7Internet Engineering Task Force (IETF) P. Saint-Andre
8Request for Comments: 7613 &yet
9Obsoletes: 4013 A. Melnikov
10Category: Standards Track Isode Ltd
11ISSN: 2070-1721 August 2015
14 Preparation, Enforcement, and Comparison of Internationalized Strings
15 Representing Usernames and Passwords
19 This document describes updated methods for handling Unicode strings
20 representing usernames and passwords. The previous approach was
21 known as SASLprep (RFC 4013) and was based on stringprep (RFC 3454).
22 The methods specified in this document provide a more sustainable
23 approach to the handling of internationalized usernames and
24 passwords. The preparation, enforcement, and comparison of
25 internationalized strings (PRECIS) framework, RFC 7564, obsoletes RFC
26 3454, and this document obsoletes RFC 4013.
30 This is an Internet Standards Track document.
32 This document is a product of the Internet Engineering Task Force
33 (IETF). It represents the consensus of the IETF community. It has
34 received public review and has been approved for publication by the
35 Internet Engineering Steering Group (IESG). Further information on
36 Internet Standards is available in Section 2 of RFC 5741.
38 Information about the current status of this document, any errata,
39 and how to provide feedback on it may be obtained at
40 http://www.rfc-editor.org/info/rfc7613.
58Saint-Andre & Melnikov Standards Track [Page 1]
60RFC 7613 PRECIS: Usernames and Passwords August 2015
65 Copyright (c) 2015 IETF Trust and the persons identified as the
66 document authors. All rights reserved.
68 This document is subject to BCP 78 and the IETF Trust's Legal
69 Provisions Relating to IETF Documents
70 (http://trustee.ietf.org/license-info) in effect on the date of
71 publication of this document. Please review these documents
72 carefully, as they describe your rights and restrictions with respect
73 to this document. Code Components extracted from this document must
74 include Simplified BSD License text as described in Section 4.e of
75 the Trust Legal Provisions and are provided without warranty as
76 described in the Simplified BSD License.
114Saint-Andre & Melnikov Standards Track [Page 2]
116RFC 7613 PRECIS: Usernames and Passwords August 2015
121 1. Introduction ....................................................4
122 2. Terminology .....................................................5
123 3. Usernames .......................................................6
124 3.1. Definition .................................................6
125 3.2. UsernameCaseMapped Profile .................................7
126 3.2.1. Preparation .........................................7
127 3.2.2. Enforcement .........................................7
128 3.2.3. Comparison ..........................................8
129 3.3. UsernameCasePreserved Profile ..............................8
130 3.3.1. Preparation .........................................8
131 3.3.2. Enforcement .........................................8
132 3.3.3. Comparison ..........................................9
133 3.4. Case Mapping vs. Case Preservation .........................9
134 3.5. Application-Layer Constructs ..............................10
135 3.6. Examples ..................................................11
136 4. Passwords ......................................................13
137 4.1. Definition ................................................13
138 4.2. OpaqueString Profile ......................................14
139 4.2.1. Preparation ........................................14
140 4.2.2. Enforcement ........................................14
141 4.2.3. Comparison .........................................15
142 4.3. Examples ..................................................15
143 5. Use in Application Protocols ...................................16
144 6. Migration ......................................................16
145 6.1. Usernames .................................................17
146 6.2. Passwords .................................................18
147 7. IANA Considerations ............................................19
148 7.1. UsernameCaseMapped Profile ................................19
149 7.2. UsernameCasePreserved Profile .............................20
150 7.3. OpaqueString Profile ......................................20
151 7.4. Stringprep Profile ........................................21
152 8. Security Considerations ........................................21
153 8.1. Password/Passphrase Strength ..............................21
154 8.2. Identifier Comparison .....................................21
155 8.3. Reuse of PRECIS ...........................................21
156 8.4. Reuse of Unicode ..........................................22
157 9. References .....................................................22
158 9.1. Normative References ......................................22
159 9.2. Informative References ....................................23
160 Appendix A. Differences from RFC 4013 .............................26
161 Acknowledgements ..................................................27
162 Authors' Addresses ................................................27
170Saint-Andre & Melnikov Standards Track [Page 3]
172RFC 7613 PRECIS: Usernames and Passwords August 2015
177 Usernames and passwords are widely used for authentication and
178 authorization on the Internet, either directly when provided in
179 plaintext (as in the PLAIN Simple Authentication and Security Layer
180 (SASL) mechanism [RFC4616] and the HTTP Basic scheme
181 [HTTP-BASIC-AUTH]) or indirectly when provided as the input to a
182 cryptographic algorithm such as a hash function (as in the Salted
183 Challenge Response Authentication Mechanism (SCRAM) SASL mechanism
184 [RFC5802] and the HTTP Digest scheme [HTTP-DIGEST-AUTH]).
186 To increase the likelihood that the input and comparison of usernames
187 and passwords will work in ways that make sense for typical users
188 throughout the world, this document defines rules for preparing,
189 enforcing, and comparing internationalized strings that represent
190 usernames and passwords. Such strings consist of characters from the
191 Unicode character set [Unicode], with special attention to characters
192 outside the ASCII range [RFC20]. The rules for handling such strings
193 are specified through profiles of the string classes defined in the
194 preparation, enforcement, and comparison of internationalized strings
195 (PRECIS) framework specification [RFC7564].
197 Profiles of the PRECIS framework enable software to handle Unicode
198 characters outside the ASCII range in an automated way, so that such
199 characters are treated carefully and consistently in application
200 protocols. In large measure, these profiles are designed to protect
201 application developers from the potentially negative consequences of
202 supporting the full range of Unicode characters. For instance, in
203 almost all application protocols it would be dangerous to treat the
204 Unicode character SUPERSCRIPT ONE (U+00B9) as equivalent to DIGIT ONE
205 (U+0031), because that would result in false positives during
206 comparison, authentication, and authorization (e.g., an attacker
207 could easy spoof an account "user1@example.com").
209 Whereas a naive use of Unicode would make such attacks trivially
210 easy, the PRECIS profile defined here for usernames generally
211 protects applications from inadvertently causing such problems.
212 (Similar considerations apply to passwords, although here it is
213 desirable to support a wider range of characters so as to maximize
214 entropy for purposes of authentication.)
216 The methods defined here might be applicable wherever usernames or
217 passwords are used. However, the methods are not intended for use in
218 preparing strings that are not usernames (e.g., Lightweight Directory
219 Access Protocol (LDAP) distinguished names), nor in cases where
220 identifiers or secrets are not strings (e.g., keys and certificates)
221 or require specialized handling.
226Saint-Andre & Melnikov Standards Track [Page 4]
228RFC 7613 PRECIS: Usernames and Passwords August 2015
231 This document obsoletes RFC 4013 (the SASLprep profile of stringprep
232 [RFC3454]) but can be used by technologies other than SASL [RFC4422],
233 such as HTTP authentication as specified in [HTTP-BASIC-AUTH] and
236 This document does not modify the handling of internationalized
237 strings in usernames and passwords as prescribed by existing
238 application protocols that use SASLprep. If the community that uses
239 such an application protocol wishes to modernize its handling of
240 internationalized strings to use PRECIS instead of stringprep, it
241 needs to explicitly update the existing application protocol
242 definition (one example is [XMPP-ADDR], which is intended to obsolete
243 [RFC6122]). Non-coordinated updates to protocol implementations are
244 discouraged because they can have a negative impact on
245 interoperability and security.
249 Many important terms used in this document are defined in [RFC5890],
250 [RFC6365], [RFC7564], and [Unicode]. The term "non-ASCII space"
251 refers to any Unicode code point having a Unicode general category of
252 "Zs", with the exception of U+0020 (here called "ASCII space").
254 As used here, the term "password" is not literally limited to a word;
255 i.e., a password could be a passphrase consisting of more than one
256 word, perhaps separated by spaces, punctuation, or other
257 non-alphanumeric characters.
259 Some SASL mechanisms (e.g., CRAM-MD5, DIGEST-MD5, and SCRAM) specify
260 that the authentication identity used in the context of such
261 mechanisms is a "simple user name" (see Section 2 of [RFC4422] as
262 well as [RFC4013]). Various application technologies also assume
263 that the identity of a user or account takes the form of a username
264 (e.g., authentication for the Hypertext Transfer Protocol as
265 specified in [HTTP-BASIC-AUTH] and [HTTP-DIGEST-AUTH]), whether or
266 not they use SASL. Note well that the exact form of a username in
267 any particular SASL mechanism or application technology is a matter
268 for implementation and deployment, and that a username does not
269 necessarily map to any particular application identifier (such as the
270 localpart of an email address).
272 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
273 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
274 "OPTIONAL" in this document are to be interpreted as described in
282Saint-Andre & Melnikov Standards Track [Page 5]
284RFC 7613 PRECIS: Usernames and Passwords August 2015
291 This document specifies that a username is a string of Unicode code
292 points [Unicode], encoded using UTF-8 [RFC3629], and structured as an
293 ordered sequence of "userparts". A userpart is allowed to contain
294 only code points that are in turn allowed by the PRECIS
295 IdentifierClass defined in Section 4.2 of [RFC7564], and thus
296 consists almost exclusively of letters and digits. A username can
297 consist of a single userpart or a space-separated sequence of
300 The syntax for a username is defined as follows, using the Augmented
301 Backus-Naur Form (ABNF) [RFC5234].
303 username = userpart *(1*SP userpart)
304 userpart = 1*(idbyte)
306 ; an "idbyte" is a byte used to represent a
307 ; UTF-8 encoded Unicode code point that can be
308 ; contained in a string that conforms to the
309 ; PRECIS IdentifierClass
312 All code points and blocks not explicitly allowed in the PRECIS
313 IdentifierClass are disallowed; this includes private use characters,
314 surrogate code points, and the other code points and blocks that were
315 defined as "Prohibited Output" in [RFC4013]. In addition, common
316 constructions such as "user@example.com" (e.g., the Network Access
317 Identifier from [RFC7542]) are allowed as usernames under this
318 specification, as they were under [RFC4013].
320 Implementation Note: The username construct defined in this
321 document does not necessarily match what all deployed applications
322 might refer to as a "username" or "userid" but instead provides a
323 relatively safe subset of Unicode characters that can be used in
324 existing SASL mechanisms and in application protocols that use
325 SASL, and even in most application protocols that do not currently
328 A username MUST NOT be zero bytes in length. This rule is to be
329 enforced after any normalization and mapping of code points.
331 In protocols that provide usernames as input to a cryptographic
332 algorithm such as a hash function, the client will need to perform
333 enforcement of the rules for the UsernameCaseMapped or
334 UsernameCasePreserved profile before applying the algorithm.
338Saint-Andre & Melnikov Standards Track [Page 6]
340RFC 7613 PRECIS: Usernames and Passwords August 2015
343 This specification defines two profiles for usernames: one that
344 performs case mapping and one that performs case preservation (see
345 further discussion under Section 3.4).
3473.2. UsernameCaseMapped Profile
349 The definition of the UsernameCaseMapped profile of the
350 IdentifierClass is provided in the following sections, including
351 detailed information about preparation, enforcement, and comparison
352 (for details on the distinction between these actions, refer to
357 An entity that prepares a string according to this profile MUST first
358 map fullwidth and halfwidth characters to their decomposition
359 mappings (see Unicode Standard Annex #11 [UAX11]). This is necessary
360 because the PRECIS "HasCompat" category specified in Section 9.17 of
361 [RFC7564] would otherwise forbid fullwidth and halfwidth characters.
362 After applying this width-mapping rule, the entity then MUST ensure
363 that the string consists only of Unicode code points that conform to
364 the PRECIS IdentifierClass defined in Section 4.2 of [RFC7564]. In
365 addition, the entity then MUST encode the string as UTF-8 [RFC3629].
369 An entity that performs enforcement according to this profile MUST
370 prepare a string as described in Section 3.2.1 and MUST also apply
371 the rules specified below for the UsernameCaseMapped profile (these
372 rules MUST be applied in the order shown):
374 1. Width-Mapping Rule: Applied as part of preparation (see above).
376 2. Additional Mapping Rule: There is no additional mapping rule.
378 3. Case-Mapping Rule: Uppercase and titlecase characters MUST be
379 mapped to their lowercase equivalents, preferably using Unicode
380 Default Case Folding as defined in the Unicode Standard [Unicode]
381 (at the time of this writing, the algorithm is specified in
382 Chapter 3 of [Unicode7.0], but the chapter number might change in
383 a future version of the Unicode Standard); see further discussion
386 4. Normalization Rule: Unicode Normalization Form C (NFC) MUST be
387 applied to all characters.
394Saint-Andre & Melnikov Standards Track [Page 7]
396RFC 7613 PRECIS: Usernames and Passwords August 2015
399 5. Directionality Rule: Applications MUST apply the "Bidi Rule"
400 defined in [RFC5893] to strings that contain right-to-left
401 characters (i.e., each of the six conditions of the Bidi Rule
406 An entity that performs comparison of two strings according to this
407 profile MUST prepare each string as specified in Section 3.2.1 and
408 then enforce the rules specified in Section 3.2.2. The two strings
409 are to be considered equivalent if they are an exact octet-for-octet
410 match (sometimes called "bit-string identity").
4123.3. UsernameCasePreserved Profile
414 The definition of the UsernameCasePreserved profile of the
415 IdentifierClass is provided in the following sections, including
416 detailed information about preparation, enforcement, and comparison
417 (for details on the distinction between these actions, refer to
422 An entity that prepares a string according to this profile MUST first
423 map fullwidth and halfwidth characters to their decomposition
424 mappings (see Unicode Standard Annex #11 [UAX11]). This is necessary
425 because the PRECIS "HasCompat" category specified in Section 9.17 of
426 [RFC7564] would otherwise forbid fullwidth and halfwidth characters.
427 After applying this width-mapping rule, the entity then MUST ensure
428 that the string consists only of Unicode code points that conform to
429 the PRECIS IdentifierClass defined in Section 4.2 of [RFC7564]. In
430 addition, the entity then MUST encode the string as UTF-8 [RFC3629].
434 An entity that performs enforcement according to this profile MUST
435 prepare a string as described in Section 3.3.1 and MUST also apply
436 the rules specified below for the UsernameCasePreserved profile
437 (these rules MUST be applied in the order shown):
439 1. Width-Mapping Rule: Applied as part of preparation (see above).
441 2. Additional Mapping Rule: There is no additional mapping rule.
443 3. Case-Mapping Rule: Uppercase and titlecase characters MUST NOT be
444 mapped to their lowercase equivalents; see further discussion in
450Saint-Andre & Melnikov Standards Track [Page 8]
452RFC 7613 PRECIS: Usernames and Passwords August 2015
455 4. Normalization Rule: Unicode Normalization Form C (NFC) MUST be
456 applied to all characters.
458 5. Directionality Rule: Applications MUST apply the "Bidi Rule"
459 defined in [RFC5893] to strings that contain right-to-left
460 characters (i.e., each of the six conditions of the Bidi Rule
465 An entity that performs comparison of two strings according to this
466 profile MUST prepare each string as specified in Section 3.3.1 and
467 then enforce the rules specified in Section 3.3.2. The two strings
468 are to be considered equivalent if they are an exact octet-for-octet
469 match (sometimes called "bit-string identity").
4713.4. Case Mapping vs. Case Preservation
473 In order to accommodate the widest range of username constructs in
474 applications, this document defines two username profiles:
475 UsernameCaseMapped and UsernameCasePreserved. These two profiles
476 differ only in the Case-Mapping Rule and are otherwise identical.
478 Case mapping is a matter for the application protocol, protocol
479 implementation, or end deployment. In general, this document
480 suggests that it is preferable to apply the UsernameCaseMapped
481 profile and therefore perform case mapping, because not doing so can
482 lead to false positives during authentication and authorization (as
483 described in [RFC6943]) and can result in confusion among end users,
484 given the prevalence of case mapping in many existing protocols and
485 applications. However, there can be good reasons to apply the
486 UsernameCasePreserved profile and thus not perform case mapping, such
487 as backward compatibility with deployed infrastructure.
491 o SASL mechanisms that follow the recommendations in this document
492 MUST specify whether and when case mapping is to be applied to
493 authentication identifiers. SASL mechanisms SHOULD delay any case
494 mapping to the last possible moment, such as when doing a lookup
495 by username, performing username comparisons, or generating a
496 cryptographic salt from a username (if the last possible moment
497 happens on the server, then decisions about case mapping can be a
498 matter of deployment policy). In keeping with [RFC4422], SASL
499 mechanisms are not to apply this or any other profile to
500 authorization identifiers, only to authentication identifiers.
506Saint-Andre & Melnikov Standards Track [Page 9]
508RFC 7613 PRECIS: Usernames and Passwords August 2015
511 o Application protocols that use SASL (such as IMAP [RFC3501] and
512 the Extensible Messaging and Presence Protocol (XMPP) [RFC6120])
513 and that directly reuse this profile MUST specify whether or not
514 case mapping is to be applied to authorization identifiers. Such
515 "SASL application protocols" SHOULD delay any case-mapping of
516 authorization identifiers to the last possible moment, which
517 happens to necessarily be on the server side (this enables
518 decisions about case mapping to be a matter of deployment policy).
519 In keeping with [RFC4422], SASL application protocols are not to
520 apply this or any other profile to authentication identifiers,
521 only to authorization identifiers.
523 o Application protocols that do not use SASL (such as HTTP
524 authentication with the HTTP Basic and Digest schemes as specified
525 in [HTTP-BASIC-AUTH] and [HTTP-DIGEST-AUTH]) but that directly
526 reuse this profile MUST specify whether and when case mapping is
527 to be applied to authentication identifiers or authorization
528 identifiers, or both. Such "non-SASL application protocols"
529 SHOULD delay any case mapping to the last possible moment, such as
530 when doing a lookup by username, performing username comparisons,
531 or generating a cryptographic salt from a username (if the last
532 possible moment happens on the server, then decisions about case
533 mapping can be a matter of deployment policy).
535 If the specification for a SASL mechanism, SASL application protocol,
536 or non-SASL application protocol uses the UsernameCaseMapped profile,
537 it MUST clearly describe whether case mapping is to be applied at the
538 level of the protocol itself, implementations thereof, or service
539 deployments (each of these approaches can be legitimate, depending on
540 the application in question).
5423.5. Application-Layer Constructs
544 Both the UsernameCaseMapped and UsernameCasePreserved profiles enable
545 an application protocol, implementation, or deployment to create
546 application-layer constructs such as a username that is a space-
547 separated set of userparts like "Firstname Middlename Lastname".
548 Although such a construct is not a profile of the PRECIS
549 IdentifierClass (because U+0020 SPACE is not allowed in the
550 IdentifierClass), it can be created at the application layer because
551 U+0020 SPACE can be used as a separator between instances of the
552 PRECIS IdentifierClass (e.g., userparts as defined in this
562Saint-Andre & Melnikov Standards Track [Page 10]
564RFC 7613 PRECIS: Usernames and Passwords August 2015
569 The following examples illustrate a small number of userparts (not
570 usernames) that are consistent with the format defined above (note
571 that the characters "<" and ">" are used here to delineate the actual
572 userparts and are not part of the userpart strings).
574 +--------------------------+---------------------------------+
575 | # | Userpart | Notes |
576 +--------------------------+---------------------------------+
577 | 1 | <juliet@example.com> | The at-sign is allowed in the |
578 | | | PRECIS IdentifierClass |
579 +--------------------------+---------------------------------+
581 +--------------------------+---------------------------------+
582 | 3 | <fußball> | The third character is LATIN |
583 | | | SMALL LETTER SHARP S (U+00DF) |
584 +--------------------------+---------------------------------+
585 | 4 | <π> | A userpart of GREEK SMALL |
586 | | | LETTER PI (U+03C0) |
587 +--------------------------+---------------------------------+
588 | 5 | <Σ> | A userpart of GREEK CAPITAL |
589 | | | LETTER SIGMA (U+03A3) |
590 +--------------------------+---------------------------------+
591 | 6 | <σ> | A userpart of GREEK SMALL |
592 | | | LETTER SIGMA (U+03C3) |
593 +--------------------------+---------------------------------+
594 | 7 | <ς> | A userpart of GREEK SMALL |
595 | | | LETTER FINAL SIGMA (U+03C2) |
596 +--------------------------+---------------------------------+
598 Table 1: A Sample of Legal Userparts
600 Several points are worth noting. Regarding examples 2 and 3:
601 although in German the character eszett (LATIN SMALL LETTER SHARP S
602 (U+00DF)) can mostly be used interchangeably with the two characters
603 "ss", the userparts in these examples are different and (if desired)
604 a server would need to enforce a registration policy that disallows
605 one of them if the other is registered. Regarding examples 5, 6, and
606 7: optional case-mapping of GREEK CAPITAL LETTER SIGMA (U+03A3) to
607 lowercase (i.e., to GREEK SMALL LETTER SIGMA (U+03C3)) during
608 comparison would result in matching the userparts in examples 5 and
609 6; however, because the PRECIS mapping rules do not account for the
610 special status of GREEK SMALL LETTER FINAL SIGMA (U+03C2), the
611 userparts in examples 5 and 7 or examples 6 and 7 would not be
612 matched during comparison.
618Saint-Andre & Melnikov Standards Track [Page 11]
620RFC 7613 PRECIS: Usernames and Passwords August 2015
623 The following examples illustrate strings that are not valid
624 userparts (not usernames) because they violate the format defined
627 +--------------------------+---------------------------------+
628 | # | Non-Userpart String | Notes |
629 +--------------------------+---------------------------------+
630 | 8 | <foo bar> | Space (U+0020) is disallowed in |
632 +--------------------------+---------------------------------+
633 | 9 | <> | Zero-length userpart |
634 +--------------------------+---------------------------------+
635 | 10| <henryⅣ> | The sixth character is ROMAN |
636 | | | NUMERAL FOUR (U+2163) |
637 +--------------------------+---------------------------------+
638 | 11| <♚> | A localpart of BLACK CHESS KING |
640 +--------------------------+---------------------------------+
642 Table 2: A Sample of Strings That Violate the Userpart Rule
644 Here again, several points are worth noting. Regarding example 8:
645 although this is not a valid userpart, it is a valid username because
646 it is a space-separated sequence of userparts. Regarding example 10:
647 the Unicode character ROMAN NUMERAL FOUR (U+2163) has a compatibility
648 equivalent of the string formed of LATIN CAPITAL LETTER I (U+0049)
649 and LATIN CAPITAL LETTER V (U+0056), but characters with
650 compatibility equivalents are not allowed in the PRECIS
651 IdentifierClass. Regarding example 11: symbol characters such as
652 BLACK CHESS KING (U+265A) are not allowed in the PRECIS
674Saint-Andre & Melnikov Standards Track [Page 12]
676RFC 7613 PRECIS: Usernames and Passwords August 2015
683 This document specifies that a password is a string of Unicode code
684 points [Unicode], encoded using UTF-8 [RFC3629], and conformant to
685 the OpaqueString profile (specified below) of the PRECIS
686 FreeformClass defined in Section 4.3 of [RFC7564].
688 The syntax for a password is defined as follows, using the Augmented
689 Backus-Naur Form (ABNF) [RFC5234].
691 password = 1*(freebyte)
693 ; a "freebyte" is a byte used to represent a
694 ; UTF-8 encoded Unicode code point that can be
695 ; contained in a string that conforms to the
696 ; PRECIS FreeformClass
699 All code points and blocks not explicitly allowed in the PRECIS
700 FreeformClass are disallowed; this includes private use characters,
701 surrogate code points, and the other code points and blocks defined
702 as "Prohibited Output" in Section 2.3 of RFC 4013 (when corrected per
705 A password MUST NOT be zero bytes in length. This rule is to be
706 enforced after any normalization and mapping of code points.
708 Note: Some existing systems allow an empty string in places where
709 a password would be expected (e.g., command-line tools that might
710 be called from an automated script, or servers that might need to
711 be restarted without human intervention). From the perspective of
712 this document (and RFC 4013 before it), these empty strings are
713 not passwords but are workarounds for the practical difficulty of
714 using passwords in certain scenarios. The prohibition of
715 zero-length passwords is not a recommendation regarding password
716 strength (because a password of only one byte is highly insecure)
717 but is meant to prevent applications from mistakenly omitting a
718 password entirely; such an outcome is possible when
719 internationalized characters are accepted, because a non-empty
720 sequence of characters can result in a zero-length password after
730Saint-Andre & Melnikov Standards Track [Page 13]
732RFC 7613 PRECIS: Usernames and Passwords August 2015
735 In protocols that provide passwords as input to a cryptographic
736 algorithm such as a hash function, the client will need to perform
737 enforcement of the rules for the OpaqueString profile before applying
738 the algorithm, because the password is not available to the server in
7414.2. OpaqueString Profile
743 The definition of the OpaqueString profile is provided in the
744 following sections, including detailed information about preparation,
745 enforcement, and comparison (for details on the distinction between
746 these actions, refer to [RFC7564]).
750 An entity that prepares a string according to this profile MUST
751 ensure that the string consists only of Unicode code points that
752 conform to the FreeformClass base string class defined in [RFC7564].
753 In addition, the entity MUST encode the string as UTF-8 [RFC3629].
757 An entity that performs enforcement according to this profile MUST
758 prepare a string as described in Section 4.2.1 and MUST also apply
759 the rules specified below for the OpaqueString profile (these rules
760 MUST be applied in the order shown):
762 1. Width-Mapping Rule: Fullwidth and halfwidth characters MUST NOT
763 be mapped to their decomposition mappings (see Unicode Standard
766 2. Additional Mapping Rule: Any instances of non-ASCII space MUST be
767 mapped to ASCII space (U+0020); a non-ASCII space is any Unicode
768 code point having a Unicode general category of "Zs" (with the
769 exception of U+0020).
771 3. Case-Mapping Rule: Uppercase and titlecase characters MUST NOT be
772 mapped to their lowercase equivalents.
774 4. Normalization Rule: Unicode Normalization Form C (NFC) MUST be
775 applied to all characters.
777 5. Directionality Rule: There is no directionality rule. The "Bidi
778 Rule" (defined in [RFC5893]) and similar rules are unnecessary
779 and inapplicable to passwords, because they can reduce the range
780 of characters that are allowed in a string and therefore reduce
781 the amount of entropy that is possible in a password. Such rules
782 are intended to minimize the possibility that the same string
786Saint-Andre & Melnikov Standards Track [Page 14]
788RFC 7613 PRECIS: Usernames and Passwords August 2015
791 will be displayed differently on a layout system set for
792 right-to-left display and a layout system set for left-to-right
793 display; however, passwords are typically not displayed at all
794 and are rarely meant to be interoperable across different layout
795 systems in the way that non-secret strings like domain names and
796 usernames are. Furthermore, it is perfectly acceptable for
797 opaque strings other than passwords to be presented differently
798 in different layout systems, as long as the presentation is
799 consistent in any given layout system.
803 An entity that performs comparison of two strings according to this
804 profile MUST prepare each string as specified in Section 4.2.1 and
805 then enforce the rules specified in Section 4.2.2. The two strings
806 are to be considered equivalent if they are an exact octet-for-octet
807 match (sometimes called "bit-string identity").
811 The following examples illustrate a small number of passwords that
812 are consistent with the format defined above (note that the
813 characters "<" and ">" are used here to delineate the actual
814 passwords and are not part of the password strings).
816 +------------------------------------+------------------------------+
817 | # | Password | Notes |
818 +------------------------------------+------------------------------+
819 | 12| <correct horse battery staple> | ASCII space is allowed |
820 +------------------------------------+------------------------------+
821 | 13| <Correct Horse Battery Staple> | Differs by case from |
823 +------------------------------------+------------------------------+
824 | 14| <πßå> | Non-ASCII letters are OK |
825 | | | (e.g., GREEK SMALL LETTER |
827 +------------------------------------+------------------------------+
828 | 15| <Jack of ♦s> | Symbols are OK (e.g., BLACK |
829 | | | DIAMOND SUIT (U+2666)) |
830 +------------------------------------+------------------------------+
831 | 16| <foo bar> | OGHAM SPACE MARK (U+1680) is |
832 | | | mapped to U+0020, and thus |
833 | | | the full string is mapped to |
835 +------------------------------------+------------------------------+
837 Table 3: A Sample of Legal Passwords
842Saint-Andre & Melnikov Standards Track [Page 15]
844RFC 7613 PRECIS: Usernames and Passwords August 2015
847 The following example illustrates a string that is not a valid
848 password because it violates the format defined above.
850 +------------------------------------+------------------------------+
851 | # | Password | Notes |
852 +------------------------------------+------------------------------+
853 | 17| <my cat is a 	by> | Controls are disallowed |
854 +------------------------------------+------------------------------+
856 Table 4: A String That Violates the Password Rules
8585. Use in Application Protocols
860 This specification defines only the PRECIS-based rules for the
861 handling of strings conforming to the UsernameCaseMapped and
862 UsernameCasePreserved profiles of the PRECIS IdentifierClass, and
863 strings conforming to the OpaqueString profile of the PRECIS
864 FreeformClass. It is the responsibility of an application protocol
865 to specify the protocol slots in which such strings can appear, the
866 entities that are expected to enforce the rules governing such
867 strings, and at what points during protocol processing or interface
868 handling the rules need to be enforced. See Section 6 of [RFC7564]
869 for guidelines on using PRECIS profiles in applications.
871 Above and beyond the PRECIS-based rules specified here, application
872 protocols can also define application-specific rules governing such
873 strings (rules regarding minimum or maximum length, further
874 restrictions on allowable characters or character ranges, safeguards
875 to mitigate the effects of visually similar characters, etc.),
876 application-layer constructs (see Section 3.5), and related matters.
878 Some PRECIS profile definitions encourage entities that enforce the
879 rules to be liberal in what they accept. However, for usernames and
880 passwords such a policy can be problematic, because it can lead to
881 false positives. An in-depth discussion can be found in [RFC6943].
885 The rules defined in this specification differ slightly from those
886 defined by the SASLprep specification [RFC4013]. The following
887 sections describe these differences, along with their implications
888 for migration, in more detail.
898Saint-Andre & Melnikov Standards Track [Page 16]
900RFC 7613 PRECIS: Usernames and Passwords August 2015
905 Deployments that currently use SASLprep for handling usernames might
906 need to scrub existing data when they migrate to the rules defined in
907 this specification. In particular:
909 o SASLprep specified the use of Unicode Normalization Form KC
910 (NFKC), whereas the UsernameCaseMapped and UsernameCasePreserved
911 profiles employ Unicode Normalization Form C (NFC). In practice,
912 this change is unlikely to cause significant problems, because
913 NFKC provides methods for mapping Unicode code points with
914 compatibility equivalents to those equivalents, whereas the PRECIS
915 IdentifierClass entirely disallows Unicode code points with
916 compatibility equivalents (i.e., during comparison, NFKC is more
917 "aggressive" about finding matches than NFC). A few examples
918 might suffice to indicate the nature of the problem:
920 1. LATIN SMALL LETTER LONG S (U+017F) is compatibility equivalent
921 to LATIN SMALL LETTER S (U+0073).
923 2. ROMAN NUMERAL FOUR (U+2163) is compatibility equivalent to
924 LATIN CAPITAL LETTER I (U+0049) and LATIN CAPITAL LETTER V
927 3. LATIN SMALL LIGATURE FI (U+FB01) is compatibility equivalent
928 to LATIN SMALL LETTER F (U+0066) and LATIN SMALL LETTER I
931 Under SASLprep, the use of NFKC also handled the mapping of
932 fullwidth and halfwidth code points to their decomposition
935 For migration purposes, operators might want to search their
936 database of usernames for names containing Unicode code points
937 with compatibility equivalents and, where there is no conflict,
938 map those code points to their equivalents. Naturally, it is
939 possible that during this process the operator will discover
940 conflicting usernames (e.g., HENRYIV with the last two characters
941 being LATIN CAPITAL LETTER I (U+0049) and LATIN CAPITAL LETTER V
942 (U+0056) vs. "HENRYIV" with the last character being ROMAN NUMERAL
943 FOUR (U+2163), which is compatibility equivalent to U+0049 and
944 U+0056); in these cases, the operator will need to determine how
945 to proceed -- for instance, by disabling the account whose name
946 contains a Unicode code point with a compatibility equivalent.
947 Such cases are probably rare, but it is important for operators to
954Saint-Andre & Melnikov Standards Track [Page 17]
956RFC 7613 PRECIS: Usernames and Passwords August 2015
959 o SASLprep mapped the "characters commonly mapped to nothing" from
960 Appendix B.1 of [RFC3454]) to nothing, whereas the PRECIS
961 IdentifierClass entirely disallows most of these characters, which
962 correspond to the code points from the PRECIS "M" category defined
963 under Section 9.13 of [RFC7564] (with the exception of MONGOLIAN
964 TODO SOFT HYPHEN (U+1806), which was "commonly mapped to nothing"
965 in Unicode 3.2 but at the time of this writing does not have a
966 derived property of Default_Ignorable_Code_Point in Unicode 7.0).
967 For migration purposes, the operator might want to remove from
968 usernames any code points contained in the PRECIS "M" category
969 (e.g., SOFT HYPHEN (U+00AD)). Because these code points would
970 have been "mapped to nothing" in stringprep, in practice a user
971 would not notice the difference if, upon migration to PRECIS, the
972 code points are removed.
974 o SASLprep allowed uppercase and titlecase characters, whereas the
975 UsernameCaseMapped profile maps uppercase and titlecase characters
976 to their lowercase equivalents (by contrast, the
977 UsernameCasePreserved profile matches SASLprep in this regard).
978 For migration purposes, the operator can use either the
979 UsernameCaseMapped profile (thus losing the case information) or
980 the UsernameCasePreserved profile (thus ignoring case difference
981 when comparing usernames).
985 Depending on local service policy, migration from RFC 4013 to this
986 specification might not involve any scrubbing of data (because
987 passwords might not be stored in the clear anyway); however, service
988 providers need to be aware of possible issues that might arise during
989 migration. In particular:
991 o SASLprep specified the use of Unicode Normalization Form KC
992 (NFKC), whereas the OpaqueString profile employs Unicode
993 Normalization Form C (NFC). Because NFKC is more aggressive about
994 finding matches than NFC, in practice this change is unlikely to
995 cause significant problems and indeed has the security benefit of
996 probably resulting in fewer false positives when comparing
997 passwords. A few examples might suffice to indicate the nature of
1000 1. LATIN SMALL LETTER LONG S (U+017F) is compatibility equivalent
1001 to LATIN SMALL LETTER S (U+0073).
1003 2. ROMAN NUMERAL FOUR (U+2163) is compatibility equivalent to
1004 LATIN CAPITAL LETTER I (U+0049) and LATIN CAPITAL LETTER V
1010Saint-Andre & Melnikov Standards Track [Page 18]
1012RFC 7613 PRECIS: Usernames and Passwords August 2015
1015 3. LATIN SMALL LIGATURE FI (U+FB01) is compatibility equivalent
1016 to LATIN SMALL LETTER F (U+0066) and LATIN SMALL LETTER I
1019 Under SASLprep, the use of NFKC also handled the mapping of
1020 fullwidth and halfwidth code points to their decomposition
1021 mappings. Although it is expected that code points with
1022 compatibility equivalents are rare in existing passwords, some
1023 passwords that matched when SASLprep was used might no longer work
1024 when the rules in this specification are applied.
1026 o SASLprep mapped the "characters commonly mapped to nothing" from
1027 Appendix B.1 of [RFC3454]) to nothing, whereas the PRECIS
1028 FreeformClass entirely disallows such characters, which correspond
1029 to the code points from the PRECIS "M" category defined under
1030 Section 9.13 of [RFC7564] (with the exception of MONGOLIAN TODO
1031 SOFT HYPHEN (U+1806), which was commonly mapped to nothing in
1032 Unicode 3.2 but at the time of this writing is allowed by
1033 Unicode 7.0). In practice, this change will probably have no
1034 effect on comparison, but user-oriented software might reject such
1035 code points instead of ignoring them during password preparation.
10377. IANA Considerations
1039 IANA has made the updates described below.
10417.1. UsernameCaseMapped Profile
1043 IANA has added the following entry to the "PRECIS Profiles" registry.
1045 Name: UsernameCaseMapped.
1047 Base Class: IdentifierClass.
1049 Applicability: Usernames in security and application protocols.
1051 Replaces: The SASLprep profile of stringprep.
1053 Width-Mapping Rule: Map fullwidth and halfwidth characters to their
1054 decomposition mappings.
1056 Additional Mapping Rule: None.
1058 Case-Mapping Rule: Map uppercase and titlecase characters to
1061 Normalization Rule: NFC.
1066Saint-Andre & Melnikov Standards Track [Page 19]
1068RFC 7613 PRECIS: Usernames and Passwords August 2015
1071 Directionality Rule: The "Bidi Rule" defined in RFC 5893 applies.
1073 Enforcement: To be defined by security or application protocols that
1076 Specification: RFC 7613 (this document), Section 3.2.
10787.2. UsernameCasePreserved Profile
1080 IANA has added the following entry to the "PRECIS Profiles" registry.
1082 Name: UsernameCasePreserved.
1084 Base Class: IdentifierClass.
1086 Applicability: Usernames in security and application protocols.
1088 Replaces: The SASLprep profile of stringprep.
1090 Width-Mapping Rule: Map fullwidth and halfwidth characters to their
1091 decomposition mappings.
1093 Additional Mapping Rule: None.
1095 Case-Mapping Rule: None.
1097 Normalization Rule: NFC.
1099 Directionality Rule: The "Bidi Rule" defined in RFC 5893 applies.
1101 Enforcement: To be defined by security or application protocols that
1104 Specification: RFC 7613 (this document), Section 3.3.
11067.3. OpaqueString Profile
1108 IANA has added the following entry to the "PRECIS Profiles" registry.
1112 Base Class: FreeformClass.
1114 Applicability: Passwords and other opaque strings in security and
1115 application protocols.
1117 Replaces: The SASLprep profile of stringprep.
1122Saint-Andre & Melnikov Standards Track [Page 20]
1124RFC 7613 PRECIS: Usernames and Passwords August 2015
1127 Width-Mapping Rule: None.
1129 Additional Mapping Rule: Map non-ASCII space characters to ASCII
1132 Case-Mapping Rule: None.
1134 Normalization Rule: NFC.
1136 Directionality Rule: None.
1138 Enforcement: To be defined by security or application protocols that
1141 Specification: RFC 7613 (this document), Section 4.2.
11437.4. Stringprep Profile
1145 The stringprep specification [RFC3454] did not provide for entries in
1146 the "Stringprep Profiles" registry to have any state except "Current"
1147 or "Not Current". Because this document obsoletes RFC 4013, which
1148 registered the SASLprep profile of stringprep, IANA has marked that
1149 profile as "Not Current" and cited this document as an additional
11528. Security Considerations
11548.1. Password/Passphrase Strength
1156 The ability to include a wide range of characters in passwords and
1157 passphrases can increase the potential for creating a strong password
1158 with high entropy. However, in practice, the ability to include such
1159 characters ought to be weighed against the possible need to reproduce
1160 them on various devices using various input methods.
11628.2. Identifier Comparison
1164 The process of comparing identifiers (such as SASL simple user names,
1165 authentication identifiers, and authorization identifiers) can lead
1166 to either false negatives or false positives, both of which have
1167 security implications. A more detailed discussion can be found in
1172 The security considerations described in [RFC7564] apply to the
1173 IdentifierClass and FreeformClass base string classes used in this
1174 document for usernames and passwords, respectively.
1178Saint-Andre & Melnikov Standards Track [Page 21]
1180RFC 7613 PRECIS: Usernames and Passwords August 2015
11838.4. Reuse of Unicode
1185 The security considerations described in [UTS39] apply to the use of
1186 Unicode characters in usernames and passwords.
11909.1. Normative References
1192 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
1193 Requirement Levels", BCP 14, RFC 2119,
1194 DOI 10.17487/RFC2119, March 1997,
1195 <http://www.rfc-editor.org/info/rfc2119>.
1197 [RFC3629] Yergeau, F., "UTF-8, a transformation format of
1198 ISO 10646", STD 63, RFC 3629, DOI 10.17487/RFC3629,
1199 November 2003, <http://www.rfc-editor.org/info/rfc3629>.
1201 [RFC5234] Crocker, D., Ed., and P. Overell, "Augmented BNF for
1202 Syntax Specifications: ABNF", STD 68, RFC 5234,
1203 DOI 10.17487/RFC5234, January 2008,
1204 <http://www.rfc-editor.org/info/rfc5234>.
1206 [RFC5890] Klensin, J., "Internationalized Domain Names for
1207 Applications (IDNA): Definitions and Document Framework",
1208 RFC 5890, DOI 10.17487/RFC5890, August 2010,
1209 <http://www.rfc-editor.org/info/rfc5890>.
1211 [RFC6365] Hoffman, P. and J. Klensin, "Terminology Used in
1212 Internationalization in the IETF", BCP 166, RFC 6365,
1213 DOI 10.17487/RFC6365, September 2011,
1214 <http://www.rfc-editor.org/info/rfc6365>.
1216 [RFC7564] Saint-Andre, P. and M. Blanchet, "PRECIS Framework:
1217 Preparation, Enforcement, and Comparison of
1218 Internationalized Strings in Application Protocols",
1219 RFC 7564, DOI 10.17487/RFC7564, May 2015,
1220 <http://www.rfc-editor.org/info/rfc7564>.
1222 [UAX11] Unicode Standard Annex #11, "East Asian Width", edited by
1223 Ken Lunde. An integral part of The Unicode Standard,
1224 <http://unicode.org/reports/tr11/>.
1234Saint-Andre & Melnikov Standards Track [Page 22]
1236RFC 7613 PRECIS: Usernames and Passwords August 2015
1239 [Unicode] The Unicode Consortium, "The Unicode Standard",
1240 <http://www.unicode.org/versions/latest/>.
1243 The Unicode Consortium, "The Unicode Standard,
1244 Version 7.0.0", (Mountain View, CA: The Unicode
1245 Consortium, 2014 ISBN 978-1-936213-09-2),
1246 <http://www.unicode.org/versions/Unicode7.0.0/>.
12489.2. Informative References
1250 [Err1812] RFC Errata, Erratum ID 1812, RFC 4013,
1251 <http://www.rfc-editor.org>.
1254 Reschke, J., "The 'Basic' HTTP Authentication Scheme",
1255 Work in Progress, draft-ietf-httpauth-basicauth-update-07,
1259 Shekh-Yusef, R., Ed., Ahrens, D., and S. Bremer, "HTTP
1260 Digest Access Authentication", Work in Progress,
1261 draft-ietf-httpauth-digest-19, April 2015.
1263 [RFC20] Cerf, V., "ASCII format for network interchange", STD 80,
1264 RFC 20, DOI 10.17487/RFC0020, October 1969,
1265 <http://www.rfc-editor.org/info/rfc20>.
1267 [RFC3454] Hoffman, P. and M. Blanchet, "Preparation of
1268 Internationalized Strings ("stringprep")", RFC 3454,
1269 DOI 10.17487/RFC3454, December 2002,
1270 <http://www.rfc-editor.org/info/rfc3454>.
1272 [RFC3501] Crispin, M., "INTERNET MESSAGE ACCESS PROTOCOL -
1273 VERSION 4rev1", RFC 3501, DOI 10.17487/RFC3501,
1274 March 2003, <http://www.rfc-editor.org/info/rfc3501>.
1276 [RFC4013] Zeilenga, K., "SASLprep: Stringprep Profile for User Names
1277 and Passwords", RFC 4013, DOI 10.17487/RFC4013,
1278 February 2005, <http://www.rfc-editor.org/info/rfc4013>.
1280 [RFC4422] Melnikov, A., Ed., and K. Zeilenga, Ed., "Simple
1281 Authentication and Security Layer (SASL)", RFC 4422,
1282 DOI 10.17487/RFC4422, June 2006,
1283 <http://www.rfc-editor.org/info/rfc4422>.
1290Saint-Andre & Melnikov Standards Track [Page 23]
1292RFC 7613 PRECIS: Usernames and Passwords August 2015
1295 [RFC4616] Zeilenga, K., Ed., "The PLAIN Simple Authentication and
1296 Security Layer (SASL) Mechanism", RFC 4616,
1297 DOI 10.17487/RFC4616, August 2006,
1298 <http://www.rfc-editor.org/info/rfc4616>.
1300 [RFC5802] Newman, C., Menon-Sen, A., Melnikov, A., and N. Williams,
1301 "Salted Challenge Response Authentication Mechanism
1302 (SCRAM) SASL and GSS-API Mechanisms", RFC 5802,
1303 DOI 10.17487/RFC5802, July 2010,
1304 <http://www.rfc-editor.org/info/rfc5802>.
1306 [RFC5891] Klensin, J., "Internationalized Domain Names in
1307 Applications (IDNA): Protocol", RFC 5891,
1308 DOI 10.17487/RFC5891, August 2010,
1309 <http://www.rfc-editor.org/info/rfc5891>.
1311 [RFC5893] Alvestrand, H., Ed., and C. Karp, "Right-to-Left Scripts
1312 for Internationalized Domain Names for Applications
1313 (IDNA)", RFC 5893, DOI 10.17487/RFC5893, August 2010,
1314 <http://www.rfc-editor.org/info/rfc5893>.
1316 [RFC5894] Klensin, J., "Internationalized Domain Names for
1317 Applications (IDNA): Background, Explanation, and
1318 Rationale", RFC 5894, DOI 10.17487/RFC5894, August 2010,
1319 <http://www.rfc-editor.org/info/rfc5894>.
1321 [RFC6120] Saint-Andre, P., "Extensible Messaging and Presence
1322 Protocol (XMPP): Core", RFC 6120, DOI 10.17487/RFC6120,
1323 March 2011, <http://www.rfc-editor.org/info/rfc6120>.
1325 [RFC6122] Saint-Andre, P., "Extensible Messaging and Presence
1326 Protocol (XMPP): Address Format", RFC 6122,
1327 DOI 10.17487/RFC6122, March 2011,
1328 <http://www.rfc-editor.org/info/rfc6122>.
1330 [RFC6943] Thaler, D., Ed., "Issues in Identifier Comparison for
1331 Security Purposes", RFC 6943, DOI 10.17487/RFC6943,
1332 May 2013, <http://www.rfc-editor.org/info/rfc6943>.
1334 [RFC7542] DeKok, A., "The Network Access Identifier", RFC 7542,
1335 DOI 10.17487/RFC7542, May 2015,
1336 <http://www.rfc-editor.org/info/rfc7542>.
1346Saint-Andre & Melnikov Standards Track [Page 24]
1348RFC 7613 PRECIS: Usernames and Passwords August 2015
1351 [UTS39] Unicode Technical Standard #39, "Unicode Security
1352 Mechanisms", edited by Mark Davis and Michel Suignard,
1353 <http://unicode.org/reports/tr39/>.
1356 Saint-Andre, P., "Extensible Messaging and Presence
1357 Protocol (XMPP): Address Format", Work in Progress,
1358 draft-ietf-xmpp-6122bis-24, June 2015.
1402Saint-Andre & Melnikov Standards Track [Page 25]
1404RFC 7613 PRECIS: Usernames and Passwords August 2015
1407Appendix A. Differences from RFC 4013
1409 This document builds upon the PRECIS framework defined in [RFC7564],
1410 which differs fundamentally from the stringprep technology [RFC3454]
1411 used in SASLprep [RFC4013]. The primary difference is that
1412 stringprep profiles allowed all characters except those characters
1413 that were explicitly disallowed, whereas PRECIS profiles disallow all
1414 characters except those characters that are explicitly allowed (this
1415 "inclusion model" was originally used for internationalized domain
1416 names in [RFC5891]; see [RFC5894] for further discussion). It is
1417 important to keep this distinction in mind when comparing the
1418 technology defined in this document to SASLprep [RFC4013].
1420 The following substantive modifications were made from RFC 4013.
1422 o A single SASLprep algorithm was replaced by three separate
1423 algorithms: one for usernames with case mapping, one for usernames
1424 with case preservation, and one for passwords.
1426 o The new preparation algorithms use PRECIS instead of a stringprep
1427 profile. The new algorithms work independently of Unicode
1430 o As recommended in the PRECIS framework, changed the Unicode
1431 normalization form from NFKC to NFC.
1433 o Some Unicode code points that were mapped to nothing in RFC 4013
1434 are simply disallowed by PRECIS.
1458Saint-Andre & Melnikov Standards Track [Page 26]
1460RFC 7613 PRECIS: Usernames and Passwords August 2015
1465 This document borrows some text from [RFC4013] and [RFC6120].
1467 The following individuals provided helpful feedback on this document:
1468 Marc Blanchet, Ben Campbell, Alan DeKok, Joe Hildebrand, Jeffrey
1469 Hutzelman, Simon Josefsson, Jonathan Lennox, James Manger, Matt
1470 Miller, Chris Newman, Yutaka OIWA, Pete Resnick, Andrew Sullivan,
1471 Nico Williams, and Yoshiro YONEYA. Nico Williams in particular
1472 deserves special recognition for providing text that was used in
1473 Section 3.4. Thanks also to Takahiro NEMOTO and Yoshiro YONEYA for
1474 implementation feedback.
1476 Robert Sparks and Derek Atkins reviewed the document on behalf of the
1477 General Area Review Team and the Security Directorate, respectively.
1479 Benoit Claise and Stephen Farrell provided helpful input during IESG
1482 Thanks to Matt Miller as document shepherd, Marc Blanchet and Yoshiro
1483 YONEYA as working group chairs, and Pete Resnick and Barry Leiba as
1486 Peter Saint-Andre wishes to acknowledge Cisco Systems, Inc., for
1487 employing him during his work on earlier draft versions of this
1495 Email: peter@andyet.com
1496 URI: https://andyet.com/
1501 5 Castle Business Village
1503 Hampton, Middlesex TW12 2BX
1506 Email: Alexey.Melnikov@isode.com
1514Saint-Andre & Melnikov Standards Track [Page 27]