idnits 2.17.00 (12 Aug 2021)
/tmp/idnits2054/draft-ietf-idnabis-rationale-09.txt:
Checking boilerplate required by RFC 5378 and the IETF Trust (see
https://trustee.ietf.org/license-info):
----------------------------------------------------------------------------
** The document seems to lack a License Notice according IETF Trust
Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009
Section 6.b -- however, there's a paragraph with a matching beginning.
Boilerplate error?
(You're using the IETF Trust Provisions' Section 6.b License Notice from
12 Feb 2009 rather than one of the newer Notices. See
https://trustee.ietf.org/license-info/.)
Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt:
----------------------------------------------------------------------------
No issues found here.
Checking nits according to https://www.ietf.org/id-info/checklist :
----------------------------------------------------------------------------
** The document seems to lack a both a reference to RFC 2119 and the
recommended RFC 2119 boilerplate, even if it appears to use RFC 2119
keywords.
RFC 2119 keyword, line 946: '... CONTEXTUAL RULE REQUIRED category and...'
RFC 2119 keyword, line 1805: '...esented directly in ASCII MUST use the...'
RFC 2119 keyword, line 1806: '...sion to A-labels MUST be performed pri...'
RFC 2119 keyword, line 2197: '... MUST, SHOULD, or MAY) and small ...'
Miscellaneous warnings:
----------------------------------------------------------------------------
== The copyright year in the IETF Trust and authors Copyright Line does not
match the current year
-- The document seems to contain a disclaimer for pre-RFC5378 work, and may
have content which was first submitted before 10 November 2008. The
disclaimer is necessary when there are original authors that you have
been unable to contact, or if some do not wish to grant the BCP78 rights
to the IETF Trust. If you are able to get all authors (current and
original) to grant those rights, you can and should remove the
disclaimer; otherwise, the disclaimer is needed and you can ignore this
comment. (See the Legal Provisions document at
https://trustee.ietf.org/license-info for more information.)
-- The document date (March 9, 2009) is 4820 days in the past. Is this
intentional?
Checking references for intended status: Informational
----------------------------------------------------------------------------
== Unused Reference: 'ASCII' is defined on line 1946, but no explicit
reference was found in the text
== Unused Reference: 'RFC1034' is defined on line 2023, but no explicit
reference was found in the text
== Unused Reference: 'RFC1035' is defined on line 2026, but no explicit
reference was found in the text
== Unused Reference: 'RFC2673' is defined on line 2048, but no explicit
reference was found in the text
== Unused Reference: 'Unicode-UTR36' is defined on line 2096, but no
explicit reference was found in the text
== Outdated reference: draft-ietf-idnabis-defs has been published as RFC
5890
== Outdated reference: draft-ietf-idnabis-tables has been published as RFC
5892
** Obsolete normative reference: RFC 3490 (Obsoleted by RFC 5890, RFC 5891)
-- Obsolete informational reference (is this intentional?): RFC 810
(Obsoleted by RFC 952)
-- Obsolete informational reference (is this intentional?): RFC 2535
(Obsoleted by RFC 4033, RFC 4034, RFC 4035)
-- Obsolete informational reference (is this intentional?): RFC 2671
(Obsoleted by RFC 6891)
-- Obsolete informational reference (is this intentional?): RFC 2673
(Obsoleted by RFC 6891)
-- Obsolete informational reference (is this intentional?): RFC 3454
(Obsoleted by RFC 7564)
-- Obsolete informational reference (is this intentional?): RFC 3491
(Obsoleted by RFC 5891)
Summary: 3 errors (**), 0 flaws (~~), 8 warnings (==), 8 comments (--).
Run idnits with the --verbose option for more detailed information about
the items above.
--------------------------------------------------------------------------------
2 Network Working Group J. Klensin
3 Internet-Draft March 9, 2009
4 Intended status: Informational
5 Expires: September 10, 2009
7 Internationalized Domain Names for Applications (IDNA): Background,
8 Explanation, and Rationale
9 draft-ietf-idnabis-rationale-09.txt
11 Status of this Memo
13 This Internet-Draft is submitted to IETF in full conformance with the
14 provisions of BCP 78 and BCP 79. This document may contain material
15 from IETF Documents or IETF Contributions published or made publicly
16 available before November 10, 2008. The person(s) controlling the
17 copyright in some of this material may not have granted the IETF
18 Trust the right to allow modifications of such material outside the
19 IETF Standards Process. Without obtaining an adequate license from
20 the person(s) controlling the copyright in such materials, this
21 document may not be modified outside the IETF Standards Process, and
22 derivative works of it may not be created outside the IETF Standards
23 Process, except to format it for publication as an RFC or to
24 translate it into languages other than English.
26 Internet-Drafts are working documents of the Internet Engineering
27 Task Force (IETF), its areas, and its working groups. Note that
28 other groups may also distribute working documents as Internet-
29 Drafts.
31 Internet-Drafts are draft documents valid for a maximum of six months
32 and may be updated, replaced, or obsoleted by other documents at any
33 time. It is inappropriate to use Internet-Drafts as reference
34 material or to cite them other than as "work in progress."
36 The list of current Internet-Drafts can be accessed at
37 http://www.ietf.org/ietf/1id-abstracts.txt.
39 The list of Internet-Draft Shadow Directories can be accessed at
40 http://www.ietf.org/shadow.html.
42 This Internet-Draft will expire on September 10, 2009.
44 Copyright Notice
46 Copyright (c) 2009 IETF Trust and the persons identified as the
47 document authors. All rights reserved.
49 This document is subject to BCP 78 and the IETF Trust's Legal
50 Provisions Relating to IETF Documents in effect on the date of
51 publication of this document (http://trustee.ietf.org/license-info).
52 Please review these documents carefully, as they describe your rights
53 and restrictions with respect to this document.
55 Abstract
57 Several years have passed since the original protocol for
58 Internationalized Domain Names (IDNs) was completed and deployed.
59 During that time, a number of issues have arisen, including the need
60 to update the system to deal with newer versions of Unicode. Some of
61 these issues require tuning of the existing protocols and the tables
62 on which they depend. This document provides an overview of a
63 revised system and provides explanatory material for its components.
65 Table of Contents
67 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4
68 1.1. Context and Overview . . . . . . . . . . . . . . . . . . . 4
69 1.2. Discussion Forum . . . . . . . . . . . . . . . . . . . . . 4
70 1.3. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5
71 1.3.1. Documents and Standards . . . . . . . . . . . . . . . 5
72 1.3.2. DNS "Name" Terminology . . . . . . . . . . . . . . . . 5
73 1.3.3. New Terminology and Restrictions . . . . . . . . . . . 6
74 1.4. Objectives . . . . . . . . . . . . . . . . . . . . . . . . 6
75 1.5. Applicability and Function of IDNA . . . . . . . . . . . . 7
76 1.6. Comprehensibility of IDNA Mechanisms and Processing . . . 8
77 2. Processing in IDNA2008 . . . . . . . . . . . . . . . . . . . . 9
78 3. Permitted Characters: An Inclusion List . . . . . . . . . . . 10
79 3.1. A Tiered Model of Permitted Characters and Labels . . . . 10
80 3.1.1. PROTOCOL-VALID . . . . . . . . . . . . . . . . . . . . 11
81 3.1.2. Characters Valid Only in Context With Others . . . . . 11
82 3.1.2.2. Rules and Their Application . . . . . . . . . . . 12
83 3.1.3. DISALLOWED . . . . . . . . . . . . . . . . . . . . . . 12
84 3.1.4. UNASSIGNED . . . . . . . . . . . . . . . . . . . . . . 13
85 3.2. Registration Policy . . . . . . . . . . . . . . . . . . . 14
86 3.3. Layered Restrictions: Tables, Context, Registration,
87 Applications . . . . . . . . . . . . . . . . . . . . . . . 14
88 4. Issues that Constrain Possible Solutions . . . . . . . . . . . 15
89 4.1. Display and Network Order . . . . . . . . . . . . . . . . 15
90 4.2. Entry and Display in Applications . . . . . . . . . . . . 16
91 4.3. Linguistic Expectations: Ligatures, Digraphs, and
92 Alternate Character Forms . . . . . . . . . . . . . . . . 17
93 4.4. Case Mapping and Related Issues . . . . . . . . . . . . . 20
94 4.5. Right to Left Text . . . . . . . . . . . . . . . . . . . . 20
95 5. IDNs and the Robustness Principle . . . . . . . . . . . . . . 21
96 6. Front-end and User Interface Processing for Lookup . . . . . . 22
97 7. Migration from IDNA2003 and Unicode Version Synchronization . 25
98 7.1. Design Criteria . . . . . . . . . . . . . . . . . . . . . 25
99 7.1.1. General IDNA Validity Criteria . . . . . . . . . . . . 25
100 7.1.2. Labels in Registration . . . . . . . . . . . . . . . . 26
101 7.1.3. Labels in Lookup . . . . . . . . . . . . . . . . . . . 27
102 7.2. Changes in Character Interpretations . . . . . . . . . . . 28
103 7.3. More Flexibility in User Agents . . . . . . . . . . . . . 30
104 7.4. The Question of Prefix Changes . . . . . . . . . . . . . . 31
105 7.4.1. Conditions Requiring a Prefix Change . . . . . . . . . 32
106 7.4.2. Conditions Not Requiring a Prefix Change . . . . . . . 32
107 7.4.3. Implications of Prefix Changes . . . . . . . . . . . . 33
108 7.5. Stringprep Changes and Compatibility . . . . . . . . . . . 33
109 7.6. The Symbol Question . . . . . . . . . . . . . . . . . . . 34
110 7.7. Migration Between Unicode Versions: Unassigned Code
111 Points . . . . . . . . . . . . . . . . . . . . . . . . . . 35
112 7.8. Other Compatibility Issues . . . . . . . . . . . . . . . . 37
113 8. Name Server Considerations . . . . . . . . . . . . . . . . . . 38
114 8.1. Processing Non-ASCII Strings . . . . . . . . . . . . . . . 38
115 8.2. DNSSEC Authentication of IDN Domain Names . . . . . . . . 38
116 8.3. Root and other DNS Server Considerations . . . . . . . . . 39
117 9. Internationalization Considerations . . . . . . . . . . . . . 39
118 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 39
119 10.1. IDNA Character Registry . . . . . . . . . . . . . . . . . 40
120 10.2. IDNA Context Registry . . . . . . . . . . . . . . . . . . 40
121 10.3. IANA Repository of IDN Practices of TLDs . . . . . . . . . 40
122 11. Security Considerations . . . . . . . . . . . . . . . . . . . 40
123 11.1. General Security Issues with IDNA . . . . . . . . . . . . 40
124 12. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 41
125 13. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 41
126 14. References . . . . . . . . . . . . . . . . . . . . . . . . . . 42
127 14.1. Normative References . . . . . . . . . . . . . . . . . . . 42
128 14.2. Informative References . . . . . . . . . . . . . . . . . . 43
129 Appendix A. Change Log . . . . . . . . . . . . . . . . . . . . . 45
130 A.1. Changes between Version -00 and Version -01 of
131 draft-ietf-idnabis-rationale . . . . . . . . . . . . . . . 45
132 A.2. Version -02 . . . . . . . . . . . . . . . . . . . . . . . 46
133 A.3. Version -03 . . . . . . . . . . . . . . . . . . . . . . . 46
134 A.4. Version -04 . . . . . . . . . . . . . . . . . . . . . . . 46
135 A.5. Version -05 . . . . . . . . . . . . . . . . . . . . . . . 47
136 A.6. Version -06 . . . . . . . . . . . . . . . . . . . . . . . 47
137 A.7. Version -07 . . . . . . . . . . . . . . . . . . . . . . . 48
138 A.8. Version -08 . . . . . . . . . . . . . . . . . . . . . . . 48
139 A.9. Version -09 . . . . . . . . . . . . . . . . . . . . . . . 48
140 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 49
142 1. Introduction
144 1.1. Context and Overview
146 The original standards for Internationalized Domain Names (IDNs) were
147 completed and deployed starting in 2003. Those standards are known
148 as Internationalized Domain Names in Applications (IDNA), taken from
149 the name of the highest level standard within the group, RFC 3490
150 [RFC3490]. After those standards were deployed, a number of issues
151 arose that led to a call for a new version of the IDNA protocol and
152 the associated tables, including a subset of those described in a
153 recent IAB report [RFC4690] and the need to update the system to deal
154 with newer versions of Unicode. This document further explains the
155 issues that have been encountered when they are important to
156 understanding of the revised protocols. It also provides an overview
157 of the new IDNA model and explanatory material for it. Additional
158 explanatory material for the specific components of the proposals
159 appears with the associated documents.
161 This document and the associated ones are written from the
162 perspective of an IDNA-aware user, application, or implementation.
163 While they may reiterate fundamental DNS rules and requirements for
164 the convenience of the reader, they make no attempt to be
165 comprehensive about DNS principles and should not be considered as a
166 substitute for a thorough understanding of the DNS protocols and
167 specifications.
169 A good deal of the background material that appeared in RFC 3490
170 [RFC3490] has been removed from this update. That material is either
171 of historical interest only or has been covered from a more recent
172 perspective in RFC 4690 [RFC4690].
174 This document is not normative. The information it provides is
175 intended to make the rules, tables, and protocol easier to understand
176 and to provide overview information and suggestions for zone
177 administrators and others who need to make policy, deployment, and
178 similar decisions about IDNs.
180 1.2. Discussion Forum
182 [[ RFC Editor: please remove this section. ]]
184 IDNA2008 is being discussed in the IETF "idnabis" Working Group and
185 on the mailing list idna-update@alvestrand.no
187 1.3. Terminology
189 Terminology that is critical for understanding this document and the
190 rest of the documents that make up IDNA2008, appears in
191 [IDNA2008-Defs]. That document also contains roadmap to the IDNA2008
192 document collection. No attempt should be made to understand this
193 document without the definitions and concepts that appear there.
195 1.3.1. Documents and Standards
197 This document uses the term "IDNA2003" to refer to the set of
198 standards that make up and support the version of IDNA published in
199 2003, i.e., those commonly known as the IDNA base specification
200 [RFC3490], Nameprep [RFC3491], Punycode [RFC3492], and Stringprep
201 [RFC3454]. In this document, those names are used to refer,
202 conceptually, to the individual documents, with the base IDNA
203 specification called just "IDNA".
205 The term "IDNA2008" is used to refer to a new version of IDNA as
206 described in this document and in the documents described in the
207 document listing of [IDNA2008-Defs]. IDNA2008 is not dependent on
208 any of the IDNA2003 specifications other than the one for Punycode
209 encoding. References to "these specifications" or "these documents"
210 are to the entire IDNA2008 set.
212 1.3.2. DNS "Name" Terminology
214 These documents depart from historical DNS terminology and usage in
215 one important respect. Over the years, the community has talked very
216 casually about "names" in the DNS, beginning with calling it "the
217 domain name system". That terminology is fine in the very precise
218 sense that the identifiers of the DNS do provide names for objects
219 and addresses. But, in the context of IDNs, the term has introduced
220 some confusion, confusion that has increased further as people have
221 begun to speak of DNS labels in terms of the words or phrases of
222 various natural languages.
224 Historically, many, perhaps most, of the "names" in the DNS have been
225 mnemonics to identify some particular concept, object, or
226 organization. They are typically derived from, or rooted in, some
227 language because most people think in language-based ways. But,
228 because they are mnemonics, they need not obey the orthographic
229 conventions of any language: it is not a requirement that it be
230 possible for them to be "words".
232 This distinction is important because the reasonable goal of an IDN
233 effort is not to be able to write the great Klingon (or language of
234 one's choice) novel in DNS labels but to be able to form a usefully
235 broad range of mnemonics in ways that are as natural as possible in a
236 very broad range of scripts.
238 1.3.3. New Terminology and Restrictions
240 These documents introduce new terminology, and precise definitions,
241 for the terms "U-label", "A-Label", LDH-label (to which all valid
242 pre-IDNA host names conformed), Reserved-LDH-label (R-LDH-label), XN-
243 label, Fake-A-Label, and Non-Reserved-LDH-label (NR-LDH-label).
245 In addition, the term "putative label" has been adopted to refer to a
246 label that may appear to meet certain definitional constraints but
247 has not yet been sufficiently tested for validity.
249 These definitions are illustrated in Figure 1 of the Definitions
250 Document [IDNA2008-Defs]. R-LDH-labels contain "--" in the third and
251 fourth character from the beginning of the label. In IDNA-aware
252 applications, only a subset of these reserved labels is permitted to
253 be used, namely the A-label subset. A-labels are a subset of the
254 R-LDH-labels that begin with the case-insensitive string "xn--".
255 Labels that bear this prefix but which are not otherwise valid fall
256 into the "Fake-A-label" category. The non-reserved labels (NR-LDH-
257 labels) are implicitly valid since they do not trigger any
258 resemblance to IDNA-landr NR-LDH-labels.
260 The creation of the Reserved-LDH category is required for three
261 reasons:
263 o to prevent confusion with pre-IDNA coding forms;
265 o to permit future extensions that would require changing the
266 prefix, no matter how unlikely those might be (see Section 7.4);
267 and
269 o to reduce the opportunities for attacks via the Punycode encoding
270 algorithm itself.
272 1.4. Objectives
274 The intent of the IDNA revision effort, and hence of this document
275 and the associated ones, is to increase the usability and
276 effectiveness of internationalized domain names (IDNs) while
277 preserving or strengthening the integrity of references that use
278 them. The original "hostname" character definitions (see, e.g.,
279 [RFC0810]) struck a balance between the creation of useful mnemonics
280 and the introduction of parsing problems or general confusion in the
281 contexts in which domain names are used. The objective of IDNA2008
282 is to preserve that balance while expanding the character repertoire
283 to include extended versions of Roman-derived scripts and scripts
284 that are not Roman in origin. No work of this sort is able to
285 completely eliminate sources of visual or textual confusion: such
286 confusion is possible even under the original host naming rules where
287 only ASCII characters were permitted. However, through the
288 application of different techniques at different points (see
289 Section 3.3), it should be possible to keep problems to an acceptable
290 minimum. One consequence of this general objective is that the
291 desire of some user or marketing community to use a particular string
292 --whether the reason is to try to write sentences of particular
293 languages in the DNS, to express a facsimile of the symbol for a
294 brand, or for some other purpose-- is not a primary goal within the
295 context of applications in the domain name space.
297 1.5. Applicability and Function of IDNA
299 The IDNA specification solves the problem of extending the repertoire
300 of characters that can be used in domain names to include a large
301 subset of the Unicode repertoire.
303 IDNA does not extend the service offered by DNS to the applications.
304 Instead, the applications (and, by implication, the users) continue
305 to see an exact-match lookup service. Either there is a single
306 exactly-matching (subject to the base DNS requirement of case-
307 insensitive ASCII matching) name or there is no match. This model
308 has served the existing applications well, but it requires, with or
309 without internationalized domain names, that users know the exact
310 spelling of the domain names that are to be typed into applications
311 such as web browsers and mail user agents. The introduction of the
312 larger repertoire of characters potentially makes the set of
313 misspellings larger, especially given that in some cases the same
314 appearance, for example on a business card, might visually match
315 several Unicode code points or several sequences of code points.
317 The IDNA standard does not require any applications to conform to it,
318 nor does it retroactively change those applications. An application
319 can elect to use IDNA in order to support IDN while maintaining
320 interoperability with existing infrastructure. If an application
321 wants to use non-ASCII characters in domain names, IDNA is the only
322 currently-defined option. Adding IDNA support to an existing
323 application entails changes to the application only, and leaves room
324 for flexibility in front-end processing and more specifically in the
325 user interface (see Section 6).
327 A great deal of the discussion of IDN solutions has focused on
328 transition issues and how IDNs will work in a world where not all of
329 the components have been updated. Proposals that were not chosen by
330 the original IDN Working Group would have depended on updating of
331 user applications, DNS resolvers, and DNS servers in order for a user
332 to apply an internationalized domain name in any form or coding
333 acceptable under that method. While processing must be performed
334 prior to or after access to the DNS, IDNA requires no changes to the
335 DNS protocol or any DNS servers or the resolvers on user's computers.
337 IDNA allows the graceful introduction of IDNs not only by avoiding
338 upgrades to existing infrastructure (such as DNS servers and mail
339 transport agents), but also by allowing some rudimentary use of IDNs
340 in applications by using the ASCII-encoded representation of the
341 labels containing non-ASCII characters. While such names are user-
342 unfriendly to read and type, and hence not optimal for user input,
343 they can be used as a last resort to allow rudimentary IDN usage.
344 For example, they might be the best choice for display if it were
345 known that relevant fonts were not available on the user's computer.
346 In order to allow user-friendly input and output of the IDNs and
347 acceptance of some characters as equivalent to those to be processed
348 according to the protocol, the applications need to be modified to
349 conform to this specification.
351 This version of IDNA uses the Unicode character repertoire, for
352 continuity with the original version of IDNA.
354 1.6. Comprehensibility of IDNA Mechanisms and Processing
356 One of the major goals of this work is to improve the general
357 understanding of how IDNA works and what characters are permitted and
358 what happens to them. Comprehensibility and predictability to users
359 and registrants are themselves important motivations and design goals
360 for this effort. The effort includes some new terminology and a
361 revised and extended model, both covered in this section, and some
362 more specific protocol, processing, and table modifications. Details
363 of the latter appear in other documents (see [IDNA2008-Defs]).
365 Several issues are inherent in the application of IDNs and, indeed,
366 almost any other system that tries to handle international characters
367 and concepts. They range from the apparently trivial --e.g., one
368 cannot display a character for which one does not have a font
369 available locally-- to the more complex and subtle. Many people have
370 observed that internationalization is just a tool to enable effective
371 localization while permitting some global uniformity. Issues of
372 display, of exactly how various strings and characters are entered,
373 and so on are inherently issues about localization and user interface
374 design.
376 A protocol such as IDNA can only assume that such operations as data
377 entry and reconciliation of differences in character forms are
378 possible. It may make some recommendations about how display might
379 work when characters and fonts are not available, but they can only
380 be general recommendations and, because display functions are rarely
381 controlled by the types of applications that would call upon IDNA,
382 will rarely be very effective.
384 However, shifting responsibility for character mapping and other
385 adjustments from the protocol (where it was located in IDNA2003) to
386 the user interface or processing before invoking IDNA raises issues
387 about both what that processing should do and about compatibility for
388 references prepared in an IDNA2003 context. Those issues are
389 discussed in Section 6.
391 Operations for converting between local character sets and normalized
392 Unicode are part of this general set of user interface issues. The
393 conversion is obviously not required at all in a Unicode-native
394 system that maintains all strings in Normalization Form C (NFC).
395 (See [Unicode-UAX15] for precise definitions of NFC and NFKC if
396 needed.) It may, however, involve some complexity in a system that
397 is not Unicode-native, especially if the elements of the local
398 character set do not map exactly and unambiguously into Unicode
399 characters or do so in a way that is not completely stable over time.
400 Perhaps more important, if a label being converted to a local
401 character set contains Unicode characters that have no correspondence
402 in that character set, the application may have to apply special,
403 locally-appropriate, methods to avoid or reduce loss of information.
405 Depending on the system involved, the major difficulty may not lie in
406 the mapping but in accurately identifying the incoming character set
407 and then applying the correct conversion routine. If a local
408 operating system uses one of the ISO 8859 character sets or an
409 extensive national or industrial system such as GB18030 [GB18030] or
410 BIG5 [BIG5], one must correctly identify the character set in use
411 before converting to Unicode even though those character coding
412 systems are substantially or completely Unicode-compatible (i.e., all
413 of the code points in them have an exact and unique mapping to
414 Unicode code points). It may be even more difficult when the
415 character coding system in local use is based on conceptually
416 different assumptions than those used by Unicode about, e.g., font
417 encodings used for publications in some Indic scripts. Those
418 differences may not easily yield unambiguous conversions or
419 interpretations even if each coding system is internally consistent
420 and adequate to represent the local language and script.
422 2. Processing in IDNA2008
424 These specifications separate Domain Name Registration and Lookup in
425 the protocol specification. Doing so reflects current practice in
426 which per-registry restrictions and special processing are applied at
427 registration time but not during lookup. Even more important in the
428 longer term, it facilitates incremental addition of permitted
429 character groups to avoid freezing on one particular version of
430 Unicode.
432 The actual registration and lookup protocols for IDNA2008 are
433 specified in [IDNA2008-Protocol].
435 3. Permitted Characters: An Inclusion List
437 This section provides an overview of the model used to establish the
438 algorithm and character lists of [IDNA2008-Tables] and describes the
439 names and applicability of the categories used there. Note that the
440 inclusion of a character in the first category group (Section 3.1.1)
441 does not imply that it can be used indiscriminately; some characters
442 are associated with contextual rules that must be applied as well.
444 The information given in this section is provided to make the rules,
445 tables, and protocol easier to understand. The normative generating
446 rules that correspond to this informal discussion appear in
447 [IDNA2008-Tables] and the rules that actually determine what labels
448 can be registered or looked up are in [IDNA2008-Protocol].
450 3.1. A Tiered Model of Permitted Characters and Labels
452 Moving to an inclusion model requires respecifying the list of
453 characters that are permitted in IDNs. In IDNA2003, the role and
454 utility of characters are independent of context and fixed forever
455 (or until the standard is replaced). Making completely context-
456 independent rules globally has proven impractical because some
457 characters, especially those that are called "Join_Controls" in
458 Unicode, are needed to make reasonable use of some scripts but have
459 no visible effect(s) in others. IDNA2003 prohibited those types of
460 characters entirely. But the restrictions led to a consensus that
461 under some conditions, these "joiner" characters were legitimately
462 needed to allow useful mnemonics for some languages and scripts. The
463 requirement to support those characters but limit their use to very
464 specific contexts was reinforced by the observation that handling of
465 particular characters across the languages that use a script, or the
466 use of similar or identical-looking characters in different scripts,
467 is more complex than many people believed it was several years ago.
469 Independently of the characters chosen (see next subsection), the
470 approach is to divide the characters that appear in Unicode into
471 three categories:
473 3.1.1. PROTOCOL-VALID
475 Characters identified as "PROTOCOL-VALID" (often abbreviated
476 "PVALID") are, in general, permitted by IDNA for all uses in IDNs.
477 Their use may be restricted by rules about the context in which they
478 appear or by other rules that apply to the entire label in which they
479 are to be embedded. For example, any label that contains a character
480 in this category that has a "right-to-left" property must be used in
481 context with the "Bidi" rules (see [IDNA2008-Bidi]).
483 The term "PROTOCOL-VALID" is used to stress the fact that the
484 presence of a character in this category does not imply that a given
485 registry need accept registrations containing any of the characters
486 in the category. Registries are still expected to apply judgment
487 about labels they will accept and to maintain rules consistent with
488 those judgments (see [IDNA2008-Protocol] and Section 3.3).
490 Characters that are placed in the "PROTOCOL-VALID" category are
491 expected to never be removed from it or reclassified. While
492 theoretically characters could be removed from Unicode, such removal
493 would be inconsistent with the Unicode stability principles (see
494 [Unicode51], Appendix F) and hence should never occur.
496 3.1.2. Characters Valid Only in Context With Others
498 Some characters may be unsuitable for general use in IDNs but
499 necessary for the plausible support of some scripts. The two most
500 commonly-cited examples are the zero-width joiner and non-joiner
501 characters (ZWJ, U+200D and ZWNJ, U+200C), but provisions for
502 unambiguous labels may require that other characters be restricted to
503 particular contexts. For example, the ASCII hyphen is not permitted
504 to start or end a label, whether that label contains non-ASCII
505 characters or not.
507 3.1.2.1. Contextual Restrictions
509 These characters must not appear in IDNs without additional
510 restrictions, typically because they have no visible consequences in
511 most scripts but affect format or presentation in a few others or
512 because they are combining characters that are safe for use only in
513 conjunction with particular characters or scripts. In order to
514 permit them to be used at all, they are specially identified as
515 "CONTEXTUAL RULE REQUIRED" and, when adequately understood,
516 associated with a rule. In addition, the rule will define whether it
517 is to be applied on lookup as well as registration. A distinction is
518 made between characters that indicate or prohibit joining (known as
519 "CONTEXT-JOINER" or "CONTEXTJ") and other characters requiring
520 contextual treatment ("CONTEXT-OTHER" or "CONTEXTO"). Only the
521 former require full testing at lookup time.
523 It is important to note that these contextual rules cannot prevent
524 all uses of the relevant characters that might be confusing or
525 problematic. What they are expected do is to confine applicability
526 of the characters to scripts (and narrower contexts) where zone
527 administrators are knowledgeable enough about the use of those
528 characters to be prepared to deal with them appropriately. For
529 example, a registry dealing with an Indic script that requires ZWJ
530 and/or ZWNJ as part of the writing system is expected to understand
531 where the characters have visible effect and where they do not and to
532 make registration rules accordingly. By contrast, a registry dealing
533 with Latin or Cyrillic script might not be actively aware that the
534 characters exist, much less about the consequences of embedding them
535 in labels drawn from those scripts.
537 3.1.2.2. Rules and Their Application
539 The actual rules may be DEFINED or NULL. If present, they may have
540 values of "True" (character may be used in any position in any
541 label), "False" (character may not be used in any label), or may be a
542 set of procedural rules that specify the context in which the
543 character is permitted.
545 Examples of descriptions of typical rules, stated informally and in
546 English, include "Must follow a character from Script XYZ", "Must
547 occur only if the entire label is in Script ABC", "Must occur only if
548 the previous and subsequent characters have the DFG property".
550 Because it is easier to identify these characters than to know that
551 they are actually needed in IDNs or how to establish exactly the
552 right rules for each one, a rule may have a null value in a given
553 version of the tables. Characters associated with null rules are not
554 permitted to appear in putative labels for either registration or
555 lookup. Of course, a later version of the tables might contain a
556 non-null rule.
558 The description of the syntax of the rules, and the rules themselves,
559 appears in [IDNA2008-Tables]. [[anchor11: ??? Section number would
560 be good here.]]
562 3.1.3. DISALLOWED
564 Some characters are inappropriate for use in IDNs and are thus
565 excluded for both registration and lookup (i.e., IDNA-conforming
566 applications performing name lookup should verify that these
567 characters are absent; if they are present, the label strings should
568 be rejected rather than converted to A-labels and looked up. Some of
569 these characters are problematic for use in IDNs (such as the
570 FRACTION SLASH character, U+2044), while some of them (such as the
571 various HEART symbols, e.g., U+2665, U+2661, and U+2765, see
572 Section 7.6) simply fall outside the conventions for typical
573 identifiers (basically letters and numbers).
575 Of course, this category would include code points that had been
576 removed entirely from Unicode should such removals ever occur.
578 Characters that are placed in the "DISALLOWED" category are expected
579 to never be removed from it or reclassified. If a character is
580 classified as "DISALLOWED" in error and the error is sufficiently
581 problematic, the only recourse would be either to introduce a new
582 code point into Unicode and classify it as "PROTOCOL-VALID" or for
583 the IETF to accept the considerable costs of an incompatible change
584 and replace the relevant RFC with one containing appropriate
585 exceptions.
587 There is provision for exception cases but, in general, characters
588 are placed into "DISALLOWED" if they fall into one or more of the
589 following groups:
591 o The character is a compatibility equivalent for another character.
592 In slightly more precise Unicode terms, application of
593 normalization method NFKC to the character yields some other
594 character.
596 o The character is an upper-case form or some other form that is
597 mapped to another character by Unicode casefolding.
599 o The character is a symbol or punctuation form or, more generally,
600 something that is not a letter, digit, or a mark that is used to
601 form a letter or digit.
603 3.1.4. UNASSIGNED
605 For convenience in processing and table-building, code points that do
606 not have assigned values in a given version of Unicode are treated as
607 belonging to a special UNASSIGNED category. Such code points are
608 prohibited in labels to be registered or looked up. The category
609 differs from DISALLOWED in that code points are moved out of it by
610 the simple expedient of being assigned in a later version of Unicode
611 (at which point, they are classified into one of the other categories
612 as appropriate).
614 The rationale for restricting the processing of UNASSIGNED characters
615 is simply that if such characters were permitted to be looked up, for
616 example, and were later assigned, but subject to some set of
617 contextual rules, un-updated instances of IDNA-aware software might
618 permit lookup of labels containing the previously-unassigned
619 characters while updated versions of IDNA-aware software might
620 restrict their use in lookup, depending on the contextual rules. It
621 should be clear that under no circumstance should an UNASSIGNED
622 character be permitted in a label to be registered as part of a
623 domain name.
625 3.2. Registration Policy
627 While these recommendations cannot and should not define registry
628 policies, registries should develop and apply additional restrictions
629 as needed to reduce confusion and other problems. For example, it is
630 generally believed that labels containing characters from more than
631 one script are a bad practice although there may be some important
632 exceptions to that principle. Some registries may choose to restrict
633 registrations to characters drawn from a very small number of
634 scripts. For many scripts, the use of variant techniques such as
635 those as described in RFC 3843 [RFC3743] and RFC 4290 [RFC4290], and
636 illustrated for Chinese by the tables described in RFC 4713 [RFC4713]
637 may be helpful in reducing problems that might be perceived by users.
639 In general, users will benefit if registries only permit characters
640 from scripts that are well-understood by the registry or its
641 advisers. If a registry decides to reduce opportunities for
642 confusion by constructing policies that disallow characters used in
643 historic writing systems or characters whose use is restricted to
644 specialized, highly technical contexts, some relevant information may
645 be found in Section 2.4 "Specific Character Adjustments", Table 4
646 "Candidate Characters for Exclusion from Identifiers" of
647 [Unicode-UAX31] and Section 3.1. "General Security Profile for
648 Identifiers" in [Unicode-Security].
650 It is worth stressing that these principles of policy development and
651 application apply at all levels of the DNS, not only, e.g., TLD or
652 SLD registrations and that even a trivial, "anything permitted that
653 is valid under the protocol" policy is helpful in that it helps users
654 and application developers know what to expect.
656 3.3. Layered Restrictions: Tables, Context, Registration, Applications
658 The essence of the character rules in IDNA2008 is based on the
659 realization that there is no single magic bullet for any of the
660 issues associated with a multiscript DNS. Instead, the
661 specifications define a variety of approaches that, together,
662 constitute multiple lines of defense against ambiguity in identifiers
663 and loss of referential integrity. The actual character tables are
664 the first mechanism, protocol rules about how those characters are
665 applied or restricted in context are the second, and those two in
666 combination constitute the limits of what can be done by a protocol
667 alone. As discussed in the previous section (Section 3.2),
668 registries are expected to restrict what they permit to be
669 registered, devising and using rules that are designed to optimize
670 the balance between confusion and risk on the one hand and maximum
671 expressiveness in mnemonics on the other.
673 In addition, there is an important role for user agents in warning
674 against label forms that appear problematic given their knowledge of
675 local contexts and conventions. Of course, no approach based on
676 naming or identifiers alone can protect against all threats.
678 4. Issues that Constrain Possible Solutions
680 4.1. Display and Network Order
682 The correct treatment of domain names requires a clear distinction
683 between Network Order (the order in which the code points are sent in
684 protocols) and Display Order (the order in which the code points are
685 displayed on a screen or paper). The order of labels in a domain
686 name that contains characters that are normally written right to left
687 is discussed in [IDNA2008-Bidi]. In particular, there are questions
688 about the order in which labels are displayed if left to right and
689 right to left labels are adjacent to each other, especially if there
690 are also multiple consecutive appearances of one of the types. The
691 decision about the display order is ultimately under the control of
692 user agents --including web browsers, mail clients, and the like--
693 which may be highly localized. Even when formats are specified by
694 protocols, the full composition of an Internationalized Resource
695 Identifier (IRI) [RFC3987] or Internationalized Email address
696 contains elements other than the domain name. For example, IRIs
697 contain protocol identifiers and field delimiter syntax such as
698 "http://" or "mailto:" while email addresses contain the "@" to
699 separate local parts from domain names. User agents are not required
700 to use those protocol-based forms directly but often do so. While
701 display, parsing, and processing within a label is specified by the
702 normative documents in the IDNA2008 collection, the relationship
703 between fully-qualified domain names and internationalized labels is
704 unchanged from the base DNS specifications. Comments in this
705 document about such full domain names are explanatory or examples of
706 what might be done and must not be considered normative.
708 Questions remain about protocol constraints implying that the overall
709 direction of these strings will always be left to right (or right to
710 left) for an IRI or email address, or if they even should conform to
711 such rules. These questions also have several possible answers.
713 Should a domain name abc.def, in which both labels are represented in
714 scripts that are written right to left, be displayed as fed.cba or
715 cba.fed? An IRI for clear text web access would, in network order,
716 begin with "http://" and the characters will appear as
717 "http://abc.def" -- but what does this suggest about the display
718 order? When entering a URI to many browsers, it may be possible to
719 provide only the domain name and leave the "http://" to be filled in
720 by default, assuming no tail (an approach that does not work for
721 protocols other than HTTP or whatever is chosen as the default). The
722 natural display order for the typed domain name on a right to left
723 system is fed.cba. Does this change if a protocol identifier, tail,
724 and the corresponding delimiters are specified?
726 While logic, precedent, and reality suggest that these are questions
727 for user interface design, not IETF protocol specifications,
728 experience in the 1980s and 1990s with mixing systems in which domain
729 name labels were read in network order (left to right) and those in
730 which those labels were read right to left would predict a great deal
731 of confusion, and heuristics that sometimes fail, if each
732 implementation of each application makes its own decisions on these
733 issues.
735 Any version of IDNA, including the current one, must be written in
736 terms of the network (transmission on the wire) order of characters
737 in labels and for the labels in complete (fully-qualified) domain
738 names and must be quite precise about those relationships. While
739 some strong suggestions about display order would be desirable to
740 reduce the chances for inconsistent transcription of domain names
741 from printed form, such suggestions are beyond the scope of these
742 specifications.
744 4.2. Entry and Display in Applications
746 Applications can accept domain names using any character set or sets
747 desired by the application developer, specified by the operating
748 system, or dictated by other constraints, and can display domain
749 names in any character set or character coding system. That is, the
750 IDNA protocol does not affect the interface between users and
751 applications.
753 An IDNA-aware application can accept and display internationalized
754 domain names in two formats: the internationalized character set(s)
755 supported by the application (i.e., an appropriate local
756 representation of a U-label), and as an A-label. Applications may
757 allow the display of A-labels, but are encouraged to not do so except
758 as an interface for special purposes, possibly for debugging, or to
759 cope with display limitations. In general, they should allow, but
760 not encourage, user input of that label form. A-labels are opaque
761 and ugly and malicious variations on them are not easily detected by
762 users. Where possible, they should thus only be exposed to users and
763 in contexts in which they are absolutely needed. Because IDN labels
764 can be rendered either as A-labels or U-labels, the application may
765 reasonably have an option for the user to select the preferred method
766 of display; if it does, rendering the U-label should normally be the
767 default.
769 Domain names are often stored and transported in many places. For
770 example, they are part of documents such as mail messages and web
771 pages. They are transported in many parts of many protocols, such as
772 both the control commands of SMTP and associated the message body
773 parts, and in the headers and the body content in HTTP. It is
774 important to remember that domain names appear both in domain name
775 slots and in the content that is passed over protocols.
777 In protocols and document formats that define how to handle
778 specification or negotiation of charsets, labels can be encoded in
779 any charset allowed by the protocol or document format. If a
780 protocol or document format only allows one charset, the labels must
781 be given in that charset. Of course, not all charsets can properly
782 represent all labels. If a U-label cannot be displayed in its
783 entirety, the only choice (without loss of information) may be to
784 display the A-label.
786 In any place where a protocol or document format allows transmission
787 of the characters in internationalized labels, labels should be
788 transmitted using whatever character encoding and escape mechanism
789 the protocol or document format uses at that place. This provision
790 is intended to prevent situations in which, e.g., UTF-8 domain names
791 appear embedded in text that is otherwise in some other character
792 coding.
794 All protocols that use domain name slots (See Section 2.3.1.6
795 [[anchor14: ?? Verify this]] in [IDNA2008-Defs]) already have the
796 capacity for handling domain names in the ASCII charset. Thus,
797 A-labels can inherently be handled by those protocols.
799 4.3. Linguistic Expectations: Ligatures, Digraphs, and Alternate
800 Character Forms
802 [[anchor15: There is some internal redundancy and repetition in the
803 material in this section. Specific suggestions about to reduce or
804 eliminate redundant text would be appreciated. If no such
805 suggestions are received before -07 is posted, this note will be
806 removed.]]
808 Users often have expectations about character matching or equivalence
809 that are based on their own languages and the orthography of those
810 languages. These expectations may not be consistent with forms or
811 actions that can be naturally accommodated in a character coding
812 system, especially if multiple languages are written using the same
813 script but using different conventions. A Norwegian user might
814 expect a label with the ae-ligature to be treated as the same label
815 as one using the Swedish spelling with a-diaeresis even though
816 applying that mapping to English would be astonishing to users. A
817 user in German might expect a label with an o-umlaut and a label that
818 had "oe" substituted, but was otherwise the same, treated as
819 equivalent even though that substitution would be a clear error in
820 Swedish. A Chinese user might expect automatic matching of
821 Simplified and Traditional Chinese characters, but applying that
822 matching for Korean or Japanese text would create considerable
823 confusion. For that matter, an English user might expect "theater"
824 and "theatre" to match.
826 Related issues arise because there are a number of languages written
827 with alphabetic scripts in which single phonemes are written using
828 two characters, termed a "digraph", for example, the "ph" in
829 "pharmacy" and "telephone". (Note that characters paired in this
830 manner can also appear consecutively without forming a digraph, as in
831 "tophat".) Certain digraphs are normally indicated typographically
832 by setting the two characters closer together than they would be if
833 used consecutively to represent different phonemes. Some digraphs
834 are fully joined as ligatures (strictly designating setting totally
835 without intervening white space, although the term is sometimes
836 applied to close set pairs). An example of this may be seen when the
837 word "encyclopaedia" is set with a U+00E6 LATIN SMALL LIGATURE AE
838 (and some would not consider that word correctly spelled unless the
839 ligature form was used or the "a" was dropped entirely). When these
840 ligature and digraph forms have the same interpretation across all
841 languages that use a given script, application of Unicode
842 normalization generally resolves the differences and causes them to
843 match. When they have different interpretations, any requirements
844 for matching must utilize other methods, presumably at the registry
845 level, or users must be educated to understand that matching will not
846 occur.
848 Difficulties arise from the fact that a given ligature may be a
849 completely optional typographic convenience for representing a
850 digraph in one language (as in the above example with some spelling
851 conventions), while in another language it is a single character that
852 may not always be correctly representable by a two-letter sequence
853 (as in the above example with different spelling conventions). This
854 can be illustrated by many words in the Norwegian language, where the
855 "ae" ligature is the 27th letter of a 29-letter extended Latin
856 alphabet. It is equivalent to the 28th letter of the Swedish
857 alphabet (also containing 29 letters), U+00E4 LATIN SMALL LETTER A
858 WITH DIAERESIS, for which an "ae" cannot be substituted according to
859 current orthographic standards.
861 That character (U+00E4) is also part of the German alphabet where,
862 unlike in the Nordic languages, the two-character sequence "ae" is
863 usually treated as a fully acceptable alternate orthography for the
864 "umlauted a" character. The inverse is however not true, and those
865 two characters cannot necessarily be combined into an "umlauted a".
866 This also applies to another German character, the "umlauted o"
867 (U+00F6 LATIN SMALL LETTER O WITH DIAERESIS) which, for example,
868 cannot be used for writing the name of the author "Goethe". It is
869 also a letter in the Swedish alphabet where, like the "a with
870 diaeresis", it cannot be correctly represented as "oe" and in the
871 Norwegian alphabet, where it is represented, not as "o with
872 diaeresis", but as "slashed o", U+00F8.
874 Some of the ligatures that have explicit code points in Unicode were
875 given special handling in IDNA2003 and now pose additional problems
876 in transition. See Section 7.2.
878 Additional cases with alphabets written right to left are described
879 in Section 4.5.
881 Whether ligatures and digraphs are to be treated as a sequence of
882 characters or as a single standalone one constitute a problem that
883 cannot be resolved solely by operating on scripts. They are,
884 however, a key concern in the IDN context. Their satisfactory
885 resolution will require support in policies set by registries, which
886 therefore need to be particularly mindful not just of this specific
887 issue, but of all other related matters that cannot be dealt with on
888 an exclusively algorithmic and global basis.
890 Just as with the examples of different-looking characters that may be
891 assumed to be the same, it is in general impossible to deal with
892 these situations in a system such as IDNA -- or with Unicode
893 normalization generally -- since determining what to do requires
894 information about the language being used, context, or both.
895 Consequently, these specifications make no attempt to treat these
896 combined characters in any special way. However, their existence
897 provides a prime example of a situation in which a registry that is
898 aware of the language context in which labels are to be registered,
899 and where that language sometimes (or always) treats the two-
900 character sequences as equivalent to the combined form, should give
901 serious consideration to applying a "variant" model [RFC3743]
902 [RFC4290], or to prohibiting registration of one the forms entirely,
903 to reduce the opportunities for user confusion and fraud that would
904 result from the related strings being registered to different
905 parties.
907 [[anchor16: Placeholder: A discussion of the Arabic digit issue
908 should go here once it is resolved in some appropriate way.]]
910 4.4. Case Mapping and Related Issues
912 In the DNS, ASCII letters are stored with their case preserved.
913 Matching during the query process is case-independent, but none of
914 the information that might be represented by choices of case has been
915 lost. That model has been accidentally helpful because, as people
916 have created DNS labels by catenating words (or parts of words) to
917 form labels, case has often been used to distinguish among components
918 and make the labels more memorable.
920 The solution of keeping the characters separate but doing matching
921 independent of case is not feasible with IDNA or any IDNA-like model
922 because the matching would then have to be done on the server rather
923 than have characters mapped on the client. That situation was
924 recognized in IDNA2003 and nothing in these specifications
925 fundamentally changes it or could do so. In IDNA2003, all characters
926 are case-folded and mapped. That results in upper-case characters
927 being mapped to lower-case ones and in some other transformations of
928 alternate forms of characters, especially those that do not have (or
929 did not have) upper-case forms. For example, Greek Final Form Sigma
930 (U+03C2) is mapped to the medial form (U+03C3) and Eszett (German
931 Sharp S, U+00DF) is mapped to "ss". Neither of these mappings is
932 reversible because the upper case of U+03C3 is the Upper Case Sigma
933 (U+03A3) and "ss" is an ASCII string. IDNA2008 permits, at the risk
934 of some incompatibility, slightly more flexibility in this area by
935 avoid case folding and treating these characters as themselves.
936 Approaches to handling that incompatibility are discussed in
937 Section 7.2. Although information is lost in IDNA2003's ToASCII
938 operation so that, in some sense, neither Final Sigma nor Eszett can
939 be represented in an IDN at all, its guarantee of mapping when those
940 characters are used as input can be interpreted as violating one of
941 the conditions discussed in Section 7.4.1 and hence requiring a
942 prefix change. The consensus was to not make a prefix change in
943 spite of this issue. Of course, had a prefix change been made (at
944 the costs discussed in Section 7.4.3) there would have been several
945 options, including, if desired, assignment of the character to the
946 CONTEXTUAL RULE REQUIRED category and requiring that it only be used
947 in carefully-selected contexts.
949 4.5. Right to Left Text
951 In order to be sure that the directionality of right to left text is
952 unambiguous, IDNA2003 required that any label in which right to left
953 characters appear both starts and ends with them, not include any
954 characters with strong left to right properties (which excludes other
955 alphabetic characters but permits European digits), and rejects any
956 other string that contains a right to left character. This is one of
957 the few places where the IDNA algorithms (both in IDNA2003 and in
958 IDAN2008) are required to examine an entire label, not just
959 individual characters. The algorithmic model used in IDNA2003
960 rejects the label when the final character in a right to left string
961 requires a combining mark in order to be correctly represented.
963 That prohibition is not acceptable for writing systems for languages
964 written with consonantal alphabets to which diacritical vocalic
965 systems are applied, and for languages with orthographies derived
966 from them where the combining marks may have different functionality.
967 In both cases the combining marks can be essential components of the
968 orthography. Examples of this are Yiddish, written with an extended
969 Hebrew script, and Dhivehi (the official language of Maldives) which
970 is written in the Thaana script (which is, in turn, derived from the
971 Arabic script). IDNA2008 removes the restriction on final combining
972 characters with a new set of rules for right to left scripts and
973 their characters. Those new rules are specified in [IDNA2008-Bidi].
975 5. IDNs and the Robustness Principle
977 The model of IDNs described in this document can be seen as a
978 particular instance of the "Robustness Principle" that has been so
979 important to other aspects of Internet protocol design. This
980 principle is often stated as "Be conservative about what you send and
981 liberal in what you accept" (See, e.g., Section 1.2.2 of the
982 applications-layer Host Requirements specification [RFC1123]). For
983 IDNs to work well, not only must the protocol be carefully designed
984 and implemented, but zone administrators (registries) must have and
985 require sensible policies about what is registered -- conservative
986 policies -- and implement and enforce them.
988 Conversely, lookup applications are expected to reject labels that
989 clearly violate global (protocol) rules (no one has ever seriously
990 claimed that being liberal in what is accepted requires being
991 stupid). However, once one gets past such global rules and deals
992 with anything sensitive to script or locale, it is necessary to
993 assume that garbage has not been placed into the DNS, i.e., one must
994 be liberal about what one is willing to look up in the DNS rather
995 than guessing about whether it should have been permitted to be
996 registered.
998 As mentioned elsewhere, if a string cannot be successfully found in
999 the DNS after the lookup processing described here, it makes no
1000 difference whether it simply wasn't registered or was prohibited by
1001 some rule at the registry. Applications should, however, be
1002 sensitive to the fact that, because of the possibility of DNS
1003 wildcards, the ability to successfully resolve a name does not
1004 guarantee that it was actually registered.
1006 If lookup applications, as a user interface (UI) or other local
1007 matter, decide to warn about some strings that are valid under the
1008 global rules but that they perceive as dangerous, that is their
1009 prerogative and we can only hope that the market (and maybe
1010 regulators) will reinforce the good choices and discourage the poor
1011 ones. In this context, a lookup application that decides a string
1012 that is valid under the protocol is dangerous and refuses to look it
1013 up is in violation of the protocols; one that is willing to look
1014 something up, but warns against it, is exercising a local choice.
1016 6. Front-end and User Interface Processing for Lookup
1018 Domain names may be identified and processed in many contexts. They
1019 may be typed in by users either by themselves or embedded in an
1020 identifier structured for a particular protocol or class of protocols
1021 such a email addresses, URIs, or IRIs. They may occur in running
1022 text or be processed by one system after being provided in another.
1023 Systems may wish to try to normalize URLs so as to determine (or
1024 guess) whether a reference is valid or two references point to the
1025 same object without actually looking the objects up and comparing
1026 them (that is necessary, not just a choice, for URI types that are
1027 not intended to be resolved). Some of these goals may be more easily
1028 and reliably satisfied than others. While there are strong arguments
1029 for any domain name that is placed "on the wire" -- transmitted
1030 between systems -- to be in the zero-ambiguity forms of A-labels, it
1031 is inevitable that programs that process domain names will encounter
1032 U-labels or variant forms.
1034 One source of such forms will be labels created under IDNA2003
1035 because that protocol allowed labels that were transformed from
1036 native-character format by mapping some characters into others before
1037 conversion into ACE ("xn--...") format. One consequence of the
1038 transformations was that, when the ToUnicode and ToASCII operations
1039 of IDNA2003 were applied, ToUnicode(ToASCII(original-label)) often
1040 did not produce the original label. IDNA2008 explicitly defines
1041 A-labels and U-labels as different forms of the same abstract label,
1042 forms that are stable when conversions are performed between them
1043 (without mappings). A different way of explaining this is that there
1044 are, today, domain names in files on the Internet that use characters
1045 that cannot be represented directly in, or recovered from, (A-label)
1046 domain names but for which interpretations are provided by IDNA2003.
1048 There are two major categories of such characters, those that are
1049 removed by NFKC normalization and those upper-case characters that
1050 are mapped to lower-case (there are also a few characters that are
1051 given special-case mapping treatment in Stringprep, including lower-
1052 case characters that are case-folded into other lower-case characters
1053 or strings).
1055 Other issues in domain name identification and processing arise
1056 because IDNA2003 specified that several other characters be treated
1057 as equivalent to the ASCII period (dot, full stop) character used as
1058 a label separator. If a string that might be a domain name appears
1059 in an arbitrary context (such as running text), it is difficult, even
1060 with only ASCII characters, to know whether an actual domain name (or
1061 a protocol parameter like a URI) is present and where it starts and
1062 ends. When using Unicode, this gets even more difficult if treatment
1063 of certain special characters (like the dot that separates labels in
1064 a domain name) depends on context (e.g., prior knowledge of whether
1065 the string represents a domain name or not). That knowledge is not
1066 available if the primary heuristic for identifying the presence of
1067 domain names in strings depends on the presence of dots separating
1068 groups of characters with no intervening spaces.
1070 As discussed elsewhere in this document, the IDNA2008 model removes
1071 all of these mappings and interpretations, including the equivalence
1072 of different forms of dots, from the protocol, discouraging such
1073 mappings and leaving them, when necessary, to local processing. This
1074 should not be taken to imply that local processing is optional or can
1075 be avoided entirely, even if doing so might have been desirable in a
1076 world without IDNA2003 IDNs in files and archives. Instead, unless
1077 the program context is such that it is known that any IDNs that
1078 appear will contain either U-label or A-label forms, or that other
1079 forms can safely be rejected, some local processing of apparent
1080 domain name strings will be required, both to maintain compatibility
1081 with IDNA2003 and to prevent user astonishment. Such local
1082 processing, while not specified in this document or the associated
1083 ones, will generally take one of two forms:
1085 o Generic Preprocessing.
1086 When the context in which the program or system that processes
1087 domain names operates is global, a reasonable balance must be
1088 found that is sensitive to the broad range of local needs and
1089 assumptions while, at the same time, not sacrificing the needs of
1090 one language, script, or user population to those of another.
1092 For this case, the best practice will usually be to apply NFKC and
1093 case-mapping (or, perhaps better yet, Stringprep itself), plus
1094 dot-mapping where appropriate, to the domain name string prior to
1095 applying IDNA. That practice will not only yield a reasonable
1096 compromise of user experience with protocol requirements but will
1097 be almost completely compatible with the various forms permitted
1098 by IDNA2003.
1100 o Highly Localized Preprocessing.
1101 Unlike the case above, there will be some situations in which
1102 software will be highly localized for a particular environment and
1103 carefully adapted to the expectations of users in that
1104 environment. The many discussions about using the Internet to
1105 preserve and support local cultures suggest that these cases may
1106 be more common in the future than they have been so far.
1108 In these cases, we should avoid trying to tell implementers what
1109 they should accept, if only because they are quite likely (and for
1110 good reason) to ignore us. We would assume that they would map
1111 characters that the intuitions of their users would suggest be
1112 mapped and would hope that they would do that mapping as early as
1113 possible, storing A-label or U-label forms in files and
1114 transporting only those forms between systems. One can imagine
1115 switches about whether some sorts of mappings occur, warnings
1116 before applying them or, in a slightly more extreme version of the
1117 approach taken in Internet Explorer version 7 (IE7), systems that
1118 utterly refuse to handle "strange" characters at all if they
1119 appear in U-label form. None of those local decisions are a
1120 threat to interoperability as long as (i) only U-labels and
1121 A-labels are used in interchange with systems outside the local
1122 environment, (ii) no character that would be valid in a U-label as
1123 itself is mapped to something else, (iii) any local mappings are
1124 applied as a preprocessing step (or, for conversions from U-labels
1125 or A-labels to presentation forms, postprocessing), not as part of
1126 IDNA processing proper, and (iv) appropriate consideration is
1127 given to labels that might have entered the environment in
1128 conformance to IDNA2003.
1130 In either case, it is vital that user interface designs and, where
1131 the interfaces are not sufficient, users, be aware that the only
1132 forms of domain names that this protocol anticipates will resolve
1133 globally or compare equal when crude methods (i.e., those not
1134 conforming to the strict definition of label equivalence given in
1135 [IDNA2008-Defs]) are used are those in which all native-script labels
1136 are in U-label form. Forms that assume mapping will occur,
1137 especially forms that were not valid under IDNA2003, may or may not
1138 function in predictable ways across all implementations.
1140 User interfaces involving Latin-based scripts should take special
1141 care when considering how to handle case mapping because small
1142 differences in label strings may cause behavior that is astonishing
1143 to users. Because case-insensitive comparison is done for ASCII
1144 strings by DNS-servers, an all-ASCII label is treated as case-
1145 insensitive. However, if even one of the characters of that string
1146 is replaced by one that requires the label to be given IDN treatment
1147 (e.g., by adding a diacritical mark), then the label effectively
1148 becomes case-sensitive because only lower-case characters are
1149 permitted in IDNs. This suggests that case mapping for Latin-based
1150 scripts (and possibly other scripts with case distinctions) as a
1151 preprocessing matter in applications may be wise to prevent user
1152 astonishment, but, since all applications may not do this and
1153 ambiguity in transport is not desirable, the that case-dependent
1154 forms should not be stored in files.
1156 The comments above apply only in operations that look up names or
1157 interpret files. There are several reasons why registration
1158 activities should require final names and verification of those names
1159 by the would-be registrant.
1161 7. Migration from IDNA2003 and Unicode Version Synchronization
1163 7.1. Design Criteria
1165 As mentioned above and in RFC 4690, two key goals of the IDNA2008
1166 design are to enable applications to be agnostic about whether they
1167 are being run in environments supporting any Unicode version from 3.2
1168 onward and to permit incrementally adding new characters, character
1169 groups, scripts, and other character collections as they are
1170 incorporated into Unicode, without disruption and, in the long term,
1171 without "heavy" processes such as those involving IETF consensus.
1172 (An IETF consensus process is required by the IDNA2008 specifications
1173 and is expected to be required and used until significant experience
1174 accumulates with IDNA operations and new versions of Unicode.) The
1175 mechanisms that support this are outlined above and elsewhere in the
1176 IDNA2008 document set, but this section reviews them in a context
1177 that may be more helpful to those who need to understand the approach
1178 and make plans for it.
1180 7.1.1. General IDNA Validity Criteria
1182 The general criteria for a putative label, and the collection of
1183 characters that make it up, to be considered IDNA-valid are (the
1184 actual rules are rigorously defined in the "Protocol" and "Tables"
1185 documents):
1187 o The characters are "letters", marks needed to form letters,
1188 numerals, or other code points used to write words in some
1189 language. Symbols, drawing characters, and various notational
1190 characters are intended to be permanently excluded -- some because
1191 they are harmful in URI, IRI, or similar contexts (e.g.,
1192 characters that appear to be slashes or other reserved URI
1193 punctuation) and others because there is no evidence that they are
1194 important enough to Internet operations or internationalization to
1195 justify expansion of domain names beyond the general principle of
1196 "letters, digits, and hyphen" and the complexities that would come
1197 with it (additional discussion and rationale for the symbol
1198 decision appears in Section 7.6).
1200 o Other than in very exceptional cases, e.g., where they are needed
1201 to write substantially any word of a given language, punctuation
1202 characters are excluded as well. The fact that a word exists is
1203 not proof that it should be usable in a DNS label and DNS labels
1204 are not expected to be usable for multiple-word phrases (although
1205 they are certainly not prohibited if the conventions and
1206 orthography of a particular language cause that to be possible).
1207 Even for English, very common constructions -- contractions like
1208 "don't" or "it's", names that are written with apostrophes such as
1209 "O'Reilly", or characters for which apostrophes are common
1210 substitutes cannot be represented in DNS labels. Words in English
1211 whose usually-preferred spellings include diacritical marks cannot
1212 be represented under the original hostname rules, but most can be
1213 represented if treated as IDNs.
1215 o Characters that are unassigned (have no character assignment at
1216 all) in the version of Unicode being used by the registry or
1217 application are not permitted, even on lookup. The issues
1218 involved in this decision are discussed in Section 7.7.
1220 o Any character that is mapped to another character by a current
1221 version of NFKC is prohibited as input to IDNA (for either
1222 registration or lookup). With a few exceptions, this principle
1223 excludes any character mapped to another by Nameprep [RFC3491].
1225 Tables used to identify the characters that are IDNA-valid are
1226 expected to be driven by the principles above, principles that are
1227 specified exactly in [IDNA2008-Tables]). The rules given there are
1228 normative, rather than being just an interpretation of the tables.
1230 7.1.2. Labels in Registration
1232 Anyone entering a label into a DNS zone must properly validate that
1233 label -- i.e., be sure that the criteria for that label are met -- in
1234 order for applications to work as intended. This principle is not
1235 new. For example, since the DNS was first deployed, zone
1236 administrators have been expected to verify that names meet
1237 "hostname" [RFC0952] where necessary for the expected applications.
1238 Later addition of special service location formats [RFC2782] imposed
1239 new requirements on zone administrators for the use of labels that
1240 conform to the requirements of those formats. For zones that will
1241 contain IDNs, support for Unicode version-independence requires
1242 restrictions on all strings placed in the zone. In particular, for
1243 such zones:
1245 o Any label that appears to be an A-label, i.e., any label that
1246 starts in "xn--", must be IDNA-valid, i.e., they must be valid
1247 A-labels, as discussed in Section 2 above.
1249 o The Unicode tables (i.e., tables of code points, character
1250 classes, and properties) and IDNA tables (i.e., tables of
1251 contextual rules such as those that appear in the Tables
1252 document), must be consistent on the systems performing or
1253 validating labels to be registered. Note that this does not
1254 require that tables reflect the latest version of Unicode, only
1255 that all tables used on a given system are consistent with each
1256 other.
1258 Under this model, a registry (or entity communicating with a registry
1259 to accomplish name registrations) will need to update its tables --
1260 both the Unicode-associated tables and the tables of permitted IDN
1261 characters -- to enable a new script or other set of new characters.
1262 It will not be affected by newer versions of Unicode, or newly-
1263 authorized characters, until and unless it wishes to make those
1264 registrations. The zone administrator is also responsible -- under
1265 the protocol and to registrants and users -- for both checking as
1266 required by the protocol and verification that whatever policies it
1267 develops are complied with, whether those policies are for minimizing
1268 risks due to confusable characters and sequences, for preserving
1269 language or script integrity, or for other purposes. Those checking
1270 and verification procedures are more extensive than those that are is
1271 expected of applications systems that look names up.
1273 Systems looking up or resolving DNS labels, especially IDN DNS
1274 labels, must be able to assume that applicable registration rules
1275 were followed for names entered into the DNS.
1277 7.1.3. Labels in Lookup
1279 Anyone looking up a label in a DNS zone is required to
1281 o Maintain a consistent set of tables, as discussed above. As with
1282 registration, the tables need not reflect the latest version of
1283 Unicode but they must be consistent.
1285 o Validate the characters in labels to be looked up only to the
1286 extent of determining that the U-label does not contain either
1287 code points prohibited by IDNA (categorized as "DISALLOWED") or
1288 code points that are unassigned in its version of Unicode.
1290 o Validate the label itself for conformance with a small number of
1291 whole-label rules, notably verifying that there are no leading
1292 combining marks, that the "bidi" conditions are met if right to
1293 left characters appear, that any required contextual rules are
1294 available and that, if such rules are associated with Joiner
1295 Controls, they are tested.
1297 o Avoid validating other contextual rules about characters,
1298 including mixed-script label prohibitions, although such rules may
1299 be used to influence presentation decisions in the user interface.
1300 [[anchor20: Check this, and all similar statements, against
1301 Protocol when that is finished.]]
1303 By avoiding applying its own interpretation of which labels are valid
1304 as a means of rejecting lookup attempts, the lookup application
1305 becomes less sensitive to version incompatibilities with the
1306 particular zone registry associated with the domain name.
1308 An application or client that processes names according to this
1309 protocol and then resolves them in the DNS will be able to locate any
1310 name that is validly registered, as long as its version of the
1311 Unicode-associated tables is sufficiently up-to-date to interpret all
1312 of the characters in the label. Messages to users should distinguish
1313 between "label contains an unallocated code point" and other types of
1314 lookup failures. A failure on the basis of an old version of Unicode
1315 may lead the user to a desire to upgrade to a newer version, but will
1316 have no other ill effects (this is consistent with behavior in the
1317 transition to the DNS when some hosts could not yet handle some forms
1318 of names or record types).
1320 7.2. Changes in Character Interpretations
1322 [[anchor21: Note in Draft: This subsection is completely new in
1323 version -04 and has been further tuned in -05 and -06 of this
1324 document. It could almost certainly use improvement, although this
1325 note will be removed if there are not significant suggestions about
1326 the -06 version. It also contains some material that is redundant
1327 with material in other sections. I have not tried to remove that
1328 material and will not do so until the WG concludes that this section
1329 is relatively stable, but would appreciate help in identifying what
1330 should be removed or how this might be enhanced to contain more of
1331 that other material. --JcK]]
1333 In those scripts that make case distinctions, there are a few
1334 characters for which an obvious and unique upper case character has
1335 not historically been available to match a lower case one or vice
1336 versa. For those characters, the mappings used in constructing the
1337 Stringprep tables for IDNA2003, performed using the Unicode CaseFold
1338 operation (See Section 5.8 of the Unicode Standard [Unicode51]),
1339 generate different characters or sets of characters. Those
1340 operations are not reversible and lose even more information than
1341 traditional upper case or lower case transformations, but are more
1342 useful than those transformations for comparison purposes. Two
1343 notable characters of this type are the German character Eszett
1344 (Sharp S, U+00DF) and the Greek Final Form Sigma (U+03C2). The
1345 former is case-folded to the ASCII string "ss", the latter to a
1346 medial (Lower Case) Sigma (U+03C3).
1348 The decision to eliminate mappings, including case folding, from the
1349 IDNA2008 protocol in order to make A-labels and U-labels idempotent
1350 made these characters problematic. If they were to be disallowed,
1351 important words and mnemonics could not be written in
1352 orthographically reasonable ways. If they were to be permitted as
1353 characters distinct from the forms produced by case folding, there
1354 would be no information loss and registries would have maximum
1355 flexibility, but labels using those characters that were looked up
1356 according to IDNA2003 rules would be transformed into A-labels using
1357 their case-mapped variations while lookup according to IDNA2008 rules
1358 would be based on different A-labels that represented the actual
1359 characters.
1361 With the understanding that there would be incompatibility either way
1362 but a judgment that the incompatibility was not significant enough to
1363 justify a prefix change, the WG concluded that Eszett and Final Form
1364 Sigma should be treated as distinct and Protocol-Valid characters.
1366 The decision faces registries, especially registries maintaining
1367 zones for third parties, with a variation on what has become a
1368 familiar problem: how to introduce a new service in a way that does
1369 not create confusion or significantly weaken or invalidate existing
1370 identifiers.
1372 There have traditionally been several approaches to problems of this
1373 type. Without any preference or claim to completeness, these are:
1375 o Do not permit use of the newly-available character at the registry
1376 level. This might cause lookup failures if a domain name were to
1377 be written with the expectation of the IDNA2003 mapping behavior,
1378 but would eliminate any possibility of false matches.
1380 o Hold a "sunrise"-like arrangement in which holders of labels that
1381 might have resulted from previous mapping (labels containing "ss"
1382 in the Eszett case or ones containing Lower Case Sigma in the
1383 Final Sigma case) are given priority (and perhaps other benefits)
1384 for registering the corresponding string containing the newly-
1385 available characters.
1387 o Adopt some sort of "variant" approach in which registrants either
1388 obtained labels with both character forms or one of them was
1389 blocked from registration by anyone but the registrant of the
1390 other form.
1392 In principle, lookup applications could also compensate for the
1393 difference in interpretation by looking up the string according to
1394 the interpretation specified in these documents and then, if that
1395 failed, doing the lookup with the mapping, simulating the IDNA2003
1396 interpretation. The risk of false positives is such that this is
1397 generally to be discouraged unless the application is able to engage
1398 in a "is this what you meant" dialogue with the end user.
1400 7.3. More Flexibility in User Agents
1402 These documents do not specify mappings between one character or code
1403 point and others for any reason. Instead, they prohibit the
1404 characters that would be mapped to others by normalization, upper
1405 case to lower case changes, or other rules. As examples, while
1406 mathematical characters based on Latin ones are accepted as input to
1407 IDNA2003, they are prohibited in IDNA2008. Similarly, double-width
1408 characters and other variations are prohibited as IDNA input.
1410 Since the rules in [IDNA2008-Tables] have the effect that only
1411 strings that are not transformed by NFKC are valid, if an application
1412 chooses to perform NFKC normalization before lookup, that operation
1413 is safe since this will never make the application unable to look up
1414 any valid string. However, as discussed above, the application
1415 cannot guarantee that any other application will perform that
1416 mapping, so it should be used only with caution and for informed
1417 users.
1419 In many cases these prohibitions should have no effect on what the
1420 user can type as input to the lookup process. It is perfectly
1421 reasonable for systems that support user interfaces to perform some
1422 character mapping that is appropriate to the local environment. This
1423 would normally be done prior to actual invocation of IDNA. At least
1424 conceptually, the mapping would be part of the Unicode conversions
1425 discussed above and in [IDNA2008-Protocol]. However, those changes
1426 will be local ones only -- local to environments in which users will
1427 clearly understand that the character forms are equivalent. For use
1428 in interchange among systems, it appears to be much more important
1429 that U-labels and A-labels can be mapped back and forth without loss
1430 of information.
1432 One specific, and very important, instance of this strategy arises
1433 with case-folding. In the ASCII-only DNS, names are looked up and
1434 matched in a case-independent way, but no actual case-folding occurs.
1435 Names can be placed in the DNS in either upper or lower case form (or
1436 any mixture of them) and that form is preserved, returned in queries,
1437 and so on. IDNA2003 simulated that behavior for non-ASCII strings by
1438 performing case-folding at registration time (resulting in only
1439 lower-case IDNs in the DNS) and when names were looked up.
1441 As suggested earlier in this section, it appears to be desirable to
1442 do as little character mapping as possible consistent with having
1443 Unicode work correctly (e.g., NFC mapping to resolve different
1444 codings for the same character is still necessary although the
1445 specifications require that it be performed prior to invoking the
1446 protocol) and to make the mapping between A-labels and U-labels
1447 idempotent. Case-mapping is not an exception to this principle. If
1448 only lower case characters can be registered in the DNS (i.e., be
1449 present in a U-label), then IDNA2008 should prohibit upper-case
1450 characters as input (and therefore does so). Some other
1451 considerations reinforce this conclusion. For example, an essential
1452 element of the ASCII case-mapping functions is that, for individual
1453 characters, uppercase(character) must be equal to
1454 uppercase(lowercase(character)). That requirement may not be
1455 satisfied with IDNs. For example, there are some characters in
1456 scripts that use case distinction that do not have counterparts in
1457 one case or the other. The relationship between upper case and lower
1458 case may even be language-dependent, with different languages (or
1459 even the same language in different areas) expecting different
1460 mappings. Of course, the expectations of users who are accustomed to
1461 a case-insensitive DNS environment will probably be well-served if
1462 user agents perform case folding prior to IDNA processing, but the
1463 IDNA procedures themselves should neither require such mapping nor
1464 expect them when they are not natural to the localized environment.
1466 7.4. The Question of Prefix Changes
1468 The conditions that would require a change in the IDNA ACE prefix
1469 ("xn--" for the version of IDNA specified in [RFC3490]) have been a
1470 great concern to the community. A prefix change would clearly be
1471 necessary if the algorithms were modified in a manner that would
1472 create serious ambiguities during subsequent transition in
1473 registrations. This section summarizes our conclusions about the
1474 conditions under which changes in prefix would be necessary and the
1475 implications of such a change.
1477 7.4.1. Conditions Requiring a Prefix Change
1479 An IDN prefix change is needed if a given string would be looked up
1480 or otherwise interpreted differently depending on the version of the
1481 protocol or tables being used. Consequently, work to update IDNs
1482 would require a prefix change if, and only if, one of the following
1483 four conditions were met:
1485 1. The conversion of an A-label to Unicode (i.e., a U-label) yields
1486 one string under IDNA2003 (RFC3490) and a different string under
1487 IDNA2008.
1489 2. An input string that is valid under IDNA2003 and also valid under
1490 IDNA2008 yields two different A-labels with the different
1491 versions of IDNA. This condition is believed to be essentially
1492 equivalent to the one above except for a very small number of
1493 edge cases which may not, pragmatically, justify a prefix change
1494 (See Section 7.2).
1496 Note, however, that if the input string is valid under one
1497 version and not valid under the other, this condition does not
1498 apply. See the first item in Section 7.4.2, below.
1500 3. A fundamental change is made to the semantics of the string that
1501 is inserted in the DNS, e.g., if a decision were made to try to
1502 include language or specific script information in that string,
1503 rather than having it be just a string of characters.
1505 4. A sufficiently large number of characters is added to Unicode so
1506 that the Punycode mechanism for block offsets no longer has
1507 enough capacity to reference the higher-numbered planes and
1508 blocks. This condition is unlikely even in the long term and
1509 certain not to arise in the next few years.
1511 7.4.2. Conditions Not Requiring a Prefix Change
1513 In particular, as a result of the principles described above, none of
1514 the following changes require a new prefix:
1516 1. Prohibition of some characters as input to IDNA. This may make
1517 names that are now registered inaccessible, but does not require
1518 a prefix change.
1520 2. Adjustments in IDNA tables or actions, including normalization
1521 definitions, that affect characters that were already invalid
1522 under IDNA2003.
1524 3. Changes in the style of the IDNA definition that does not alter
1525 the actions performed by IDNA.
1527 7.4.3. Implications of Prefix Changes
1529 While it might be possible to make a prefix change, the costs of such
1530 a change are considerable. Even if they wanted to do so, registries
1531 could not convert all IDNA2003 ("xn--") registrations to a new form
1532 at the same time and synchronize that change with applications
1533 supporting lookup. Unless all existing registrations were simply to
1534 be declared invalid (and perhaps even then) systems that needed to
1535 support both labels with old prefixes and labels with new ones would
1536 first process a putative label under the IDNA2008 rules and try to
1537 look it up and then, if it were not found, would process the label
1538 under IDNA2003 rules and look it up again. That process could
1539 significantly slow down all processing that involved IDNs in the DNS
1540 especially since, in principle, a fully-qualified name could contain
1541 a mixture of labels that were registered with the old and new
1542 prefixes, a situation that would make the use of DNS caching very
1543 difficult. In addition, looking up the same input string as two
1544 separate A-labels would create some potential for confusion and
1545 attacks, since they could, in principle, map to different targets and
1546 then resolve to different entries in the DNS.
1548 Consequently, a prefix change is to be avoided if at all possible,
1549 even if it means accepting some IDNA2003 decisions about character
1550 distinctions as irreversible and/or giving special treatment to edge
1551 cases.
1553 7.5. Stringprep Changes and Compatibility
1555 The Nameprep [RFC3491] specification, a key part of IDNA2003, is a
1556 profile of Stringprep [RFC3454]. While Nameprep is a Stringprep
1557 profile specific to IDNA, Stringprep is used by a number of other
1558 protocols. Concerns have been expressed about problems for non-DNS
1559 uses of Stringprep being caused by changes to the specification
1560 intended to improve the handling of IDNs, most notably as this might
1561 affect identification and authentication protocols. The proposed new
1562 inclusion tables [IDNA2008-Tables], the reduction in the number of
1563 characters permitted as input for registration or lookup (Section 3),
1564 and even the proposed changes in handling of right to left strings
1565 [IDNA2008-Bidi] either give interpretations to strings prohibited
1566 under IDNA2003 or prohibit strings that IDNA2003 permitted. The
1567 IDNA2008 protocol does not use either Nameprep or Stringprep at all,
1568 so there are no side-effect changes to other protocols.
1570 It is particularly important to keep IDNA processing separate from
1571 processing for various security protocols because some of the
1572 constraints that are necessary for smooth and comprehensible use of
1573 IDNs may be unwanted or undesirable in other contexts. For example,
1574 the criteria for good passwords or passphrases are very different
1575 from those for desirable IDNs: passwords should be hard to guess,
1576 while domain names should normally be easily memorable. Similarly,
1577 internationalized SCSI identifiers and other protocol components are
1578 likely to have different requirements than IDNs.
1580 7.6. The Symbol Question
1582 One of the major differences between this specification and the
1583 original version of IDNA is that the original version permitted non-
1584 letter symbols of various sorts, including punctuation and line-
1585 drawing symbols, in the protocol. They were always discouraged in
1586 practice. In particular, both the "IESG Statement" about IDNA and
1587 all versions of the ICANN Guidelines specify that only language
1588 characters be used in labels. This specification disallows symbols
1589 entirely. There are several reasons for this, which include:
1591 o As discussed elsewhere, the original IDNA specification assumed
1592 that as many Unicode characters as possible should be permitted,
1593 directly or via mapping to other characters, in IDNs. This
1594 specification operates on an inclusion model, extrapolating from
1595 the LDH rules -- which have served the Internet very well -- to a
1596 Unicode base rather than an ASCII base.
1598 o Most Unicode names for letters are, in most cases, fairly
1599 intuitive, unambiguous and recognizable to users of the relevant
1600 script. Symbol names are more problematic because there may be no
1601 general agreement on whether a particular glyph matches a symbol;
1602 there are no uniform conventions for naming; variations such as
1603 outline, solid, and shaded forms may or may not exist; and so on.
1604 As just one example, consider a "heart" symbol as it might appear
1605 in a logo that might be read as "I love...". While the user might
1606 read such a logo as "I love..." or "I heart...", considerable
1607 knowledge of the coding distinctions made in Unicode is needed to
1608 know that there more than one "heart" character (e.g., U+2665,
1609 U+2661, and U+2765) and how to describe it. These issues are of
1610 particular importance if strings are expected to be understood or
1611 transcribed by the listener after being read out loud.
1612 [[anchor22: The above paragraph remains controversial as to
1613 whether it is valid. The WG will need to make a decision if this
1614 section is not dropped entirely.]]
1616 o Consider the case of a screen reader used by blind Internet users
1617 who must listen to renderings of IDN domain names and possibly
1618 reproduce them on the keyboard.
1620 o As a simplified example of this, assume one wanted to use a
1621 "heart" or "star" symbol in a label. This is problematic because
1622 those names are ambiguous in the Unicode system of naming (the
1623 actual Unicode names require far more qualification). A user or
1624 would-be registrant has no way to know -- absent careful study of
1625 the code tables -- whether it is ambiguous (e.g., where there are
1626 multiple "heart" characters) or not. Conversely, the user seeing
1627 the hypothetical label doesn't know whether to read it -- try to
1628 transmit it to a colleague by voice -- as "heart", as "love", as
1629 "black heart", or as any of the other examples below.
1631 o The actual situation is even worse than this. There is no
1632 possible way for a normal, casual, user to tell the difference
1633 between the hearts of U+2665 and U+2765 and the stars of U+2606
1634 and U+2729 or the without somehow knowing to look for a
1635 distinction. We have a white heart (U+2661) and few black hearts.
1636 Consequently, describing a label as containing a heart hopelessly
1637 ambiguous: we can only know that it contains one of several
1638 characters that look like hearts or have "heart" in their names.
1639 In cities where "Square" is a popular part of a location name, one
1640 might well want to use a square symbol in a label as well and
1641 there are far more squares of various flavors in Unicode than
1642 there are hearts or stars.
1644 o The consequence of these ambiguities of description and
1645 dependencies on distinctions that were, or were not, made in
1646 Unicode codings is that symbols are a very poor basis for reliable
1647 communication. Consistent with this conclusion, the Unicode
1648 standard recommends that strings used in identifiers not contain
1649 symbols or punctuation [Unicode-UAX31]. Of course, these
1650 difficulties with symbols do not arise with actual pictographic
1651 languages and scripts which would be treated like any other
1652 language characters; the two should not be confused.
1654 7.7. Migration Between Unicode Versions: Unassigned Code Points
1656 In IDNA2003, labels containing unassigned code points are looked up
1657 on the assumption that, if they appear in labels and can be mapped
1658 and then resolved, the relevant standards must have changed and the
1659 registry has properly allocated only assigned values.
1661 In the protocol as described in these documents, strings containing
1662 unassigned code points must not be either looked up or registered.
1663 There are several reasons for this, with the most important ones
1664 being:
1666 o It cannot be known in advance, and with sufficient reliability,
1667 that a code point that was not previously assigned will not be
1668 assigned to a compatibility character or one that would be
1669 otherwise disallowed by the rules in [IDNA2008-Tables]. In
1670 IDNA2003, since there is no direct dependency on NFKC (many of the
1671 entries in Stringprep's tables are based on NFKC, but IDNA2003
1672 depends only on Stringprep), allocation of a compatibility
1673 character might produce some odd situations, but it would not be a
1674 problem. In IDNA2008, where compatibility characters are assigned
1675 to DISALLOWED unless character-specific exceptions are made,
1676 permitting strings containing unassigned characters to be looked
1677 up would permit violating the principle that characters in
1678 DISALLOWED are not looked up.
1680 o The Unicode Standard specifies that an unassigned code point
1681 normalizes (and, where relevant, case folds) to itself. If the
1682 code point is later assigned to a character, and particularly if
1683 the newly-assigned code point has a combining class that
1684 determines its placement relative to other combining characters,
1685 it could normalize to some other code point or sequence, creating
1686 confusion and/or violating other rules listed here.
1688 o Tests involving the context of characters (e.g., some characters
1689 being permitted only adjacent to ones of specific types but
1690 otherwise invisible or very problematic for other reasons) and
1691 integrity tests on complete labels are needed. Unassigned code
1692 points cannot be permitted because one cannot determine whether
1693 particular code points will require contextual rules (and what
1694 those rules should be) before characters are assigned to them and
1695 the properties of those characters fully understood.
1697 o More generally, the status of an unassigned character with regard
1698 to the DISALLOWED and PROTOCOL-VALID categories, and whether
1699 contextual rules are required with the latter, cannot be evaluated
1700 until a character is actually assigned and known. By contrast,
1701 characters that are actually DISALLOWED are placed in that
1702 category only as a consequence of rules applied to known
1703 properties or per-character evaluation.
1705 Another way to look at this is that permitting an unassigned
1706 character to be looked up is nearly equivalent to reclassifying a
1707 character from DISALLOWED to PROTOCOL-VALID since different systems
1708 will interpret the character in different ways.
1710 It is possible to argue that the issues above are not important and
1711 that, as a consequence, it is better to retain the principle of
1712 looking up labels even if they contain unassigned characters because
1713 all of the important scripts and characters have been coded as of
1714 Unicode 5.1 and hence unassigned code points will be assigned only to
1715 obscure characters or archaic scripts. Unfortunately, that does not
1716 appear to be a safe assumption for at least two reasons. First, much
1717 the same claim of completeness has been made for earlier versions of
1718 Unicode. The reality is that a script that is obscure to much of the
1719 world may still be very important to those who use it. Cultural and
1720 linguistic preservation principles make it inappropriate to declare
1721 the script of no importance in IDNs. Second, we already have
1722 counterexamples in, e.g., the relationships associated with new Han
1723 characters being added (whether in the BMP or in Unicode Plane 2).
1725 Independent of the technical transition issues identified above, it
1726 can be observed that any addition of characters to an existing script
1727 to make it easier to use or to better accommodate particular
1728 languages may lead to transition issues. Such changes may change the
1729 preferred form for writing a particular string, changes that may be
1730 reflected, e.g., in keyboard transition modules that would
1731 necessarily be different from those for earlier versions of Unicode
1732 where the newer characters may not exist. This creates an inherent
1733 transition problem because attempts to access labels may use either
1734 the old or the new conventions, requiring registry action whether the
1735 older conventions were used in labels or not. The need to consider
1736 transition mechanisms is inherent to evolution of Unicode to better
1737 accommodate writing systems and is independent of how IDNs are
1738 represented in the DNS or how transitions among versions of those
1739 mechanisms occur. The requirement for transitions of this type is
1740 illustrated by the addition of Malayalam Chillu in Unicode 5.1.0.
1742 7.8. Other Compatibility Issues
1744 The 2003 IDNA model includes several odd artifacts of the context in
1745 which it was developed. Many, if not all, of these are potential
1746 avenues for exploits, especially if the registration process permits
1747 "source" names (names that have not been processed through IDNA and
1748 Nameprep) to be registered. As one example, since the character
1749 Eszett, used in German, is mapped by IDNA2003 into the sequence "ss"
1750 rather than being retained as itself or prohibited, a string
1751 containing that character but that is otherwise in ASCII is not
1752 really an IDN (in the U-label sense defined above) at all. After
1753 Nameprep maps the Eszett out, the result is an ASCII string and so
1754 does not get an xn-- prefix, but the string that can be displayed to
1755 a user appears to be an IDN. The newer version of the protocol
1756 eliminates this artifact. A character is either permitted as itself
1757 or it is prohibited; special cases that make sense only in a
1758 particular linguistic or cultural context can be dealt with as
1759 localization matters where appropriate.
1761 8. Name Server Considerations
1763 8.1. Processing Non-ASCII Strings
1765 Existing DNS servers do not know the IDNA rules for handling non-
1766 ASCII forms of IDNs, and therefore need to be shielded from them.
1767 All existing channels through which names can enter a DNS server
1768 database (for example, master files (as described in RFC 1034) and
1769 DNS update messages [RFC2136]) are IDN-unaware because they predate
1770 IDNA. Other sections of this document provide the needed shielding
1771 by ensuring that internationalized domain names entering DNS server
1772 databases through such channels have already been converted to their
1773 equivalent ASCII A-label forms.
1775 Because of the distinction made between the algorithms for
1776 Registration and Lookup in [IDNA2008-Protocol] (a domain name
1777 containing only ASCII codepoints can not be converted to an A-label),
1778 there can not be more than one A-label form for any given U-label.
1780 As specified in RFC 2181 [RFC2181], the DNS protocol explicitly
1781 allows domain labels to contain octets beyond the ASCII range
1782 (0000..007F), and this document does not change that. However,
1783 although the interpretation of octets 0080..00FF is well-defined in
1784 the DNS, many application protocols support only ASCII labels and
1785 there is no defined interpretation of these non-ASCII octets as
1786 characters and, in particular, no interpretation of case-independent
1787 matching for them (see, e.g., [RFC4343]). If labels containing these
1788 octets are returned to applications, unpredictable behavior could
1789 result. The A-label form, which cannot contain those characters, is
1790 the only standard representation for internationalized labels in the
1791 DNS protocol.
1793 8.2. DNSSEC Authentication of IDN Domain Names
1795 DNS Security (DNSSEC) [RFC2535] is a method for supplying
1796 cryptographic verification information along with DNS messages.
1797 Public Key Cryptography is used in conjunction with digital
1798 signatures to provide a means for a requester of domain information
1799 to authenticate the source of the data. This ensures that it can be
1800 traced back to a trusted source, either directly or via a chain of
1801 trust linking the source of the information to the top of the DNS
1802 hierarchy.
1804 IDNA specifies that all internationalized domain names served by DNS
1805 servers that cannot be represented directly in ASCII MUST use the
1806 A-label form. Conversion to A-labels MUST be performed prior to a
1807 zone being signed by the private key for that zone. Because of this
1808 ordering, it is important to recognize that DNSSEC authenticates a
1809 domain name containing A-labels or conventional LDH-labels, not
1810 U-labels. In the presence of DNSSEC, no form of a zone file or query
1811 response that contains a U-label may be signed or the signature
1812 validated.
1814 One consequence of this for sites deploying IDNA in the presence of
1815 DNSSEC is that any special purpose proxies or forwarders used to
1816 transform user input into IDNs must be earlier in the lookup flow
1817 than DNSSEC authenticating nameservers for DNSSEC to work.
1819 8.3. Root and other DNS Server Considerations
1821 IDNs in A-label form will generally be somewhat longer than current
1822 domain names, so the bandwidth needed by the root servers is likely
1823 to go up by a small amount. Also, queries and responses for IDNs
1824 will probably be somewhat longer than typical queries historically,
1825 so EDNS0 [RFC2671] support may be more important (otherwise, queries
1826 and responses may be forced to go to TCP instead of UDP).
1828 9. Internationalization Considerations
1830 DNS labels and fully-qualified domain names provide mnemonics that
1831 assist in identifying and referring to resources on the Internet.
1832 IDNs expand the range of those mnemonics to include those based on
1833 languages and character sets other than Western European and Roman-
1834 derived ones. But domain "names" are not, in general, words in any
1835 language. The recommendations of the IETF policy on character sets
1836 and languages, BCP 18 [RFC2277] are applicable to situations in which
1837 language identification is used to provide language-specific
1838 contexts. The DNS is, by contrast, global and international and
1839 ultimately has nothing to do with languages. Adding languages (or
1840 similar context) to IDNs generally, or to DNS matching in particular,
1841 would imply context dependent matching in DNS, which would be a very
1842 significant change to the DNS protocol itself. It would also imply
1843 that users would need to identify the language associated with a
1844 particular label in order to look that label up, a decision that
1845 would be impossible in many or most cases.
1847 10. IANA Considerations
1849 This section gives an overview of registries required for IDNA. The
1850 actual definitions of the first two appear in [IDNA2008-Tables].
1852 10.1. IDNA Character Registry
1854 The distinction among the three major categories "UNASSIGNED",
1855 "DISALLOWED", and "PROTOCOL-VALID" is made by special categories and
1856 rules that are integral elements of [IDNA2008-Tables]. Convenience
1857 in programming and validation requires a registry of characters and
1858 scripts and their categories, updated for each new version of Unicode
1859 and the characters it contains. The details of this registry are
1860 specified in [IDNA2008-Tables].
1862 10.2. IDNA Context Registry
1864 For characters that are defined in the IDNA Character Registry list
1865 as PROTOCOL-VALID but requiring a contextual rule (i.e., the types of
1866 rule described in Section 3.1.2), IANA will create and maintain a
1867 list of approved contextual rules. The details for those rules
1868 appear in [IDNA2008-Tables].
1870 10.3. IANA Repository of IDN Practices of TLDs
1872 This registry, historically described as the "IANA Language Character
1873 Set Registry" or "IANA Script Registry" (both somewhat misleading
1874 terms) is maintained by IANA at the request of ICANN. It is used to
1875 provide a central documentation repository of the IDN policies used
1876 by top level domain (TLD) registries who volunteer to contribute to
1877 it and is used in conjunction with ICANN Guidelines for IDN use.
1879 It is not an IETF-managed registry and, while the protocol changes
1880 specified here may call for some revisions to the tables, these
1881 specifications have no direct effect on that registry and no IANA
1882 action is required as a result.
1884 11. Security Considerations
1886 11.1. General Security Issues with IDNA
1888 This document in the IDNA2008 series is purely explanatory and
1889 informational and consequently introduces no new security issues. It
1890 would, of course, be a poor idea for someone to try to implement from
1891 it; such an attempt would almost certainly lead to interoperability
1892 problems and might lead to security ones. A discussion of security
1893 issues with IDNA, including some relevant history, appears in
1894 [IDNA2008-Defs].
1896 12. Acknowledgments
1898 The editor and contributors would like to express their thanks to
1899 those who contributed significant early (pre-WG) review comments,
1900 sometimes accompanied by text, especially Mark Davis, Paul Hoffman,
1901 Simon Josefsson, and Sam Weiler. In addition, some specific ideas
1902 were incorporated from suggestions, text, or comments about sections
1903 that were unclear supplied by Vint Cerf, Frank Ellerman, Michael
1904 Everson, Asmus Freytag, Erik van der Poel, Michel Suignard, and Ken
1905 Whistler, although, as usual, they bear little or no responsibility
1906 for the conclusions the editor and contributors reached after
1907 receiving their suggestions. Thanks are also due to Vint Cerf,
1908 Debbie Garside, and Jefsey Morfin for conversations that led to
1909 considerable improvements in the content of this document.
1911 A meeting was held on 30 January 2008 to attempt to reconcile
1912 differences in perspective and terminology about this set of
1913 specifications between the design team and members of the Unicode
1914 Technical Consortium. The discussions at and subsequent to that
1915 meeting were very helpful in focusing the issues and in refining the
1916 specifications. The active participants at that meeting were (in
1917 alphabetic order as usual) Harald Alvestrand, Vint Cerf, Tina Dam,
1918 Mark Davis, Lisa Dusseault, Patrik Faltstrom (by telephone), Cary
1919 Karp, John Klensin, Warren Kumari, Lisa Moore, Erik van der Poel,
1920 Michel Suignard, and Ken Whistler. We express our thanks to Google
1921 for support of that meeting and to the participants for their
1922 contributions.
1924 Useful comments and text on the WG versions of the draft were
1925 received from many participants in the IETF "IDNABIS" WG and a number
1926 of document changes resulted from mailing list discussions made by
1927 that group. Marcos Sanz provided specific analysis and suggestions
1928 that were exceptionally helpful in refining the text, as did Vint
1929 Cerf, Mark Davis, Martin Duerst, Andrew Sullivan, and Ken Whistler.
1931 13. Contributors
1933 While the listed editor held the pen, the core of this document and
1934 the initial WG version represents the joint work and conclusions of
1935 an ad hoc design team consisting of the editor and, in alphabetic
1936 order, Harald Alvestrand, Tina Dam, Patrik Faltstrom, and Cary Karp.
1937 In addition, there were many specific contributions and helpful
1938 comments from those listed in the Acknowledgments section and others
1939 who have contributed to the development and use of the IDNA
1940 protocols.
1942 14. References
1944 14.1. Normative References
1946 [ASCII] American National Standards Institute (formerly United
1947 States of America Standards Institute), "USA Code for
1948 Information Interchange", ANSI X3.4-1968, 1968.
1950 ANSI X3.4-1968 has been replaced by newer versions with
1951 slight modifications, but the 1968 version remains
1952 definitive for the Internet.
1954 [IDNA2008-Bidi]
1955 Alvestrand, H. and C. Karp, "An updated IDNA criterion for
1956 right to left scripts", July 2008, .
1959 [IDNA2008-Defs]
1960 Klensin, J., "Internationalized Domain Names for
1961 Applications (IDNA): Definitions and Document Framework",
1962 November 2008, .
1965 [IDNA2008-Protocol]
1966 Klensin, J., "Internationalized Domain Names in
1967 Applications (IDNA): Protocol", November 2008, .
1970 [IDNA2008-Tables]
1971 Faltstrom, P., "The Unicode Code Points and IDNA",
1972 July 2008, .
1975 A version of this document is available in HTML format at
1976 http://stupid.domain.name/idnabis/
1977 draft-ietf-idnabis-tables-02.html
1979 [RFC3490] Faltstrom, P., Hoffman, P., and A. Costello,
1980 "Internationalizing Domain Names in Applications (IDNA)",
1981 RFC 3490, March 2003.
1983 [RFC3492] Costello, A., "Punycode: A Bootstring encoding of Unicode
1984 for Internationalized Domain Names in Applications
1985 (IDNA)", RFC 3492, March 2003.
1987 [Unicode-UAX15]
1988 The Unicode Consortium, "Unicode Standard Annex #15:
1989 Unicode Normalization Forms", March 2008,
1990 .
1992 [Unicode51]
1993 The Unicode Consortium, "The Unicode Standard, Version
1994 5.1.0", 2008.
1996 defined by: The Unicode Standard, Version 5.0, Boston, MA,
1997 Addison-Wesley, 2007, ISBN 0-321-48091-0, as amended by
1998 Unicode 5.1.0
1999 (http://www.unicode.org/versions/Unicode5.1.0/).
2001 14.2. Informative References
2003 [BIG5] Institute for Information Industry of Taiwan, "Computer
2004 Chinese Glyph and Character Code Mapping Table, Technical
2005 Report C-26", 1984.
2007 There are several forms and variations and a closely-
2008 related standard, CNS 11643. See the discussion in
2009 Chapter 3 of Lunde, K., CJKV Information Processing,
2010 O'Reilly & Associates, 1999
2012 [GB18030] "Chinese National Standard GB 18030-2000: Information
2013 Technology -- Chinese ideograms coded character set for
2014 information interchange -- Extension for the basic set.",
2015 2000.
2017 [RFC0810] Feinler, E., Harrenstien, K., Su, Z., and V. White, "DoD
2018 Internet host table specification", RFC 810, March 1982.
2020 [RFC0952] Harrenstien, K., Stahl, M., and E. Feinler, "DoD Internet
2021 host table specification", RFC 952, October 1985.
2023 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities",
2024 STD 13, RFC 1034, November 1987.
2026 [RFC1035] Mockapetris, P., "Domain names - implementation and
2027 specification", STD 13, RFC 1035, November 1987.
2029 [RFC1123] Braden, R., "Requirements for Internet Hosts - Application
2030 and Support", STD 3, RFC 1123, October 1989.
2032 [RFC2136] Vixie, P., Thomson, S., Rekhter, Y., and J. Bound,
2033 "Dynamic Updates in the Domain Name System (DNS UPDATE)",
2034 RFC 2136, April 1997.
2036 [RFC2181] Elz, R. and R. Bush, "Clarifications to the DNS
2037 Specification", RFC 2181, July 1997.
2039 [RFC2277] Alvestrand, H., "IETF Policy on Character Sets and
2040 Languages", BCP 18, RFC 2277, January 1998.
2042 [RFC2535] Eastlake, D., "Domain Name System Security Extensions",
2043 RFC 2535, March 1999.
2045 [RFC2671] Vixie, P., "Extension Mechanisms for DNS (EDNS0)",
2046 RFC 2671, August 1999.
2048 [RFC2673] Crawford, M., "Binary Labels in the Domain Name System",
2049 RFC 2673, August 1999.
2051 [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for
2052 specifying the location of services (DNS SRV)", RFC 2782,
2053 February 2000.
2055 [RFC3454] Hoffman, P. and M. Blanchet, "Preparation of
2056 Internationalized Strings ("stringprep")", RFC 3454,
2057 December 2002.
2059 [RFC3491] Hoffman, P. and M. Blanchet, "Nameprep: A Stringprep
2060 Profile for Internationalized Domain Names (IDN)",
2061 RFC 3491, March 2003.
2063 [RFC3743] Konishi, K., Huang, K., Qian, H., and Y. Ko, "Joint
2064 Engineering Team (JET) Guidelines for Internationalized
2065 Domain Names (IDN) Registration and Administration for
2066 Chinese, Japanese, and Korean", RFC 3743, April 2004.
2068 [RFC3987] Duerst, M. and M. Suignard, "Internationalized Resource
2069 Identifiers (IRIs)", RFC 3987, January 2005.
2071 [RFC4290] Klensin, J., "Suggested Practices for Registration of
2072 Internationalized Domain Names (IDN)", RFC 4290,
2073 December 2005.
2075 [RFC4343] Eastlake, D., "Domain Name System (DNS) Case Insensitivity
2076 Clarification", RFC 4343, January 2006.
2078 [RFC4690] Klensin, J., Faltstrom, P., Karp, C., and IAB, "Review and
2079 Recommendations for Internationalized Domain Names
2080 (IDNs)", RFC 4690, September 2006.
2082 [RFC4713] Lee, X., Mao, W., Chen, E., Hsu, N., and J. Klensin,
2083 "Registration and Administration Recommendations for
2084 Chinese Domain Names", RFC 4713, October 2006.
2086 [Unicode-Security]
2087 The Unicode Consortium, "Unicode Technical Standard #39:
2088 Unicode Security Mechanisms", August 2008,
2089 .
2091 [Unicode-UAX31]
2092 The Unicode Consortium, "Unicode Standard Annex #31:
2093 Unicode Identifier and Pattern Syntax", March 2008,
2094 .
2096 [Unicode-UTR36]
2097 The Unicode Consortium, "Unicode Technical Report #36:
2098 Unicode Security Considerations", July 2008,
2099 .
2101 Appendix A. Change Log
2103 [[ RFC Editor: Please remove this appendix. ]]
2105 A.1. Changes between Version -00 and Version -01 of
2106 draft-ietf-idnabis-rationale
2108 o Clarified the U-label definition to note that U-labels must
2109 contain at least one non-ASCII character. Also clarified the
2110 relationship among label types.
2112 o Rewrote the discussion of Labels in Registration (Section 7.1.2)
2113 and related text about IDNA-validity (in the "Defs" document as of
2114 -04 of this one) to narrow its focus and remove more general
2115 restrictions. Added a temporary note in line to explain the
2116 situation.
2118 o Changed the "IDNA uses Unicode" statement to focus on
2119 compatibility with IDNA2003 and avoid more general or
2120 controversial assertions.
2122 o Added a discussion of examples to Section 7.1
2124 o Made a number of other small editorial changes and corrections
2125 suggested by Mark Davis.
2127 o Added several more discussion anchors and notes and expanded or
2128 updated some existing ones.
2130 A.2. Version -02
2132 o Trimmed change log, removing information about pre-WG drafts.
2134 o Adjusted discussion of Contextual Rules to match the new location
2135 of the tables and some conceptual material.
2137 o Rewrote the material on preprocessing somewhat.
2139 o Moved the material about relationships with IDNA2003 to be part of
2140 a single section on transitions.
2142 o Removed several placeholders and made editorial changes in
2143 accordance with decisions made at IETF 72 in Dublin and not
2144 disputed on the mailing list.
2146 A.3. Version -03
2148 This special update to the Rationale document is intended to try to
2149 get the discussion of what is normative or not under control. While
2150 the IETF does not normally annotate individual sections of documents
2151 with whether they are normative or not, concerns that we don't know
2152 which is which, claims that some material is normative that would be
2153 problematic if so classified, etc., argue that we should at least be
2154 able to have a clear discussion on the subject.
2156 Two annotations have been applied to sections that might reasonably
2157 be considered normative. One annotation is based on the list of
2158 sections in Mark Davis's note of 29 September (http://
2159 www.alvestrand.no/pipermail/idna-update/2008-September/002667.html).
2160 The other is based on an elaboration of John Klensin's response on 7
2161 October (http://www.alvestrand.no/pipermail/idna-update/2008-October/
2162 002691.html). These should just be considered two suggestions to
2163 illuminate and, one hopes, advance the Working Group's discussions.
2165 Some additional editorial changes have been made, but they are
2166 basically trivial. In the editor's judgment, it is not possible to
2167 make significantly more progress with this document until the matter
2168 of document organization is settled.
2170 A.4. Version -04
2172 o Definitional and other normative material moved to new document
2173 (draft-ietf-idnabis-defs). Version -03 annotations removed.
2175 o Material on differences between IDNA2003 and IDNA2008 moved to an
2176 appendix in Protocol.
2178 o Material left over from the origins of this document as a
2179 preliminary proposal has been removed or rewritten.
2181 o Changes made to reflect consensus call results, including removing
2182 several placeholder notes for discussion.
2184 o Added more material, including discussion of historic scripts, to
2185 Section 3.2 on registration policies.
2187 o Added a new section (Section 7.2) to contain specific discussion
2188 of handling of characters that are interpreted differently in
2189 input to IDNA2003 and 2008.
2191 o Some material, including this section/appendix, rearranged.
2193 A.5. Version -05
2195 o Many small editorial changes, including changes to eliminate the
2196 last vestiges of what appeared to be 2119 language (upper-case
2197 MUST, SHOULD, or MAY) and small adjustments to terminology.
2199 A.6. Version -06
2201 o Removed Security Considerations material and pointed to Defs,
2202 where it now appears as of version 05.
2204 o Started changing uses of "IDNA2008" in running text to "in these
2205 specifications" or the equivalent. These documents are titled
2206 simply "IDNA"; once they are standardized, "the current version"
2207 may be a more appropriate reference than one containing a year.
2208 As discussed on the mailing list, we can and should discuss how to
2209 refer to these documents at an appropriate time (e.g., when we
2210 know when we will be finished) but, in the interim, it seems
2211 appropriate to simply start getting rid of the version-specific
2212 terminology where it can naturally be removed.
2214 o Additional discussion of mappings, etc., especially for case-
2215 sensitivity.
2217 o Clarified relationship to base DNS specifications.
2219 o Consolidated discussion of lookup of unassigned characters.
2221 o More editorial fine-tuning.
2223 A.7. Version -07
2225 o Revised terminology by adding terms: NR-LDH-label, Invalid-A-label
2226 (or False-A-label), R-LDH-label, valid IDNA-label in
2227 Section 1.3.3.
2229 o Moved the "name server considerations" material to this document
2230 from Protocol because it is non-normative and not part of the
2231 protocol itself.
2233 o To improve clarity, redid discussion of the reasons why looking up
2234 unassigned code points is prohibited.
2236 o Editorial and other non-substantive corrections to reflect earlier
2237 errors as well as new definitions and terminology.
2239 A.8. Version -08
2241 o Slight revision to "contextual" discussion (Section 3.1.2) and
2242 moving it to a separate subsection, rather than under "PVALID",
2243 for better parallelism with Tables. Also reflected Mark's
2244 comments about the limitations of the approach.
2246 o Added placeholder notes as reminders of where references to the
2247 other documents need Section numbers. More of these will be added
2248 as needed (feel free to identify relevant places), but the actual
2249 section numbers will not be inserted until the documents are
2250 completely stable, i.e., on their way to the RFC Editor.
2252 A.9. Version -09
2254 o Small editorial changes to clarify transition possibilities.
2256 o Small clarification to the description of DNS "exact match".
2258 o Added discussion of adding characters to an existing script to the
2259 discussion of unassigned code point transitions in Section 7.7.
2261 o Tightened up the discussion of non-ASCII string processing
2262 (Section 8.1) slightly.
2264 o Removed some placeholders and comments that have been around long
2265 enough to be considered acceptable or that no longer seem
2266 necessary for other reasons.
2268 Author's Address
2270 John C Klensin
2271 1770 Massachusetts Ave, Ste 322
2272 Cambridge, MA 02140
2273 USA
2275 Phone: +1 617 245 1457
2276 Email: john+ietf@jck.com