Hacker News new | ask | show | jobs
by djrogers 1980 days ago
This is a bad link, not sure how it got upvoted when following it fails (there’s a trailing . after the domain). That’s kinda fishy...

Correct link should be https://bugs.xdavidhu.me/google/2021/01/18/the-embedded-yout...

5 comments

The trailing dot is fine.

http://www.dns-sd.org/trailingdotsindomainnames.html

Though now I'm curious why it fails for you.

Fails in Safari for me. Works fine in Firefox and Chrome on the same machine.
Sounds like a bug in Safari.
It works for me, and having a dot after the TLD is perfectly valid.[1]

1. Page 7 of RFC 1034 https://www.ietf.org/rfc/rfc1034.txt

Interestingly enough, I can follow it without any errors.

Firefox 84.0 on Arch Linux.

Working in Mozilla, Edge, Chrome 87....
Adding the dot after the tld actually works on some sites to bypass paywalls.
And in the old days people were accustomed to be able to resolve local hostnames, without having to supply any domain part. The "search" keyword in resolv.conf is a remnant of this. By adding a final dot, you ensured that the domain part you wrote referred to the global DNS root, not any local one. Unless of course someone had changed "ndots" which controls how many dots are needed to disable the search feature.