idnits 2.17.1 draft-ietf-avtcore-multi-party-rtt-mix-16.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: ---------------------------------------------------------------------------- == There is 1 instance of lines with non-ascii characters in the document. 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 document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). (Using the creation date from RFC4103, updated by this document, for RFC5378 checks: 2003-11-21) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (1 May 2021) is 1084 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) == Missing Reference: 'Bob' is mentioned on line 1507, but not defined -- Possible downref: Non-RFC (?) normative reference: ref. 'T140' -- Possible downref: Non-RFC (?) normative reference: ref. 'T140ad1' Summary: 0 errors (**), 0 flaws (~~), 4 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 AVTCore G. Hellstrom 3 Internet-Draft Gunnar Hellstrom Accessible Communication 4 Updates: 4103 (if approved) 1 May 2021 5 Intended status: Standards Track 6 Expires: 2 November 2021 8 RTP-mixer formatting of multi-party Real-time text 9 draft-ietf-avtcore-multi-party-rtt-mix-16 11 Abstract 13 Enhancements for RFC 4103 real-time text mixing are provided in this 14 document, suitable for a centralized conference model that enables 15 source identification and rapidly interleaved transmission of text 16 from different sources. The intended use is for real-time text 17 mixers and participant endpoints capable of providing an efficient 18 presentation or other treatment of a multi-party real-time text 19 session. The specified mechanism builds on the standard use of the 20 CSRC list in the RTP packet for source identification. The method 21 makes use of the same "text/t140" and "text/red" formats as for two- 22 party sessions. 24 Solutions using multiple RTP streams in the same RTP session are 25 briefly mentioned, as they could have some benefits over the RTP- 26 mixer model. The possibility to implement the solution in a wide 27 range of existing RTP implementations made the RTP-mixer model be 28 selected to be fully specified in this document. 30 A capability exchange is specified so that it can be verified that a 31 mixer and a participant can handle the multi-party coded real-time 32 text stream using the RTP-mixer method. The capability is indicated 33 by use of an SDP media attribute "rtt-mixer". 35 The document updates RFC 4103 "RTP Payload for Text Conversation". 37 A specification of how a mixer can format text for the case when the 38 endpoint is not multi-party aware is also provided. 40 Status of This Memo 42 This Internet-Draft is submitted in full conformance with the 43 provisions of BCP 78 and BCP 79. 45 Internet-Drafts are working documents of the Internet Engineering 46 Task Force (IETF). Note that other groups may also distribute 47 working documents as Internet-Drafts. The list of current Internet- 48 Drafts is at https://datatracker.ietf.org/drafts/current/. 50 Internet-Drafts are draft documents valid for a maximum of six months 51 and may be updated, replaced, or obsoleted by other documents at any 52 time. It is inappropriate to use Internet-Drafts as reference 53 material or to cite them other than as "work in progress." 55 This Internet-Draft will expire on 2 November 2021. 57 Copyright Notice 59 Copyright (c) 2021 IETF Trust and the persons identified as the 60 document authors. All rights reserved. 62 This document is subject to BCP 78 and the IETF Trust's Legal 63 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 64 license-info) in effect on the date of publication of this document. 65 Please review these documents carefully, as they describe your rights 66 and restrictions with respect to this document. Code Components 67 extracted from this document must include Simplified BSD License text 68 as described in Section 4.e of the Trust Legal Provisions and are 69 provided without warranty as described in the Simplified BSD License. 71 Table of Contents 73 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 74 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 6 75 1.2. Selected solution and considered alternatives . . . . . . 6 76 1.3. Intended application . . . . . . . . . . . . . . . . . . 9 77 2. Overview of the two specified solutions and selection of 78 method . . . . . . . . . . . . . . . . . . . . . . . . . 10 79 2.1. The RTP-mixer based solution for multi-party aware 80 endpoints . . . . . . . . . . . . . . . . . . . . . . . . 10 81 2.2. Mixing for multi-party unaware endpoints . . . . . . . . 10 82 2.3. Offer/answer considerations . . . . . . . . . . . . . . . 11 83 2.4. Actions depending on capability negotiation result . . . 12 84 3. Details for the RTP-mixer based mixing method for multi-party 85 aware endpoints . . . . . . . . . . . . . . . . . . . . . 13 86 3.1. Use of fields in the RTP packets . . . . . . . . . . . . 13 87 3.2. Initial transmission of a BOM character . . . . . . . . . 13 88 3.3. Keep-alive . . . . . . . . . . . . . . . . . . . . . . . 14 89 3.4. Transmission interval . . . . . . . . . . . . . . . . . . 14 90 3.5. Only one source per packet . . . . . . . . . . . . . . . 14 91 3.6. Do not send received text to the originating source . . . 14 92 3.7. Clean incoming text . . . . . . . . . . . . . . . . . . . 15 93 3.8. Redundant transmission principles . . . . . . . . . . . . 15 94 3.9. Interleaving text from different sources . . . . . . . . 15 95 3.10. Text placement in packets . . . . . . . . . . . . . . . . 15 96 3.11. Empty T140blocks . . . . . . . . . . . . . . . . . . . . 16 97 3.12. Creation of the redundancy . . . . . . . . . . . . . . . 16 98 3.13. Timer offset fields . . . . . . . . . . . . . . . . . . . 17 99 3.14. Other RTP header fields . . . . . . . . . . . . . . . . . 17 100 3.15. Pause in transmission . . . . . . . . . . . . . . . . . . 17 101 3.16. RTCP considerations . . . . . . . . . . . . . . . . . . . 17 102 3.17. Reception of multi-party contents . . . . . . . . . . . . 18 103 3.18. Performance considerations . . . . . . . . . . . . . . . 20 104 3.19. Security for session control and media . . . . . . . . . 20 105 3.20. SDP offer/answer examples . . . . . . . . . . . . . . . . 21 106 3.21. Packet sequence example from interleaved transmission . . 22 107 3.22. Maximum character rate "CPS" . . . . . . . . . . . . . . 25 108 4. Presentation level considerations . . . . . . . . . . . . . . 25 109 4.1. Presentation by multi-party aware endpoints . . . . . . . 26 110 4.2. Multi-party mixing for multi-party unaware endpoints . . 28 111 5. Relation to Conference Control . . . . . . . . . . . . . . . 34 112 5.1. Use with SIP centralized conferencing framework . . . . . 34 113 5.2. Conference control . . . . . . . . . . . . . . . . . . . 34 114 6. Gateway Considerations . . . . . . . . . . . . . . . . . . . 34 115 6.1. Gateway considerations with Textphones (e.g. TTYs). . . 34 116 6.2. Gateway considerations with WebRTC. . . . . . . . . . . . 35 117 7. Updates to RFC 4103 . . . . . . . . . . . . . . . . . . . . . 36 118 8. Congestion considerations . . . . . . . . . . . . . . . . . . 36 119 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 36 120 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 36 121 10.1. Registration of the "rtt-mixer" SDP media attribute . . 36 122 11. Security Considerations . . . . . . . . . . . . . . . . . . . 37 123 12. Change history . . . . . . . . . . . . . . . . . . . . . . . 38 124 12.1. Changes included in 125 draft-ietf-avtcore-multi-party-rtt-mix-16 . . . . . . . 38 126 12.2. Changes included in 127 draft-ietf-avtcore-multi-party-rtt-mix-15 . . . . . . . 38 128 12.3. Changes included in 129 draft-ietf-avtcore-multi-party-rtt-mix-14 . . . . . . . 38 130 12.4. Changes included in 131 draft-ietf-avtcore-multi-party-rtt-mix-13 . . . . . . . 39 132 12.5. Changes included in 133 draft-ietf-avtcore-multi-party-rtt-mix-12 . . . . . . . 39 134 12.6. Changes included in 135 draft-ietf-avtcore-multi-party-rtt-mix-11 . . . . . . . 39 136 12.7. Changes included in 137 draft-ietf-avtcore-multi-party-rtt-mix-10 . . . . . . . 39 138 12.8. Changes included in 139 draft-ietf-avtcore-multi-party-rtt-mix-09 . . . . . . . 40 140 12.9. Changes included in 141 draft-ietf-avtcore-multi-party-rtt-mix-08 . . . . . . . 40 142 12.10. Changes included in 143 draft-ietf-avtcore-multi-party-rtt-mix-07 . . . . . . . 40 144 12.11. Changes included in 145 draft-ietf-avtcore-multi-party-rtt-mix-06 . . . . . . . 40 147 12.12. Changes included in 148 draft-ietf-avtcore-multi-party-rtt-mix-05 . . . . . . . 41 149 12.13. Changes included in 150 draft-ietf-avtcore-multi-party-rtt-mix-04 . . . . . . . 41 151 12.14. Changes included in 152 draft-ietf-avtcore-multi-party-rtt-mix-03 . . . . . . . 41 153 12.15. Changes included in 154 draft-ietf-avtcore-multi-party-rtt-mix-02 . . . . . . . 42 155 12.16. Changes to draft-ietf-avtcore-multi-party-rtt-mix-01 . . 42 156 12.17. Changes from 157 draft-hellstrom-avtcore-multi-party-rtt-source-03 to 158 draft-ietf-avtcore-multi-party-rtt-mix-00 . . . . . . . 42 159 12.18. Changes from 160 draft-hellstrom-avtcore-multi-party-rtt-source-02 to 161 -03 . . . . . . . . . . . . . . . . . . . . . . . . . . 43 162 12.19. Changes from 163 draft-hellstrom-avtcore-multi-party-rtt-source-01 to 164 -02 . . . . . . . . . . . . . . . . . . . . . . . . . . 43 165 12.20. Changes from 166 draft-hellstrom-avtcore-multi-party-rtt-source-00 to 167 -01 . . . . . . . . . . . . . . . . . . . . . . . . . . 44 168 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 44 169 13.1. Normative References . . . . . . . . . . . . . . . . . . 44 170 13.2. Informative References . . . . . . . . . . . . . . . . . 46 171 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 46 173 1. Introduction 175 "RTP Payload for Text Conversation" [RFC4103] specifies use of RTP 176 [RFC3550] for transmission of real-time text (RTT) and the "text/ 177 t140" format. It also specifies a redundancy format "text/red" for 178 increased robustness. The "text/red" format is registered in 179 [RFC4102]. 181 Real-time text is usually provided together with audio and sometimes 182 with video in conversational sessions. 184 A requirement related to multi-party sessions from the presentation 185 level standard T.140 [T140] for real-time text is: "The display of 186 text from the members of the conversation should be arranged so that 187 the text from each participant is clearly readable, and its source 188 and the relative timing of entered text is visualized in the 189 display." 191 Another requirement is that the mixing procedure must not introduce 192 delays in the text streams that are experienced to be disturbing the 193 real-time experience of the receiving users. 195 Use of RTT is increasing, and specifically, use in emergency calls is 196 increasing. Emergency call use requires multi-party mixing. RFC 197 4103 "RTP Payload for Text Conversation" mixer implementations can 198 use traditional RTP functions for source identification, but the 199 performance of the mixer when giving turns for the different sources 200 to transmit is limited when using the default transmission 201 characteristics with redundancy. 203 The redundancy scheme of [RFC4103] enables efficient transmission of 204 earlier transmitted redundant text in packets together with new text. 205 However the redundancy header format has no source indicators for the 206 redundant transmissions. The redundant parts in a packet must 207 therefore be from the same source as the new text. The recommended 208 transmission is one new and two redundant generations of text 209 (T140blocks) in each packet and the recommended transmission interval 210 for two-party use is 300 ms. 212 Real-time text mixers for multi-party sessions need to include the 213 source with each transmitted group of text from a conference 214 participant so that the text can be transmitted interleaved with text 215 groups from different sources in the rate they are created. This 216 enables the text groups to be presented by endpoints in suitable 217 grouping with other text from the same source. 219 The presentation can then be arranged so that text from different 220 sources can be presented in real-time and easily read. At the same 221 time it is possible for a reading user to perceive approximately when 222 the text was created in real time by the different parties. The 223 transmission and mixing is intended to be done in a general way so 224 that presentation can be arranged in a layout decided by the 225 endpoint. 227 There are existing implementations of RFC 4103 in endpoints without 228 the updates from this document. These will not be able to receive 229 and present real-time text mixed for multi-party aware endpoints. 231 A negotiation mechanism is therefore needed for verification if the 232 parties are able to handle a common method for multi-party 233 transmission and agreeing on using that method. 235 A fall-back mixing procedure is also needed for cases when the 236 negotiation result indicates that a receiving endpoint is not capable 237 of handling the mixed format. Multi-party unaware endpoints would 238 possibly otherwise present all received multi-party mixed text as if 239 it came from the same source regardless of any accompanying source 240 indication coded in fields in the packet. Or they may have any other 241 undesirable way of acting on the multi-party content. The fall-back 242 method is called the mixing procedure for multi-party unaware 243 endpoints. The fall-back method is naturally not expected to meet 244 all performance requirements placed on the mixing procedure for 245 multi-party aware endpoints. 247 The document updates [RFC4103] by introducing an attribute for 248 indicating capability for the RTP-mixer based multi-party mixing case 249 and rules for source indications and interleaving of text from 250 different sources. 252 1.1. Terminology 254 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 255 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 256 "OPTIONAL" in this document are to be interpreted as described in BCP 257 14 [RFC2119] [RFC8174] when, and only when, they appear in all 258 capitals, as shown above. 260 The terms SDES, CNAME, NAME, SSRC, CSRC, CSRC list, CC, RTCP, RTP- 261 mixer, RTP-translator are defined in [RFC3550]. 263 The term "T140block" is defined in [RFC4103] to contain one or more 264 T.140 code elements. 266 "TTY" stands for a text telephone type used in North America. 268 "WebRTC" stands for web based communication specified by W3C and 269 IETF. See [RFC8825]. 271 "DTLS-SRTP" stands for security specified in [RFC5764]. 273 "multi-party aware" stands for an endpoint receiving real-time text 274 from multiple sources through a common conference mixer being able to 275 present the text in real-time separated by source and presented so 276 that a user can get an impression of the approximate relative timing 277 of text from different parties. 279 "multi-party unaware" stands for an endpoint not itself being able to 280 separate text from different sources when received through a common 281 conference mixer. 283 1.2. Selected solution and considered alternatives 285 A number of alternatives were considered when searching an efficient 286 and easily implemented multi-party method for real-time text. This 287 section explains a few of them briefly. 289 Multiple RTP streams, one per participant. 290 One RTP stream per source would be sent in the same RTP session 291 with the "text/red" format. From some points of view, use of 292 multiple RTP streams, one for each source, sent in the same RTP 293 session would be efficient, and would use exactly the same packet 294 format as [RFC4103] and the same payload type. A couple of 295 relevant scenarios using multiple RTP-streams are specified in 296 "RTP Topologies" [RFC7667]. One possibility of special interest 297 is the Selective Forwarding Middlebox (SFM) topology specified in 298 RFC 7667 section 3.7 that could enable end to end encryption. In 299 contrast to audio and video, real-time text is only transmitted 300 when the users actually transmit information. Thus an SFM 301 solution would not need to exclude any party from transmission 302 under normal conditions. In order to allow the mixer to convey 303 the packets with the payload preserved and encrypted, an SFM 304 solution would need to act on some specific characteristics of the 305 "text/red" format. The redundancy headers are part of the 306 payload, so the receiver would need to just assume that the 307 payload type number in the redundancy header is for "text/t140". 308 The characters per second parameter (CPS) would need to act per 309 stream. The relation between the SSRC and the source would need 310 to be conveyed in some specified way, e.g. in the CSRC. Recovery 311 and loss detection would preferably be based on sequence number 312 gap detection. Thus sequence number gaps in the incoming stream 313 to the mixer would need to be reflected in the stream to the 314 participant and no new gaps created by the mixer. However, the 315 RTP implementation in both mixers and endpoints need to support 316 multiple streams in the same RTP session in order to use this 317 mechanism. For best deployment opportunity, it should be possible 318 to upgrade existing endpoint solutions to be multi-party aware 319 with a reasonable effort. There is currently a lack of support 320 for multi-stream RTP in certain implementation technologies. This 321 fact made this solution only briefly mentioned in this document as 322 an option for further study. 324 RTP-mixer based method for multi-party aware endpoints. 325 The "text/red" format in RFC 4103 is sent with shorter 326 transmission interval with the RTP-mixer method and indicating 327 source in CSRC. The "text/red" format with "text/t140" payload in 328 a single RTP stream can be sent when text is available from the 329 call participants instead of at the regular 300 ms. The source is 330 indicated in the CSRC field. Transmission of packets with text 331 from different sources can then be done smoothly while 332 simultaneous transmission occurs as long as it is not limited by 333 the maximum character rate "CPS". With ten participants sending 334 text simultaneously, the switching and transmission performance is 335 good. With more simultaneously sending participants, and 336 receivers with default capacity there will be a noticeable 337 jerkiness and delay in text presentation. The jerkiness will be 338 more expressed the more participants who send text simultaneously. 339 Two seconds jerkiness will be noticeable and slightly unpleasant, 340 but corresponds in time to what typing humans often cause by 341 hesitation or changing position while typing. A benefit of this 342 method is that no new packet format needs to be introduced and 343 implemented. Since simultaneous typing by more than two parties 344 is very rare, this method can be used successfully with good 345 performance. Recovery of text in case of packet loss is based on 346 analysis of timestamps of received redundancy versus earlier 347 received text. Negotiation is based on a new SDP media attribute 348 "rtt-mixer". This method is selected to be the main one specified 349 in this document. 351 Multiple sources per packet. 352 A new "text" media subtype would be specified with up to 15 353 sources in each packet. The mechanism would make use of the RTP 354 mixer model specified in RTP [RFC3550]. Text from up to 15 355 sources can be included in each packet. Packets are normally sent 356 every 300 ms. The mean delay will be 150 ms. The sources are 357 indicated in strict order in the CSRC list of the RTP packets. A 358 new redundancy packet format is specified. This method would 359 result in good performance, but would require standardisation and 360 implementation of new releases in the target technologies that 361 would take more time than desirable to complete. It was therefore 362 not selected to be included in this document. 364 Mixing for multi-party unaware endpoints 365 Presentation of text from multiple parties is prepared by the 366 mixer in one single stream. It is desirable to have a method that 367 does not require any modifications in existing user devices 368 implementing RFC 4103 for RTT without explicit support of multi- 369 party sessions. This is possible by having the mixer insert a new 370 line and a text formatted source label before each switch of text 371 source in the stream. Switch of source can only be done in places 372 in the text where it does not disturb the perception of the 373 contents. Text from only one source can be presented in real time 374 at a time. The delay will therefore be varying. The method also 375 has other limitations, but is included in this document as a 376 fallback method. In calls where parties take turns properly by 377 ending their entries with a new line, the limitations will have 378 limited influence on the user experience. while only two parties 379 send text, these two will see the text in real time with no delay. 380 This method is specified as a fallback method in this document. 382 RTT transport in WebRTC 383 Transport of real-time text in the WebRTC technology is specified 384 to use the WebRTC data channel in [RFC8865]. That specification 385 contains a section briefly describing its use in multi-party 386 sessions. The focus of this document is RTP transport. 387 Therefore, even if the WebRTC transport provides good multi-party 388 performance, it is just mentioned in this document in relation to 389 providing gateways with multi-party capabilities between RTP and 390 WebRTC technologies. 392 1.3. Intended application 394 The method for multi-party real-time text specified in this document 395 is primarily intended for use in transmission between mixers and 396 endpoints in centralised mixing configurations. It is also 397 applicable between mixers. An often mentioned application is for 398 emergency service calls with real-time text and voice, where a 399 calltaker wants to make an attended handover of a call to another 400 agent, and stay observing the session. Multimedia conference 401 sessions with support for participants to contribute in text is 402 another application. Conferences with central support for speech-to- 403 text conversion is yet another mentioned application. 405 In all these applications, normally only one participant at a time 406 will send long text utterances. In some cases, one other participant 407 will occasionally contribute with a longer comment simultaneously. 408 That may also happen in some rare cases when text is interpreted to 409 text in another language in a conference. Apart from these cases, 410 other participants are only expected to contribute with very brief 411 utterings while others are sending text. 413 Users expect that the text they send is presented in real-time in a 414 readable way to the other participants even if they send 415 simultaneously with other users and even when they make brief edit 416 operations of their text by backspacing and correcting their text. 418 Text is supposed to be human generated, by some text input means, 419 such as typing on a keyboard or using speech-to-text technology. 420 Occasional small cut-and-paste operations may appear even if that is 421 not the initial purpose of real-time text. 423 The real-time characteristics of real-time text is essential for the 424 participants to be able to contribute to a conversation. If the text 425 is too much delayed from typing a letter to its presentation, then, 426 in some conference situations, the opportunity to comment will be 427 gone and someone else will grab the turn. A delay of more than one 428 second in such situations is an obstacle for good conversation. 430 2. Overview of the two specified solutions and selection of method 432 This section contains a brief introduction of the two methods 433 specified in this document. 435 2.1. The RTP-mixer based solution for multi-party aware endpoints 437 This method specifies negotiated use of the RFC 4103 format for 438 multi-party transmission in a single RTP stream. The main purpose of 439 this document is to specify a method for true multi-party real-time 440 text mixing for multi-party aware endpoints that can be widely 441 deployed. The RTP-mixer based method makes use of the current format 442 for real-time text in [RFC4103]. It is an update of RFC 4103 by a 443 clarification on one way to use it in the multi-party situation. 444 That is done by completing a negotiation for this kind of multi-party 445 capability and by interleaving packets from different sources. The 446 source is indicated in the CSRC element in the RTP packets. Specific 447 considerations are made to be able to recover text after packet loss. 449 The detailed procedures for the RTP-mixer based multi-party aware 450 case are specified in Section 3. 452 Please use [RFC4103] as reference when reading the specification. 454 2.2. Mixing for multi-party unaware endpoints 456 A method is also specified in this document for cases when the 457 endpoint participating in a multi-party call does not itself 458 implement any solution, or not the same, as the mixer. The method 459 requires the mixer to insert text dividers and readable labels and 460 only send text from one source at a time until a suitable point 461 appears for source change. This solution is a fallback method with 462 functional limitations. It acts on the presentation level. 464 A party acting as a mixer, which has not negotiated any method for 465 true multi-party RTT handling, but negotiated a "text/red" or "text/ 466 t140" format in a session with a participant SHOULD in order to 467 maintain interoperability, if nothing else is specified for the 468 application, format transmitted text to that participant to be 469 suitable to present on a multi-party unaware endpoint as further 470 specified in Section 4.2. 472 2.3. Offer/answer considerations 474 RTP Payload for Text Conversation [RFC4103] specifies use of RTP 475 [RFC3550], and a redundancy format "text/red" for increased 476 robustness of real-time text transmission. This document updates 477 [RFC4103] by introducing a capability negotiation for handling multi- 478 party real-time text, a way to indicate the source of transmitted 479 text, and rules for efficient timing of the transmissions interleaved 480 from different sources. 482 The capability negotiation for the "RTP-mixer based multi-party 483 method" is based on use of the SDP media attribute "rtt-mixer". 485 The syntax is as follows: 486 "a=rtt-mixer" 488 If any other method for RTP-based multi-party real-time text gets 489 specified by additional work, it is assumed that it will be 490 recognized by some specific SDP feature exchange. 492 2.3.1. Initial offer 494 A party intending to set up a session and being willing to use the 495 RTP-mixer based method of this specification for sending or receiving 496 or both sending and receiving real-time text SHALL include the "rtt- 497 mixer" SDP attribute in the corresponding "text" media section in the 498 initial offer. 500 The party MAY indicate capability for both the RTP-mixer based method 501 of this specification and other methods. 503 When the offeror has sent the offer including the "rtt-mixer" 504 attribute, it MUST be prepared to receive and handle real-time text 505 formatted according to both the method for multi-party aware parties 506 specified in Section 3 in this specification and two-party formatted 507 real-time text. 509 2.3.2. Answering the offer 511 A party receiving an offer containing the "rtt-mixer" SDP attribute 512 and being willing to use the RTP-mixer based method of this 513 specification for sending or receiving or both sending and receiving 514 SHALL include the "rtt-mixer" SDP attribute in the corresponding 515 "text" media section in the answer. 517 If the offer did not contain the "rtt-mixer" attribute, the answer 518 MUST NOT contain the "rtt-mixer" attribute. 520 An answer MUST NOT include acceptance of more than one method for 521 multi-party real-time text in the same RTP session. 523 When the answer including acceptance is transmitted, the answerer 524 MUST be prepared to act on received text in the negotiated session 525 according to the method for multi-party aware parties specified in 526 Section 3 of this specification. Reception of text for a two-party 527 session SHALL also be supported. 529 2.3.3. Offeror processing the answer 531 When the answer is processed by the offeror, it MUST act as specified 532 in Section 2.4 534 2.3.4. Modifying a session 536 A session MAY be modified at any time by any party offering a 537 modified SDP with or without the "rtt-mixer" SDP attribute expressing 538 a desired change in the support of multi-party real-time text. 540 If the modified offer adds indication of support for multi-party 541 real-time text by including the "rtt-mixer" SDP attribute, the 542 procedures specified in the previous subsections SHALL be applied. 544 If the modified offer deletes indication of support for multi-party 545 real-time text by excluding the "rtt-mixer" SDP attribute, the answer 546 MUST NOT contain the "rtt-mixer" attribute, and both parties SHALL 547 after processing the SDP exchange NOT send real-time text formatted 548 for multi-party aware parties according to this specification. 550 2.4. Actions depending on capability negotiation result 552 A transmitting party SHALL send text according to the RTP-mixer based 553 multi-party method only when the negotiation for that method was 554 successful and when it conveys text for another source. In all other 555 cases, the packets SHALL be populated and interpreted as for a two- 556 party session. 558 A party which has negotiated the "rtt-mixer" SDP media attribute MUST 559 populate the CSRC-list and format the packets according to Section 3 560 if it acts as an rtp-mixer and sends multi-party text. 562 A party which has negotiated the "rtt-mixer" SDP media attribute MUST 563 interpret the contents of the "CC" field, the CSRC-list and the 564 packets according to Section 3 in received RTP packets in the 565 corresponding RTP stream. 567 A party which has not successfully completed the negotiation of the 568 "rtt-mixer" SDP media attribute MUST NOT transmit packets interleaved 569 from different sources in the same RTP stream as specified in 570 Section 3. If the party is a mixer and did declare the "rtt-mixer" 571 SDP media attribute, it SHOULD perform the procedure for multi-party 572 unaware endpoints. If the party is not a mixer, it SHOULD transmit 573 as in a two-party session according to [RFC4103]. 575 3. Details for the RTP-mixer based mixing method for multi-party aware 576 endpoints 578 3.1. Use of fields in the RTP packets 580 The CC field SHALL show the number of members in the CSRC list, which 581 SHALL be one (1) in transmissions from a mixer when conveying text 582 from other sources in a multi-party session, and otherwise 0. 584 When text is conveyed by a mixer during a multi-party session, a CSRC 585 list SHALL be included in the packet. The single member in the CSRC- 586 list SHALL contain the SSRC of the source of the T140blocks in the 587 packet. 589 When redundancy is used, the RECOMMENDED level of redundancy is to 590 use one primary and two redundant generations of T140blocks. In some 591 cases, a primary or redundant T140block is empty, but is still 592 represented by a member in the redundancy header. 594 From other aspects, the contents of the RTP packets are equal to what 595 is specified in [RFC4103]. 597 3.2. Initial transmission of a BOM character 599 As soon as a participant is known to participate in a session with 600 another entity and is available for text reception, a Unicode BOM 601 character SHALL be sent to it by the other entity according to the 602 procedures in this section. If the transmitter is a mixer, then the 603 source of this character SHALL be indicated to be the mixer itself. 605 Note that the BOM character SHALL be transmitted with the same 606 redundancy procedures as any other text. 608 3.3. Keep-alive 610 After that, the transmitter SHALL send keep-alive traffic to the 611 receiver(s) at regular intervals when no other traffic has occurred 612 during that interval, if that is decided for the actual connection. 613 It is RECOMMENDED to use the keep-alive solution from [RFC6263]. The 614 consent check of [RFC7675] is a possible alternative if it is used 615 anyway for other reasons. 617 3.4. Transmission interval 619 A "text/red" or "text/t140" transmitter in a mixer SHALL send packets 620 distributed in time as long as there is something (new or redundant 621 T140blocks) to transmit. The maximum transmission interval SHALL 622 then be 330 ms, when no other limitations cause a longer interval to 623 be temporarily used. It is RECOMMENDED to send the next packet to a 624 receiver as soon as new text to that receiver is available, as long 625 as the maximum character rate ("CPS") to the receiver is not exceeded 626 during any 10 second interval. The intention of these time intervals 627 is to keep the latency low and network load limited while keeping a 628 good protection against text loss in bursty packet loss conditions. 629 The main purpose of the 330 ms interval is for timing of redundant 630 transmission, when no new text from the same source is available. 632 If the "CPS" value is reached, longer transmission intervals SHALL be 633 applied and only part of the text queued for transmission sent at end 634 of each transmission interval, until the transmission rate falls 635 under the "CPS" value again. See also Section 8 637 For a transmitter not acting in a mixer, the transmission interval 638 principles from [RFC4103] apply, and the transmission interval SHALL 639 be 300 ms. 641 3.5. Only one source per packet 643 New text and redundant copies of earlier text from one source SHALL 644 be transmitted in the same packet if available for transmission at 645 the same time. Text from different sources MUST NOT be transmitted 646 in the same packet. 648 3.6. Do not send received text to the originating source 650 Text received by a mixer from a participant SHOULD NOT be included in 651 transmission from the mixer to that participant, because the normal 652 behavior of the endpoint is to present locally produced locally. 654 3.7. Clean incoming text 656 A mixer SHALL handle reception, recovery from packet loss, deletion 657 of superfluous redundancy, marking of possible text loss and deletion 658 of 'BOM' characters from each participant before queueing received 659 text for transmission to receiving participants. 661 3.8. Redundant transmission principles 663 A transmitting party using redundancy SHALL send redundant 664 repetitions of T140blocks already transmitted in earlier packets. 666 The number of redundant generations of T140blocks to include in 667 transmitted packets SHALL be deduced from the SDP negotiation. It 668 SHALL be set to the minimum of the number declared by the two parties 669 negotiating a connection. It is RECOMMENDED to declare and transmit 670 one original and two redundant generations of the T140blocks, because 671 that provides good protection against text loss in case of packet 672 loss, and low overhead. 674 3.9. Interleaving text from different sources 676 When text from more than one source is available for transmission 677 from a mixer, the mixer SHALL let the sources take turns in having 678 their text transmitted. 680 The source with the oldest text received in the mixer or oldest 681 redundant text SHALL be next in turn to get all its available unsent 682 text transmitted. Any redundant repetitions of earlier transmitted 683 text not yet sent the intended number of times SHALL be included as 684 redundant retransmission in the transmission. 686 3.10. Text placement in packets 688 The mixer SHALL compose and transmit an RTP packet to a receiver when 689 one of the following conditions has occurred: 691 * There is unsent text available for transmission to that receiver. 693 * 330 ms has passed since already transmitted text was queued for 694 transmission as redundant text. 696 At time of transmission, the mixer SHALL populate the RTP packet with 697 all T140blocks queued for transmission originating from the source in 698 turn for transmission as long as this is not in conflict with the 699 allowed number of characters per second ("CPS") or the maximum packet 700 size. In this way, the latency of the latest received text is kept 701 low even in moments of simultaneous transmission from many sources. 703 Redundant text SHALL also be included. See Section 3.12 705 The SSRC of the source SHALL be placed as the only member in the 706 CSRC-list. 708 Note: The CSRC-list in an RTP packet only includes the participant 709 whose text is included in text blocks. It is not the same as the 710 total list of participants in a conference. With audio and video 711 media, the CSRC-list would often contain all participants who are not 712 muted whereas text participants that don't type are completely silent 713 and thus are not represented in RTP packet CSRC-lists. 715 3.11. Empty T140blocks 717 If no unsent T140blocks were available for a source at the time of 718 populating a packet, but T140blocks are available which have not yet 719 been sent the full intended number of redundant transmissions, then 720 the primary T140block for that source is composed of an empty 721 T140block, and populated (without taking up any length) in a packet 722 for transmission. The corresponding SSRC SHALL be placed as usual in 723 its place in the CSRC-list. 725 The first packet in the session, the first after a source switch and 726 the first after a pause SHALL be poulated with the available 727 T140blocks for the source in turn to be sent as primary, and empty 728 T140blocks for the agreed number of redundancy generations. 730 3.12. Creation of the redundancy 732 The primary T140block from a source in the latest transmitted packet 733 is saved for populating the first redundant T140block for that source 734 in next transmission of text from that source. The first redundant 735 T140block for that source from the latest transmission is saved for 736 populating the second redundant T140block in next transmission of 737 text from that source. 739 Usually this is the level of redundancy used. If a higher number of 740 redundancy is negotiated, then the procedure SHALL be maintained 741 until all available redundant levels of T140blocks are placed in the 742 packet. If a receiver has negotiated a lower number of "text/red" 743 generations, then that level SHALL be the maximum used by the 744 transmitter. 746 The T140blocks saved for transmission as redundant data are assigned 747 a planned transmission time 330 ms after the current time, but SHOULD 748 be transmitted earlier if new text for the same source gets in turn 749 for transmission before that time. 751 3.13. Timer offset fields 753 The timestamp offset values SHALL be inserted in the redundancy 754 header, with the time offset from the RTP timestamp in the packet 755 when the corresponding T140block was sent as primary. 757 The timestamp offsets are expressed in the same clock tick units as 758 the RTP timestamp. 760 The timestamp offset values for empty T140blocks have no relevance 761 but SHOULD be assigned realistic values. 763 3.14. Other RTP header fields 765 The number of members in the CSRC list ( 0 or 1) SHALL be placed in 766 the "CC" header field. Only mixers place value 1 in the "CC" field. 767 A value of "0" indicates that the source is the transmitting device 768 itself and that the source is indicated by the SSRC field. This 769 value is used by endpoints, and by mixers sending data that it is 770 source of itself. 772 The current time SHALL be inserted in the timestamp. 774 The SSRC of the mixer for the RTT session SHALL be inserted in the 775 SSRC field of the RTP header. 777 The M-bit SHALL be handled as specified in [RFC4103]. 779 3.15. Pause in transmission 781 When there is no new T140block to transmit, and no redundant 782 T140block that has not been retransmitted the intended number of 783 times from any source, the transmission process SHALL be stopped 784 until either new T140blocks arrive, or a keep-alive method calls for 785 transmission of keep-alive packets. 787 3.16. RTCP considerations 789 A mixer SHALL send RTCP reports with SDES, CNAME and NAME information 790 about the sources in the multi-party call. This makes it possible 791 for participants to compose a suitable label for text from each 792 source. 794 Integrity SHALL be considered when composing these fields. They 795 contain name and address information that may be sensitive to 796 transmit in its entirety e.g. to unauthenticated participants. 797 Similar considerations SHALL be taken as for other media. 799 3.17. Reception of multi-party contents 801 The "text/red" receiver included in an endpoint with presentation 802 functions will receive RTP packets in the single stream from the 803 mixer, and SHALL distribute the T140blocks for presentation in 804 presentation areas for each source. Other receiver roles, such as 805 gateways or chained mixers are also feasible, and requires 806 consideration if the stream shall just be forwarded, or distributed 807 based on the different sources. 809 3.17.1. Acting on the source of the packet contents 811 If the "CC" field value of a received packet is 1, it indicates that 812 the text is conveyed from a source indicated in the single member in 813 the CSRC-list, and the receiver MUST act on the source according to 814 its role. If the CC value is 0, the source is indicated in the SSRC 815 field. 817 3.17.2. Detection and indication of possible text loss 819 The RTP sequence numbers of the received packets SHALL be monitored 820 for gaps and packets out of order. If a sequence number gap appears 821 and still exists after some defined short time for jitter resolution, 822 the packets in the gap SHALL be regarded as lost. 824 If it is known that only one source is active in the RTP session, 825 then it is likely that a gap equal to or larger than the agreed 826 number of redundancy generations (including the primary) causes text 827 loss. In that case a t140block SHALL be created with a marker for 828 possible text loss [T140ad1] and assigned to the source and inserted 829 in the reception buffer for that source. 831 If it is known that more than one source is active in the RTP 832 session, then it is not possible in general to evaluate if text was 833 lost when packets were lost. With two active sources and the 834 recommended number of redundancy generations (3), it can take a gap 835 of five consecutive lost packets until any text may be lost, but text 836 loss can also appear if three non-consecutive packets are lost when 837 they contained consecutive data from the same source. A simple 838 method to decide when there is risk for resulting text loss is to 839 evaluate if three or more packets were lost within one second. If 840 this simple method is used, then a t140block SHOULD be created with a 841 marker for possible text loss [T140ad1] and assigned to the SSRC of 842 the transmitter as a general input from the mixer. 844 Implementations MAY apply more refined methods for more reliable 845 detection of if text was lost or not. Any refined method SHALL 846 prefer marking possible loss rather than not marking when it is 847 uncertain if there was loss. 849 3.17.3. Extracting text and handling recovery 851 When applying the following procedures, the effects MUST be 852 considered of possible timestamp wrap around and the RTP session 853 possibly changing SSRC. 855 When a packet is received in an RTP session using the packetization 856 for multi-party aware endpoints, its T140blocks SHALL be extracted in 857 the following way. The description is adapted to the default 858 redundancy case using the original and two redundant generations. 860 The source SHALL be extracted from the CSRC-list if available, 861 otherwise from the SSRC. 863 If the received packet is the first packet received from the source, 864 then all T140blocks in the packet SHALL be retrieved and assigned to 865 a receive buffer for the source beginning with the second generation 866 redundancy, continuing with the first generation redundancy and 867 finally the primary. 869 Note: The normal case is that in the first packet, only the primary 870 data has contents. The redundant data has contents in the first 871 received packet from a source only after initial packet loss. 873 If the packet is not the first packet from a source, then if the 874 second generation redundant data is available, its timestamp SHALL be 875 created by subtracting its timestamp offset from the RTP timestamp. 876 If the resulting timestamp is later than the latest retrieved data 877 from the same source, then the redundant data SHALL be retrieved and 878 appended to the receive buffer. The process SHALL be continued in 879 the same way for the first generation redundant data. After that, 880 the primary data SHALL be retrieved from the packet and appended to 881 the receive buffer for the source. 883 3.17.4. Delete 'BOM' 885 Unicode character 'BOM' is used as a start indication and sometimes 886 used as a filler or keep alive by transmission implementations. 887 These SHALL be deleted after extraction from received packets. 889 3.18. Performance considerations 891 This solution has good performance with low text delays as long as 892 the sum of characters per second during any 10 second interval sent 893 from a number of simultaneously sending participants to a receiving 894 participant does not reach the 'CPS' value. At higher numbers of 895 characters per second sent, a jerkiness is visible in the 896 presentation of text. The solution is therefore suitable for 897 emergency service use, relay service use, and small or well-managed 898 larger multimedia conferences. Only in large unmanaged conferences 899 with a high number of participants there may on very rare occasions 900 appear situations when many participants happen to send text 901 simultaneously, resulting in unpleasantly jerky presentation of text 902 from each sending participant. It should be noted that it is only 903 the number of users sending text within the same moment that causes 904 jerkiness, not the total number of users with RTT capability. 906 3.19. Security for session control and media 908 Security SHOULD be applied when possible regarding the capabilities 909 of the participating devices by use of SIP over TLS by default 910 according to [RFC5630] section 3.1.3 on session control level and by 911 default using DTLS-SRTP [RFC5764] on media level. In applications 912 where legacy endpoints without security may exist, a negotiation 913 SHOULD be performed to decide if security by encryption on media 914 level will be applied. If no other security solution is mandated for 915 the application, then OSRTP [RFC8643] is a suitable method be applied 916 to negotiate SRTP media security with DTLS. Most SDP examples below 917 are for simplicity expressed without the security additions. The 918 principles (but not all details) for applying DTLS-SRTP [RFC5764] 919 security is shown in a couple of the following examples. 921 This document contains two mixing procedures which imply different 922 security levels. The mixing for conference-unaware endpoints has 923 lower security level than the mixing method for conference-aware 924 endpoints, because there may be an opportunity for a malicious mixer 925 or a middleman to masquerade the source labels accompanying the text 926 streams in text format. This is especially true if support of un- 927 encrypted SIP and media is supported because of lack of such support 928 in the target endpoints. However, the mixing for conference-aware 929 endpoints as specified here also requires that the mixer can be 930 trusted. End to end encryption would require further work and could 931 be based on WebRTC as specified in Section 1.2. 933 3.20. SDP offer/answer examples 935 This section shows some examples of SDP for session negotiation of 936 the real-time text media in SIP sessions. Audio is usually provided 937 in the same session, and sometimes also video. The examples only 938 show the part of importance for the real-time text media. The 939 examples relate to the single RTP stream mixing for multi-party aware 940 endpoints and for multi-party unaware endpoints. 942 Note: Multi-party RTT MAY also be provided through other methods, 943 e.g. by a Selective Forwarding Middlebox (SFM). In that case, the 944 SDP of the offer will include something specific for that method, and 945 an answer acknowledging the use of that method would accept it by 946 something specific included in the SDP. The offer may contain also 947 the "rtt-mixer" SDP media attribute for the main RTT media when the 948 offeror has capability for both multi-party methods, while an answer, 949 selecting to use SFM will not include the "rtt-mixer" SDP media 950 attribute. 952 Offer example for "text/red" format and multi-party support: 954 m=text 11000 RTP/AVP 100 98 955 a=rtpmap:98 t140/1000 956 a=rtpmap:100 red/1000 957 a=fmtp:100 98/98/98 958 a=rtt-mixer 960 Answer example from a multi-party capable device 961 m=text 14000 RTP/AVP 100 98 962 a=rtpmap:98 t140/1000 963 a=rtpmap:100 red/1000 964 a=fmtp:100 98/98/98 965 a=rtt-mixer 967 Offer example for "text/red" format including multi-party 968 and security: 969 a=fingerprint: (fingerprint1) 970 m=text 11000 RTP/AVP 100 98 971 a=rtpmap:98 t140/1000 972 a=rtpmap:100 red/1000 973 a=fmtp:100 98/98/98 974 a=rtt-mixer 976 The "fingerprint" is sufficient to offer DTLS-SRTP, with the media 977 line still indicating RTP/AVP. 979 Note: For brevity, the entire value of the SDP fingerprint attribute 980 is not shown in this and the following example. 982 Answer example from a multi-party capable device with security 983 a=fingerprint: (fingerprint2) 984 m=text 16000 RTP/AVP 100 98 985 a=rtpmap:98 t140/1000 986 a=rtpmap:100 red/1000 987 a=fmtp:100 98/98/98 988 a=rtt-mixer 990 With the "fingerprint" the device acknowledges use of SRTP/DTLS. 992 Answer example from a multi-party unaware device that also 993 does not support security: 995 m=text 12000 RTP/AVP 100 98 996 a=rtpmap:98 t140/1000 997 a=rtpmap:100 red/1000 998 a=fmtp:100 98/98/98 1000 3.21. Packet sequence example from interleaved transmission 1002 This example shows a symbolic flow of packets from a mixer including 1003 loss and recovery. The sequence includes interleaved transmission of 1004 text from two RTT sources A and B. P indicates primary data. R1 is 1005 first redundant generation data and R2 is the second redundant 1006 generation data. A1, B1, A2 etc are text chunks (T140blocks) 1007 received from the respective sources and sent on to the receiver by 1008 the mixer. X indicates dropped packet between the mixer and a 1009 receiver. The session is assumed to use original and two redundant 1010 generations of RTT. 1012 |-----------------------| 1013 |Seq no 101, Time=20400 | 1014 |CC=1 | 1015 |CSRC list A | 1016 |R2: A1, Offset=600 | 1017 |R1: A2, Offset=300 | 1018 |P: A3 | 1019 |-----------------------| 1021 Assuming that earlier packets ( with text A1 and A2) were received in 1022 sequence, text A3 is received from packet 101 and assigned to 1023 reception area A. The mixer is now assumed to have received text 1024 from source B 100 ms after packet 101 and will send that text. 1025 Transmission of A2 and A3 as redundancy is planned for 330 ms after 1026 packet 101 if no new text from A is ready to be sent before that. 1028 |-----------------------| 1029 |Seq no 102, Time=20500 | 1030 |CC=1 | 1031 |CSRC list B | 1032 |R2 Empty, Offset=600 | 1033 |R1: Empty, Offset=300 | 1034 |P: B1 | 1035 |-----------------------| 1036 Packet 102 is received. 1037 B1 is retrieved from this packet. Redundant transmission of 1038 B1 is planned 330 ms after packet 102. 1040 X------------------------| 1041 X Seq no 103, Timer=20730| 1042 X CC=1 | 1043 X CSRC list A | 1044 X R2: A2, Offset=630 | 1045 X R1: A3, Offset=330 | 1046 X P: Empty | 1047 X------------------------| 1048 Packet 103 is assumed to be lost due to network problems. 1049 It contains redundancy for A. Sending A3 as second level 1050 redundancy is planned for 330 ms after packet 103. 1052 X------------------------| 1053 X Seq no 104, Timer=20830| 1054 X CC=1 | 1055 X CSRC list B | 1056 X R2: Empty, Offset=600 | 1057 X R1: B1, Offset=300 | 1058 X P: B2 | 1059 X------------------------| 1060 Packet 104 contains text from B, including new B2 and 1061 redundant B1. It is assumed dropped in network 1062 problems. 1063 The mixer has A3 redundancy to send but no new text 1064 appears from A and therefore the redundancy is sent 1065 330 ms after the previous packet with text from A. 1067 |------------------------| 1068 | Seq no 105, Timer=21060| 1069 | CC=1 | 1070 | CSRC list A | 1071 | R2: A3, Offset=660 | 1072 | R1: Empty, Offset=330 | 1073 | P: Empty | 1074 |------------------------| 1075 Packet 105 is received. 1076 A gap for lost 103, and 104 is detected. 1077 Assume that no other loss was detected the last second. 1078 Then it can be concluded that nothing was totally lost. 1080 R2 is checked. Its original time was 21040-660=20400. 1081 A packet with text from A was received with that 1082 timestamp, so nothing needs to be recovered. 1084 B1 and B2 still needs to be transmitted as redundancy. 1085 This is planned 330 ms after packet 105. That 1086 would be at 21150. 1088 |-----------------------| 1089 |Seq no 106, Timer=21160| 1090 |CC=1 | 1091 |CSRC list B | 1092 | R2: B1, Offset=660 | 1093 | R1: B2, Offset=330 | 1094 | P: Empty | 1095 |-----------------------| 1097 Packet 106 is received. 1099 The second level redundancy in packet 106 is B1 and has timestamp 1100 offset 660 ms. The timestamp of packet 106 minus 660 is 20500 which 1101 is the timestamp of packet 102 THAT was received. So B1 does not 1102 need to be retrieved. The first level redundancy in packet 106 has 1103 offset 330. The timestamp of packet 106 minus 330 is 20830. That is 1104 later than the latest received packet with source B. Therefore B2 is 1105 retrieved and assigned to the input buffer for source B. No primary 1106 is available in packet 106. 1108 After this sequence, A3 and B1 and B2 have been received. In this 1109 case no text was lost. 1111 3.22. Maximum character rate "CPS" 1113 The default maximum rate of reception of "text/t140" real-time text 1114 is in [RFC4103] specified to be 30 characters per second. The value 1115 MAY be modified in the "CPS" parameter of the FMTP attribute in the 1116 media section for the "text/t140" media. A mixer combining real-time 1117 text from a number of sources may occasionally have a higher combined 1118 flow of text coming from the sources. Endpoints SHOULD therefore 1119 specify a suitable higher value for the "CPS" parameter, 1120 corresponding to its real reception capability. A value for "CPS" of 1121 90 SHALL be the default for the "text/t140" stream in the "text/red" 1122 format when multi-party real-time text is negotiated. See [RFC4103] 1123 for the format and use of the "CPS" parameter. The same rules apply 1124 for the multi-party case except for the default value. 1126 4. Presentation level considerations 1128 "Protocol for multimedia application text conversation" [T140] 1129 provides the presentation level requirements for the [RFC4103] 1130 transport. Functions for erasure and other formatting functions and 1131 are specified in [T140] which has the following general statement for 1132 the presentation: 1134 "The display of text from the members of the conversation should be 1135 arranged so that the text from each participant is clearly readable, 1136 and its source and the relative timing of entered text is visualized 1137 in the display. Mechanisms for looking back in the contents from the 1138 current session should be provided. The text should be displayed as 1139 soon as it is received." 1141 Strict application of [T140] is of essence for the interoperability 1142 of real-time text implementations and to fulfill the intention that 1143 the session participants have the same information of the text 1144 contents of the conversation without necessarily having the exact 1145 same layout of the conversation. 1147 [T140] specifies a set of presentation control codes to include in 1148 the stream. Some of them are optional. Implementations MUST be able 1149 to ignore optional control codes that they do not support. 1151 There is no strict "message" concept in real-time text. The Unicode 1152 Line Separator character SHALL be used as a separator allowing a part 1153 of received text to be grouped in presentation. The characters 1154 "CRLF" may be used by other implementations as replacement for Line 1155 Separator. The "CRLF" combination SHALL be erased by just one 1156 erasing action, just as the Line Separator. Presentation functions 1157 are allowed to group text for presentation in smaller groups than the 1158 line separators imply and present such groups with source indication 1159 together with text groups from other sources (see the following 1160 presentation examples). Erasure has no specific limit by any 1161 delimiter in the text stream. 1163 4.1. Presentation by multi-party aware endpoints 1165 A multi-party aware receiving party, presenting real-time text MUST 1166 separate text from different sources and present them in separate 1167 presentation fields. The receiving party MAY separate presentation 1168 of parts of text from a source in readable groups based on other 1169 criteria than line separator and merge these groups in the 1170 presentation area when it benefits the user to most easily find and 1171 read text from the different participants. The criteria MAY e.g. be 1172 a received comma, full stop, or other phrase delimiters, or a long 1173 pause. 1175 When text is received from multiple original sources, the 1176 presentation SHALL provide a view where text is added in multiple 1177 presentation fields. 1179 If the presentation presents text from different sources in one 1180 common area, the presenting endpoint SHOULD insert text from the 1181 local user ended at suitable points merged with received text to 1182 indicate the relative timing for when the text groups were completed. 1183 In this presentation mode, the receiving endpoint SHALL present the 1184 source of the different groups of text. This presentation style is 1185 called the "chat" style here and provides a possibility to follow 1186 text arriving from multiple parties and the approximate relative time 1187 that text is received related to text from the local user. 1189 A view of a three-party RTT call in chat style is shown in this 1190 example . 1192 _________________________________________________ 1193 | |^| 1194 |[Alice] Hi, Alice here. |-| 1195 | | | 1196 |[Bob] Bob as well. | | 1197 | | | 1198 |[Eve] Hi, this is Eve, calling from Paris. | | 1199 | I thought you should be here. | | 1200 | | | 1201 |[Alice] I am coming on Thursday, my | | 1202 | performance is not until Friday morning.| | 1203 | | | 1204 |[Bob] And I on Wednesday evening. | | 1205 | | | 1206 |[Alice] Can we meet on Thursday evening? | | 1207 | | | 1208 |[Eve] Yes, definitely. How about 7pm. | | 1209 | at the entrance of the restaurant | | 1210 | Le Lion Blanc? | | 1211 |[Eve] we can have dinner and then take a walk |-| 1212 |______________________________________________|v| 1213 | But I need to be back to |^| 1214 | the hotel by 11 because I need |-| 1215 | | | 1216 | I wou |-| 1217 |______________________________________________|v| 1218 | of course, I underst | 1219 |________________________________________________| 1221 Figure 3: Example of a three-party RTT call presented in chat style 1222 seen at participant 'Alice's endpoint. 1224 Other presentation styles than the chat style MAY be arranged. 1226 This figure shows how a coordinated column view MAY be presented. 1228 _____________________________________________________________________ 1229 | Bob | Eve | Alice | 1230 |____________________|______________________|_______________________| 1231 | | |I will arrive by TGV. | 1232 |My flight is to Orly| |Convenient to the main | 1233 | |Hi all, can we plan |station. | 1234 | |for the seminar? | | 1235 |Eve, will you do | | | 1236 |your presentation on| | | 1237 |Friday? |Yes, Friday at 10. | | 1238 |Fine, wo | |We need to meet befo | 1239 |___________________________________________________________________| 1240 Figure 4: An example of a coordinated column-view of a three-party 1241 session with entries ordered vertically in approximate time-order. 1243 4.2. Multi-party mixing for multi-party unaware endpoints 1245 When the mixer has indicated RTT multi-party capability in an SDP 1246 negotiation, but the multi-party capability negotiation fails with an 1247 endpoint, then the agreed "text/red" or "text/t140" format SHALL be 1248 used and the mixer SHOULD compose a best-effort presentation of 1249 multi-party real-time text in one stream intended to be presented by 1250 an endpoint with no multi-party awareness, when that is desired in 1251 the actual implementation. The following specifies a procedure which 1252 MAY be applied in that situation. 1254 This presentation format has functional limitations and SHOULD be 1255 used only to enable participation in multi-party calls by legacy 1256 deployed endpoints implementing only RFC 4103 without any multi-party 1257 extensions specified in this document. 1259 The principles and procedures below do not specify any new protocol 1260 elements. They are instead composed from the information in [T140] 1261 and an ambition to provide a best effort presentation on an endpoint 1262 which has functions only for two-party calls. 1264 The mixer mixing for multi-party unaware endpoints SHALL compose a 1265 simulated limited multi-party RTT view suitable for presentation in 1266 one presentation area. The mixer SHALL group text in suitable groups 1267 and prepare for presentation of them by inserting a new line between 1268 them if the transmitted text did not already end with a new line. A 1269 presentable label SHALL be composed and sent for the source initially 1270 in the session and after each source switch. With this procedure the 1271 time for switching from transmission of text from one source to 1272 transmission of text from another source is depending on the actions 1273 of the users. In order to expedite source switch, a user can for 1274 example end its turn with a new line. 1276 4.2.1. Actions by the mixer at reception from the call participants 1278 When text is received by the mixer from the different participants, 1279 the mixer SHALL recover text from redundancy if any packets are lost. 1280 The mark for lost text [T140ad1] SHALL be inserted in the stream if 1281 unrecoverable loss appears. Any Unicode "BOM" characters, possibly 1282 used for keep-alive SHALL be deleted. The time of creation of text 1283 (retrieved from the RTP timestamp) SHALL be stored together with the 1284 received text from each source in queues for transmission to the 1285 recipients in order to be able to evaluate text loss. 1287 4.2.2. Actions by the mixer for transmission to the recipients 1289 The following procedure SHALL be applied for each multi-party unaware 1290 recipient of multi-party text from the mixer. 1292 The text for transmission SHALL be formatted by the mixer for each 1293 receiving user for presentation in one single presentation area. 1294 Text received from a participant SHOULD NOT be included in 1295 transmission to that participant because it is usually presented 1296 locally at transmission time. When there is text available for 1297 transmission from the mixer to a receiving party from more than one 1298 participant, the mixer SHALL switch between transmission of text from 1299 the different sources at suitable points in the transmitted stream. 1301 When switching source, the mixer SHALL insert a line separator if the 1302 already transmitted text did not end with a new line (line separator 1303 or CRLF). A label SHALL be composed from information in the CNAME 1304 and NAME fields in RTCP reports from the participant to have its text 1305 transmitted, or from other session information for that user. The 1306 label SHALL be delimited by suitable characters (e.g. '[ ]') and 1307 transmitted. The CSRC SHALL indicate the selected source. Then text 1308 from that selected participant SHALL be transmitted until a new 1309 suitable point for switching source is reached. 1311 Integrity considerations SHALL be taken when composing the label. 1313 Seeking a suitable point for switching source SHALL be done when 1314 there is older text waiting for transmission from any party than the 1315 age of the last transmitted text. Suitable points for switching are: 1317 * A completed phrase ended by comma 1319 * A completed sentence 1321 * A new line (line separator or CRLF) 1323 * A long pause (e.g. > 10 seconds) in received text from the 1324 currently transmitted source 1326 * If text from one participant has been transmitted with text from 1327 other sources waiting for transmission for a long time (e.g. > 1 1328 minute) and none of the other suitable points for switching has 1329 occurred, a source switch MAY be forced by the mixer at next word 1330 delimiter, and also if even a word delimiter does not occur within 1331 a time (e.g. 15 seconds) after the scan for word delimiter 1332 started. 1334 When switching source, the source which has the oldest text in queue 1335 SHALL be selected to be transmitted. A character display count SHALL 1336 be maintained for the currently transmitted source, starting at zero 1337 after the label is transmitted for the currently transmitted source. 1339 The status SHALL be maintained for the latest control code for Select 1340 Graphic Rendition (SGR) from each source. If there is an SGR code 1341 stored as the status for the current source before the source switch 1342 is done, a reset of SGR SHALL be sent by the sequence SGR 0 [009B 1343 0000 006D] after the new line and before the new label during a 1344 source switch. See SGR below for an explanation. This transmission 1345 does not influence the display count. 1347 If there is an SGR code stored for the new source after the source 1348 switch, that SGR code SHALL be transmitted to the recipient before 1349 the label. This transmission does not influence the display count. 1351 4.2.3. Actions on transmission of text 1353 Text from a source sent to the recipient SHALL increase the display 1354 count by one per transmitted character. 1356 4.2.4. Actions on transmission of control codes 1358 The following control codes specified by T.140 require specific 1359 actions. They SHALL cause specific considerations in the mixer. 1360 Note that the codes presented here are expressed in UCS-16, while 1361 transmission is made in UTF-8 transform of these codes. 1363 BEL 0007 Bell Alert in session, provides for alerting during an 1364 active session. The display count SHALL NOT be altered. 1366 NEW LINE 2028 Line separator. Check and perform a source switch if 1367 appropriate. Increase display count by 1. 1369 CR LF 000D 000A A supported, but not preferred way of requesting a 1370 new line. Check and perform a source switch if appropriate. 1371 Increase display count by 1. 1373 INT ESC 0061 Interrupt (used to initiate mode negotiation 1374 procedure). The display count SHALL NOT be altered. 1376 SGR 009B Ps 006D Select graphic rendition. Ps is rendition 1377 parameters specified in ISO 6429. The display count SHALL NOT be 1378 altered. The SGR code SHOULD be stored for the current source. 1380 SOS 0098 Start of string, used as a general protocol element 1381 introducer, followed by a maximum 256 bytes string and the ST. 1382 The display count SHALL NOT be altered. 1384 ST 009C String terminator, end of SOS string. The display count 1385 SHALL NOT be altered. 1387 ESC 001B Escape - used in control strings. The display count SHALL 1388 NOT be altered for the complete escape code. 1390 Byte order mark "BOM" (U+FEFF) "Zero width, no break space", used 1391 for synchronization and keep-alive, SHALL be deleted from incoming 1392 streams. It SHALL also be sent first after session establishment 1393 to the recipient. The display count SHALL NOT be altered. 1395 Missing text mark (U+FFFD) "Replacement character", represented as a 1396 question mark in a rhombus, or if that is not feasible, replaced 1397 by an apostrophe ', marks place in stream of possible text loss. 1398 This mark SHALL be inserted by the reception procedure in case of 1399 unrecoverable loss of packets. The display count SHALL be 1400 increased by one when sent as for any other character. 1402 SGR If a control code for selecting graphic rendition (SGR), other 1403 than reset of the graphic rendition (SGR 0) is sent to a 1404 recipient, that control code SHALL also be stored as status for 1405 the source in the storage for SGR status. If a reset graphic 1406 rendition (SGR 0) originated from a source is sent, then the SGR 1407 status storage for that source SHALL be cleared. The display 1408 count SHALL NOT be increased. 1410 BS (U+0008) Back Space, intended to erase the last entered character 1411 by a source. Erasure by backspace cannot always be performed as 1412 the erasing party intended. If an erasing action erases all text 1413 up to the end of the leading label after a source switch, then the 1414 mixer MUST NOT transmit more backspaces. Instead it is 1415 RECOMMENDED that a letter "X" is inserted in the text stream for 1416 each backspace as an indication of the intent to erase more. A 1417 new line is usually coded by a Line Separator, but the character 1418 combination "CRLF" MAY be used instead. Erasure of a new line is 1419 in both cases done by just one erasing action (Backspace). If the 1420 display count has a positive value it SHALL be decreased by one 1421 when the BS is sent. If the display count is at zero, it SHALL 1422 NOT not altered. 1424 4.2.5. Packet transmission 1426 A mixer transmitting to a multi-party unaware terminal SHALL send 1427 primary data only from one source per packet. The SSRC SHALL be the 1428 SSRC of the mixer. The CSRC list SHALL contain one member and be the 1429 SSRC of the source of the primary data. 1431 4.2.6. Functional limitations 1433 When a multi-party unaware endpoint presents a conversation in one 1434 display area in a chat style, it inserts source indications for 1435 remote text and local user text as they are merged in completed text 1436 groups. When an endpoint using this layout receives and presents 1437 text mixed for multi-party unaware endpoints, there will be two 1438 levels of source indicators for the received text; one generated by 1439 the mixer and inserted in a label after each source switch, and 1440 another generated by the receiving endpoint and inserted after each 1441 switch between local and remote source in the presentation area. 1442 This will waste display space and look inconsistent to the reader. 1444 New text can be presented only from one source at a time. Switch of 1445 source to be presented takes place at suitable places in the text, 1446 such as end of phrase, end of sentence, line separator and 1447 inactivity. Therefore the time to switch to present waiting text 1448 from other sources may become long and will vary and depend on the 1449 actions of the currently presented source. 1451 Erasure can only be done up to the latest source switch. If a user 1452 tries to erase more text, the erasing actions will be presented as 1453 letter X after the label. 1455 Text loss because of network errors may hit the label between entries 1456 from different parties, causing risk for misunderstanding from which 1457 source a piece of text is. 1459 These facts make it strongly RECOMMENDED to implement multi-party 1460 awareness in RTT endpoints. The use of the mixing method for multi- 1461 party-unaware endpoints should be left for use with endpoints which 1462 are impossible to upgrade to become multi-party aware. 1464 4.2.7. Example views of presentation on multi-party unaware endpoints 1466 The following pictures are examples of the view on a participant's 1467 display for the multi-party-unaware case. 1469 _________________________________________________ 1470 | Conference | Alice | 1471 |________________________|_________________________| 1472 | |I will arrive by TGV. | 1473 |[Bob]:My flight is to |Convenient to the main | 1474 |Orly. |station. | 1475 |[Eve]:Hi all, can we | | 1476 |plan for the seminar. | | 1477 | | | 1478 |[Bob]:Eve, will you do | | 1479 |your presentation on | | 1480 |Friday? | | 1481 |[Eve]:Yes, Friday at 10.| | 1482 |[Bob]: Fine, wo |We need to meet befo | 1483 |________________________|_________________________| 1485 Figure 5: Alice who has a conference-unaware client is receiving the 1486 multi-party real-time text in a single-stream. 1488 This figure shows how a coordinated column view MAY be presented on 1489 Alice's device in a view with two-columns. The mixer inserts labels 1490 to show how the sources alternate in the column with received text. 1491 The mixer alternates between the sources at suitable points in the 1492 text exchange so that text entries from each party can be 1493 conveniently read. 1495 _________________________________________________ 1496 | |^| 1497 |(Alice) Hi, Alice here. |-| 1498 | | | 1499 |(mix)[Bob)] Bob as well. | | 1500 | | | 1501 |[Eve] Hi, this is Eve, calling from Paris | | 1502 | I thought you should be here. | | 1503 | | | 1504 |(Alice) I am coming on Thursday, my | | 1505 | performance is not until Friday morning.| | 1506 | | | 1507 |(mix)[Bob] And I on Wednesday evening. | | 1508 | | | 1509 |[Eve] we can have dinner and then walk | | 1510 | | | 1511 |[Eve] But I need to be back to | | 1512 | the hotel by 11 because I need | | 1513 | |-| 1514 |______________________________________________|v| 1515 | of course, I underst | 1516 |________________________________________________| 1518 Figure 6: An example of a view of the multi-party unaware 1519 presentation in chat style. Alice is the local user. 1521 In this view, there is a tradition in receiving applications to 1522 include a label showing the source of the text, here shown with 1523 parenthesis "()". The mixer also inserts source labels for the 1524 multi-party call participants, here shown with brackets "[]". 1526 5. Relation to Conference Control 1528 5.1. Use with SIP centralized conferencing framework 1530 The SIP conferencing framework, mainly specified in [RFC4353], 1531 [RFC4579] and [RFC4575] is suitable for coordinating sessions 1532 including multi-party RTT. The RTT stream between the mixer and a 1533 participant is one and the same during the conference. Participants 1534 get announced by notifications when participants are joining or 1535 leaving, and further user information may be provided. The SSRC of 1536 the text to expect from joined users MAY be included in a 1537 notification. The notifications MAY be used both for security 1538 purposes and for translation to a label for presentation to other 1539 users. 1541 5.2. Conference control 1543 In managed conferences, control of the real-time text media SHOULD be 1544 provided in the same way as other for media, e.g. for muting and 1545 unmuting by the direction attributes in SDP [RFC8866]. 1547 Note that floor control functions may be of value for RTT users as 1548 well as for users of other media in a conference. 1550 6. Gateway Considerations 1552 6.1. Gateway considerations with Textphones (e.g. TTYs). 1554 Multi-party RTT sessions may involve gateways of different kinds. 1555 Gateways involved in setting up sessions SHALL correctly reflect the 1556 multi-party capability or unawareness of the combination of the 1557 gateway and the remote endpoint beyond the gateway. 1559 One case that may occur is a gateway to PSTN for communication with 1560 textphones (e.g. TTYs). Textphones are limited devices with no 1561 multi-party awareness, and it SHOULD therefore be suitable for the 1562 gateway to not indicate multi-party awareness for that case. Another 1563 solution is that the gateway indicates multi-party capability towards 1564 the mixer, and includes the multi-party mixer function for multi- 1565 party unaware endpoints itself. This solution makes it possible to 1566 make adaptations for the functional limitations of the textphone 1567 (TTY). 1569 More information on gateways to textphones (TTYs) is found in 1570 [RFC5194] 1572 6.2. Gateway considerations with WebRTC. 1574 Gateway operation to real-time text in WebRTC may also be required. 1575 In WebRTC, RTT is specified in [RFC8865]. 1577 A multi-party bridge may have functionality for communicating by RTT 1578 both in RTP streams with RTT and WebRTC T.140 data channels. Other 1579 configurations may consist of a multi-party bridge with either 1580 technology for RTT transport and a separate gateway for conversion of 1581 the text communication streams between RTP and T.140 data channel. 1583 In WebRTC, it is assumed that for a multi-party session, one T.140 1584 data channel is established for each source from a gateway or bridge 1585 to each participant. Each participant also has a data channel with 1586 two-way connection with the gateway or bridge. 1588 The t140 channel used both ways is for text from the WebRTC user and 1589 from the bridge or gateway itself to the WebRTC user. The label 1590 parameter of this t140 channel is used as NAME field in RTCP to 1591 participants on the RTP side. The other t140 channels are only for 1592 text from other participants to the WebRTC user. 1594 When a new participant has entered the session with RTP transport of 1595 RTT, a new T.140 channel SHOULD be established to WebRTC users with 1596 the label parameter composed from the NAME field in RTCP on the RTP 1597 side. 1599 When a new participant has entered the multi-party session with RTT 1600 transport in a WebRTC T.140 data channel, the new participant SHOULD 1601 be announced by a notification to RTP users. The label parameter 1602 from the WebRTC side SHOULD be used as the NAME RTCP field on the RTP 1603 side, or other available session information. 1605 When a participant on the RTP side disappears, the corresponding 1606 T.140 data channel(s) SHOULD be closed. 1608 When a WebRTC user of T.140 data channels disconnects from the mixer, 1609 the corresponding RTP streams or sources in an RTP-mixed stream 1610 SHOULD be closed. 1612 T.140 data channels MAY be opened and closed by negotiation or 1613 renegotiation of the session or by any other valid means as specified 1614 in section 1 of [RFC8865]. 1616 7. Updates to RFC 4103 1618 This document updates [RFC4103] by introducing an SDP media attribute 1619 "rtt-mixer" for negotiation of multi-party mixing capability with the 1620 [RFC4103] format, and by specifying the rules for packets when multi- 1621 party capability is negotiated and in use. 1623 8. Congestion considerations 1625 The congestion considerations and recommended actions from [RFC4103] 1626 are valid also in multi-party situations. 1628 The first action in case of congestion SHALL be to temporarily 1629 increase the transmission interval up to two seconds. 1631 If the very unlikely situation appears that many participants in a 1632 conference send text simultaneously for a long period, a delay may 1633 build up for presentation of text at the receivers if the limitation 1634 in characters per second("CPS") to be transmitted to the participants 1635 is exceeded. More delay than 7 seconds can cause confusion in the 1636 session. It is therefore RECOMMENDED that an RTP-mixer based mixer 1637 discards such text in excess and inserts a general indication of 1638 possible text loss [T140ad1] in the session. If the main text 1639 contributor is indicated in any way, the mixer MAY avoid deleting 1640 text from that participant. It should however be noted that human 1641 creation of text normally contains pauses, when the transmission can 1642 catch up, so that the transmission overload situations are expected 1643 to be very rare. 1645 9. Acknowledgements 1647 James Hamlin for format and performance aspects. 1649 10. IANA Considerations 1651 10.1. Registration of the "rtt-mixer" SDP media attribute 1653 [RFC EDITOR NOTE: Please replace all instances of RFCXXXX with the 1654 RFC number of this document.] 1655 IANA is asked to register the new SDP attribute "rtt-mixer". 1657 Contact name: IESG 1659 Contact email: iesg@ietf.org 1661 Attribute name: rtt-mixer 1663 Attribute semantics: See RFCXXXX Section 2.3 1665 Attribute value: none 1667 Usage level: media 1669 Purpose: Indicate support by mixer and endpoint of multi-party 1670 mixing for real-time text transmission, using a common RTP-stream 1671 for transmission of text from a number of sources mixed with one 1672 source at a time and the source indicated in a single CSRC-list 1673 member. 1675 Charset Dependent: no 1677 O/A procedure: See RFCXXXX Section 2.3 1679 Mux Category: normal 1681 Reference: RFCXXXX 1683 11. Security Considerations 1685 The RTP-mixer model requires the mixer to be allowed to decrypt, pack 1686 and encrypt secured text from the conference participants. Therefore 1687 the mixer needs to be trusted. This is similar to the situation for 1688 central mixers of audio and video. 1690 The requirement to transfer information about the user in RTCP 1691 reports in SDES, CNAME and NAME fields, and in conference 1692 notifications, for creation of labels may have privacy concerns as 1693 already stated in RFC 3550 [RFC3550], and may be restricted for 1694 privacy reasons. The receiving user will then get a more symbolic 1695 label for the source. 1697 Participants with malicious intentions may appear and e.g. disturb 1698 the multi-party session by a continuous flow of text, or masquerade 1699 as text from other participants. Counteractions should be to require 1700 secure signaling, media and authentication, and to provide higher 1701 level conference functions e.g. for blocking and expelling 1702 participants. 1704 Further security considerations specific for this application are 1705 specified in section Section 3.19. 1707 12. Change history 1709 [RFC Editor: Please remove this section prior to publication.] 1711 12.1. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-16 1713 Improvements in the offer/answer considerations section by adding 1714 subsections for each phase in the negotiation as requested by IANA 1715 expert review. 1717 12.2. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-15 1719 Actions on review comments from Jurgen Schonwalder: 1721 A bit more about congestion situations and that they are expected to 1722 be very rare. 1724 Explanation of differences in security between the conference-aware 1725 and the conference-unaware case added in security section. 1727 Presentation examples with suource labels made less confusing, and 1728 explained. 1730 Reference to T.140 inserted at first mentioning of T.140. 1732 Reference to RFC 8825 inserted to explain WebRTC 1734 Nit in wording in terminology section adjusted. 1736 12.3. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-14 1738 Changes from comments by Murray Cucherawy during AD review. 1740 Many SHOULD in section 4.2 on multi-party unaware mixing changed to 1741 SHALL, and the whole section instead specified to be optional 1742 depending on the application. 1744 Some SHOULD in section 3 either explained or changed to SHALL. 1746 In order to have explainable conditions behind SHOULDs, the 1747 transmission interval in 3.4 is changed to as soon as text is 1748 available as a main principle. The call participants send with 300 1749 ms interval so that will create realistic load conditions anyway. 1751 12.4. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-13 1753 Changed year to 2021. 1755 Changed reference to draft on RTT in WebRTC to recently published RFC 1756 8865. 1758 Changed label brackets in example from "[]" to "()" to avoid nits 1759 comment. 1761 Changed reference "RFC 4566" to recently published "RFC 8866" 1763 12.5. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-12 1765 Changes according to responses on comments from Brian Rosen in 1766 Avtcore list on 2020-12-05 and -06. 1768 Changes according to responses to comments by Bernard Aboba in 1769 avtcore list 2020-12-06. 1771 Introduction of an optiona RTP multi-stream mixing method for further 1772 study as proposed by Bernard Aboba. 1774 Changes clarifying how to open and close T.140 data channels included 1775 in 6.2 after comments by Lorenzo Miniero. 1777 Changes to satisfy nits check. Some "not" changed to "NOT" in 1778 normative wording combinations. Some lower case normative words 1779 changed to upper case. A normative reference deleted from the 1780 abstract. Two informative documents moved from normative references 1781 to informative references. 1783 12.6. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-11 1785 Timestamps and timestamp offsets added to the packet examples in 1786 section 3.23, and the description corrected. 1788 A number of minor corrections added in sections 3.10 - 3.23. 1790 12.7. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-10 1792 The packet composition was modified for interleaving packets from 1793 different sources. 1795 The packet reception was modified for the new interleaving method. 1797 The packet sequence examples was adjusted for the new interleaving 1798 method. 1800 Modifications according to responses to Brian Rosen of 2020-11-03 1802 12.8. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-09 1804 Changed name on the SDP media attribute to "rtt-mixer" 1806 Restructure of section 2 for balance between aware and unaware cases. 1808 Moved conference control to own section. 1810 Improved clarification of recovery and loss in the packet sequence 1811 example. 1813 A number of editorial corrections and improvements. 1815 12.9. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-08 1817 Deleted the method requiring a new packet format "text/rex" because 1818 of the longer standardization and implementation period it needs. 1820 Focus on use of RFC 4103 text/red format with shorter transmission 1821 interval, and source indicated in CSRC. 1823 12.10. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-07 1825 Added a method based on the "text/red" format and single source per 1826 packet, negotiated by the "rtt-mixer" SDP attribute. 1828 Added reasoning and recommendation about indication of loss. 1830 The highest number of sources in one packet is 15, not 16. Changed. 1832 Added in information on update to RFC 4103 that RFC 4103 explicitly 1833 allows addition of FEC method. The redundancy is a kind of forward 1834 error correction.. 1836 12.11. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-06 1838 Improved definitions list format. 1840 The format of the media subtype parameters is made to match the 1841 requirements. 1843 The mapping of media subtype parameters to SDP is included. 1845 The "CPS" parameter belongs to the t140 subtype and does not need to 1846 be registered here. 1848 12.12. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-05 1850 nomenclature and editorial improvements 1852 "this document" used consistently to refer to this document. 1854 12.13. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-04 1856 'Redundancy header' renamed to 'data header'. 1858 More clarifications added. 1860 Language and figure number corrections. 1862 12.14. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-03 1864 Mention possible need to mute and raise hands as for other media. 1865 ---done ---- 1867 Make sure that use in two-party calls is also possible and explained. 1868 - may need more wording - 1870 Clarify the RTT is often used together with other media. --done-- 1872 Tell that text mixing is N-1. A users own text is not received in 1873 the mix. -done- 1875 In 3. correct the interval to: A "text/rex" transmitter SHOULD send 1876 packets distributed in time as long as there is something (new or 1877 redundant T140blocks) to transmit. The maximum transmission interval 1878 SHOULD then be 300 ms. It is RECOMMENDED to send a packet to a 1879 receiver as soon as new text to that receiver is available, as long 1880 as the time after the latest sent packet to the same receiver is more 1881 than 150 ms, and also the maximum character rate to the receiver is 1882 not exceeded. The intention is to keep the latency low while keeping 1883 a good protection against text loss in bursty packet loss conditions. 1884 -done- 1886 In 1.3 say that the format is used both ways. -done- 1888 In 13.1 change presentation area to presentation field so that reader 1889 does not think it shall be totally separated. -done- 1891 In Performance and intro, tell the performance in number of 1892 simultaneous sending users and introduced delay 16, 150 vs 1893 requirements 5 vs 500. -done -- 1895 Clarify redundancy level per connection. -done- 1896 Timestamp also for the last data header. To make it possible for all 1897 text to have time offset as for transmission from the source. Make 1898 that header equal to the others. -done- 1900 Mixer always use the CSRC list, even for its own BOM. -done- 1902 Combine all talk about transmission interval (300 ms vs when text has 1903 arrived) in section 3 in one paragraph or close to each other. -done- 1905 Documents the goal of good performance with low delay for 5 1906 simultaneous typers in the introduction. -done- 1908 Describe better that only primary text shall be sent on to receivers. 1909 Redundancy and loss must be resolved by the mixer. -done- 1911 12.15. Changes included in draft-ietf-avtcore-multi-party-rtt-mix-02 1913 SDP and better description and visibility of security by OSRTP RFC 1914 8634 needed. 1916 The description of gatewaying to WebRTC extended. 1918 The description of the data header in the packet is improved. 1920 12.16. Changes to draft-ietf-avtcore-multi-party-rtt-mix-01 1922 2,5,6 More efficient format "text/rex" introduced and attribute 1923 a=rtt-mix deleted. 1925 3. Brief about use of OSRTP for security included- More needed. 1927 4. Brief motivation for the solution and why not rtp-translator is 1928 used added to intro. 1930 7. More limitations for the multi-party unaware mixing method 1931 inserted. 1933 8. Updates to RFC 4102 and 4103 more clearly expressed. 1935 9. Gateway to WebRTC started. More needed. 1937 12.17. Changes from draft-hellstrom-avtcore-multi-party-rtt-source-03 1938 to draft-ietf-avtcore-multi-party-rtt-mix-00 1940 Changed file name to draft-ietf-avtcore-multi-party-rtt-mix-00 1942 Replaced CDATA in IANA registration table with better coding. 1944 Converted to xml2rfc version 3. 1946 12.18. Changes from draft-hellstrom-avtcore-multi-party-rtt-source-02 1947 to -03 1949 Changed company and e-mail of the author. 1951 Changed title to "RTP-mixer formatting of multi-party Real-time text" 1952 to better match contents. 1954 Check and modification where needed of use of RFC 2119 words SHALL 1955 etc. 1957 More about the CC value in sections on transmitters and receivers so 1958 that 1-to-1 sessions do not use the mixer format. 1960 Enhanced section on presentation for multi-party-unaware endpoints 1962 A paragraph recommending CPS=150 inserted in the performance section. 1964 12.19. Changes from draft-hellstrom-avtcore-multi-party-rtt-source-01 1965 to -02 1967 In Abstract and 1. Introduction: Introduced wording about regulatory 1968 requirements. 1970 In section 5: The transmission interval is decreased to 100 ms when 1971 there is text from more than one source to transmit. 1973 In section 11 about SDP negotiation, a SHOULD-requirement is 1974 introduced that the mixer should make a mix for multi-party unaware 1975 endpoints if the negotiation is not successful. And a reference to a 1976 later chapter about it. 1978 The presentation considerations chapter 14 is extended with more 1979 information about presentation on multi-party aware endpoints, and a 1980 new section on the multi-party unaware mixing with low functionality 1981 but SHOULD a be implemented in mixers. Presentation examples are 1982 added. 1984 A short chapter 15 on gateway considerations is introduced. 1986 Clarification about the text/t140 format included in chapter 10. 1988 This sentence added to the chapter 10 about use without redundancy. 1989 "The text/red format SHOULD be used unless some other protection 1990 against packet loss is utilized, for example a reliable network or 1991 transport." 1992 Note about deviation from RFC 2198 added in chapter 4. 1994 In chapter 9. "Use with SIP centralized conferencing framework" the 1995 following note is inserted: Note: The CSRC-list in an RTP packet only 1996 includes participants who's text is included in one or more text 1997 blocks. It is not the same as the list of participants in a 1998 conference. With audio and video media, the CSRC-list would often 1999 contain all participants who are not muted whereas text participants 2000 that don't type are completely silent and so don't show up in RTP 2001 packet CSRC-lists. 2003 12.20. Changes from draft-hellstrom-avtcore-multi-party-rtt-source-00 2004 to -01 2006 Editorial cleanup. 2008 Changed capability indication from fmtp-parameter to SDP attribute 2009 "rtt-mix". 2011 Swapped order of redundancy elements in the example to match reality. 2013 Increased the SDP negotiation section 2015 13. References 2017 13.1. Normative References 2019 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2020 Requirement Levels", BCP 14, RFC 2119, 2021 DOI 10.17487/RFC2119, March 1997, 2022 . 2024 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 2025 Jacobson, "RTP: A Transport Protocol for Real-Time 2026 Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, 2027 July 2003, . 2029 [RFC4102] Jones, P., "Registration of the text/red MIME Sub-Type", 2030 RFC 4102, DOI 10.17487/RFC4102, June 2005, 2031 . 2033 [RFC4103] Hellstrom, G. and P. Jones, "RTP Payload for Text 2034 Conversation", RFC 4103, DOI 10.17487/RFC4103, June 2005, 2035 . 2037 [RFC5630] Audet, F., "The Use of the SIPS URI Scheme in the Session 2038 Initiation Protocol (SIP)", RFC 5630, 2039 DOI 10.17487/RFC5630, October 2009, 2040 . 2042 [RFC5764] McGrew, D. and E. Rescorla, "Datagram Transport Layer 2043 Security (DTLS) Extension to Establish Keys for the Secure 2044 Real-time Transport Protocol (SRTP)", RFC 5764, 2045 DOI 10.17487/RFC5764, May 2010, 2046 . 2048 [RFC6263] Marjou, X. and A. Sollaud, "Application Mechanism for 2049 Keeping Alive the NAT Mappings Associated with RTP / RTP 2050 Control Protocol (RTCP) Flows", RFC 6263, 2051 DOI 10.17487/RFC6263, June 2011, 2052 . 2054 [RFC7675] Perumal, M., Wing, D., Ravindranath, R., Reddy, T., and M. 2055 Thomson, "Session Traversal Utilities for NAT (STUN) Usage 2056 for Consent Freshness", RFC 7675, DOI 10.17487/RFC7675, 2057 October 2015, . 2059 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2060 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2061 May 2017, . 2063 [RFC8825] Alvestrand, H., "Overview: Real-Time Protocols for 2064 Browser-Based Applications", RFC 8825, 2065 DOI 10.17487/RFC8825, January 2021, 2066 . 2068 [RFC8865] Holmberg, C. and G. Hellström, "T.140 Real-Time Text 2069 Conversation over WebRTC Data Channels", RFC 8865, 2070 DOI 10.17487/RFC8865, January 2021, 2071 . 2073 [RFC8866] Begen, A., Kyzivat, P., Perkins, C., and M. Handley, "SDP: 2074 Session Description Protocol", RFC 8866, 2075 DOI 10.17487/RFC8866, January 2021, 2076 . 2078 [T140] ITU-T, "Recommendation ITU-T T.140 (02/1998), Protocol for 2079 multimedia application text conversation", February 1998, 2080 . 2082 [T140ad1] ITU-T, "Recommendation ITU-T.140 Addendum 1 - (02/2000), 2083 Protocol for multimedia application text conversation", 2084 February 2000, 2085 . 2087 13.2. Informative References 2089 [RFC4353] Rosenberg, J., "A Framework for Conferencing with the 2090 Session Initiation Protocol (SIP)", RFC 4353, 2091 DOI 10.17487/RFC4353, February 2006, 2092 . 2094 [RFC4575] Rosenberg, J., Schulzrinne, H., and O. Levin, Ed., "A 2095 Session Initiation Protocol (SIP) Event Package for 2096 Conference State", RFC 4575, DOI 10.17487/RFC4575, August 2097 2006, . 2099 [RFC4579] Johnston, A. and O. Levin, "Session Initiation Protocol 2100 (SIP) Call Control - Conferencing for User Agents", 2101 BCP 119, RFC 4579, DOI 10.17487/RFC4579, August 2006, 2102 . 2104 [RFC5194] van Wijk, A., Ed. and G. Gybels, Ed., "Framework for Real- 2105 Time Text over IP Using the Session Initiation Protocol 2106 (SIP)", RFC 5194, DOI 10.17487/RFC5194, June 2008, 2107 . 2109 [RFC7667] Westerlund, M. and S. Wenger, "RTP Topologies", RFC 7667, 2110 DOI 10.17487/RFC7667, November 2015, 2111 . 2113 [RFC8643] Johnston, A., Aboba, B., Hutton, A., Jesske, R., and T. 2114 Stach, "An Opportunistic Approach for Secure Real-time 2115 Transport Protocol (OSRTP)", RFC 8643, 2116 DOI 10.17487/RFC8643, August 2019, 2117 . 2119 Author's Address 2121 Gunnar Hellstrom 2122 Gunnar Hellstrom Accessible Communication 2123 SE-13670 Vendelso 2124 Sweden 2126 Email: gunnar.hellstrom@ghaccess.se