# HTTP 404
HTTP 404 (Not Found) is a client-error status code of the Hypertext Transfer Protocol that tells the client the origin server could not locate a current representation of the target resource, or declines to reveal whether one is present.[1](https://datatracker.ietf.org/doc/html/rfc9110/) It is the code a browser receives when a requested URL does not correspond to any resource the server can or will return.
| Key fact | Detail |
|---|---|
| Meaning | The origin server did not find a current representation for the target resource, or is not willing to disclose that one exists[1](https://datatracker.ietf.org/doc/html/rfc9110/) |
| Class | 4xx (Client Error): the request contains bad syntax or cannot be fulfilled; the first digit of a status code defines its class[2](https://www.rfc-editor.org/rfc/rfc9110.html) |
| Caching | 404 is one of the status codes defined as heuristically cacheable, so a cache may reuse it with heuristic expiration unless explicit cache controls say otherwise[2](https://www.rfc-editor.org/rfc/rfc9110.html) |
| Reason phrase | "Not Found" is only a recommendation; it can be replaced by local equivalents or left out without affecting the protocol[2](https://www.rfc-editor.org/rfc/rfc9110.html) |
| Related code | 410 Gone signals that the resource was deliberately removed and is unlikely to be available again, a distinction site owners use to tell search engines a URL is really gone[3](https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status) |
| Soft 404s | In a study of 11 million redirecting URIs, 62,000 custom 404 URIs were identified, almost half of them soft 404s, which can compromise SEO and user experience[4](https://arxiv.org/abs/2507.22019) |
## Definition in the HTTP standard
The [Internet Engineering Task Force](https://www.edgepedia.org/internet-engineering-task-force) defines HTTP semantics in RFC 9110 (June 2022), which lists 404 Not Found among the status codes in section 15.5.4.[2](https://www.rfc-editor.org/rfc/rfc9110.html) A status code is a three-digit integer describing the result of the request; all valid codes fall between 100 and 599, and the first digit alone determines the class, so a client that does not recognize a specific code must still treat it as equivalent to the x00 code of its class.[2](https://www.rfc-editor.org/rfc/rfc9110.html) The 4xx range covers client errors, meaning the request contains bad syntax or cannot be fulfilled.[2](https://www.rfc-editor.org/rfc/rfc9110.html)
The standard's wording carries an important qualification: a 404 does not tell the client whether the resource ever existed. The server may be concealing a resource it does have, and the code covers both cases.[1](https://datatracker.ietf.org/doc/html/rfc9110/) The accompanying reason phrase is advisory only; servers may substitute their own text or omit it entirely.[2](https://www.rfc-editor.org/rfc/rfc9110.html)
## Neighboring codes and how 404 differs
404 sits in a family of client-error codes defined by RFC 9110, including 400 Bad Request, 401 Unauthorized, 403 Forbidden, 405 Method Not Allowed, and 410 Gone.[2](https://www.rfc-editor.org/rfc/rfc9110.html) The distinctions matter in practice. A 403 Forbidden means the client lacks access rights but the resource exists; a 404 means the server cannot find the requested resource, which in a browser means the URL is not recognized.[3](https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status)
## Soft 404s and web practice
A **soft 404** is a page that returns HTTP 200 (success) with content that effectively says the page does not exist, or otherwise mishandles the missing-resource case. In a 2025 analysis of 11 million redirecting URIs, researchers identified 62,000 custom 404 URIs, almost half of which were soft 404s, a pattern the authors note can compromise SEO and user experience.[4](https://arxiv.org/abs/2507.22019)
404 responses also matter in web archiving. Because 404 is heuristically cacheable, caching a 404 response can prevent a browser's repeated requests for a missing page from reaching a web archive server, eliminating unnecessary archival replay requests.[5](https://arxiv.org/abs/2212.00760) For readers who hit a 404, research on link rot has explored rediscovering missing pages by extracting the page's title and generating a lexical signature from it, or from the page's link neighborhood, to feed search engines.[6](https://arxiv.org/abs/0907.2268)
## Origin of the number
The three-digit scheme predates the specific code. [Tim Berners-Lee](https://www.edgepedia.org/tim-berners-lee) and colleagues defined HTTP's status codes in the early 1990s at CERN, following a logic borrowed from older protocols such as FTP and NNTP: a three-digit code in which the first digit indicates the category of the response.[7](https://pulseit.novaorbit.com.br/origem-erro-404-sala-cern/) The Web's core software, including the first server, was developed at CERN around 1990, and info.cern.ch was the address of the world's first web server.[8](https://timeline-upgrade.web.cern.ch/timeline-header/90)
A popular legend holds that 404 refers to a room 404 in the CERN building where the Web was born, supposedly housing the first servers or Berners-Lee's office. This story is a myth: the code is simply an item in the status table the protocol's designers worked out, not a reference to a physical room.[7](https://pulseit.novaorbit.com.br/origem-erro-404-sala-cern/)
## References
1. [RFC 9110: HTTP Semantics, section on 404 (Not Found), IETF Datatracker](https://datatracker.ietf.org/doc/html/rfc9110/)
2. [RFC 9110: HTTP Semantics, RFC Editor](https://www.rfc-editor.org/rfc/rfc9110.html)
3. [HTTP response status codes, MDN Web Docs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status)
4. [Not Here, Go There: Analyzing Redirection Patterns on the Web, arXiv:2507.22019](https://arxiv.org/abs/2507.22019)
5. [Caching HTTP 404 Responses Eliminates Unnecessary Archival Replay Requests, arXiv:2212.00760](https://arxiv.org/abs/2212.00760)
6. [Evaluating Methods to Rediscover Missing Web Pages from the Web Infrastructure, arXiv:0907.2268](https://arxiv.org/abs/0907.2268)
7. [A origem do erro 404: a sala do CERN que nunca existiu, PulseIT](https://pulseit.novaorbit.com.br/origem-erro-404-sala-cern/)
8. [The birth of the World Wide Web, CERN timeline](https://timeline-upgrade.web.cern.ch/timeline-header/90)
---
*Topic: Encyclopedia › Technology and the built world › Computing and digital systems › Networks and security › HTTP and web communication protocols*
*Initially written Sep 17, 2026 · Reviewed: Sep 30, 2026; Oct 15, 2026 · Edited: Oct 15, 2026 · Last review: Oct 15, 2026*
*Copyright 2026 EdgeChat AI, a subsidiary of Biostate AI.*
License: Edgepedia Community License 1.0, https://www.edgepedia.org/edgepedia/license