Hacker News Reader

Comment by dheera | Hacker News Reader

Even better yet, it would be interesting if we just abandoned TLDs. There is no reason why http://google/ can't work, really, with a bunch of infrastructure changes.

View on HN

Infrastructure changes aren't even needed; Google already owns the .google TLD. All they would need to do is attach an A record to the root.

The used to have an A record on some of the other TLDs they own.

I maintain a list at captnemo.in/tld-a-record

Do you remember the details?

Yes, I did a blog post when it happened: https://captnemo.in/blog/2018/06/02/google-tld-no-more-a-rec....

Google was pointing its TLDs to 127.0.0.53, and once they started using the TLDs actively, they removed the records.

ICANN doesn't allow A records on the DNS root on gTLDs. If they did, we already would've.

http://cm./ used to resolve

https://news.ycombinator.com/item?id=974111

http://to./ used to be a URL shortener.

Either the rules on ccTLDs are different, or it's a recentish crackdown.

There are no rules for ccTLDs, which their countries like to claim as sovereign property (see https://meetings.icann.org/en/dublin54/schedule/wed-ccnso-me... and https://gac.icann.org/principles-and-guidelines/public/princ... for some of the views)

Some ccTLDS voluntarily agree to abide by the same rules and operational requirements as the gTLDs - https://www.icann.org/resources/pages/cctld-2012-02-25-en has more details.

This means, for example, than ccTLDs aren’t bound by the findings from SAC053 at https://www.icann.org/en/system/files/files/sac-053-en.pdf which have been included in the gTLD contracts.

Wow, that's a glimpse into the past

.cm is a ccTLD, not a gTLD. All two-letter TLDs are ccTLDs and all 3+ letter TLDs are gTLDs. Countries have wide latitude to do whatever they want with their ccTLDs.

http://ai./ still works

What happens if I have a computer on my network named ai? Wouldn't it cause a conflict?

I called my FreeNAS box freenas because I didn't feel very creative, and I can browse to it as http://freenas or http://freenas./ but if they took a domain of 'freenas' suddenly that means I have to do extra work and rename my stuff. Doesn't seem fair!

You should have used a reserved TLD like `.local'. If you set it as the local domain name `ai' and `ai.local' would work. `ai.' would not work, as the dot at the end tells it to look at the top level. (like the slash at the beginning of an absolute path)

Ah yes I forgot that, .local does work so I guess that fixes that problem.

This explains so much. I'd always wondered why e.g. CERN's website was served from home.cern instead of just cern.

Sure, but why would they?

Can you imagine all the search advertising Google would lose if they allowed just 'nike' and 'att' to resolve the websites in Chrome, rather than search results?

Did we just find the secret to beating google?

With the super shady sale of .org it may be time to consider who these people are selling essentially nothing.

I was pondering about the idea of a browser extension for a sort of local DNS. Paying for domains could be ditched if everyone had a local copy of a database that mapped any character sequence (even emojis) to a server address that would get filled in by the extension in the url bar. Not sure if it would extend to mobile or all websites, but could work similar to a social network in the sense that you need to get the extension (i.e. access) first, and then you can find someone just by using a string that they've provided. Combined with something like Github pages it would offer a full free solution.

Seems like the hosts files of old and one kind of Alternative root. https://en.m.wikipedia.org/wiki/Alternative_DNS_root

I remember seeing a NIC that did this with their TLD. But I can't remember which one :/. I would love if someone finds it.

I think it's pretty common. Our national NIC does it: http://dk redirects to their website.

Does not load for me

"dk’s server IP address could not be found."

Another one is http://ai

Perhaps that will work.

This works if you add the trailing .: http://ai.

This redirected to https://dk.com for me. Is this my ISP's DNS messing with me?

Firefox will sometimes append .com onto the URL.

What a nasty non-feature. Can this be disabled?

Surprisingly, going to http://dk redirects to https://www.penguin.com.au/browse/dk for me. How strange (I've never visited the penguin website before this).

DNS suffix, to name one.

Reply on news.ycombinator.com