Hacker News Reader

Comment by swiley | Hacker News Reader

My understanding is that running resolvers at the kind of scale they need to handle attacks is actually difficult and expensive. The registries themselves are just a couple of machines with databases on them.

Maintaining namespaces is just hard and the DNS, with groups of people running TLDs, is the best we’ve come up with so far. I personally thought namecoin was a creative alternative and I’m a little disappointed it hasn’t become more popular but until something like that does we’re going to continue seeing things like this.

View on HN

To be clear, the previous owners of .org didn't actually do the technical parts of running their TLD, they subcontracted it out: https://www.reddit.com/r/programming/comments/bi6gg5/help_ke...

> DNS, with groups of people running TLDs, is the best we’ve come up with so far.

Is it? I'm not very familiar with the domain, so I'm actually asking. Generally, I'm very cautious when someone says that current state of affairs is the best we have come up with, because it doesn't acknowledge the inertia to keep things as they are.

No, there are properly decentralised domain name projects like namecoin which uses blockchain to control ownership of domains and zone edits in a more secure way.

There are a few other projects that decentralise domains and fix security issues, but our current federated system has become a heavily established billion dollar industry and the people in control of it have no incentive to give up that control.

Any move to a new system will have to come from end users who don't know there are issues, and don't care enough to spend the effort understanding the issue and solution.

As if a den of evangelical blockchain shills would run the domain name system any more ethically, responsibly, and transparently than the current bunch of old school shysters running it, if only they'd hand over the reigns of control to the new blockchain shysters and make them God. Slapping the "blockchain" buzzword on everything doesn't solve any problems, it just makes you look like yet another con-man.

https://en.wikipedia.org/wiki/Namecoin

>A 2015 study found that of the 120,000 domain names registered on Namecoin, only 28 were in use.

An empirical study of Namecoin and lessons for decentralized namespace design

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.698...

>Based on all the empirical evidence we present, we are left to conclude that the Namecoin ecosystem is dysfunctional. The vast majority of registered names represent squatting and there is little evidence of a secondary market for names. While there could be many factors that explain the lack of adoption, there appears to be clear room for improvements in the design to minimize squatting and other problems. To this end, in Section 6, we explore the design space of decentralized namespaces and make recommendations.

Why Namecoin Didn't Take Off: A Cautionary Tale

https://cointelegraph.com/news/why-namecoin-didnt-take-off-a...

>Cointelegraph: You used to be one of Namecoin's biggest cheerleaders. What happened?

>Michael Dean: So here's my opinion, which is really going to get me hated, but I think Namecoin as a decentralized DNS-type system is dead.

>[...] The biggest flaw is the clunky wallet that takes ten minutes to open. The other biggest flaw is that the near-zero expense of registering domains encouraged domain squatting. All the good names are taken. Someone registered every noun in the dictionary as Dot-Bit.

>And these squatters didn't even do it right. When you go to any of these squatted domains, they don't go to a place holder with a page that says "to buy your site in Dot-Bit, email me here and make me an offer." The sites just don't resolve! These people are unclear on the semi-scummy concept of domain squatting. In Namecoin, even that gets done wrong!

>Also, Dot-Bit is actually vulnerable to a hostile takeover from ICANN. If ICANN decided to support Dot-Bit and decide where things would resolve contrary to where the Namecoin wallet was resolving things, and corporate DNS servers sided with the ICANN (which they largely would), it would create consumer confidence chaos with the Namecoin system.

You are getting a bit evangelical there.

I'm not an advocate for blockchain it has only one very narrow use case global distributed consensus on ownership of a virtual token that's perfect for money and domain names and nothing else.

I'm sorry if you bought some flim flam mans marketing spin but I only ever look at the tech.

All of these issues with people not using their domains just backs up my argument, that people don't care enough to use it, hell even the squatters don't care enough to use it.

Ultimately ICANN could have given them the TLD but didn't because they have no incentive to give up that control.

Thank you for backing up my argument with links :)

>I'm sorry if you bought some flim flam mans marketing spin but I only ever look at the tech.

What made you think I believed any bullshit, or that I'm evangelizing? What company did I endorse? And the links prove you're evangelizing Namecoin as a "properly decentralised domain name project", but it's actually a complete failure, which you conveniently neglected to mention. Your statements are the exact opposite of the truth. Speak for youself, and stop projecting.

You claimed that Namecoin was "properly decentralised" and "uses blockchain to control ownership of domains and zone edits in a more secure way". Yeah, all 28 domains in use. You're the one evangelizing a spectacular failure.

> You claimed that Namecoin was "properly decentralised" and "uses blockchain to control ownership of domains and zone edits in a more secure way". Yeah, all 28 domains in use.

Yes that was my argument, we are both in agreement why are you so upset?

I'm not at all upset (and you're projecting again) -- I'm being sarcastic, and it whooshed right over your head. Look up: All 28 domains. Why didn't you point out what a failure it was in your argument (which has already been severely downvoted), unless you were being disingenuous when you evangelized Namecoin as a properly decentralised more secure solution, and not a failed improper insecure vulnerable dead dysfunctional boondoggle, as my quotes and citations proved?

Read dmos62's comment and my response again.

I mention namecoin as a single example of decentralised dns and mention that there are other projects (thats not advocating or evangelising namecoin).

I then point out the major problem with all of these alternative dns, no one uses them because few have incentive to do so.

You responded by calling me a con-man (sounds reasonable) because I had the audacity to mention a technology in one of the only use case's its good at.

> I'm not at all upset (and you're projecting again) -- I'm being sarcastic, and it whooshed right over your head.

I got the sarcasm, I got the name calling, it's childish. If I got more upvotes would you consider my words instead of the con-man image you are projecting? I don't think so.

Handshake (https://handshake.org/) is another attempt at a decentralized domain name system. It’s compatible with the old system so I’m more hopeful that it can gain some adoption. Let’s see.

The cost of running this infrastructure doesn't go up or down with each additional domain name registered.

Surely resolvers for a TLD with 100K domains and 1MM lookups per day are cheaper to run than those with 100MM domains and 1BB lookups per day.

It may not have an obvious change with a single new domain registered, but surely there are significant variable costs.

(To be clear, I oppose this transaction, just not on the grounds that root name servers are all fixed costs.)

> Surely resolvers for a TLD with 100K domains and 1MM lookups per day are cheaper to run than those with 100MM domains and 1BB lookups per day.

This on the face of it sounds reasonable, but the cost difference is likely marginal.

If you're going to provide a highly available DDoS resistant infrastructure to deal with a large scale of customers, the majority of the costs are going to come from building that distributed infrastructure across many parts of the world.

The costs may go up between 1 million lookups vs 1 billion lookups, but if there is an increase, the order of increase will be closer to 10x than it is 1000x.

Can confirm. Our TLDs scaled by about two orders of magnitude in total registration number following our launch of .app, but the associated costs that rose were very marginal. Most of the cost is not strictly related to volume.

The increase in cost is sufficiently sub-linear that the cost of each new .org registration should be going down as the number of them goes up, if it was truly run for public benefit while breaking even.

As another comment says, most of the cost will be in building and maintaining a robust, scalable, attack-resistant distributed system. The linear scaling costs (such as bandwidth and processors) will be small in comparison.

(Of course at much higher numbers, mathematically O(N) starts to dominate again, but 1 billion lookups per day isn't enough for that.)

Reply on news.ycombinator.com