Section 1.1.1 of RFC 3986 defines URI syntax as "a federated and extensible naming system wherein each scheme's specification may further restrict the syntax and semantics of identifiers using that scheme." In other words, the structure of a URI is defined by its scheme. While it is common for schemes to further delegate their substructure to the URI's owner, publishing independent standards that mandate particular forms of substructure in URIs is often problematic. This document provides guidance on the specification of URI substructure in standards. This document obsoletes RFC 7320 and updates RFC 3986.

This memo documents an Internet Best Current Practice. This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Further information on BCPs is available in Section 2 of RFC 7841. Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc8820.

Copyright (c) 2020 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.

URIs [RFC 3986] very often include structured application data. This might include artifacts from filesystems (often occurring in the path component) and user information (often in the query component). In some cases, there can even be application-specific data in the authority component (e.g., some applications are spread across several hostnames to enable a form of partitioning or dispatch). Implementations can impose further constraints upon the structure of URIs; for example, many web servers use the filename extension of the last path segment to determine the media type of the response. Likewise, prepackaged applications often have highly structured URIs that can only be changed in limited ways (often, just the hostname and port on which they are deployed). Because the owner of the URI (as defined in [webarch], Section 2.2.2.1) is choosing to use the server or the application, this can be seen as reasonable delegation of authority. However, when such conventions are mandated by a party other than the owner, it can have several potentially detrimental effects:

• Collisions - As more ad hoc conventions for URI structure become standardized, it becomes more likely that there will be collisions between them (especially considering that servers, applications, and individual deployments will have their own conventions).
• Dilution - When the information added to a URI is ephemeral, this dilutes its utility by reducing its stability (see [webarch], Section 3.5.1) and can cause several alternate forms of the URI to exist (see [webarch], Section 2.3.1).
• Rigidity - Fixed URI syntax often interferes with desired deployment patterns. For example, if an authority wishes to offer several applications on a single hostname, it becomes difficult to impossible to do if their URIs do not allow the required flexibility.
• Operational Difficulty - Supporting some URI conventions can be difficult in some implementations. For example, specifying that a particular query parameter be used with "http" URIs can preclude the use of web servers that serve the response from a filesystem. Likewise, an application that fixes a base path for its operation (e.g., "/v1") makes it impossible to deploy other applications with the same prefix on the same host.
• Client Assumptions - When conventions are standardized, some clients will inevitably assume that the standards are in use when those conventions are seen. This can lead to interoperability problems; for example, if a specification documents that the "sig" URI query parameter indicates that its payload is a cryptographic signature for the URI, it can lead to undesirable behavior.

Publishing a standard that constrains an existing URI structure in ways that aren't explicitly allowed by [RFC 3986] (usually, by updating the URI scheme definition) is therefore sometimes problematic, both for these reasons and because the structure of a URI needs to be firmly under the control of its owner. This document explains some best current practices for establishing URI structures, conventions, and formats in standards. It also offers strategies for specifications in Section 3.

This document's guidelines and requirements target the authors of specifications that constrain the syntax or structure of URIs or parts of them. Two classes of such specifications are called out specifically:

• Protocol Extensions ("Extensions") - specifications that offer new capabilities that could apply to any identifier or to a large subset of possible identifiers, e.g., a new signature mechanism for "http" URIs, metadata for any URI, or a new format.
• Applications Using URIs ("Applications") - specifications that use URIs to meet specific needs, e.g., an HTTP interface to particular information on a host.

Requirements that target the generic class "Specifications" apply to all specifications, including both those enumerated above and others. Note that this specification ought not be interpreted as preventing the allocation of control of URIs by parties that legitimately own them or have delegated that ownership; for example, a specification might legitimately define the semantics of a URI on IANA's web site as part of the establishment of a registry. There may be existing IETF specifications that already deviate from the guidance in this document. In these cases, it is up to the relevant communities (i.e., those of the URI scheme as well as any relevant community that produced the specification in question) to determine an appropriate outcome, e.g., updating the scheme definition or changing the specification.

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC 2119] [RFC 8174] when, and only when, they appear in all capitals, as shown here.

This section updates [RFC 3986] by advising Specifications how they should define structure and semantics within URIs. Best practices differ, depending on the URI component in question, as described below.

Applications and Extensions can require the use of one or more specific URI schemes; for example, it is perfectly acceptable to require that an Application support "http" and "https" URIs. However, Applications ought not preclude the use of other URI schemes in the future, unless they are clearly only usable with the nominated schemes. A Specification that defines substructure for URI schemes overall (e.g., a prefix or suffix for URI scheme names) MUST do so by modifying [BCP35] (an exceptional circumstance).

Scheme definitions define the presence, format, and semantics of an authority component in URIs; all other Specifications MUST NOT constrain or define the structure or the semantics for URI authorities, unless they update the scheme registration itself or the structures it relies upon (e.g., DNS name syntax, as defined in Section 3.5 of RFC 1034). For example, an Extension or Application cannot say that the "foo" prefix in "https://foo_app.example.com" is meaningful or triggers special handling in URIs, unless they update either the "http" URI scheme or the DNS hostname syntax. Applications can nominate or constrain the port they use, when applicable. For example, BarApp could run over port nnnn (provided that it is properly registered).

Scheme definitions define the presence, format, and semantics of a path component in URIs, although these are often delegated to the Application(s) in a given deployment. To avoid collisions, rigidity, and erroneous client assumptions, Specifications MUST NOT define a fixed prefix for their URI paths -- for example, "/myapp" -- unless allowed by the scheme definition. One such exception to this requirement is registered "well-known" URIs, as specified by [RFC 8615]. See that document for a description of the applicability of that mechanism. Note that this does not apply to Applications defining a structure of a URI's path "under" a resource controlled by the server. Because the prefix is under control of the party deploying the Application, collisions and rigidity are avoided, and the risk of erroneous client assumptions is reduced. For example, an Application might define "app_root" as a deployment-controlled URI prefix. Application-defined resources might then be assumed to be present at "{app_root}/foo" and "{app_root}/bar". Extensions MUST NOT define a structure within individual URI components (e.g., a prefix or suffix), again to avoid collisions and erroneous client assumptions.

The presence, format, and semantics of the query component of URIs are dependent upon many factors and can be constrained by a scheme definition. Often, they are determined by the implementation of a resource itself. Applications can specify the syntax of queries for the resources under their control. However, doing so can cause operational difficulties for deployments that do not support a particular form of a query. For example, a site may wish to support an Application using "static" files that do not support query parameters. Extensions MUST NOT constrain the format or semantics of queries, to avoid collisions and erroneous client assumptions. For example, an Extension that indicates that all query parameters with the name "sig" indicate a cryptographic signature would collide with potentially preexisting query parameters on sites and lead clients to assume that any matching query parameter is a signature. Per the "Form submission" section of [HTML5], HTML constrains the syntax of query strings used in form submission. New form languages are encouraged to allow creation of a broader variety of URIs (e.g., by allowing the form to create new path components, and so forth).

Section 3.5 of RFC 3986 specifies fragment identifiers' syntax and semantics as being dependent upon the media type of a potentially retrieved resource. As a result, other Specifications MUST NOT define structure within the fragment identifier, unless they are explicitly defining one for reuse by media types in their definitions (for example, as JSON Pointer [RFC 6901] does). An Application that defines common fragment identifiers across media types not controlled by it would engender interoperability problems with handlers for those media types (because the new, non-standard syntax is not expected).

Given the issues described in Section 1, the most successful strategy for Applications and Extensions that wish to use URIs is to use them in the fashion for which they were designed: as links that are exchanged as part of the protocol, rather than statically specified syntax. Several existing specifications can aid in this. [RFC 8288] specifies relation types for web links. By providing a framework for linking on the Web, where every link has a relation type, context, and target, it allows Applications to define a link's semantics and connectivity. [RFC 6570] provides a standard syntax for URI Templates that can be used to dynamically insert Application-specific variables into a URI to enable such Applications while avoiding impinging upon URI owners' control of them. [RFC 8615] allows specific paths to be "reserved" for standard use on URI schemes that opt into that mechanism ("http" and "https" by default). Note, however, that this is not a general "escape valve" for Applications that need structured URIs; see that specification for more information. Specifying more elaborate structures in an attempt to avoid collisions is not an acceptable solution and does not address the issues described in Section 1. For example, prefixing query parameters with "myapp_" does not help, because the prefix itself is subject to the risk of collision (since it is not "reserved").

This document does not introduce new protocol artifacts with security considerations. It prohibits some practices that might lead to vulnerabilities; for example, if a security-sensitive mechanism is introduced by assuming that a URI path component or query string has a particular meaning, false positives might be encountered (due to sites that already use the chosen string). See also [RFC 6943].

This document has no IANA actions.

[RFC2119]

S. Bradner, "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.

[RFC3986]

T. Berners-Lee, R. Fielding, and L. Masinter, "Uniform Resource Identifier (URI): Generic Syntax", STD 66, RFC 3986, DOI 10.17487/RFC3986, January 2005,
<https://www.rfc-editor.org/info/rfc3986>.

[RFC8174]

B. Leiba, "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017,
<https://www.rfc-editor.org/info/rfc8174>.

[webarch]

I. Jacobs, and N. Walsh, "Architecture of the World Wide Web, Volume One", December 2004,
<https://www.w3.org/TR/2004/REC-webarch-20041215>.

[BCP35]

D. Thaler, T. Hansen, and T. Hardie, "Guidelines and Registration Procedures for New URI Schemes", BCP 35, RFC 7595, June 2015,
<https://www.rfc-editor.org/info/bcp35>.

[HTML5]

WHATWG, "HTML - Living Standard", June 2020,
<https://html.spec.whatwg.org/#form-submission>.

[RFC1034]

P.V. Mockapetris, "Domain names - concepts and facilities", STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987,
<https://www.rfc-editor.org/info/rfc1034>.

[RFC6570]

J. Gregorio, R. Fielding, M. Hadley, M. Nottingham, and D. Orchard, "URI Template", RFC 6570, DOI 10.17487/RFC6570, March 2012,
<https://www.rfc-editor.org/info/rfc6570>.

[RFC6901]

P. Bryan, K. Zyp, and M. Nottingham, "JavaScript Object Notation (JSON) Pointer", RFC 6901, DOI 10.17487/RFC6901, April 2013,
<https://www.rfc-editor.org/info/rfc6901>.

[RFC6943]

D. Thaler, "Issues in Identifier Comparison for Security Purposes", RFC 6943, DOI 10.17487/RFC6943, May 2013,
<https://www.rfc-editor.org/info/rfc6943>.

[RFC8288]

M. Nottingham, "Web Linking", RFC 8288, DOI 10.17487/RFC8288, October 2017,
<https://www.rfc-editor.org/info/rfc8288>.

[RFC8615]

M. Nottingham, "Well-Known Uniform Resource Identifiers (URIs)", RFC 8615, DOI 10.17487/RFC8615, May 2019,
<https://www.rfc-editor.org/info/rfc8615>.

Many of the requirements of RFC 7320 were removed, in the spirit of making this BCP guidance rather than rules.

Thanks to David Booth, Dave Crocker, Tim Bray, Anne van Kesteren, Martin Thomson, Erik Wilde, Dave Thaler, and Barry Leiba for their suggestions and feedback.

Mark Nottingham