idnits 2.17.1 draft-gellens-mmusic-negotiating-human-language-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The exact meaning of the all-uppercase expression 'NOT REQUIRED' is not defined in RFC 2119. If it is intended as a requirements expression, it should be rewritten using one of the combinations defined in RFC 2119; otherwise it should not be all-uppercase. -- The document date (February 13, 2014) is 3718 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'I-D.iab-privacy-considerations' is defined on line 873, but no explicit reference was found in the text ** Obsolete normative reference: RFC 4566 (Obsoleted by RFC 8866) -- Obsolete informational reference (is this intentional?): RFC 3066 (Obsoleted by RFC 4646, RFC 4647) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 MMUSIC Working Group R. Gellens 3 Internet-Draft Qualcomm Technologies Inc. 4 Intended status: Standards Track February 13, 2014 5 Expires: August 17, 2014 7 Negotiating Human Language in Real-Time Communications 8 draft-gellens-mmusic-negotiating-human-language-02 10 Abstract 12 Users have various human (natural) language needs, abilities, and 13 preferences regarding spoken, written, and signed languages. When 14 establishing interactive communication "calls" there needs to be a 15 way to communicate and ideally match (i.e., negotiate) the caller's 16 language preferences with the capabilities of the called party. This 17 is especially important with emergency calls, where a call can be 18 routed to a Public Safety Answering Point (PSAP) or call taker 19 capable of communicating with the user, or a translator or relay 20 operator can be bridged into the call during setup, but this applies 21 to non-emergency calls as well (as an example, when calling a company 22 call center). 24 This document describes the need and expected use, and describes a 25 solution using new SDP stream attributes plus an optional SIP "hint." 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at http://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on August 17, 2014. 44 Copyright Notice 46 Copyright (c) 2014 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 51 (http://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 62 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 63 3. Expected Use . . . . . . . . . . . . . . . . . . . . . . . . 5 64 4. Example Use Cases . . . . . . . . . . . . . . . . . . . . . . 5 65 4.1. Emergency Call from English Speaker in Spain . . . . . . 5 66 4.2. Emergency Call from Spanish/English Speaker in France . . 6 67 4.3. Call to Call Center from Russian Speaker in U.S. . . . . 6 68 4.4. Emergency Call from speech-impaired caller in the U.S. . 6 69 4.5. Emergency Call from deaf caller in the U.S. . . . . . . . 7 70 5. Desired Semantics . . . . . . . . . . . . . . . . . . . . . . 7 71 6. The existing 'lang' attribute . . . . . . . . . . . . . . . . 7 72 7. Proposed Solution . . . . . . . . . . . . . . . . . . . . . . 9 73 7.1. New 'humintlang-send' and 'humintlang-recv' attributes . 9 74 7.2. Advisory vs Required . . . . . . . . . . . . . . . . . . 10 75 7.3. SIP "hint" . . . . . . . . . . . . . . . . . . . . . . . 11 76 7.4. Silly States . . . . . . . . . . . . . . . . . . . . . . 11 77 8. Alternative Proposal: Caller-prefs . . . . . . . . . . . . . 12 78 8.1. Use of Caller Preferences Without Additions . . . . . . . 12 79 8.2. Additional Caller Preferences for Asymmetric Needs . . . 14 80 8.2.1. Caller Preferences for Asymmetric Modality Needs . . 14 81 8.2.2. Caller Preferences for Asymmetric Language Tags . . . 15 82 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 16 83 10. Security Considerations . . . . . . . . . . . . . . . . . . . 16 84 11. Changes from Previous Versions . . . . . . . . . . . . . . . 16 85 11.1. Changes from draft-gellens-...-00 to -01 . . . . . . . . 16 86 11.2. Changes from draft-gellens-...-01 to -02 . . . . . . . . 17 87 11.3. Changes from draft-gellens-...-02 to draft-gellens- 88 mmusic-...-00 . . . . . . . . . . . . . . . . . . . . . 17 89 11.4. Changes from draft-gellens-mmusic-...-00 to -01 . . . . 18 90 11.5. Changes from draft-gellens-mmusic-...-01 to -02 . . . . 18 91 12. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 18 92 13. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 19 93 14. References . . . . . . . . . . . . . . . . . . . . . . . . . 19 94 14.1. Normative References . . . . . . . . . . . . . . . . . . 19 95 14.2. Informational References . . . . . . . . . . . . . . . . 19 96 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 20 98 1. Introduction 100 When setting up interactive communication sessions (using SIP or 101 other protocols), human (natural) language negotiation may be needed. 102 When the caller and callee know each other or where context or out of 103 band information implies the language, such negotiation is typically 104 not needed. In other cases, there is a need for spoken, signed, or 105 written languages to be negotiated based on the caller's preferences 106 and the callee's capabilities. This need applies to both emergency 107 and non-emergency calls. For various reasons, including the ability 108 to establish multiple streams using different media (e.g., voice, 109 text, video), it makes sense to use a per-stream negotiation 110 mechanism, in this case, SDP. 112 This approach has a number of benefits, including that it is generic 113 (applies to all interactive communications negotiated using SDP) and 114 not limited to emergency calls. In some cases such a facility isn't 115 needed, because the language is known from the context (such as when 116 a caller places a call to a sign language relay center, to a friend, 117 or colleague). But it is clearly useful in many other cases. For 118 example, someone calling a company call center or a Public Safety 119 Answering Point (PSAP) should be able to indicate if one or more 120 specific signed, written, and/or spoken languages are preferred, the 121 callee should be able to indicate its capabilities in this area, and 122 the call proceed using in-common language(s) and media forms. 124 Since this is a protocol mechanism, the user equipment (UE client) 125 needs to know the user's preferred languages; a reasonable technique 126 could include a configuration mechanism with a default of the 127 language of the user interface. In some cases, a UE could tie 128 language and media preferences, such as a preference for a video 129 stream using a signed language and/or a text or audio stream using a 130 written/spoken language. 132 Including the user's human (natural) language preferences in the 133 session establishment negotiation is independent of the use of a 134 relay service and is transparent to a voice service provider. For 135 example, assume a user within the United States who speaks Spanish 136 but not English places a voice call using an IMS device. It doesn't 137 matter if the call is an emergency call or not (e.g., to an airline 138 reservation desk). The language information is transparent to the 139 IMS carrier, but is part of the session negotiation between the UE 140 and the terminating entity. In the case of a call to e.g., an 141 airline, the call can be automatically routed to a Spanish-speaking 142 agent. In the case of an emergency call, the Emergency Services IP 143 network (ESInet) and the PSAP may choose to take the language and 144 media preferences into account when determining how to route and 145 process the call (i.e., language and media needs may be considered 146 within policy-based routing (PBR)). 148 By treating language as another attribute that is negotiated along 149 with other aspects of a media stream, it becomes possible to 150 accommodate a range of users' needs and called party facilities. For 151 example, some users may be able to speak several languages, but have 152 a preference. Some called parties may support some of those 153 languages internally but require the use of a translation service for 154 others, or may have a limited number of call takers able to use 155 certain languages. Another example would be a user who is able to 156 speak but is deaf or hard-of-hearing and requires a voice stream plus 157 a text stream (known as voice carry over). Making language a media 158 attribute allows the standard session negotiation mechanism to handle 159 this by providing the information and mechanism for the endpoints to 160 make appropriate decisions. 162 Regarding relay services, in the case of an emergency call requiring 163 sign language such as ASL, there are two common approaches: the 164 caller initiates the call to a relay center, or the caller places the 165 call to emergency services (e.g., 911 in the U.S. or 112 in Europe). 166 In the former case, the language need is ancillary and supplemental. 167 In the latter case, the ESInet and/or PSAP may take the need for sign 168 language into account and bridge in a relay center. In this case, 169 the ESInet and PSAP have all the standard information available (such 170 as location) but are able to bridge the relay sooner in the call 171 processing. 173 By making this facility part of the end-to-end negotiation, the 174 question of which entity provides or engages the relay service 175 becomes separate from the call processing mechanics; if the caller 176 directs the call to a relay service then the human language 177 negotiation facility provides extra information to the relay service 178 but calls will still function without it; if the caller directs the 179 call to emergency services, then the ESInet/PSAP are able to take the 180 user's human language needs into account, e.g., by routing to a 181 particular PSAP or call taker or bridging a relay service or 182 translator. 184 The term "negotiation" is used here rather than "indication" because 185 human language (spoken/written/signed) is something that can be 186 negotiated in the same way as which forms of media (audio/text/video) 187 or which codecs. For example, if we think of non-emergency calls, 188 such as a user calling an airline reservation center, the user may 189 have a set of languages he or she speaks, with perhaps preferences 190 for one or a few, while the airline reservation center will support a 191 fixed set of languages. Negotiation should select the user's most 192 preferred language that is supported by the call center. Both sides 193 should be aware of which language was negotiated. This is 194 conceptually similar to the way other aspects of each media stream 195 are negotiated using SDP (e.g., media type and codecs). 197 2. Terminology 199 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 200 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 201 document are to be interpreted as described in RFC 2119 [RFC2119]. 203 3. Expected Use 205 This facility is expected to be used by NENA and 3GPP. NENA is 206 likely to reference it in NENA 08-01 (i3 Stage 3) in describing 207 attributes of calls presented to an ESInet, and in that or other 208 documents describing Policy-Based Routing (PBR) capabilities within a 209 Policy-Based Routing Function. 3GPP is expected to reference this 210 mechanism in general call handling and emergency call handling. 211 Recent CRs introduced in SA1 have anticipated this functionality 212 being provided within SDP. 214 4. Example Use Cases 216 4.1. Emergency Call from English Speaker in Spain 218 Someone who speaks only English is visiting Spain and places an 219 emergency (112) call. The call offers an audio stream using English. 220 The ESInet and PSAP have policy-based routing rules that take into 221 account the SDP language request when deciding how to route and 222 process the call. The ESInet routes the call to a PSAP within Spain 223 where an English-speaking call taker is available, and the PSAP 224 selects an English-speaking call taker to handle the call. The PSAP 225 answers the offer with an audio stream using English. The call is 226 established with an audio stream; the caller and call taker 227 communicate in English. 229 Alternatively, the ESInet routes the call to a cooperating PSAP 230 within the U.K. The PSAP answers the offer with an audio stream 231 using English. The call is established with an audio stream; the 232 caller and call taker communicate in English. (This approach is 233 similar to that envisioned in REACH112 Total Conversation.) 235 4.2. Emergency Call from Spanish/English Speaker in France 237 Someone who speaks both Spanish and English (but prefers Spanish) is 238 visiting France and places an emergency (112) call. The call offers 239 an audio stream listing first Spanish (meaning most preferred) and 240 then English. The ESInet and PSAP have policy-based routing rules 241 that take into account the SDP language request when deciding how to 242 route and process the call. The ESInet routes the call to a PSAP 243 within France where a Spanish-speaking call taker is available, and 244 the PSAP selects a Spanish-speaking call taker to handle the call. 245 The PSAP answers the offer with an audio stream listing Spanish. The 246 call is established with an audio stream; the caller and call taker 247 communicate in Spanish. 249 Alternatively, the ESInet routes the call to a cooperating PSAP in 250 Spain or England. (This approach is similar to that envisioned in 251 REACH112 Total Conversation.) 253 Alternatively, there is no ESInet or the ESInet does not take 254 language into account in its PBR. The call is routed to a PSAP in 255 France. The PSAP ignores the language information in the SDP offer, 256 and answers the offer with an audio stream with no language or with 257 French. The UE continues the call anyway. The call taker answers in 258 French, the user tries speaking Spanish and perhaps English. The 259 call taker bridges in a translation service or transfers the call to 260 a multilingual call taker. 262 4.3. Call to Call Center from Russian Speaker in U.S. 264 A Russian speaker is visiting the U.S. and places a call to her 265 airline reservation desk to inquire about her return flight. The 266 airline call processing system takes into account the SDP language 267 request and decides to route the call to its call center within 268 Russia. 270 Alternatively, if the airline call processing system does not look at 271 SDP, it uses the SIP "hint" if present. 273 4.4. Emergency Call from speech-impaired caller in the U.S. 275 Someone who uses English but is speech-impaired places an emergency 276 (911) call. The call offers an audio stream listing English and a 277 real-time text stream also using English. The ESInet and PSAP have 278 policy-based routing rules that take into account the SDP language 279 and media requests when deciding how to route and process the call. 280 The ESInet routes the call to a PSAP with real-time text 281 capabilities. The PSAP answers the offer with an audio stream 282 listing English and a real-time text stream listing English. The 283 call is established with an audio and a real-time text stream; the 284 caller and call taker communicate in English using voice from the 285 call-taker to the caller and text from the caller to the call taker. 286 The audio stream is two-way, allowing the call taker to hear 287 background sounds. 289 4.5. Emergency Call from deaf caller in the U.S. 291 A deaf caller who uses American Sign Language (ASL) places an 292 emergency (911) call. The call offers a video stream listing ASL and 293 an audio stream with no language indicated. The ESInet and PSAP have 294 policy-based routing rules that take into account the SDP language 295 and media needs when deciding how to route and process the call. The 296 ESInet routes the call to a PSAP. The PSAP answers the offer with an 297 audio stream listing English and a video stream listing ASL. The 298 PSAP bridges in a sign language interpreter. The call is established 299 with an audio and a video stream. 301 5. Desired Semantics 303 The desired solution is a media attribute that may be used within an 304 offer to indicate the preferred language of each media stream, and 305 within an answer to indicate the accepted language. The semantics of 306 including multiple values for a media stream within an offer is that 307 the languages are listed in order of preference. 309 (While it is true that a conversation among multilingual people often 310 involves multiple languages, the usefulness of providing a way to 311 negotiate this as a general facility is outweighed by the complexity 312 of the desired semantics of the SDP attribute to allow negotiation of 313 multiple simultaneous languages within an interactive media stream.) 315 6. The existing 'lang' attribute 317 RFC 4566 specifies an attribute 'lang' which sounds similar to what 318 is needed here, the difference being that it specifies that 'a=lang' 319 is declarative with the semantics of multiple 'lang' attributes being 320 that all of them are used, while we want a means to negotiate which 321 one is used in each stream. This difference means that the existing 322 'lang' attribute can't be used and we need to define a new attribute. 324 The text from RFC 4566 [RFC4566] is: 326 a=lang: 328 This can be a session-level attribute or a media-level attribute. 329 As a session-level attribute, it specifies the default language 330 for the session being described. As a media- level attribute, it 331 specifies the language for that media, overriding any session- 332 level language specified. Multiple lang attributes can be 333 provided either at session or media level if the session 334 description or media use multiple languages, in which case the 335 order of the attributes indicates the order of importance of the 336 various languages in the session or media from most important to 337 least important. 339 The "lang" attribute value must be a single [RFC3066] language tag 340 in US-ASCII [RFC3066]. It is not dependent on the charset 341 attribute. A "lang" attribute SHOULD be specified when a session 342 is of sufficient scope to cross geographic boundaries where the 343 language of recipients cannot be assumed, or where the session is 344 in a different language from the locally assumed norm. 346 Note that there are existing examples of it being used in exactly the 347 way we need. For example, draft-saintandre-sip-xmpp-chat-04 348 [I-D.saintandre-sip-xmpp-chat] contains an example where the initial 349 invitation contains two 'a=lang' entries for a media stream (for 350 English and Italian) and the OK accepts one of them (Italian), which 351 matches what we need: 353 Example: (F1) SIP user starts the session 355 INVITE sip:juliet@example.com SIP/2.0 356 To: 357 From: ;tag=576 358 Subject: Open chat with Romeo? 359 Call-ID: 742507no 360 Content-Type: application/sdp 362 c=IN IP4 s2x.example.net 363 m=message 7313 TCP/MSRP * 364 a=accept-types:text/plain 365 a=lang:en 366 a=lang:it 367 a=path:msrp://s2x.example.net:7313/ansp71weztas;tcp 369 Example: (F2) Gateway accepts session on Juliet's behalf 371 SIP/2.0 200 OK 372 To: ;tag=534 373 From: ;tag=576 374 Call-ID: 742507no 375 Content-Type: application/sdp 377 c=IN IP4 x2s.example.com 378 m=message 8763 TCP/MSRP * 379 a=accept-types:text/plain 380 a=lang:it 381 a=path:msrp://x2s.example.com:8763/lkjh37s2s20w2a;tcp 383 The example serves as evidence of the need for an SDP attribute with 384 the semantics as described in this document; unfortunately, the 385 existing 'lang' attribute is not it. 387 7. Proposed Solution 389 An SDP attribute seems the natural choice to negotiate human 390 (natural) language of an interactive media stream. The attribute 391 value should be a language tag per RFC 5646 [RFC5646] 393 7.1. New 'humintlang-send' and 'humintlang-recv' attributes 395 Rather than re-use 'lang' we define two new media-level attributes 396 starting with 'humintlang' (short for "human interactive language") 397 to negotiate which human language is used in each (interactive) media 398 stream. There are two attributes, one ending in "-send" and the 399 other in "-recv" to indicate the language used when sending and 400 receiving media: 402 a=humintlang-send: 404 a=humintlang-recv: 406 Each can appear multiple times in an offer for a media stream. 408 In an offer, the 'humintlang-send' values constitute a list in 409 preference order (first is most preferred) of the languages the 410 offerer wishes to send using the media, and the 'humintlang-recv' 411 values constitute a list in preference order of the languages the 412 offerer wishes to receive using the media. In cases where the user 413 wishes to use one media for sending and another for receiving (such 414 as a speech-impaired user who wishes to send using text and receive 415 using audio), one of the two MAY be unset. In cases where a media is 416 not primarily intended for language (for example, a video or audio 417 stream intended for background only) both SHOULD be unset. In other 418 cases, both SHOULD have the same values in the same order. The two 419 SHOULD NOT be set to languages which are difficult to match together 420 (e.g., specifying a desire to send audio in Hungarian and receive 421 audio in Portuguese will make it difficult to successfully complete 422 the call). 424 In an answer, 'humintlang-send' is the accepted language the answerer 425 will send (which in most cases is one of the languages in the offer's 426 'humintlang-recv'), and 'humintlang-recv' is the accepted language 427 the answerer expects to receive (which in most cases is one of the 428 languages in the offer's 'humintlang-send'). 430 Each value MUST be a language tag per RFC 5646 [RFC5646]. RFC 5646 431 describes mechanisms for matching language tags. While RFC 5646 432 provides a mechanism accommodating increasingly fine-grained 433 distinctions, in the interest of maximum interoperability for real- 434 time interactive communications, each 'humintlang-send' and 435 'humintlang-recv' value SHOULD be restricted to the largest 436 granularity of language tags; in other words, it is RECOMMENDED to 437 specify only a Primary-subtag and NOT to include subtags (e.g., for 438 region or dialect) unless the languages might be mutually 439 incomprehensible without them. 441 In an offer, each language tag value MAY have an asterisk appended as 442 the last character (after the registry value). The asterisk 443 indicates a request by the caller to not fail the call if there is no 444 language in common. See Section 7.2 for more information and 445 discussion. 447 When placing an emergency call, and in any other case where the 448 language cannot be assumed from context, each media stream in an 449 offer primarily intended for human language communication SHOULD 450 specify one or both 'humintlang-send' and 'humintlang-recv' 451 attributes (to avoid ambiguity). 453 Note that while signed language tags are used with a video stream to 454 indicate sign language, a spoken language tag for a video stream in 455 parallel with an audio stream with the same spoken language tag 456 indicates a request for a supplemental video stream to see the 457 speaker. 459 Clients acting on behalf of end users are expected to set one or both 460 'humintlang-send' and 'humintlang-recv' attributes on each media 461 stream primarily intended for human communication in an offer when 462 placing an outgoing session, but either ignore or take into 463 consideration the attributes when receiving incoming calls, based on 464 local configuration and capabilities. Systems acting on behalf of 465 call centers and PSAPs are expected to take into account the values 466 when processing inbound calls. 468 7.2. Advisory vs Required 470 One important consideration with this mechanism is if the call fails 471 if the callee does not support any of the languages requested by the 472 caller. 474 In order to provide for maximum likelihood of a successful 475 communication session, especially in the case of emergency calling, 476 the mechanism defined here provides a way for the caller to indicate 477 a preference for the call failing or succeeding when there is no 478 language in common. However, the callee is NOT REQUIRED to honor 479 this preference. For example, a PSAP MAY choose to attempt the call 480 even with no language in common, while a corporate call center MAY 481 choose to fail the call. 483 The mechanism for indicating this preference is that, in an offer, if 484 the last character of any of the 'humintlang-recv' or 'humintlang- 485 send' values is an asterisk, this indicates a request to not fail the 486 call (similar to SIP Accept-Language syntax). Either way, the called 487 party MAY ignore this, e.g., for the emergency services use case, a 488 PSAP will likely not fail the call. 490 7.3. SIP "hint" 492 SDP is used for stream negotiation, and emergency services based on 493 NENA i3 have the ability to reference SDP within policy-based routing 494 rules. However, it is possible that some entities wishing to take 495 the caller's language needs into account may lack this ability. 496 Accordingly, a SIP header is provided to supply a "hint" regarding 497 the caller's language needs. This is merely a hint, not actual 498 negotiation. 500 Protocols other than SIP may be used to establish interactive 501 communication sessions; this document does not provide a "hint" 502 mechanism for any protocol other than SIP. 504 TBD: The SIP header used for the "hint" is either a new header or 505 caller preferences (RFC 3841), presumably Accept-Contact (where the 506 caller sets media and language and optionally required; it's not yet 507 clear if this matches our desired semantics. 509 This SIP header is just a hint; it is RECOMMENDED to be sent; it is 510 NOT REQUIRED to be sent or to be used on receipt. 512 In the case of spoken languages using an audio stream, this "hint" 513 regarding language may be sufficient to allow the callee to optimally 514 handle the call, but since the "hint" only deals with language, not 515 media type, it is not sufficient when the caller requests non-audio 516 media such as text or video. 518 7.4. Silly States 519 It is possible to specify a "silly state" where the language 520 specified does not make sense for the media type, such as specifying 521 a signed language for an audio media stream. 523 An offer MUST NOT be created where the language does not make sense 524 for the media type. If such an offer is received, the receiver MAY 525 reject the media, ignore the language specified, or attempt to 526 interpret the intent (e.g., if American Sign Language is specified 527 for an audio media stream, this might be interpreted as a desire to 528 use spoken English). 530 A spoken language tag for a video stream in conjunction with an audio 531 stream with the same language might indicate a request for 532 supplemental video to see the speaker. 534 8. Alternative Proposal: Caller-prefs 536 An alternative proposal is to use the SIP-level Caller Preferences 537 mechanism from RFC 3840 [RFC3840] and RFC 3841 [RFC3841] rather than 538 the SDP-level mechanism described above. 540 The Caller-prefs mechanism includes a priority system by which 541 different combinations of media and languages can be assigned 542 different priorities. The evaluation and decisions on what to do 543 with the call can be done either by proxies along the call path, or 544 by the addressed UA. Evaluation of alternatives for routing is 545 described in RFC 3841 [RFC3841]. 547 8.1. Use of Caller Preferences Without Additions 549 The following is possible without adding any new registered tags: 551 Potential callers and recipients MAY include in the Contact field in 552 their SIP registrations media and language tags according to the 553 joint capabilities of the UA and the human user according to RFC 3840 554 [RFC3840]. 556 The most relevant media capability tags are "video", "text" and 557 "audio". Each tag represents a capability to use the media in two- 558 way communication. 560 Language capabilities are declared with a comma-separated list of 561 languages that can be used in the call as parameters to the tag 562 "language=". 564 This is an example of how it is used in a SIP REGISTER: 566 REGISTER user@example.net 568 Contact: audio; video; text; 569 language="en,es,ase" 571 Including this information in SIP REGISTER allows proxies to act on 572 the information. For the problem set addressed by this document, it 573 is not anticipated that proxies will do so using registration data. 574 Further, there are classes of devices (such as cellular mobile 575 phones) that are not anticipated to include this information in their 576 registrations. Hence, use in registration is OPTIONAL. 578 In a call, a list of acceptable media and language combinations is 579 declared, and a priority assigned to each combination. 581 This is done by the Accept-Contact header field, which defines 582 different combinations of media and languages and assigns priorities 583 for completing the call with the SIP URI represented by that Contact. 584 A priority is assigned to each set as a so-called "q-value" which 585 ranges from 1 (most preferred) to 0 (least preferred). 587 Using the Accept-Contact header field in INVITE requests and 588 responses allows these capabilities to be expressed and used during 589 call set-up. Clients SHOULD include this information in INVITE 590 requests and responses. 592 Example: 594 Accept-Contact: *; text; language="en"; q=0.2 596 Accept-Contact: *; video; language="ase"; q=0.8 598 This example shows the highest preference expressed by the caller is 599 to use video with American Sign Language (language code "ase"). As a 600 fallback, it is acceptable to get the call connected with only 601 English text used for human communication. Other media may of course 602 be connected as well, without expectation that it will be usable by 603 the caller for interactive communications (but may still be helpful 604 to the caller). 606 This system satisfies all the needs described in the previous 607 chapters, except that language specifications do not make any 608 distinction between spoken and written language, and that the need 609 for directionality in the specification cannot be fulfilled. 611 To some degree the lack of media specification between speech and 612 text in language tags can be compensated by only specifying the 613 important medium in the Accept-Contact field. 615 Thus, a user who wants to use English mainly for text would specify: 617 Accept-Contact: *;text;language="en";q=1.0 619 While a user who wants to use English mainly for speech but accept it 620 for text would specify: 622 Accept-Contact: *;audio;language="en";q=0.8 624 Accept-Contact: *;text;language="en";q=0.2 626 However, a user who would like to talk, but receive text back has no 627 way to do it with the existing specification. 629 8.2. Additional Caller Preferences for Asymmetric Needs 631 In order to be able to specify asymmetric preferences, there are two 632 possibilities. Either new language tags in the style of the 633 humintlang parameters described above for SDP could be registered, or 634 additional media tags describing the asymmetry could be registered. 636 8.2.1. Caller Preferences for Asymmetric Modality Needs 638 The following new media tags should be defined: 640 speech-receive 642 speech-send 644 text-receive 646 text-send 648 sign-send 650 sign-receive 652 A user who prefers to talk and get text in return in English would 653 register the following (if including this information in registration 654 data): 656 REGISTER user@example.net 658 Contact: audio;text;speech-send;text- 659 receive;language="en" 661 At call time, a user who prefers to talk and get text in return in 662 English would set the Accept-Contact header field to: 664 Accept-Contact: *; audio; text; speech-receive; text-send; 665 language="en";q=0.8 667 Accept-Contact: *; text; language="en"; q=0.2 669 Note that the directions specified here are as viewed from the callee 670 side to match what the callee has registered. 672 A bridge arranged for invoking a relay service specifically arranged 673 for captioned telephony would register the following for supporting 674 calling users: 676 REGISTER ct@ctrelay.net 678 Contact: audio; text; speech-receive; 679 text-send; language="en" 681 A bridge arranged for invoking a relay service specifically arranged 682 for captioned telephony would register the following for supporting 683 called users: 685 REGISTER ct@ctrelay.net 687 Contact: audio; text; speech-send; text- 688 receive; language="en" 690 At call time, these alternatives are included in the list of possible 691 outcome of the call routing by the SIP proxies and the proper relay 692 service is invoked. 694 8.2.2. Caller Preferences for Asymmetric Language Tags 696 An alternative is to register new language tags for the purpose of 697 asymmetric language usage. 699 Instead of using "language=", six new language tags would be 700 registered: 702 humintlang-text-recv 704 humintlang-text-send 706 humintlang-speech-recv 708 humintlang-speech-send 710 humintlang-sign-recv 712 humintlang-sign-send 714 These language tags would be used instead of the regular 715 bidirectional language tags, and users with bidirectional 716 capabilities SHOULD specify values for both directions. Services 717 specifically arranged for supporting users with asymmetric needs 718 SHOULD specify only the asymmetry they support. 720 9. IANA Considerations 722 IANA is kindly requested to add two entries to the 'att-field (media 723 level only)' table of the SDP parameters registry: 725 +------------------------------+-----------------+-----------------+ 726 | Type | Name | Reference | 727 +------------------------------+-----------------+-----------------+ 728 | att-field (media level only) | humintlang-send | (this document) | 729 | att-field (media level only) | humintlang-recv | (this document) | 730 +------------------------------+-----------------+-----------------+ 732 Table 1: att-field (media level only)' entries 734 10. Security Considerations 736 The Security Considerations of RFC 5646 [RFC5646] apply here (as a 737 use of that RFC). In addition, if the 'humintlang-send' or 738 'humintlang-recv' values are altered or deleted en route, the session 739 could fail or languages incomprehensible to the caller could be 740 selected; however, this is also a risk if any SDP parameters are 741 modified en route. 743 11. Changes from Previous Versions 745 11.1. Changes from draft-gellens-...-00 to -01 746 o Changed name of (possible) new attribute from 'humlang" to 747 "humintlang" 749 o Added discussion of silly state (language not appropriate for 750 media type) 752 o Added Voice Carry Over example 754 o Added mention of multilingual people and multiple languages 756 o Minor text clarifications 758 11.2. Changes from draft-gellens-...-01 to -02 760 o Updated text for (possible) new attribute "humintlang" to 761 reference RFC 5646 763 o Added clarifying text for (possible) re-use of existing 'lang' 764 attribute saying that the registration would be updated to reflect 765 different semantics for multiple values for interactive versus 766 non-interactive media. 768 o Added clarifying text for (possible) new attribute "humintlang" to 769 attempt to better describe the role of language tags in media in 770 an offer and an answer. 772 11.3. Changes from draft-gellens-...-02 to draft-gellens-mmusic-...-00 774 o Updated text to refer to RFC 5646 rather than the IANA language 775 subtags registry directly. 777 o Moved discussion of existing 'lang' attribute out of "Proposed 778 Solution" section and into own section now that it is not part of 779 proposal. 781 o Updated text about existing 'lang' attribute. 783 o Added example use cases. 785 o Replaced proposed single 'humintlang' attribute with 'humintlang- 786 send' and 'humintlang-recv' per Harald's request/information that 787 it was a misuse of SDP to use the same attribute for sending and 788 receiving. 790 o Added section describing usage being advisory vs required and text 791 in attribute section. 793 o Added section on SIP "hint" header (not yet nailed down between 794 new and existing header). 796 o Added text discussing usage in policy-based routing function or 797 use of SIP header "hint" if unable to do so. 799 o Added SHOULD that the value of the parameters stick to the largest 800 granularity of language tags. 802 o Added text to Introduction to be try and be more clear about 803 purpose of document and problem being solved. 805 o Many wording improvements and clarifications throughout the 806 document. 808 o Filled in Security Considerations. 810 o Filled in IANA Considerations. 812 o Added to Acknowledgments those who participated in the Orlando ad- 813 hoc discussion as well as those who participated in email 814 discussion and side one-on-one discussions. 816 11.4. Changes from draft-gellens-mmusic-...-00 to -01 818 o Relaxed language on setting -send and -receive to same values; 819 added text on leaving on empty to indicate asymmetric usage. 821 o Added text that clients on behalf of end users are expected to set 822 the attributes on outgoing calls and ignore on incoming calls 823 while systems on behalf of call centers and PSAPs are expected to 824 take the attributes into account when processing incoming calls. 826 11.5. Changes from draft-gellens-mmusic-...-01 to -02 828 o Added clarifying text on leaving attributes unset for media not 829 primarily intended for human language communication (e.g., 830 background audio or video). 832 o Added new section Section 8 ("Alternative Proposal: Caller-prefs") 833 discussing use of SIP-level Caller-prefs instead of SDP-level. 835 12. Contributors 837 Gunnar Hellstrom deserves special mention for his reviews, 838 assistance, and especially for contributing the text in Section 8 on 839 the alternative proposal using SIP-level Caller-prefs. 841 13. Acknowledgments 843 Many thanks to Bernard Aboba, Harald Alvestrand, Flemming Andreasen, 844 Francois Audet, Eric Burger, Keith Drage, Doug Ewell, Christian 845 Groves, Andrew Hutton, Hadriel Kaplan, Ari Keranen, John Klensin, 846 Paul Kyzivat, John Levine, Alexey Melnikov, James Polk, Pete Resnick, 847 Peter Saint-Andre, and Dale Worley for reviews, corrections, 848 suggestions, and participating in in-person and email discussions. 850 14. References 852 14.1. Normative References 854 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 855 Requirement Levels", BCP 14, RFC 2119, March 1997. 857 [RFC3840] Rosenberg, J., Schulzrinne, H., and P. Kyzivat, 858 "Indicating User Agent Capabilities in the Session 859 Initiation Protocol (SIP)", RFC 3840, August 2004. 861 [RFC3841] Rosenberg, J., Schulzrinne, H., and P. Kyzivat, "Caller 862 Preferences for the Session Initiation Protocol (SIP)", 863 RFC 3841, August 2004. 865 [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 866 Description Protocol", RFC 4566, July 2006. 868 [RFC5646] Phillips, A. and M. Davis, "Tags for Identifying 869 Languages", BCP 47, RFC 5646, September 2009. 871 14.2. Informational References 873 [I-D.iab-privacy-considerations] 874 Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 875 Morris, J., Hansen, M., and R. Smith, "Privacy 876 Considerations for Internet Protocols", draft-iab-privacy- 877 considerations-09 (work in progress), May 2013. 879 [I-D.saintandre-sip-xmpp-chat] 880 Saint-Andre, P., Loreto, S., Gavita, E., and N. Hossain, 881 "Interworking between the Session Initiation Protocol 882 (SIP) and the Extensible Messaging and Presence Protocol 883 (XMPP): One-to-One Text Chat", draft-saintandre-sip-xmpp- 884 chat-06 (work in progress), June 2013. 886 [RFC3066] Alvestrand, H., "Tags for the Identification of 887 Languages", RFC 3066, January 2001. 889 Author's Address 891 Randall Gellens 892 Qualcomm Technologies Inc. 893 5775 Morehouse Drive 894 San Diego, CA 92121 895 US 897 Email: rg+ietf@qti.qualcomm.com