I found a bunch of “404”/“page not found” galleries and sites. But when was the expression first used, and why those numbers in particular?
Geek equivalent of “the lights are on, but nobody’s home” used to be to roll your eyes and mutter ‘404’ to your colleague.
Be thankful when you do get a 404 error.
These days, spammers and hackers have most alternate spellings of popular domains allocated. Mistype drudgereport, amazon, or any other popular site and you’re playing Russian Roulette. There’s some nasty code traps out there.
It’s worth mentioning that response codes also happen when there is not an error - for example, response code 200 indicates that the HTTP request was received and accepted - the user doesn’t normally become aware of this, because it usually means the client (for example a web browser) can carry on and do what was asked of it (i.e. display a web page).
Yep. Wikipedia also has an article covering all the other HTTP status codes.
So that’s what she meant by 413!
You sure it wasn’t a 402? 
Being a long-time geek, I must state that “404” as a geekism must post-date the introduction of “404” as an internet error message, and therefore cannot be the origin of the term.
Does anyone actually type things into their address bar anymore? That’s what Google is for.
Google doesn’t search intranets.
Surely any decently-made intranet should include navigable linked indexes and ideally, its own search engine?
Generally you won’t get a 404 error if you try to visit an invalid domain. To get a 404 error you have to be communicating with an HTTP server, but if you try to connect to an invalid domain there is no HTTP server to talk to.
Of course. But I’m willing to bet that the non-decently made intranets outnumber the decently-made ones.
Wikipedia (the source of all approximate knowledge) gives a 2008 citation. I’m too lazy to look for more, considering how much dross the phrase 404 generates, however, apparently the error codes were established in '92
Yes, it does.
No, it doesn’t. It’s clear to anyone with more than three brain cells that psychonaut was referring to Google’s public search service, not hardware search appliances that they sell to customers.
But congratulations for realizing that a word can have different meanings in different contexts. I’m sure that feat made you feel super-l33t.
These status codes are based on a kind of defacto unixy way of doing things. You’ll see these codes in smtp, ftp, etc. Generally 100-300’s means successful/message and 400-500’s mean error. Next time you fat finger someone’s email address, read the bounce message. It’ll probably be something like ‘550 - user doesnt exist.’
When TBL was writing the HTTP spec, he was just following old unix pricinples.
Cool. Thanks.
Interesting that he then agrees there are intranets with “navigable linked indexes and ideally, its own search engine”… I guess taking the thread in context is too hard? Perhaps if you weren’t so quick to make ad hominem attacks you would have realized?
Google does in fact search intranets, and in fact there are a number of large organizations that are using Google’s tool to do just that.
I know this is hard to grasp. There’s no way to describe it in smaller words, though, sorry.