RFC 2324 was written by Larry Masinter, who describes it as a satire, saying "This has a serious purpose – it identifies many of the ways in which HTTP has been extended inappropriately."[5] The wording of the protocol made it clear that it was not entirely serious; for example, it notes that "there is a strong, dark, rich requirement for a protocol designed espressoly for the brewing of coffee".
Despite the joking nature of its origins, or perhaps because of it, the protocol has remained as a minor presence online. The editor Emacs includes a fully functional client-side implementation of it,[6] and a number of bug reports exist complaining about Mozilla's lack of support for the protocol.[7] Ten years after the publication of HTCPCP, the Web-Controlled Coffee Consortium (WC3) published a first draft of "HTCPCP Vocabulary in RDF"[8] in parody of the World Wide Web Consortium (W3C)'s "HTTP Vocabulary in RDF".[9]
On April 1, 2014, RFC 7168 extended HTCPCP to fully handle teapots.[4]
Commands and replies
HTCPCP is an extension of HTTP. HTCPCP requests are identified with the Uniform Resource Identifier (URI) scheme coffee (or the corresponding word in any other of the 29 listed languages) and contain several additions to the HTTP methods:
Method
Definition
BREW or POST
Causes the HTCPCP server to brew coffee. Using POST for this purpose is deprecated. A new HTTP request header field, "Accept-Additions", is proposed, supporting optional additions including Cream, Whole-milk, Vanilla, Raspberry, Whisky, Aquavit, etc.
The HTCPCP server is unable to provide the requested addition for some reason; the response should indicate a list of available additions. The RFC observes, "In practice, most automated coffee pots cannot currently provide additions."
408 Request Timeout
The HTCPCP server is unable to make tea for a timeout and forbidden actions.
418 I'm a teapot
The HTCPCP server is a teapot; the resulting entity body "may be short and stout" (a reference to the song "I'm a Little Teapot"). Demonstrations of this behaviour exist.[1][10]
503 Service Unavailable
According to Mozilla Developer Documentation, "A combined coffee/tea pot that is temporarily out of coffee should instead return 503", when requested to brew.[11]
Save 418 movement
On 5 August 2017, Mark Nottingham, chairman of the IETFHTTPBIS Working Group, called for the removal of status code 418 "I'm a teapot" from the Node.js platform, a code implemented in reference to the original 418 "I'm a teapot" established in Hyper Text Coffee Pot Control Protocol.[12] On 6 August 2017, Nottingham requested that references to 418 "I'm a teapot" be removed from the programming language Go[13] and subsequently from Python's Requests[14] and ASP.NET's HttpAbstractions library[15] as well.
In response, 15-year-old developer Shane Brunswick created a website, save418.com,[16] and established the "Save 418 Movement", asserting that references to 418 "I'm a teapot" in different projects serve as "a reminder that the underlying processes of computers are still made by humans". Brunswick's site went viral in the hours following its publishing, garnering thousands of upvotes on the social platform Reddit,[17] and causing the mass adoption of the "#save418" Twitter hashtag he introduced on his site. Heeding the public outcry, Node.js, Go, Python's Requests, and ASP.NET's HttpAbstractions library decided against removing 418 "I'm a teapot" from their respective projects. The unanimous support from the aforementioned projects and the general public prompted Nottingham to begin the process of having 418 marked as a reserved HTTP status code,[18] ensuring that 418 will not be replaced by an official status code for the foreseeable future.
On 5 October 2020, Python 3.9 released with an updated HTTP library including 418 IM_A_TEAPOT status code.[19] In the corresponding pull request, the Save 418 movement was directly cited in support of adoption.[20]
Usage
The status code 418 is sometimes returned by servers when blocking a request, instead of the more appropriate 403 Forbidden,[21] or 404 Not Found.[22]
^Masinter, Larry M. (April 1998), "Request for Comments 2324", Network Working Group, IETF, archived from the original on 2012-04-04, retrieved 2012-03-20
