4 ms·

Thanks for that, appreciated. I'll be honest- I'm just a 'little guy' in the food chain so I always figured that doing something like that was for the ISP level

Thanks for that, appreciated. I'll be honest- I'm just a 'little guy' in the food chain so I always figured that doing something like that was for the ISP level folks <edit to clarify, I mean connecting to a Zone 1 Resolver. I wasn't aware that one could download the Root Hints File directly (Thanks!). One quick question though - After taking a quick skim of it the list seems to be extremely 'Western-Centric' (reference link https://www.internic.net/domain/named.root https://www.internic.net/domain/named.root)

icedchai 5y ago
The root servers are anycasted. Each one of those root server IPs corresponds to N physical servers at diverse networks / locations all over the world.
tylersmith 5y ago
The canonical DNS system itself is extremely Western-Centric.
- kfrzcode 5y ago
  As are many Western inventions
  - contingencies 5y ago
    DNS[0] is only a decentralized hierarchy with caching, a class of system which pre-dates the digital era as the de-facto means of political and military organization in any human society larger than a village or town. DNS as a directory system for IP is could itself be viewed as a direct philosophical descendant of military insignia (perhaps via the then-popular branch-tangent of the telephone book, itself ex-telegraph, and postal system) and these could all be in effect traced back to at least Roman society[1], I don't think arguing this is a "western" invention is very convincing or useful. Any ancient army or polity of any size would have had an equivalent, which would then include ancient Egypt, China[2], India, Mesopotamia[3], Mesoamerica, etc. Actually, come to think of it, the comparative study of ancient postal systems would be pretty interesting.[4] [0] Original DNS RFC1035 https://datatracker.ietf.org/doc/html/rfc1035 https://datatracker.ietf.org/doc/html/rfc1035 (1987) [1] Somewhat cheekily as the inventor of DNS has a Greek surname. https://en.wikipedia.org/wiki/Paul_Mockapetris https://en.wikipedia.org/wiki/Paul_Mockapetris [2] 2000+ years ago and mature enough to have QoS+max-TTL/hop: http://libgen.rs/scimag/10.1163%2F9789004292123 http://libgen.rs/scimag/10.1163%2F9789004292123 (pp17-48) + where I write this. [3] Evidenced to 9th century BC https://www.ucl.ac.uk/sargon/essentials/governors/thekingsroad/ https://www.ucl.ac.uk/sargon/essentials/governors/thekingsro... [4] Start by fixing https://en.wikipedia.org/wiki/Timeline_of_postal_history https://en.wikipedia.org/wiki/Timeline_of_postal_history
    - kortilla 5y ago
      By that same token the internet was invented by the first person to hand gesture to another one. You can’t dilute DNS down to a directory because there were/and are already other directory protocols.
      - qeternity 5y ago
        Walking gets you from A to B just like a car, so actually the first bipeds really invented the automobile.
        contingencies 5y ago
        It's more like describing a walking robot as an original invention, which it is, but all they really did was ape apes.
    - qeternity 5y ago
      > I don't think arguing this is a "western" invention is very convincing or useful. Any ancient army or polity of any size would have had an equivalent, which would then include ancient Egypt, China[2], India, Mesopotamia[3], Mesoamerica, etc. And yet, none of these other regions and cultures actually did invent it, and thus it remains a Western invention.
      - contingencies 5y ago
        Of course. However, my point was the originality and claim to authorship is low, because if you look at store and forward networks with centrally agreed node identification and local caching using that for routing purposes, humans have literally done it globally for 3000+ years... that's clear prior art. It's like "technology" being used to describe a bash script, or "invention" used to describe a standard algorithm.
        kfrzcode 5y ago
        https://datatracker.ietf.org/doc/html/rfc883#page-72 https://datatracker.ietf.org/doc/html/rfc883#page-72
aaomidi 5y ago
They are western centric, and unfortunately, in this current state of the web they're still essentially the authority on DNS. Alternatively, you can maintain the NSes for all the TLDs you are particularly interested in, and alert yourself if they change to something you don't recognize. Finally, keep in mind that whatever you do, you need to have multiple vantage points to the internet. There's not a lot stopping your ISP from not delivering you to the right host when you try to talk to it. E.g. your ISP can fake the DNS responses.
- endymi0n 5y ago
  > They are western centric, and unfortunately, in this current state of the web they're still essentially the authority on DNS. I‘m curious to see your evidence on that or which future state you would see as a more fortunate one.
  - jka 5y ago
    Questioning why the distributed cluster runs on nodes 'a' and 'b' alone doesn't necessarily imply that nodes 'c', 'd' and 'e' are any better or worse, today or in future.
  - aaomidi 5y ago
    If I knew the answer to this I would be very rich and probably have my name on multiple textbooks of solving decentralized computing problems.
lapinot 5y ago
> I'm just a 'little guy' in the food chain so I always figured that doing something like that was for the ISP level folks A lot of people are running recursive resolvers at home (like pi-hole stuff, or most people running some custom openwrt router/modem). I'm running one on my laptop (my resolver is localhost) and it works great. > After taking a quick skim of it the list seems to be extremely 'Western-Centric' It is, but that's what the internet is. But by running your own recursive resolver you can control your cache and a lot of the data doesn't change often. If you're extra paranoid you can cache the record data (or even archive the history) for ccTLD (or even all TLDs). For stuff (domains) you're interested in you can also hard-code or otherwise program "non-standard" ways to resolve the ips (by somehow populating a local database that overrides recursive resolution), like pi-hole/safebrowsing blocklists, stuff from institutions or CDNs you trust.

icedchai 5y ago

The root servers are anycasted. Each one of those root server IPs corresponds to N physical servers at diverse networks / locations all over the world.

tylersmith 5y ago

The canonical DNS system itself is extremely Western-Centric.

kfrzcode 5y ago
As are many Western inventions
- contingencies 5y ago
  DNS[0] is only a decentralized hierarchy with caching, a class of system which pre-dates the digital era as the de-facto means of political and military organization in any human society larger than a village or town. DNS as a directory system for IP is could itself be viewed as a direct philosophical descendant of military insignia (perhaps via the then-popular branch-tangent of the telephone book, itself ex-telegraph, and postal system) and these could all be in effect traced back to at least Roman society[1], I don't think arguing this is a "western" invention is very convincing or useful. Any ancient army or polity of any size would have had an equivalent, which would then include ancient Egypt, China[2], India, Mesopotamia[3], Mesoamerica, etc. Actually, come to think of it, the comparative study of ancient postal systems would be pretty interesting.[4] [0] Original DNS RFC1035 https://datatracker.ietf.org/doc/html/rfc1035 https://datatracker.ietf.org/doc/html/rfc1035 (1987) [1] Somewhat cheekily as the inventor of DNS has a Greek surname. https://en.wikipedia.org/wiki/Paul_Mockapetris https://en.wikipedia.org/wiki/Paul_Mockapetris [2] 2000+ years ago and mature enough to have QoS+max-TTL/hop: http://libgen.rs/scimag/10.1163%2F9789004292123 http://libgen.rs/scimag/10.1163%2F9789004292123 (pp17-48) + where I write this. [3] Evidenced to 9th century BC https://www.ucl.ac.uk/sargon/essentials/governors/thekingsroad/ https://www.ucl.ac.uk/sargon/essentials/governors/thekingsro... [4] Start by fixing https://en.wikipedia.org/wiki/Timeline_of_postal_history https://en.wikipedia.org/wiki/Timeline_of_postal_history
  - kortilla 5y ago
    By that same token the internet was invented by the first person to hand gesture to another one. You can’t dilute DNS down to a directory because there were/and are already other directory protocols.
    - qeternity 5y ago
      Walking gets you from A to B just like a car, so actually the first bipeds really invented the automobile.
      - contingencies 5y ago
        It's more like describing a walking robot as an original invention, which it is, but all they really did was ape apes.
  - qeternity 5y ago
    > I don't think arguing this is a "western" invention is very convincing or useful. Any ancient army or polity of any size would have had an equivalent, which would then include ancient Egypt, China[2], India, Mesopotamia[3], Mesoamerica, etc. And yet, none of these other regions and cultures actually did invent it, and thus it remains a Western invention.
    - contingencies 5y ago
      Of course. However, my point was the originality and claim to authorship is low, because if you look at store and forward networks with centrally agreed node identification and local caching using that for routing purposes, humans have literally done it globally for 3000+ years... that's clear prior art. It's like "technology" being used to describe a bash script, or "invention" used to describe a standard algorithm.
      - kfrzcode 5y ago
        https://datatracker.ietf.org/doc/html/rfc883#page-72 https://datatracker.ietf.org/doc/html/rfc883#page-72

aaomidi 5y ago

They are western centric, and unfortunately, in this current state of the web they're still essentially the authority on DNS. Alternatively, you can maintain the NSes for all the TLDs you are particularly interested in, and alert yourself if they change to something you don't recognize. Finally, keep in mind that whatever you do, you need to have multiple vantage points to the internet. There's not a lot stopping your ISP from not delivering you to the right host when you try to talk to it. E.g. your ISP can fake the DNS responses.

endymi0n 5y ago
> They are western centric, and unfortunately, in this current state of the web they're still essentially the authority on DNS. I‘m curious to see your evidence on that or which future state you would see as a more fortunate one.
- jka 5y ago
  Questioning why the distributed cluster runs on nodes 'a' and 'b' alone doesn't necessarily imply that nodes 'c', 'd' and 'e' are any better or worse, today or in future.
- aaomidi 5y ago
  If I knew the answer to this I would be very rich and probably have my name on multiple textbooks of solving decentralized computing problems.

lapinot 5y ago

> I'm just a 'little guy' in the food chain so I always figured that doing something like that was for the ISP level folks A lot of people are running recursive resolvers at home (like pi-hole stuff, or most people running some custom openwrt router/modem). I'm running one on my laptop (my resolver is localhost) and it works great. > After taking a quick skim of it the list seems to be extremely 'Western-Centric' It is, but that's what the internet is. But by running your own recursive resolver you can control your cache and a lot of the data doesn't change often. If you're extra paranoid you can cache the record data (or even archive the history) for ccTLD (or even all TLDs). For stuff (domains) you're interested in you can also hard-code or otherwise program "non-standard" ways to resolve the ips (by somehow populating a local database that overrides recursive resolution), like pi-hole/safebrowsing blocklists, stuff from institutions or CDNs you trust.