HTTP Cats | Hacker News Reader

HTTP Cats

http.cat

408 pointsafshinmeh74 comments

The .cat domain is restricted to Catalan-speaking stuff, but this site still exists. Well played, Internet. https://en.wikipedia.org/wiki/.cat#Restrictions

Poking around, it has a catalan translation, but not a spanish or french translation. https://http.cat/?lang=cat https://http.cat/?lang=es https://http.cat/?lang=fr

Edit: http://nyan.cat/ has a catalan version too (català). Apparently you've been nyaning is has nyanyejat in catalan:

"HAS NYANYEJAT DURANT 117.6 SEGONS Tweet Your Score"

Nyan is an Onomatopœia for the sound a cat makes, commonly used in Japan. English usually uses "meow". The catalan word is "mèu".

The "to meow" verb would be "miolar", thus "Nyanyejar" is just "to nyan" turned into a catalan verb, kinda.

nyancats only nyan though...

Even more, I suspect the reading as “nyan” was actually accidental, the onomatopoeia is usually just “nya” if I recall correctly.

Which is easy to understand. The music in the background of the “Nyancat” video is actually produced with UTAU, a sort of offshoot of Vocaloid, the voice synthesis software, using an “UTAUloid” called Momo, with the song being entitled “Nyanyanyanyanyanyanya!”. It seems that “Nyan” resulted from parsing it out of the sequential “nya”s.

It’s kind of fascinating how the “poptart cat” gif and the UTAU song somehow ended up merged into this bizarre internet phenomena.

I also wonder if its at least partly (presumably, subconsciously) the adoption of the -an suffix (as adjective marker) from English: note the noun in use is almost always nyan cat / nyancat, so nyan is serving something of an adjectival role.

It's fascinating the wash back and forth across morphologically unrelated languages.

>It seems that “Nyan” resulted from parsing it out of the sequential “nya”s.

"nyanko" is a quite informal word to refer to a cat or kitten. Sorta like a child calling a cat a "meow-meow".

In that sense, is it actually being used as an adjective proper? This is amusing. I would’ve assumed it was just an accident, but it would be more amusing if it was both accidental and “sort of” grammatically correct.

Nyanko is just a name. The usual name for cat is "neko". There's nothing grammatically wrong about using nyanko in place of neko.

Sorry, I meant the “nyan” in “nyanko”, being an adjective. But thinking about it more, the added “n” syllable probably has more to do with phonetics than language semantics. I studied only basic Japanese and now am pretty rusty.

Oh, you thought it was an '-n' adjective. I get it now.

As the onomatopoeia is actually nya or nyaa, and the meme is "nyancat", there could be something to this.

And a NYAN language setting - clever.

I got NaN on iOS Safari:

HAS NYANYEJAT DURANT NaN SEGONS Tweet Your Score

You've NYANED for NaN seconds Tweet Your Score

NYAN NYAN NYAN NaN NYAN Tweet Your Score

I'm glad someone else is worried about this. For some reason Catalans are very special and love tlds. For example in BCN one is within the geometric catchment for:

  - .barcelona
  - .cat
  - .es
  - .eu

I don't think any other settlement on earth is so blessed.

For what it's worth, the registrar for .cat (Nominalia) isn't very strict

The status of Catalonia is an enormously controversial issue. In 2017 the government of Catalonia declared independence from Spain, resulting in the Catalonian government being forcibly deposed by the Spanish government, which is opposed to Catalonian independence regardless of the wishes of the Catalonians.

So it's not that surprising they would want their own, .scot is a similar situation.

Except in 2014 Scotland's referendum was to remain in the UK. Scotland is a country too, so it's really quite different.

Scotland is only a "country" by the UK's totally weird definition of the word. It's not that different from Niedersachsen or Alabama or, indeed, Catalonia in that it has its own parliament etc but it's not independent.

In the UK (and in pro football, somehow), "country" appears to mean "either a member of the UN or a federal state of the UK"

If it is a weird definition, it's a well-recognised one. Take the second sentence of the relevant Wikipedia page:

"A country may be an independent sovereign state or part of a larger state, as a non-sovereign or formerly sovereign political division, or a geographic region associated with sets of previously independent or differently associated people with distinct political characteristics." - https://en.wikipedia.org/wiki/Country

Yes, England's definition of the English word "country" includes Scotland

That's not a weird definition by definition - English is what England defines it to be.

> I don't think any other settlement on earth is so blessed.

AFAIK you can buy them anywhere, but Moscow-related (and other major Russian cities) extensions include:

.RU .SU .MOSCOW .МОСКВА .РУС (Rus) .РФ (Russian Federation)

And with second levels:

MSK.RU MSK.SU

Can you actually still get .su domains?

Absolutely. The Soviet Union may be dormant, but they'll happily give you a domain, presumably it's possible to have a website that's relevant to it after its supposed demise.

It's also in "de facto" use by the CIS[0] successor organisation.

[0] https://en.wikipedia.org/wiki/Commonwealth_of_Independent_St...

I can think of a locality that is covered by a higher number of geographic gTLDs: Cologne, Germany, has .cologne, .koeln, .nrw, .de, and .eu.

(But, granted, Barcelona actually has .bcn as well.)

> .bcn domain is intended to maintain a relationship of "cooperation and sum of efforts" with the .cat domain

I wanted to learn why this TLD exists but I didn't understand the explanation. What does cooperating with another TLD mean and can I use it for bacon stuff?

So, a website has to have Catalan content in order to be eligible to use .cat? Couldn't someone just hire translators and/or make friends with Catalan-speaking people to translate some English content to Catalan.

I'm pretty sure that is exactly the goal of the rule.

I think they want the person who owns the domain to be Catalan, and since the Spanish government doesn't want Catalonia to be recognized as an independent place, the language used is a proxy for being from there. I think they might try and limit it to people who speak Catalan.

I feel this is an appropriate place for a shameless plug about my website, http://ipkitten.com. It offers valuable IPv4 information while also offering a kitten GIF. No ads, no tracking; just cats and IP addresses. Zero lines of Javascript.

That's good. I've been using jsonip but I think I'll switch to yours, because it's still nice and minimal! It's good to have the geoip even though it's inaccurate. Right now geoip thinks I'm in Ft Lauderdale when I'm actually in Miami.

that is an oddly appropriate shameless plug.

It doesn't handle non-US countries particularly well. It found my city, and the post code is close by (2000), but then the query in the Google Maps URL is just "2000". How will that get me here? Maybe add the country code to the Google Maps URL, e.g. "<country code> <post code>"?

Otherwise, appropriate shameless plug. I only mention the above, because I could see myself making use of it.

Please refresh and see what happens.

Thanks for the quick change. That's definitely better.

Unfortunately, most European countries aren't federal states, and so dividing them by "state" (more like region, county, department or province) makes little sense. Although, that being said, I am confident that Google would find "2000 Capital Region" correctly, no matter where I search from. (Can't really test it, since Google is clever enough to base the post code on where I'm searching from.)

I am able to be neither a beggar nor a chooser with the API that I am leveraging to geo-code the IP addresses, because I am not paying for it! Either way, thank you for letting me know about non-US use-cases.

Did we break it?

Yes, thank you.

EDIT: We are back.

It seems to have been hugged to death (at least for me in Canada)

> no tracking

I wouldn't say there is no tracking at all. There might be no tracking by you and there certainly is no Javascript tracking applied, both of which is better than >90% websites out there, but you use two external resources, first bootstrapcdn for css and second giphy for the kitten gif. Those web sites get data for each request, including ip address, user agent and (possibly / if the browser supports it) cookie information.

giphy is a startup that is monetized via ads while bootstrapcdn's privacy policy explicitly states that it uses the ip address and user agent information for targeted advertisement [1]. As you are using a deeplink, you might avoid giphy's tracking which might only be js based, but one can never know. bootstrapcdn was built to be used as a CDN so they will definitely use the data that servers get for tracking.

So don't visit this website too often unless you want to see cat food ads :). Or IDK.

I don't think there is a big benefit from using a CDN for the CSS file, but self hosting the kitten picture means of course a big increase of traffic and idk how it looks like from a license point of view. I have no idea what a privacy friendly kitten picture hosting service would be.

[1]: https://www.bootstrapcdn.com/privacy-policy/

Decentraleyes neuters CDN tracking.

    This server could not prove that it is ipkitten.com;
    its security certificate is from
    cambridgecamps.paytrax.com.
    This may be caused by a misconfiguration or an
    attacker intercepting your connection.

Sad face :(

The only goal of the website is to deliver kitten GIFs and IPv4 addresses. It does not send an SSL certificate, pursuant to that goal.

But it does send a certificate if you ask for it - cambridgecamps.paytrax.com https://ipkitten.com/

I can see my inter-net number at the top and it knows I have firefox. How is that not tracking. :(

(\s. Just in case)

Gives Firefox (via HTTPS Everywhere) a sad face.

>Felkod: SSL_ERROR_BAD_CERT_DOMAIN

>cambridgecamps.paytrax.com

It gave me the face melt scene from Raiders of the Lost Ark, definitely not a kitten

[maybe NSFW and could be disturbing or funny depending on your personality]:

https://i.imgur.com/nu5jkyo.png

https://giphy.com/gifs/room-mega-ilovefuzzcom-oYbahUPK6yWfC

Where is the valuable ipv4 information? I knew my IP address and user agent already.

> 420 Enhance Your Calm

Ha, no way.

https://en.wikipedia.org/wiki/List_of_HTTP_status_codes

Unofficial Codes .... 420 Enhance Your Calm (Twitter) Returned by version 1 of the Twitter Search and Trends API when the client is being rate limited; versions 1.1 and later use the 429 Too Many Requests response code instead.[74]

https://placekitten.com/ is a genuinely useful page for generating placeholder cat images on the fly. Invaluable when prototyping an interface.

Is there a ToS or anything for this, or is it "please don't fire huge traffic loads"?

I use it regularly as a quick way to look up status codes. If some log just shows http 418, typing http.cat/418 is the fastest way I've found to get the corresponding message.

Yeah, this is our go-to way of referencing status codes at work.

> Resta prohibit explícitament per la ICANN la utilització del domini .cat per a pàgines de gats (cat en anglès), llevat que siguin en català o tinguin a veure amb la cultura catalana.

It's explicitly forbbiden by ICANN the use of .cat domain for pages about cats, unless are in catalan [language] or it has to do with catalan culture.

[1] https://ca.wikipedia.org/wiki/.cat

Pointed out by another user, but the site has a Catalan translation

https://http.cat/?lang=cat

It does have a link to the catalan version at the bottom.

This really isn't enforced. I have a .cat domain that is simply a picture of my cat and I've never had an issue.

I think if you're running something more high-profile that drives a ton of content, that rule would come into play.

Party poopers :-P

https://http.cat/451

451 Unavailable For Legal Reasons

This one is brilliant, and has a cat too. The reference is acknowledged in the RFC also: https://tools.ietf.org/html/rfc7725#page-5

Even April Fool's codes!

https://http.cat/418

I've found that one to have a useful purpose in a very specific situation, ie. an endpoint merged with another one and the behavior is sufficiently different that the service replying to the request is not what the requester had in mind. Yes versioning would have prevented that but in our setup, it was not possible.

> the service replying to the request is not what the requester had in mind

Sounds like a 400 bad request? If the service can't make heads or tails of it...

My favorite HTTP status site is http://httpstatusrappers.com

Can we have dogs as well?

You may! https://httpstatusdogs.com/

I got a status 500 error. With no cat :(

I had the same! Missed opportunity.

I immediately looked for "418 I'm a teapot", and the expectations were fulfilled :)

ha! We did a similar thing for our 404, 500, and empty pages:

- https://serpapi.com/404.htm

- https://serpapi.com/500.html

I don't think your "404.html" page is what you intended to link :P

404.html will be the one not displaying a 404 for some reason, but 404.htm does lol.

You are using rails, remove the default 404.html from your public folder and it will stop serving the default rails 404.html.

That's a good point. Not sure why we haven't already did it. Just pushed a fix. Demoing 404s are not day to day business though. :)