers and ICANN can't touch you.
Like I said, I'm switching...
--
All browsers' default homepage should read: Don't Panic...
All browsers' default homepage should read: Don't Panic...
Okay...I run BIND for where I work. Given that I'm not Earthlink and connecting massive amounts of the american populous to the net, but we've got about 5500 people.
As I read through the comments on here, I find things like Superroot.net, alternic, etc...
So, for people to get access to these rogue sites, I need to add all these other entries to my root.db and other files. And merge the entries from the various splinter groups, as I can't just dowload one groups root.db and run with it. And what is going to happen when two groups both have the same TLD listed in there?
I'm all for ICANN going away, but like it or not, there needs to be that tiny bit of control in there to keep utter anarchy (ie alt.*) from happening.
Anyone have a better way of handling this nightmare?
Buy the DNS & Bind book (if you haven't already)
Setup a nice little name server with a catchy TLD like ".slash"
Add your friends' boxes to your new TLD
Change your .sig to tell people how to modify their resolv.conf
Rinse & Repeat
Once roughly a third of the Internet is using "pirate" DNS systems some propeller-head at Yahoo or some such will have the great idea of mirroring all this at their site and ICANN will soil their trousers. You can expect ICANN to pressure ISP's into only using the "official" name servers, and a few lawsuits to settle who can run what services (named) on their own machines. Those things will likely split the Internet (again) between the haves (those who have the knowledge and will to modify their resolve.conf) and the have nots (those who must use their ISP's config.) However, that might be a Good Thing.
"Hmm... this link to l337.h4X0r seems to be broken..."
"Dammit! That darn sensorware must've blocked newdgeeks.slash"