3 ms·

The concept of the domain on today's internet is a bit anachronistic on the internet these days. People just want to get "places" on the internet, not to addres

The concept of the domain on today's internet is a bit anachronistic on the internet these days. People just want to get "places" on the internet, not to addresses.

autoexec 3y ago
What is the alternative? IP addresses? AOL Keywords? Abstracting away the whole thing and forcing people to only connect to your webserver via a mobile app? Domains are still the best thing anyone has managed to come up with. We just need better administration of them since ICANN sold out.
- Palomides 3y ago
  most people with a laptop/desktop will type "facebook" into the address bar of their browser, which will do a search in whatever search engine was provided as the default, then click on the first link they have little to no conception of what a TLD is actually, most people use an app on their phone for everything and avoid the web browser I don't say this to insult the average person, this is simply what I see non-engineers do
  - besofrrn 3y ago
    [dead]
- kalkr 3y ago
  https://www.gnunet.org/en/gns.html https://www.gnunet.org/en/gns.html
- et-al 3y ago
  SEO, unfortunately. As Palomides mentioned, most people just search for a website they will go to instead of typing in whatever cute domain you've crafted.
usrbinbash 3y ago
Sorry, but that reads a lot like "not invented here". You do realize that domains were invented so people could get to "places" instead of addresses, right? So pray tell, what would you replace domains with? And what advantage would your schema have over what exists now? Because, the addresses don't go away. Underneath whatever you come up with, will always be an IP address, and has to be, because you need a way to route packets. So whatever you want to replace domains with, has o solve the exact same problem: Wrap IP addresses in a nice, human friendly, hierarchically organized schema, that is easy to read, remember and type.
- samatman 3y ago
  Hierarchically organized is not a hard requirement. A domain could consist of nothing more than a unique Unicode string with no '/' or ' ' in it. It might do well to require a '.' be present, but even that isn't necessary. I'm quite content with what we have, however. A "unique string domain" regime has to pick between an anarchic landscape of independent registrars playing whack-a-mole with sniping and squatting, or a single central source of truth, neither of those strikes me as ideal. I don't think routing would actually be an issue though, given the power of today's servers and good algorithms for hashmaps in the many millions of keys.
  - usrbinbash 3y ago
    > Hierarchically organized is not a hard requirement Technically correct, but in practical use, it is absolutely a requirement. Entities within domains are usually hierarchically organized, this already starts at the top level by having www.domain, mail.domain, dns.domain, file.domain, ... So if we implemented a scheme without built-in hierarchy, I can 100% guarantee that one of the first things admins will do, is implement their own hierarchy naming conventions on top of it, just to keep things organized. The difference then is, that we won't have one clear way to do that, but hundreds of different ones. > I don't think routing would actually be an issue though, given the power of today's servers Anything that relies on "worse solution offset by throwing more compute at it", is an instant "denied" in my book. That's how we ended up with electron apps.
besofrrn 3y ago
[dead]
palata 3y ago
I don't think it is common to use `.us` in the US, but many other countries use their ccTLD a lot more. I personally don't like it when companies hack them (typically with `.io` or `.ai`), because I am used to assign a meaning to the ccTLDs. My point being that while it may feel anachronistic in the US these days, think that there are many countries that are not the US :-).

The concept of the domain on today's internet is a bit anachronistic on the internet these days. People just want to get "places" on the internet, not to addresses.

autoexec 3y ago

What is the alternative? IP addresses? AOL Keywords? Abstracting away the whole thing and forcing people to only connect to your webserver via a mobile app? Domains are still the best thing anyone has managed to come up with. We just need better administration of them since ICANN sold out.

Palomides 3y ago
most people with a laptop/desktop will type "facebook" into the address bar of their browser, which will do a search in whatever search engine was provided as the default, then click on the first link they have little to no conception of what a TLD is actually, most people use an app on their phone for everything and avoid the web browser I don't say this to insult the average person, this is simply what I see non-engineers do
- besofrrn 3y ago
  [dead]
kalkr 3y ago
https://www.gnunet.org/en/gns.html https://www.gnunet.org/en/gns.html
et-al 3y ago
SEO, unfortunately. As Palomides mentioned, most people just search for a website they will go to instead of typing in whatever cute domain you've crafted.

usrbinbash 3y ago

Sorry, but that reads a lot like "not invented here". You do realize that domains were invented so people could get to "places" instead of addresses, right? So pray tell, what would you replace domains with? And what advantage would your schema have over what exists now? Because, the addresses don't go away. Underneath whatever you come up with, will always be an IP address, and has to be, because you need a way to route packets. So whatever you want to replace domains with, has o solve the exact same problem: Wrap IP addresses in a nice, human friendly, hierarchically organized schema, that is easy to read, remember and type.

samatman 3y ago
Hierarchically organized is not a hard requirement. A domain could consist of nothing more than a unique Unicode string with no '/' or ' ' in it. It might do well to require a '.' be present, but even that isn't necessary. I'm quite content with what we have, however. A "unique string domain" regime has to pick between an anarchic landscape of independent registrars playing whack-a-mole with sniping and squatting, or a single central source of truth, neither of those strikes me as ideal. I don't think routing would actually be an issue though, given the power of today's servers and good algorithms for hashmaps in the many millions of keys.
- usrbinbash 3y ago
  > Hierarchically organized is not a hard requirement Technically correct, but in practical use, it is absolutely a requirement. Entities within domains are usually hierarchically organized, this already starts at the top level by having www.domain, mail.domain, dns.domain, file.domain, ... So if we implemented a scheme without built-in hierarchy, I can 100% guarantee that one of the first things admins will do, is implement their own hierarchy naming conventions on top of it, just to keep things organized. The difference then is, that we won't have one clear way to do that, but hundreds of different ones. > I don't think routing would actually be an issue though, given the power of today's servers Anything that relies on "worse solution offset by throwing more compute at it", is an instant "denied" in my book. That's how we ended up with electron apps.

besofrrn 3y ago

[dead]

palata 3y ago

I don't think it is common to use `.us` in the US, but many other countries use their ccTLD a lot more. I personally don't like it when companies hack them (typically with `.io` or `.ai`), because I am used to assign a meaning to the ccTLDs. My point being that while it may feel anachronistic in the US these days, think that there are many countries that are not the US :-).