RFC 8265






Internet Engineering Task Force (IETF)                    P. Saint-Andre
Request for Comments: 8265                                    Jabber.org
Obsoletes: 7613                                              A. Melnikov
Category: Standards Track                                      Isode Ltd
ISSN: 2070-1721                                             October 2017


Preparation, Enforcement, and Comparison of Internationalized Strings
                  Representing Usernames and Passwords

Abstract



   This document describes updated methods for handling Unicode strings
   representing usernames and passwords.  The previous approach was
   known as SASLprep (RFC 4013) and was based on Stringprep (RFC 3454).
   The methods specified in this document provide a more sustainable
   approach to the handling of internationalized usernames and
   passwords.  This document obsoletes RFC 7613.

Status of This Memo



   This is an Internet Standards Track document.

   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
   Internet Standards 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/rfc8265.

Copyright Notice



   Copyright (c) 2017 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.




Saint-Andre & Melnikov       Standards Track                    [Page 1]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


Table of Contents



   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   3
   2.  Terminology . . . . . . . . . . . . . . . . . . . . . . . . .   4
   3.  Usernames . . . . . . . . . . . . . . . . . . . . . . . . . .   5
     3.1.  Definition  . . . . . . . . . . . . . . . . . . . . . . .   5
     3.2.  Case Mapping vs. Case Preservation  . . . . . . . . . . .   6
     3.3.  UsernameCaseMapped Profile  . . . . . . . . . . . . . . .   7
       3.3.1.  Rules . . . . . . . . . . . . . . . . . . . . . . . .   7
       3.3.2.  Preparation . . . . . . . . . . . . . . . . . . . . .   8
       3.3.3.  Enforcement . . . . . . . . . . . . . . . . . . . . .   8
       3.3.4.  Comparison  . . . . . . . . . . . . . . . . . . . . .   9
     3.4.  UsernameCasePreserved Profile . . . . . . . . . . . . . .   9
       3.4.1.  Rules . . . . . . . . . . . . . . . . . . . . . . . .   9
       3.4.2.  Preparation . . . . . . . . . . . . . . . . . . . . .   9
       3.4.3.  Enforcement . . . . . . . . . . . . . . . . . . . . .  10
       3.4.4.  Comparison  . . . . . . . . . . . . . . . . . . . . .  10
     3.5.  Application-Layer Constructs  . . . . . . . . . . . . . .  11
     3.6.  Examples  . . . . . . . . . . . . . . . . . . . . . . . .  11
   4.  Passwords . . . . . . . . . . . . . . . . . . . . . . . . . .  13
     4.1.  Definition  . . . . . . . . . . . . . . . . . . . . . . .  13
     4.2.  OpaqueString Profile  . . . . . . . . . . . . . . . . . .  14
       4.2.1.  Preparation . . . . . . . . . . . . . . . . . . . . .  14
       4.2.2.  Enforcement . . . . . . . . . . . . . . . . . . . . .  14
       4.2.3.  Comparison  . . . . . . . . . . . . . . . . . . . . .  15
     4.3.  Examples  . . . . . . . . . . . . . . . . . . . . . . . .  15
   5.  Use in Application Protocols  . . . . . . . . . . . . . . . .  16
   6.  Migration . . . . . . . . . . . . . . . . . . . . . . . . . .  17
     6.1.  Usernames . . . . . . . . . . . . . . . . . . . . . . . .  17
     6.2.  Passwords . . . . . . . . . . . . . . . . . . . . . . . .  19
   7.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .  20
     7.1.  UsernameCaseMapped Profile  . . . . . . . . . . . . . . .  20
     7.2.  UsernameCasePreserved Profile . . . . . . . . . . . . . .  20
     7.3.  OpaqueString Profile  . . . . . . . . . . . . . . . . . .  21
     7.4.  Stringprep Profile  . . . . . . . . . . . . . . . . . . .  22
   8.  Security Considerations . . . . . . . . . . . . . . . . . . .  22
     8.1.  Password/Passphrase Strength  . . . . . . . . . . . . . .  22
     8.2.  Password/Passphrase Comparison  . . . . . . . . . . . . .  22
     8.3.  Identifier Comparison . . . . . . . . . . . . . . . . . .  22
     8.4.  Reuse of PRECIS . . . . . . . . . . . . . . . . . . . . .  22
     8.5.  Reuse of Unicode  . . . . . . . . . . . . . . . . . . . .  22
   9.  References  . . . . . . . . . . . . . . . . . . . . . . . . .  23
     9.1.  Normative References  . . . . . . . . . . . . . . . . . .  23
     9.2.  Informative References  . . . . . . . . . . . . . . . . .  24
   Appendix A.  Changes from RFC 7613  . . . . . . . . . . . . . . .  25
   Acknowledgements  . . . . . . . . . . . . . . . . . . . . . . . .  26
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  26




Saint-Andre & Melnikov       Standards Track                    [Page 2]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


1.  Introduction



   Usernames and passwords are widely used for authentication and
   authorization on the Internet, either directly when provided in
   plaintext (as in the PLAIN Simple Authentication and Security Layer
   (SASL) mechanism [RFC4616] and the HTTP Basic scheme [RFC7617]) or
   indirectly when provided as the input to a cryptographic algorithm
   such as a hash function (as in the Salted Challenge Response
   Authentication Mechanism (SCRAM) SASL mechanism [RFC5802] and the
   HTTP Digest scheme [RFC7616]).

   To increase the likelihood that the input and comparison of usernames
   and passwords will work in ways that make sense for typical users
   throughout the world, this document defines rules for handling
   internationalized strings that represent usernames and passwords.
   Such strings consist of code points from the Unicode coded character
   set [Unicode], with special attention to code points outside the
   ASCII range [RFC20].  The rules for handling such strings are
   specified through profiles of the string classes defined in the
   preparation, enforcement, and comparison of internationalized strings
   (PRECIS) framework specification [RFC8264].

   Profiles of the PRECIS framework enable software to handle Unicode
   code points outside the ASCII range in an automated way, so that such
   code points are treated carefully and consistently in application
   protocols.  In large measure, these profiles are designed to protect
   application developers from the potentially negative consequences of
   supporting the full range of Unicode code points.  For instance, in
   almost all application protocols it would be dangerous to treat the
   Unicode code point "¹" (SUPERSCRIPT ONE, U+00B9) as equivalent to "1"
   (DIGIT ONE, U+0031), because that would result in false accepts
   during comparison, authentication, and authorization (e.g., an
   attacker could easily spoof an account "user1@example.com").

   Whereas a naive use of Unicode would make such attacks trivially
   easy, the PRECIS profile defined here for usernames generally
   protects applications from inadvertently causing such problems.
   (Similar considerations apply to passwords, although here it is
   desirable to support a wider range of characters so as to maximize
   entropy for purposes of authentication.)

   The methods defined here might be applicable wherever usernames or
   passwords are used.  However, the methods are not intended for use in
   preparing strings that are not usernames (e.g., Lightweight Directory
   Access Protocol (LDAP) distinguished names), nor in cases where
   identifiers or secrets are not strings (e.g., keys and certificates)
   or require specialized handling.




Saint-Andre & Melnikov       Standards Track                    [Page 3]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   Although the historical predecessor of this document was the SASLprep
   profile of Stringprep [RFC3454]), the approach defined here can be
   used by technologies other than SASL [RFC4422], such as HTTP
   authentication as specified in [RFC7617] and [RFC7616].

   This document does not modify the handling of internationalized
   strings in usernames and passwords as prescribed by existing
   application protocols that use SASLprep.  If the community that uses
   such an application protocol wishes to modernize its handling of
   internationalized strings to use PRECIS instead of Stringprep, it
   needs to explicitly update the existing application protocol
   definition (one example is [RFC7622]).  Non-coordinated updates to
   protocol implementations are discouraged because they can have a
   negative impact on interoperability and security.

2.  Terminology



   A "username" or "user identifier" is a string of characters
   designating an account on a computing device or system, often but not
   necessarily for use by a person.  Although some devices and systems
   might allow a username to be part or all of a person's name and a
   person might want their account designator to be part or all of their
   name, because of the complexities involved, that outcome is not
   guaranteed for all human names on all computing devices or systems
   that follow the rules defined in this specification.  Protocol
   designers and application developers who wish to allow a wider range
   of characters are encouraged to consider a separation between more
   restrictive account identifiers and more expressive display names or
   nicknames (see [RFC8266]).

   A "password" is a string of characters that allows access to a
   computing device or system, often associated with a particular
   username.  A password is not literally limited to a word, because a
   password could be a passphrase consisting of more than one word,
   perhaps separated by spaces, punctuation, or other non-alphanumeric
   characters.

   Some SASL mechanisms (e.g., CRAM-MD5, DIGEST-MD5, and SCRAM) specify
   that the authentication identity used in the context of such
   mechanisms is a "simple username" (see Section 2 of [RFC4422] as well
   as [RFC4013]).  Various application technologies also assume that the
   identity of a user or account takes the form of a username (e.g.,
   authentication for the Hypertext Transfer Protocol as specified in
   [RFC7617] and [RFC7616]), whether or not they use SASL.  Note well
   that the exact form of a username in any particular SASL mechanism or
   application technology is a matter for implementation and deployment;
   note also that a username does not necessarily map to any particular
   application identifier.



Saint-Andre & Melnikov       Standards Track                    [Page 4]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   Many important terms used in this document are defined in [RFC5890],
   [RFC6365], [RFC8264], and [Unicode].  The term "non-ASCII space"
   refers to any Unicode code point having a Unicode general category of
   "Zs", naturally with the exception of SPACE (U+0020).

   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 [RFC2119] [RFC8174] when, and only when, they appear in all
   capitals, as shown here.

3.  Usernames



3.1.  Definition



   This document specifies that a username is a string of Unicode code
   points [Unicode] that is structured as an ordered sequence of
   "userparts" and expressed in a standard Unicode Encoding Form (such
   as UTF-8 [RFC3629]).  A userpart is allowed to contain only code
   points that are allowed by the PRECIS IdentifierClass defined in
   Section 4.2 of [RFC8264] and thus consists almost exclusively of
   letters and digits.  A username can consist of a single userpart or a
   space-separated sequence of userparts.

   The syntax for a username is defined as follows, using the Augmented
   Backus-Naur Form (ABNF) [RFC5234].

      username   = userpart *(1*SP userpart)
      userpart   = 1*(idpoint)
                   ;
                   ; an "idpoint" is a Unicode code point that
                   ; can be contained in a string conforming to
                   ; the PRECIS IdentifierClass
                   ;

   All code points and blocks not explicitly allowed in the PRECIS
   IdentifierClass are disallowed; this includes private-use code
   points, surrogate code points, and the other code points and blocks
   that were defined as "Prohibited Output" in Section 2.3 of [RFC4013]
   (when corrected per [Err1812]).  In addition, common constructions
   such as "user@example.com" (e.g., the Network Access Identifier from
   [RFC7542]) are allowed as usernames under this specification, as they
   were under [RFC4013].

      Implementation Note: The username construct defined in this
      document does not necessarily match what all deployed applications
      might refer to as a "username" or "userid" but instead provides a
      relatively safe subset of Unicode code points that can be used in



Saint-Andre & Melnikov       Standards Track                    [Page 5]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


      existing SASL mechanisms and in application protocols that use
      SASL, and even in most application protocols that do not currently
      use SASL.

   A username MUST NOT be zero bytes in length.  This rule is to be
   enforced after any normalization and mapping of code points.

   This specification defines two profiles for usernames: the
   UsernameCaseMapped profile performs case mapping, and the
   UsernameCasePreserved performs case preservation (see further
   discussion under Section 3.2).

   In protocols that provide usernames as input to a cryptographic
   algorithm such as a hash function, the client will need to perform
   enforcement of the rules for the UsernameCaseMapped or
   UsernameCasePreserved profile before applying the algorithm.

3.2.  Case Mapping vs. Case Preservation



   In order to accommodate the widest range of username constructs in
   applications, this document defines two username profiles:
   UsernameCaseMapped and UsernameCasePreserved.  These two profiles
   differ only in their use (or not) of the Case Mapping Rule and are
   otherwise identical.

   Case mapping is a matter for the application protocol, protocol
   implementation, or end deployment.  In general, this document
   suggests that it is preferable to apply the UsernameCaseMapped
   profile and therefore perform case mapping, because not doing so can
   lead to false accepts during authentication and authorization (as
   described in [RFC6943]) and can result in confusion among end users,
   given the prevalence of case mapping in many existing protocols and
   applications.  However, there can be good reasons to apply the
   UsernameCasePreserved profile and thus not perform case mapping, such
   as backward compatibility with deployed infrastructure.

   In particular:

   o  SASL mechanisms that follow the recommendations in this document
      MUST specify whether and when case mapping is to be applied to
      authentication identifiers.  Because case mapping results in
      information loss, in order to retain that information for as long
      as possible during processing, implementations SHOULD delay any
      case mapping to the last possible moment, such as when doing a
      lookup by username, performing username comparisons, or generating
      a cryptographic salt from a username (if the last possible moment
      happens on a server, then decisions about case mapping can be a
      matter of service deployment policy).  In keeping with [RFC4422],



Saint-Andre & Melnikov       Standards Track                    [Page 6]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


      SASL mechanisms are not to apply this or any other profile to
      authorization identifiers, only to authentication identifiers.

   o  Application protocols that use SASL (such as IMAP [RFC3501] and
      the Extensible Messaging and Presence Protocol (XMPP) [RFC6120])
      and that directly reuse this profile MUST specify whether or not
      case mapping is to be applied to authorization identifiers.  Such
      "SASL application protocols" SHOULD delay any case mapping of
      authorization identifiers to the last possible moment, which
      happens to necessarily be on the server side (this enables
      decisions about case mapping to be a matter of service deployment
      policy).  In keeping with [RFC4422], SASL application protocols
      are not to apply this or any other profile to authentication
      identifiers, only to authorization identifiers.

   o  Application protocols that do not use SASL (such as HTTP
      authentication with the HTTP Basic and Digest schemes as specified
      in [RFC7617] and [RFC7616]) but that directly reuse this profile
      MUST specify whether and when case mapping is to be applied to
      authentication identifiers or authorization identifiers, or both.
      Such "non-SASL application protocols" SHOULD delay any case
      mapping to the last possible moment, such as when doing a lookup
      by username, performing username comparisons, or generating a
      cryptographic salt from a username (if the last possible moment
      happens on the server, then decisions about case mapping can be a
      matter of service deployment policy).

   If the specification for a SASL mechanism, SASL application protocol,
   or non-SASL application protocol uses the UsernameCaseMapped profile,
   it MUST clearly describe whether case mapping is to be applied at the
   level of the protocol itself, implementations thereof, or service
   deployments (each of these approaches can be legitimate, depending on
   the application in question).

3.3.  UsernameCaseMapped Profile



3.3.1.  Rules



   The following rules are defined for use within the UsernameCaseMapped
   profile of the PRECIS IdentifierClass.

   1.  Width Mapping Rule: Map fullwidth and halfwidth code points to
       their decomposition mappings (see Unicode Standard Annex #11
       [UAX11]).

   2.  Additional Mapping Rule: There is no additional mapping rule.





Saint-Andre & Melnikov       Standards Track                    [Page 7]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   3.  Case Mapping Rule: Map uppercase and titlecase code points to
       their lowercase equivalents, preferably using the Unicode
       toLowerCase() operation as defined in the Unicode Standard
       [Unicode]; see further discussion in Section 3.2.

   4.  Normalization Rule: Apply Unicode Normalization Form C (NFC) to
       all strings.



   5.  Directionality Rule: Apply the "Bidi Rule" defined in [RFC5893]
       to strings that contain right-to-left code points (i.e., each of
       the six conditions of the Bidi Rule must be satisfied); for
       strings that do not contain right-to-left code points, there is
       no special processing for directionality.

3.3.2.  Preparation



   An entity that prepares an input string for subsequent enforcement
   according to this profile MUST proceed as follows (applying the steps
   in the order shown).

   1.  Apply the width mapping rule specified in Section 3.3.1.  It is
       necessary to apply the rule at this point because otherwise the
       PRECIS "HasCompat" category specified in Section 9.17 of
       [RFC8264] would forbid fullwidth and halfwidth code points.

   2.  Ensure that the string consists only of Unicode code points that
       are explicitly allowed by the PRECIS IdentifierClass defined in
       Section 4.2 of [RFC8264].

3.3.3.  Enforcement



   An entity that performs enforcement according to this profile MUST
   prepare an input string as described in Section 3.3.2 and MUST also
   apply the following rules specified in Section 3.3.1 in the order
   shown:

   1.  Case Mapping Rule

   2.  Normalization Rule

   3.  Directionality Rule

   After all of the foregoing rules have been enforced, the entity MUST
   ensure that the username is not zero bytes in length (this is done
   after enforcing the rules to prevent applications from mistakenly
   omitting a username entirely, because when internationalized strings
   are accepted, a non-empty sequence of characters can result in a
   zero-length username after canonicalization).



Saint-Andre & Melnikov       Standards Track                    [Page 8]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   The result of the foregoing operations is an output string that
   conforms to the UsernameCaseMapped profile.  Until an implementation
   produces such an output string, it MUST NOT treat the string as
   conforming (in particular, it MUST NOT assume that an input string is
   conforming before the enforcement operation has been completed).

3.3.4.  Comparison



   An entity that performs comparison of two strings according to this
   profile MUST prepare each string as specified in Section 3.3.2 and
   then MUST enforce the rules specified in Section 3.3.3.  The two
   strings are to be considered equivalent if and only if they are an
   exact octet-for-octet match (sometimes called "bit-string identity").

   Until an implementation determines whether two strings are to be
   considered equivalent, it MUST NOT treat them as equivalent (in
   particular, it MUST NOT assume that two input strings are equivalent
   before the comparison operation has been completed).

3.4.  UsernameCasePreserved Profile



3.4.1.  Rules



   The following rules are defined for use within the
   UsernameCasePreserved profile of the PRECIS IdentifierClass.

   1.  Width Mapping Rule: Map fullwidth and halfwidth code points to
       their decomposition mappings (see Unicode Standard Annex #11
       [UAX11]).

   2.  Additional Mapping Rule: There is no additional mapping rule.

   3.  Case Mapping Rule: There is no case mapping rule.

   4.  Normalization Rule: Apply Unicode Normalization Form C (NFC) to
       all strings.

   5.  Directionality Rule: Apply the "Bidi Rule" defined in [RFC5893]
       to strings that contain right-to-left code points (i.e., each of
       the six conditions of the Bidi Rule must be satisfied); for
       strings that do not contain right-to-left code points, there is
       no special processing for directionality.

3.4.2.  Preparation



   An entity that prepares a string for subsequent enforcement according
   to this profile MUST proceed as follows (applying the steps in the
   order shown).



Saint-Andre & Melnikov       Standards Track                    [Page 9]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   1.  Apply the width mapping rule specified in Section 3.4.1.  It is
       necessary to apply the rule at this point because otherwise the
       PRECIS "HasCompat" category specified in Section 9.17 of
       [RFC8264] would forbid fullwidth and halfwidth code points.

   2.  Ensure that the string consists only of Unicode code points that
       are explicitly allowed by the PRECIS IdentifierClass defined in
       Section 4.2 of [RFC8264].

3.4.3.  Enforcement



   An entity that performs enforcement according to this profile MUST
   prepare a string as described in Section 3.4.2 and MUST also apply
   the following rules specified in Section 3.4.1 in the order shown:

   1.  Normalization Rule

   2.  Directionality Rule

   After all of the foregoing rules have been enforced, the entity MUST
   ensure that the username is not zero bytes in length (this is done
   after enforcing the rules to prevent applications from mistakenly
   omitting a username entirely, because when internationalized strings
   are accepted, a non-empty sequence of characters can result in a
   zero-length username after canonicalization).

   The result of the foregoing operations is an output string that
   conforms to the UsernameCasePreserved profile.  Until an
   implementation produces such an output string, it MUST NOT treat the
   string as conforming (in particular, it MUST NOT assume that an input
   string is conforming before the enforcement operation has been
   completed).

3.4.4.  Comparison



   An entity that performs comparison of two strings according to this
   profile MUST prepare each string as specified in Section 3.4.2 and
   then MUST enforce the rules specified in Section 3.4.3.  The two
   strings are to be considered equivalent if and only if they are an
   exact octet-for-octet match (sometimes called "bit-string identity").

   Until an implementation determines whether two strings are to be
   considered equivalent, it MUST NOT treat them as equivalent (in
   particular, it MUST NOT assume that two input strings are equivalent
   before the comparison operation has been completed).






Saint-Andre & Melnikov       Standards Track                   [Page 10]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


3.5.  Application-Layer Constructs



   Both the UsernameCaseMapped and UsernameCasePreserved profiles enable
   an application protocol, implementation, or deployment to create
   application-layer constructs such as a username that is a space-
   separated set of userparts like "Firstname Middlename Lastname".
   Such a construct is not a profile of the PRECIS IdentifierClass,
   because SPACE (U+0020) is not allowed in the IdentifierClass;
   however, it can be created at the application layer because SPACE
   (U+0020) can be used as a separator between instances of the PRECIS
   IdentifierClass (e.g., userparts as defined in this specification).

3.6.  Examples



   The following examples illustrate a small number of userparts (not
   usernames) that are consistent with the format defined above (note
   that the characters "<" and ">" are used here to delineate the actual
   userparts and are not part of the userpart strings).

      +--------------------------+---------------------------------+
      | # | Userpart             | Notes                           |
      +--------------------------+---------------------------------+
      | 1 | <juliet@example.com> | The "at" sign ("@") is allowed  |
      |   |                      | in the PRECIS IdentifierClass   |
      +--------------------------+---------------------------------+
      | 2 | <fussball>           |                                 |
      +--------------------------+---------------------------------+
      | 3 | <fußball>            | The third character is LATIN    |
      |   |                      | SMALL LETTER SHARP S (U+00DF)   |
      +--------------------------+---------------------------------+
      | 4 | <π>                  | A userpart of GREEK SMALL       |
      |   |                      | LETTER PI (U+03C0)              |
      +--------------------------+---------------------------------+
      | 5 | <Σ>                  | A userpart of GREEK CAPITAL     |
      |   |                      | LETTER SIGMA (U+03A3)           |
      +--------------------------+---------------------------------+
      | 6 | <σ>                  | A userpart of GREEK SMALL       |
      |   |                      | LETTER SIGMA (U+03C3)           |
      +--------------------------+---------------------------------+
      | 7 | <ς>                  | A userpart of GREEK SMALL       |
      |   |                      | LETTER FINAL SIGMA (U+03C2)     |
      +--------------------------+---------------------------------+

                   Table 1: A Sample of Legal Userparts

   Regarding examples 2 and 3: although in German writing the character
   eszett "ß" (LATIN SMALL LETTER SHARP S, U+00DF) can mostly be used
   interchangeably with the two characters "ss", the userparts in these



Saint-Andre & Melnikov       Standards Track                   [Page 11]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   examples are different and (if desired) a server would need to
   enforce a registration policy that disallows one of them if the other
   is registered.

   Regarding examples 5, 6, and 7: optional case mapping of "Σ" (GREEK
   CAPITAL LETTER SIGMA, U+03A3) to the lowercase character "σ" (GREEK
   SMALL LETTER SIGMA, U+03C3) during comparison would result in
   matching the userparts in examples 5 and 6; however, because the
   PRECIS mapping rules do not account for the special status of the
   character "ς" (GREEK SMALL LETTER FINAL SIGMA, U+03C2), the userparts
   in examples 5 and 7 or examples 6 and 7 would not be matched during
   comparison.

   The following examples illustrate strings that are not valid
   userparts (not usernames) because they violate the format defined
   above.

      +--------------------------+---------------------------------+
      | # | Non-Userpart String  | Notes                           |
      +--------------------------+---------------------------------+
      | 8 | <foo bar>            | SPACE (U+0020) is disallowed in |
      |   |                      | the userpart                    |
      +--------------------------+---------------------------------+
      | 9 | <>                   | Zero-length userpart            |
      +--------------------------+---------------------------------+
      | 10| <henryⅣ>            | The sixth character is ROMAN    |
      |   |                      | NUMERAL FOUR (U+2163)           |
      +--------------------------+---------------------------------+
      | 11| <∞>                  | A userpart of INFINITY (U+221E) |
      +--------------------------+---------------------------------+

       Table 2: A Sample of Strings That Violate the Userpart Rules

   Regarding example 8: although this is not a valid userpart, it is a
   valid username because it is a space-separated sequence of userparts.

   Regarding example 10: the character "Ⅳ" (ROMAN NUMERAL FOUR, U+2163)
   has a compatibility equivalent of the characters "I" (LATIN CAPITAL
   LETTER I, U+0049) and "V" (LATIN CAPITAL LETTER V, U+0056), but code
   points with compatibility equivalents are not allowed in the PRECIS
   IdentifierClass.

   Regarding example 11: symbol characters such as "∞" (INFINITY,
   U+221E) are not allowed in the PRECIS IdentifierClass.







Saint-Andre & Melnikov       Standards Track                   [Page 12]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


4.  Passwords

4.1.  Definition



   This document specifies that a password is a string of Unicode code
   points [Unicode] that is conformant to the OpaqueString profile
   (specified below) of the PRECIS FreeformClass defined in Section 4.3
   of [RFC8264] and expressed in a standard Unicode Encoding Form (such
   as UTF-8 [RFC3629]).

   The syntax for a password is defined as follows, using the Augmented
   Backus-Naur Form (ABNF) [RFC5234].

      password   = 1*(freepoint)
                   ;
                   ; a "freepoint" is a Unicode code point that
                   ; can be contained in a string conforming to
                   ; the PRECIS FreeformClass
                   ;

   All code points and blocks not explicitly allowed in the PRECIS
   FreeformClass are disallowed; this includes private-use code points,
   surrogate code points, and the other code points and blocks defined
   as "Prohibited Output" in Section 2.3 of [RFC4013] (when corrected
   per [Err1812]).

   A password MUST NOT be zero bytes in length.  This rule is to be
   enforced after any normalization and mapping of code points.

      Note: Some existing systems allow an empty string in places where
      a password would be expected (e.g., command-line tools that might
      be called from an automated script, or servers that might need to
      be restarted without human intervention).  From the perspective of
      this document (and RFC 4013 before it), these empty strings are
      not passwords but are workarounds for the practical difficulty of
      using passwords in certain scenarios.

      Note: The prohibition of zero-length passwords is not a
      recommendation regarding password strength (because a password of
      only one byte is highly insecure) but is meant to prevent
      applications from mistakenly omitting a password entirely; such an
      outcome is possible when internationalized strings are accepted,
      because a non-empty sequence of characters can result in a zero-
      length password after canonicalization.

   In protocols that provide passwords as input to a cryptographic
   algorithm such as a hash function, the client will need to perform
   enforcement of the rules for the OpaqueString profile before applying



Saint-Andre & Melnikov       Standards Track                   [Page 13]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   the algorithm, because the password is not available to the server in
   plaintext form.

4.2.  OpaqueString Profile



   The definition of the OpaqueString profile is provided in the
   following sections, including detailed information about preparation,
   enforcement, and comparison (for details on the distinction between
   these actions, refer to [RFC8264]).

4.2.1.  Preparation



   An entity that prepares a string according to this profile MUST
   ensure that the string consists only of Unicode code points that are
   explicitly allowed by the FreeformClass string class defined in
   [RFC8264].

4.2.2.  Enforcement



   An entity that performs enforcement according to this profile MUST
   prepare a string as described in Section 4.2.1 and MUST also apply
   the rules specified below for the OpaqueString profile (these rules
   MUST be applied in the order shown):

   1.  Width Mapping Rule: Fullwidth and halfwidth code points MUST NOT
       be mapped to their decomposition mappings (see Unicode Standard
       Annex #11 [UAX11]).

   2.  Additional Mapping Rule: Any instances of non-ASCII space MUST be
       mapped to SPACE (U+0020); a non-ASCII space is any Unicode code
       point having a Unicode general category of "Zs", with the
       exception of SPACE (U+0020).  As was the case in RFC 4013, the
       inclusion of only SPACE (U+0020) prevents confusion with various
       non-ASCII space code points, many of which are difficult to
       reproduce across different input methods.

   3.  Case Mapping Rule: There is no case mapping rule (because mapping
       uppercase and titlecase code points to their lowercase
       equivalents would lead to false accepts and thus to reduced
       security).

   4.  Normalization Rule: Unicode Normalization Form C (NFC) MUST be
       applied to all strings.

   5.  Directionality Rule: There is no directionality rule.  The "Bidi
       Rule" (defined in [RFC5893]) and similar rules are unnecessary
       and inapplicable to passwords, because they can reduce the
       repertoire of characters that are allowed in a string and



Saint-Andre & Melnikov       Standards Track                   [Page 14]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


       therefore reduce the amount of entropy that is possible in a
       password.  Such rules are intended to minimize the possibility
       that the same string will be displayed differently on a layout
       system set for right-to-left display and a layout system set for
       left-to-right display; however, passwords are typically not
       displayed at all and are rarely meant to be interoperable across
       different layout systems in the way that non-secret strings like
       domain names and usernames are.  Furthermore, it is perfectly
       acceptable for opaque strings other than passwords to be
       presented differently in different layout systems, as long as the
       presentation is consistent in any given layout system.

   The result of the foregoing operations is an output string that
   conforms to the OpaqueString profile.  Until an implementation
   produces such an output string, it MUST NOT treat the string as
   conforming (in particular, it MUST NOT assume that an input string is
   conforming before the enforcement operation has been completed).

4.2.3.  Comparison



   An entity that performs comparison of two strings according to this
   profile MUST prepare each string as specified in Section 4.2.1 and
   then MUST enforce the rules specified in Section 4.2.2.  The two
   strings are to be considered equivalent if and only if they are an
   exact octet-for-octet match (sometimes called "bit-string identity").

   Until an implementation determines whether two strings are to be
   considered equivalent, it MUST NOT treat them as equivalent (in
   particular, it MUST NOT assume that two input strings are equivalent
   before the comparison operation has been completed).

   See Section 8.2 regarding comparison of passwords and passphrases.

4.3.  Examples



   The following examples illustrate a small number of passwords that
   are consistent with the format defined above (note that the
   characters "<" and ">" are used here to delineate the actual
   passwords and are not part of the password strings).












Saint-Andre & Melnikov       Standards Track                   [Page 15]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   +------------------------------------+------------------------------+
   | # | Password                       | Notes                        |
   +------------------------------------+------------------------------+
   | 12| <correct horse battery staple> | SPACE (U+0020) is allowed    |
   +------------------------------------+------------------------------+
   | 13| <Correct Horse Battery Staple> | Differs by case from         |
   |   |                                | example 12                   |
   +------------------------------------+------------------------------+
   | 14| <πßå>                          | Non-ASCII letters are OK     |
   |   |                                | (e.g., GREEK SMALL LETTER    |
   |   |                                | PI (U+03C0))                 |
   +------------------------------------+------------------------------+
   | 15| <Jack of ♦s>                   | Symbols are OK (e.g., BLACK  |
   |   |                                | DIAMOND SUIT (U+2666))       |
   +------------------------------------+------------------------------+
   | 16| <foo bar>                      | OGHAM SPACE MARK (U+1680) is |
   |   |                                | mapped to SPACE (U+0020);    |
   |   |                                | thus, the full string is     |
   |   |                                | mapped to <foo bar>          |
   +------------------------------------+------------------------------+

                   Table 3: A Sample of Legal Passwords

   The following examples illustrate strings that are not valid
   passwords because they violate the format defined above.

   +------------------------------------+------------------------------+
   | # | Password                       | Notes                        |
   +------------------------------------+------------------------------+
   | 17| <>                             | Zero-length passwords are    |
   |   |                                | disallowed                   |
   +------------------------------------+------------------------------+
   | 18| <my cat is a by>          | Control characters like TAB  |
   |   |                                | (U+0009) are disallowed      |
   +------------------------------------+------------------------------+

       Table 4: A Sample of Strings That Violate the Password Rules

   Note: Following the "XML Notation" used in [RFC3987], the character
   TAB (U+0009) in example 18 is represented as because otherwise
   it could not be shown in running text.

5.  Use in Application Protocols



   This specification defines only the PRECIS-based rules for the
   handling of strings conforming to the UsernameCaseMapped and
   UsernameCasePreserved profiles of the PRECIS IdentifierClass, and
   strings conforming to the OpaqueString profile of the PRECIS



Saint-Andre & Melnikov       Standards Track                   [Page 16]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   FreeformClass.  It is the responsibility of an application protocol
   to specify the protocol slots in which such strings can appear, the
   entities that are expected to enforce the rules governing such
   strings, and at what points during protocol processing or interface
   handling the rules need to be enforced.  See Section 6 of [RFC8264]
   for guidelines on using PRECIS profiles in applications.

   Above and beyond the PRECIS-based rules specified here, application
   protocols can also define application-specific rules governing such
   strings (rules regarding minimum or maximum length, further
   restrictions on allowable code points or character ranges, safeguards
   to mitigate the effects of visually similar characters, etc.),
   application-layer constructs (see Section 3.5), and related matters.

   Some PRECIS profile definitions encourage entities that enforce the
   rules to be liberal in what they accept.  However, for usernames and
   passwords such a policy can be problematic, because it can lead to
   false accepts.  An in-depth discussion can be found in [RFC6943].

   Applying the rules for any given PRECIS profile is not necessarily an
   idempotent procedure for all code points.  Therefore, an
   implementation SHOULD apply the rules repeatedly until the output
   string is stable; if the output string does not stabilize after
   reapplying the rules three (3) additional times after the first
   application, the implementation SHOULD terminate application of the
   rules and reject the input string as invalid.

6.  Migration



   The rules defined in this specification differ slightly from those
   defined by the SASLprep specification [RFC4013] (but not from
   [RFC7613]).  In order to smooth the process of migrating from
   SASLprep to the approach defined herein, the following sections
   describe these differences, along with their implications for
   migration, in more detail.

6.1.  Usernames



   Deployments that currently use SASLprep for handling usernames might
   need to scrub existing data when they migrate to the rules defined in
   this specification.  In particular:

   o  SASLprep specified the use of Unicode Normalization Form KC
      (NFKC), whereas the UsernameCaseMapped and UsernameCasePreserved
      profiles employ Unicode Normalization Form C (NFC).  In practice,
      this change is unlikely to cause significant problems, because
      NFKC provides methods for mapping Unicode code points with
      compatibility equivalents to those equivalents, whereas the PRECIS



Saint-Andre & Melnikov       Standards Track                   [Page 17]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


      IdentifierClass entirely disallows Unicode code points with
      compatibility equivalents (i.e., during comparison, NFKC is more
      "aggressive" about finding matches than NFC).  A few examples
      might suffice to indicate the nature of the problem:

      1.  "ſ" (LATIN SMALL LETTER LONG S, U+017F) is compatibility
          equivalent to "s" (LATIN SMALL LETTER S, U+0073).

      2.  "Ⅳ" (ROMAN NUMERAL FOUR, U+2163) is compatibility equivalent
          to "I" (LATIN CAPITAL LETTER I, U+0049) and "V" (LATIN CAPITAL
          LETTER V, U+0056).

      3.  "fi" (LATIN SMALL LIGATURE FI, U+FB01) is compatibility
          equivalent to "f" (LATIN SMALL LETTER F, U+0066) and "i"
          (LATIN SMALL LETTER I, U+0069).

      Under SASLprep, the use of NFKC also handled the mapping of
      fullwidth and halfwidth code points to their decomposition
      mappings.

      For migration purposes, operators might want to search their
      database of usernames for names containing Unicode code points
      with compatibility equivalents and, where there is no conflict,
      map those code points to their equivalents.  Naturally, it is
      possible that during this process the operator will discover
      conflicting usernames; for instance, "HENRYIV" with the last two
      code points being LATIN CAPITAL LETTER I (U+0049) and LATIN
      CAPITAL LETTER V (U+0056) as opposed to "HENRYⅣ" with the last
      character being "Ⅳ" (ROMAN NUMERAL FOUR, U+2163), which is
      compatibility equivalent to U+0049 and U+0056).  In these cases,
      the operator will need to determine how to proceed, for instance,
      by disabling the account whose name contains a Unicode code point
      with a compatibility equivalent.  Such cases are probably rare,
      but it is important for operators to be aware of them.

   o  SASLprep mapped the "characters commonly mapped to nothing" (from
      Appendix B.1 of [RFC3454]) to nothing, whereas the PRECIS
      IdentifierClass entirely disallows most of these code points,
      which correspond to the code points from the PRECIS "M" category
      defined under Section 9.13 of [RFC8264].  For migration purposes,
      the operator might want to remove from usernames any code points
      contained in the PRECIS "M" category (e.g., SOFT HYPHEN (U+00AD)).
      Because these code points would have been "mapped to nothing" in
      Stringprep, in practice a user would not notice the difference if,
      upon migration to PRECIS, the code points are removed.

   o  SASLprep allowed uppercase and titlecase code points, whereas the
      UsernameCaseMapped profile maps uppercase and titlecase code



Saint-Andre & Melnikov       Standards Track                   [Page 18]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


      points to their lowercase equivalents (by contrast, the
      UsernameCasePreserved profile matches SASLprep in this regard).
      For migration purposes, the operator can use either the
      UsernameCaseMapped profile (thus losing the case information) or
      the UsernameCasePreserved profile (thus ignoring case difference
      when comparing usernames).

6.2.  Passwords



   Depending on local service policy, migration from SASLprep to this
   specification might not involve any scrubbing of data (because
   passwords might not be stored in the clear anyway); however, service
   providers need to be aware of possible issues that might arise during
   migration.  In particular:

   o  SASLprep specified the use of Unicode Normalization Form KC
      (NFKC), whereas the OpaqueString profile employs Unicode
      Normalization Form C (NFC).  Because NFKC is more aggressive about
      finding matches than NFC, in practice this change is unlikely to
      cause significant problems and indeed has the security benefit of
      probably resulting in fewer false accepts when comparing
      passwords.  A few examples might suffice to indicate the nature of
      the problem:

      1.  "ſ" (LATIN SMALL LETTER LONG S, U+017F) is compatibility
          equivalent to "s" (LATIN SMALL LETTER S, U+0073).

      2.  "Ⅳ" (ROMAN NUMERAL FOUR, U+2163) is compatibility equivalent
          to "I" (LATIN CAPITAL LETTER I, U+0049) and "V" (LATIN CAPITAL
          LETTER V, U+0056).

      3.  "fi" (LATIN SMALL LIGATURE FI, U+FB01) is compatibility
          equivalent to "f" (LATIN SMALL LETTER F, U+0066) and "i"
          (LATIN SMALL LETTER I, U+0069).

      Under SASLprep, the use of NFKC also handled the mapping of
      fullwidth and halfwidth code points to their decomposition
      mappings.  Although it is expected that code points with
      compatibility equivalents are rare in existing passwords, some
      passwords that matched when SASLprep was used might no longer work
      when the rules in this specification are applied.

   o  SASLprep mapped the "characters commonly mapped to nothing" (from
      Appendix B.1 of [RFC3454]) to nothing, whereas the PRECIS
      FreeformClass entirely disallows such code points, which
      correspond to the code points from the PRECIS "M" category defined
      under Section 9.13 of [RFC8264].  In practice, this change will
      probably have no effect on comparison, but user-oriented software



Saint-Andre & Melnikov       Standards Track                   [Page 19]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


      might reject such code points instead of ignoring them during
      password preparation.

7.  IANA Considerations



   IANA has made the updates described below.

7.1.  UsernameCaseMapped Profile



   IANA has added the following entry to the "PRECIS Profiles" registry.

   Name:  UsernameCaseMapped.

   Base Class:  IdentifierClass.

   Applicability:  Usernames in security and application protocols.

   Replaces:  The SASLprep profile of Stringprep.

   Width Mapping Rule:  Map fullwidth and halfwidth code points to their
      decomposition mappings.

   Additional Mapping Rule:  None.

   Case Mapping Rule:  Map uppercase and titlecase code points to
      lowercase.

   Normalization Rule:  NFC.

   Directionality Rule:  The "Bidi Rule" defined in RFC 5893 applies.

   Enforcement:  To be defined by security or application protocols that
      use this profile.

   Specification:  Section 3.3 of RFC 8265.

7.2.  UsernameCasePreserved Profile



   IANA has added the following entry to the "PRECIS Profiles" registry.

   Name:  UsernameCasePreserved.

   Base Class:  IdentifierClass.

   Applicability:  Usernames in security and application protocols.

   Replaces:  The SASLprep profile of Stringprep.




Saint-Andre & Melnikov       Standards Track                   [Page 20]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   Width Mapping Rule:  Map fullwidth and halfwidth code points to their
      decomposition mappings.

   Additional Mapping Rule:  None.

   Case Mapping Rule:  None.

   Normalization Rule:  NFC.

   Directionality Rule:  The "Bidi Rule" defined in RFC 5893 applies.

   Enforcement:  To be defined by security or application protocols that
      use this profile.

   Specification:  Section 3.4 of RFC 8265.

7.3.  OpaqueString Profile



   IANA has added the following entry to the "PRECIS Profiles" registry.

   Name:  OpaqueString.

   Base Class:  FreeformClass.

   Applicability:  Passwords and other opaque strings in security and
      application protocols.

   Replaces:  The SASLprep profile of Stringprep.

   Width Mapping Rule:  None.

   Additional Mapping Rule:  Map non-ASCII space code points to SPACE
      (U+0020).

   Case Mapping Rule:  None.

   Normalization Rule:  NFC.

   Directionality Rule:  None.

   Enforcement:  To be defined by security or application protocols that
      use this profile.

   Specification:  Section 4.2 of RFC 8265.







Saint-Andre & Melnikov       Standards Track                   [Page 21]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


7.4.  Stringprep Profile



   The Stringprep specification [RFC3454] did not provide for entries in
   the "Stringprep Profiles" registry to have any state except "Current"
   or "Not Current".  Because RFC 7613 obsoleted RFC 4013, which
   registered the SASLprep profile of Stringprep, IANA previously marked
   that profile as "Not Current" and cited RFC 7613 as an additional
   reference.  IANA has modified the profile so that the current
   document is now cited as the additional reference.

8.  Security Considerations



8.1.  Password/Passphrase Strength



   The ability to include a wide range of characters in passwords and
   passphrases can increase the potential for creating a strong password
   with high entropy.  However, in practice, the ability to include such
   characters ought to be weighed against the possible need to reproduce
   them on various devices using various input methods.

8.2.  Password/Passphrase Comparison



   In systems that conform to modern best practices for security,
   verification of passwords during authentication will not use the
   comparison defined in Section 4.2.3.  Instead, because the system
   performs cryptographic calculations to verify the password, it will
   prepare the password as defined in Section 4.2.1 and enforce the
   rules as defined in Section 4.2.2 before performing the relevant
   calculations.

8.3.  Identifier Comparison



   The process of comparing identifiers (such as SASL simple usernames,
   authentication identifiers, and authorization identifiers) can lead
   to either false rejects or false accepts, both of which have security
   implications.  A more detailed discussion can be found in [RFC6943].

8.4.  Reuse of PRECIS



   The security considerations described in [RFC8264] apply to the
   IdentifierClass and FreeformClass string classes used in this
   document for usernames and passwords, respectively.

8.5.  Reuse of Unicode



   The security considerations described in [UTS39] apply to the use of
   Unicode code points in usernames and passwords.




Saint-Andre & Melnikov       Standards Track                   [Page 22]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


9.  References



9.1.  Normative References



   [RFC2119]  Bradner, S., "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>.

   [RFC3629]  Yergeau, F., "UTF-8, a transformation format of ISO
              10646", STD 63, RFC 3629, DOI 10.17487/RFC3629, November
              2003, <https://www.rfc-editor.org/info/rfc3629>.

   [RFC5234]  Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax
              Specifications: ABNF", STD 68, RFC 5234,
              DOI 10.17487/RFC5234, January 2008,
              <https://www.rfc-editor.org/info/rfc5234>.

   [RFC5890]  Klensin, J., "Internationalized Domain Names for
              Applications (IDNA): Definitions and Document Framework",
              RFC 5890, DOI 10.17487/RFC5890, August 2010,
              <https://www.rfc-editor.org/info/rfc5890>.

   [RFC6365]  Hoffman, P. and J. Klensin, "Terminology Used in
              Internationalization in the IETF", BCP 166, RFC 6365,
              DOI 10.17487/RFC6365, September 2011,
              <https://www.rfc-editor.org/info/rfc6365>.

   [RFC8174]  Leiba, B., "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>.

   [RFC8264]  Saint-Andre, P. and M. Blanchet, "PRECIS Framework:
              Preparation, Enforcement, and Comparison of
              Internationalized Strings in Application Protocols",
              RFC 8264, DOI 10.17487/RFC8264, October 2017,
              <https://www.rfc-editor.org/info/rfc8264>.

   [UAX11]    Unicode Standard Annex #11, "East Asian Width", edited by
              Ken Lunde.  An integral part of The Unicode Standard,
              <http://unicode.org/reports/tr11/>.

   [Unicode]  The Unicode Consortium, "The Unicode Standard",
              <http://www.unicode.org/versions/latest/>.







Saint-Andre & Melnikov       Standards Track                   [Page 23]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


9.2.  Informative References



   [Err1812]  RFC Errata, Erratum ID 1812, RFC 4013,
              <https://www.rfc-editor.org/errata/eid1812>.

   [RFC20]    Cerf, V., "ASCII format for network interchange", STD 80,
              RFC 20, DOI 10.17487/RFC0020, October 1969,
              <https://www.rfc-editor.org/info/rfc20>.

   [RFC3454]  Hoffman, P. and M. Blanchet, "Preparation of
              Internationalized Strings ("stringprep")", RFC 3454,
              DOI 10.17487/RFC3454, December 2002,
              <https://www.rfc-editor.org/info/rfc3454>.

   [RFC3501]  Crispin, M., "INTERNET MESSAGE ACCESS PROTOCOL - VERSION
              4rev1", RFC 3501, DOI 10.17487/RFC3501, March 2003,
              <https://www.rfc-editor.org/info/rfc3501>.

   [RFC3987]  Duerst, M. and M. Suignard, "Internationalized Resource
              Identifiers (IRIs)", RFC 3987, DOI 10.17487/RFC3987,
              January 2005, <https://www.rfc-editor.org/info/rfc3987>.

   [RFC4013]  Zeilenga, K., "SASLprep: Stringprep Profile for User Names
              and Passwords", RFC 4013, DOI 10.17487/RFC4013, February
              2005, <https://www.rfc-editor.org/info/rfc4013>.

   [RFC4422]  Melnikov, A., Ed. and K. Zeilenga, Ed., "Simple
              Authentication and Security Layer (SASL)", RFC 4422,
              DOI 10.17487/RFC4422, June 2006,
              <https://www.rfc-editor.org/info/rfc4422>.

   [RFC4616]  Zeilenga, K., Ed., "The PLAIN Simple Authentication and
              Security Layer (SASL) Mechanism", RFC 4616,
              DOI 10.17487/RFC4616, August 2006,
              <https://www.rfc-editor.org/info/rfc4616>.

   [RFC5802]  Newman, C., Menon-Sen, A., Melnikov, A., and N. Williams,
              "Salted Challenge Response Authentication Mechanism
              (SCRAM) SASL and GSS-API Mechanisms", RFC 5802,
              DOI 10.17487/RFC5802, July 2010,
              <https://www.rfc-editor.org/info/rfc5802>.

   [RFC5893]  Alvestrand, H., Ed. and C. Karp, "Right-to-Left Scripts
              for Internationalized Domain Names for Applications
              (IDNA)", RFC 5893, DOI 10.17487/RFC5893, August 2010,
              <https://www.rfc-editor.org/info/rfc5893>.





Saint-Andre & Melnikov       Standards Track                   [Page 24]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   [RFC6120]  Saint-Andre, P., "Extensible Messaging and Presence
              Protocol (XMPP): Core", RFC 6120, DOI 10.17487/RFC6120,
              March 2011, <https://www.rfc-editor.org/info/rfc6120>.

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

   [RFC7542]  DeKok, A., "The Network Access Identifier", RFC 7542,
              DOI 10.17487/RFC7542, May 2015,
              <https://www.rfc-editor.org/info/rfc7542>.

   [RFC7613]  Saint-Andre, P. and A. Melnikov, "Preparation,
              Enforcement, and Comparison of Internationalized Strings
              Representing Usernames and Passwords", RFC 7613,
              DOI 10.17487/RFC7613, August 2015,
              <https://www.rfc-editor.org/info/rfc7613>.

   [RFC7616]  Shekh-Yusef, R., Ed., Ahrens, D., and S. Bremer, "HTTP
              Digest Access Authentication", RFC 7616,
              DOI 10.17487/RFC7616, September 2015,
              <https://www.rfc-editor.org/info/rfc7616>.

   [RFC7617]  Reschke, J., "The 'Basic' HTTP Authentication Scheme",
              RFC 7617, DOI 10.17487/RFC7617, September 2015,
              <https://www.rfc-editor.org/info/rfc7617>.

   [RFC7622]  Saint-Andre, P., "Extensible Messaging and Presence
              Protocol (XMPP): Address Format", RFC 7622,
              DOI 10.17487/RFC7622, September 2015,
              <https://www.rfc-editor.org/info/rfc7622>.

   [RFC8266]  Saint-Andre, P., "Preparation, Enforcement, and Comparison
              of Internationalized Strings Representing Nicknames",
              RFC 8266, DOI 10.17487/RFC8266, October 2017,
              <https://www.rfc-editor.org/info/rfc8266>.

   [UTS39]    Unicode Technical Standard #39, "Unicode Security
              Mechanisms", edited by Mark Davis and Michel Suignard,
              <http://unicode.org/reports/tr39/>.

Appendix A.  Changes from RFC 7613



   The following changes were made from [RFC7613].

   o  Corrected the order of operations for the UsernameCaseMapped
      profile to ensure consistency with [RFC8264].




Saint-Andre & Melnikov       Standards Track                   [Page 25]

RFC 8265             PRECIS: Usernames and Passwords        October 2017


   o  In accordance with working group discussions and updates to
      [RFC8264], removed the use of the Unicode toCaseFold() operation
      in favor of the Unicode toLowerCase() operation.

   o  Modified the presentation (but not the content) of the rules.

   o  Removed UTF-8 as a mandatory encoding, because that is a matter
      for the application.

   o  Clarified several editorial matters.

   o  Updated references.

   See [RFC7613] for a description of the differences from [RFC4013].

Acknowledgements



   Thanks to Christian Schudt and Sam Whited for their bug reports and
   feedback.

   See [RFC7613] for acknowledgements related to the specification that
   this document supersedes.

Authors' Addresses



   Peter Saint-Andre
   Jabber.org
   P.O. Box 787
   Parker, CO  80134
   United States of America

   Phone: +1 720 256 6756
   Email: stpeter@jabber.org
   URI:   https://www.jabber.org/


   Alexey Melnikov
   Isode Ltd
   5 Castle Business Village
   36 Station Road
   Hampton, Middlesex  TW12 2BX
   United Kingdom

   Email: Alexey.Melnikov@isode.com







Saint-Andre & Melnikov       Standards Track                   [Page 26]