For me DNS lookup fails (using dig, dns-sd, dscacheutil). Maybe an ISP thing?
Edit: Ah, I see (quoted in the shortest URLs page):
> Dotless domains will not be universally reachable and the SSAC recommends strongly against their use. As a result, the SSAC also recommends that the use of DNS resource records such as A, AAAA, and MX in the apex of a TopLevel Domain (TLD) be contractually prohibited where appropriate and strongly discouraged in all cases.
Not ISP-dependent, but DNS-server dependent. The recommendation is also just that: It's not enforced. The TLD can decide to serve those records at the apex if it wishes, and any given DNS server should look them up as requested. It just so happens in your case it doesn't -- Windows DNS, for example, will not serve the TLD's A record.
Regardless, in your case, `dig pn. +noall +comments`, you'll likely see SERVFAIL.
Trying `dig pn. +noall +comments @8.8.8.8` and you'll see that you get a status of NOERROR.
(And, of course, `dig pn. +short @8.8.8.8` will give you the correct IP of 139.162.17.173)
The only browser that allows me to use any addon I want. I'll continue to use it until Mozilla fulfill their promise and allow me to load any addon I want.
I don't know if I'm more surprised by this quirk of DNS (as in, domains of the form <TLD><dot>) working, or more impressed that someone's still got a readily-available Netscape install to test it out.
How do you deal with constant SSL error messages? I tried to use Netscape for a while, it was a pleasant experience except the fact that most websites are HTTPS now and Netscape wasn't able to handle recent versions.
For a fun third Firefox (104.0.2) on Arch Linux behavior, it seems to resolve for me but warns me it can load it with TLS. I imagine that if I bypassed this it would work fine though.
I am impressed with HN gracefully degrading the handling of the URL. It doesn't know how to parse the url for the preview on the homepage, but it doesn't crash the page or prevent the link entry from rendering.
I’m not using anything fancy when I visit HN, what is it doing to preview the homepage and why would it crash or prevent anything from rendering? I don’t even understand what you’re congratulating HN for other than having hyperlinks.
I keep re-reading this and I just have to admit that I sincerely don’t understand. What runtime behavior is potentially crash prone in the first place? Are you saying stricter behavior would be more crash prone, or less? Which languages are more likely to be susceptible? I’m baffled and truly don’t mean to be dismissive.
Edit: nice article! I looked at the HN data and code, and trying to handle these as normal URLs with sitenames in parens would also let a lot of garbage through and/or be a lot of work. Given that the current submission kind of gracefully degraded (https://news.ycombinator.com/item?id=32805527), I think it's probably best not to invest the effort.
If it’s not working for you it’s probably due to the resolver you’re using or how your host’s stub resolver is configured. Some stub resolvers have rules about the number of dots in the name.
I’m using dnsmasq on my router pointed at Google DNS and it’s not currently resolving for me on my iPhone. I’ll have to check where the failure is occurring.
Edit: if I turn off WiFi on my iPhone, I can load the site just fine over Verizon cellular. So it’s my dnsmasq config most likely.
Yes. The trailing dot is typically to tell the stub resolver not to try qualifying with the local domain. Rules vary by stub resolver though. On iOS I’m getting “A server with the specified hostname could not be found. NSURLErrorDomain” regardless of browser. I’ll have to debug tomorrow when I’m not too lazy to get up off the couch from my phone.
Both of your URLs work for me. A 404 on the first. I stay logged in to HN on the second. My guess is my trouble is with my local dnsmasq config.
I used to work for a guy who registered borgosoft.com and CNAMEed it to microsoft.com. In the days before HTTP 1.1, a Host: header was not required on a request, and it would happily serve up microsoft.com's content under borgosoft.com. Eventually they started requiring a host header, and the trick stopped working. The guy I worked for claimed credit for making them do this; I have my doubts.
In 2020, someone find if you add "." in host suffix of Youtube url: http://www.youtube.com./ it will works without AD (IIRC, because AD resource file is viewed as different domain and blocked)
further context: https://news.ycombinator.com/item?id=32801359, and see also the other two-letter domains on the front page of HN today. i'm not really sure if this is what you were asking, but it at least explains what's going on to somebody from the future who finds their way back to this comment section.
69 comments
[ 3.7 ms ] story [ 146 ms ] threadEdit: Ah, I see (quoted in the shortest URLs page):
> Dotless domains will not be universally reachable and the SSAC recommends strongly against their use. As a result, the SSAC also recommends that the use of DNS resource records such as A, AAAA, and MX in the apex of a TopLevel Domain (TLD) be contractually prohibited where appropriate and strongly discouraged in all cases.
https://www.icann.org/en/announcements/details/new-gtld-dotl...
BTW, http://ai./ doesn't work for me neither.
Regardless, in your case, `dig pn. +noall +comments`, you'll likely see SERVFAIL.
Trying `dig pn. +noall +comments @8.8.8.8` and you'll see that you get a status of NOERROR.
(And, of course, `dig pn. +short @8.8.8.8` will give you the correct IP of 139.162.17.173)
https://en.wikipedia.org/wiki/Pitcairn_Islands
(ccTLD for the Pitcairn Islands, one of the most isolated communities on the planet)
Fitness for use trumps security.
Please explain the domain to us, how did you do it?
Edit: https://ibb.co/1Th7mTX
Edit: And IE: https://ibb.co/HK9pzbm
Edit: Camino, one of my favorite browsers: https://ibb.co/8x3STrj
(I work on retro websites and test in these all day long.)
Can you please share links?
Or I spent too much time reading O'Reilly's "DNS and BIND" and everyone else thinks "no, that's not the root-root zone, that's a quirk".
Though for general purpose browsing, the modern stuff like HTML5 / CSS3 / ES6+ would make it somewhere between unusable and nonfunctional.
There's some quirks that will resolve it, once its cached, but generally speaking, you need to tell systemd to resolve these names.
[0] https://wiki.archlinux.org/title/systemd-resolved#systemd-re...
Depends how strict they are at runtime
Edit: nice article! I looked at the HN data and code, and trying to handle these as normal URLs with sitenames in parens would also let a lot of garbage through and/or be a lot of work. Given that the current submission kind of gracefully degraded (https://news.ycombinator.com/item?id=32805527), I think it's probably best not to invest the effort.
I’m using dnsmasq on my router pointed at Google DNS and it’s not currently resolving for me on my iPhone. I’ll have to check where the failure is occurring.
Edit: if I turn off WiFi on my iPhone, I can load the site just fine over Verizon cellular. So it’s my dnsmasq config most likely.
For some fun: http://gs.com./ gives an error. https://news.ycombinator.com./ works, but has me logged out (probably the cookie doesn't match?).
Both of your URLs work for me. A 404 on the first. I stay logged in to HN on the second. My guess is my trouble is with my local dnsmasq config.
The HTTP host header which is sometimes (ab)used to perform a kind of not-quite-right-SNI doesn't have the luxury of being quite so well specified.
In 2020, someone find if you add "." in host suffix of Youtube url: http://www.youtube.com./ it will works without AD (IIRC, because AD resource file is viewed as different domain and blocked)
https://news.ycombinator.com/item?id=32804743
further context: https://news.ycombinator.com/item?id=32801359, and see also the other two-letter domains on the front page of HN today. i'm not really sure if this is what you were asking, but it at least explains what's going on to somebody from the future who finds their way back to this comment section.
> Apache/2.2.22 (Debian) Server at pn Port 443
Yikes!
Shortest URLs on the Internet - https://news.ycombinator.com/item?id=32801359 - Sept 2022 (168 comments)
.ai - https://news.ycombinator.com/item?id=32804189 - Sept 2022 (81 comments)