Blockchain DNS

Unbound
1

Hi

If that has been discussed previously than I was not able to trace it but being interested in whether there are plans to resolve blockchain domains with Unbound?

Thus far my understanding is that it would require:

  1. a transaction ID with the respective blockchain, and then

  2. the initial download of the entire current blockchain, and

  3. subsequent incremental downloads of the bits being added/deleted to/from the blockchain

The blockchain would contain the DNS for the domains and also the security bits for each domain and thus there is no root and no DNNSEC.

2

The reason we designed DNS is that you cannot possibly store all domains in the world locally. How do you think your solution scales if you not only need to store all worldwide domains but also their entire history ?

I have written about this in the past,

https://nohats.ca/wordpress/blog/2012/04/09/you-cant-p2p-the-dns-and-have-it-too/

Paul

3

Thank you for the link and the fine reading it makes.

You make the point - the resolver is not the judge of the trust but handles and interprets the information handed down the chain of trust. It depends on the user whether to trust the institutions that established the chain of trust in the first place.

Thus if users place their trust in a blockchain then it perhaps should not make a difference for a resolver to deal with the data provided by such a blockchain.

Scaling is certainly a differen aspect of it and perhaps depends also on the data design also considering that DNSSEC/DMARC/DKIM/TLSA/CAA records are not involved.
But is it not that blockchain transaction ID (wallet) apps are facing the same challenge and have to download the entire blockchain first and keep it then current with the bits beig removed/added?

Blockchain might not be the saviour for all mankind challenges, like it is sometimes being promoted, but it is for the moment at least something interesting to consider.

4

so you say. it's not interesting for me to consider. its hype cycle has been extreme even by 21st century internet standards. all proposed uses must be suspected of being a solution in search of a problem, as is absolutely the case for namecoin and the other blockchain naming systems. the burden of proof to show why anyone ought to spend any time considering anything related to blockchain is on whomsoever asserts its utility. that burden has not been met, or even approached, for naming.

i expect that we are off-topic for unbound, as well.