idnits 2.17.1 draft-ietf-rtcweb-jsep-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The document has examples using IPv4 documentation addresses according to RFC6890, but does not use any IPv6 documentation addresses. Maybe there should be IPv6 examples, too? Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 22, 2012) is 4197 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC4566' is defined on line 862, but no explicit reference was found in the text == Unused Reference: 'RFC5245' is defined on line 888, but no explicit reference was found in the text == Outdated reference: A later version (-01) exists of draft-rescorla-mmusic-ice-trickle-00 ** Obsolete normative reference: RFC 4566 (Obsoleted by RFC 8866) == Outdated reference: A later version (-26) exists of draft-ietf-rtcweb-rtp-usage-04 == Outdated reference: A later version (-08) exists of draft-nandakumar-rtcweb-sdp-00 -- Obsolete informational reference (is this intentional?): RFC 5245 (Obsoleted by RFC 8445, RFC 8839) Summary: 1 error (**), 0 flaws (~~), 6 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group J. Uberti 3 Internet-Draft Google 4 Intended status: Standards Track C. Jennings 5 Expires: April 25, 2013 Cisco 6 October 22, 2012 8 Javascript Session Establishment Protocol 9 draft-ietf-rtcweb-jsep-02 11 Abstract 13 This document proposes a mechanism for allowing a Javascript 14 application to fully control the signaling plane of a multimedia 15 session, and discusses how this would work with existing signaling 16 protocols. 18 Status of this Memo 20 This Internet-Draft is submitted in full conformance with the 21 provisions of BCP 78 and BCP 79. 23 Internet-Drafts are working documents of the Internet Engineering 24 Task Force (IETF). Note that other groups may also distribute 25 working documents as Internet-Drafts. The list of current Internet- 26 Drafts is at http://datatracker.ietf.org/drafts/current/. 28 Internet-Drafts are draft documents valid for a maximum of six months 29 and may be updated, replaced, or obsoleted by other documents at any 30 time. It is inappropriate to use Internet-Drafts as reference 31 material or to cite them other than as "work in progress." 33 This Internet-Draft will expire on April 25, 2013. 35 Copyright Notice 37 Copyright (c) 2012 IETF Trust and the persons identified as the 38 document authors. All rights reserved. 40 This document is subject to BCP 78 and the IETF Trust's Legal 41 Provisions Relating to IETF Documents 42 (http://trustee.ietf.org/license-info) in effect on the date of 43 publication of this document. Please review these documents 44 carefully, as they describe your rights and restrictions with respect 45 to this document. Code Components extracted from this document must 46 include Simplified BSD License text as described in Section 4.e of 47 the Trust Legal Provisions and are provided without warranty as 48 described in the Simplified BSD License. 50 Table of Contents 52 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 53 2. Other Approaches Considered . . . . . . . . . . . . . . . . . 5 54 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 6 55 4. Semantics and Syntax . . . . . . . . . . . . . . . . . . . . . 7 56 4.1. Signaling Model . . . . . . . . . . . . . . . . . . . . . 7 57 4.2. Session Descriptions and State Machine . . . . . . . . . . 7 58 4.3. Session Description Format . . . . . . . . . . . . . . . . 9 59 4.4. ICE . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 60 4.4.1. ICE Candidate Trickling . . . . . . . . . . . . . . . 10 61 4.4.1.1. ICE Candidate Format . . . . . . . . . . . . . . . 10 62 4.5. Interactions With Forking . . . . . . . . . . . . . . . . 11 63 4.5.1. Sequential Forking . . . . . . . . . . . . . . . . . . 11 64 4.5.2. Parallel Forking . . . . . . . . . . . . . . . . . . . 12 65 4.6. Session Rehydration . . . . . . . . . . . . . . . . . . . 13 66 5. Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 14 67 5.1. SDP Requirements . . . . . . . . . . . . . . . . . . . . . 14 68 5.2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . 15 69 5.2.1. createOffer . . . . . . . . . . . . . . . . . . . . . 15 70 5.2.2. createAnswer . . . . . . . . . . . . . . . . . . . . . 15 71 5.2.3. SessionDescriptionType . . . . . . . . . . . . . . . . 16 72 5.2.3.1. Creating Answers . . . . . . . . . . . . . . . . . 17 73 5.2.4. setLocalDescription . . . . . . . . . . . . . . . . . 17 74 5.2.5. setRemoteDescription . . . . . . . . . . . . . . . . . 18 75 5.2.6. localDescription . . . . . . . . . . . . . . . . . . . 18 76 5.2.7. remoteDescription . . . . . . . . . . . . . . . . . . 18 77 5.2.8. updateIce . . . . . . . . . . . . . . . . . . . . . . 18 78 5.2.9. addIceCandidate . . . . . . . . . . . . . . . . . . . 19 79 6. Configurable SDP Parameters . . . . . . . . . . . . . . . . . 20 80 7. Security Considerations . . . . . . . . . . . . . . . . . . . 21 81 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 22 82 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 23 83 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 24 84 10.1. Normative References . . . . . . . . . . . . . . . . . . . 24 85 10.2. Informative References . . . . . . . . . . . . . . . . . . 24 86 Appendix A. JSEP Implementation Examples . . . . . . . . . . . . 26 87 A.1. Example API Flows . . . . . . . . . . . . . . . . . . . . 26 88 A.1.1. Call using ROAP . . . . . . . . . . . . . . . . . . . 26 89 A.1.2. Call using XMPP . . . . . . . . . . . . . . . . . . . 27 90 A.1.3. Adding video to a call, using XMPP . . . . . . . . . . 28 91 A.1.4. Simultaneous add of video streams, using XMPP . . . . 28 92 A.1.5. Call using SIP . . . . . . . . . . . . . . . . . . . . 29 93 A.1.6. Handling early media (e.g. 1-800-GO FEDEX), using 94 SIP . . . . . . . . . . . . . . . . . . . . . . . . . 30 95 Appendix B. Change log . . . . . . . . . . . . . . . . . . . . . 32 96 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 33 98 1. Introduction 100 The thinking behind WebRTC call setup has been to fully specify and 101 control the media plane, but to leave the signaling plane up to the 102 application as much as possible. The rationale is that different 103 applications may prefer to use different protocols, such as the 104 existing SIP or Jingle call signaling protocols, or something custom 105 to the particular application, perhaps for a novel use case. In this 106 approach, the key information that needs to be exchanged is the 107 multimedia session description, which specifies the necessary 108 transport and media configuration information necessary to establish 109 the media plane. 111 The browser environment also has its own challenges that cause 112 problems for an embedded signaling state machine. One of these is 113 that the user may reload the web page at any time. If this happens, 114 and the state machine is being run at a server, the server can simply 115 push the current state back down to the page and resume the call 116 where it left off. 118 This document describes the Javascript Session Establishment Protocol 119 (JSEP) that pulls the signaling state machine out of the browser and 120 into Javascript. This mechanism effectively removes the browser 121 almost completely from the core signaling flow; the only interface 122 needed is a way for the application to pass in the local and remote 123 session descriptions negotiated by whatever signaling mechanism is 124 used, and a way to interact with the ICE state machine. 126 JSEP's handling of session descriptions is simple and 127 straightforward. Whenever an offer/answer exchange is needed, the 128 initiating side creates an offer by calling a createOffer() API. The 129 application optionally modifies that offer, and then uses it to set 130 up its local config via the setLocalDescription() API. The offer is 131 then sent off to the remote side over its preferred signaling 132 mechanism (e.g., WebSockets); upon receipt of that offer, the remote 133 party installs it using the setRemoteDescription() API. 135 When the call is accepted, the callee uses the createAnswer() API to 136 generate an appropriate answer, applies it using 137 setLocalDescription(), and sends the answer back to the initiator 138 over the signaling channel. When the offerer gets that answer, it 139 installs it using setRemoteDescription(), and initial setup is 140 complete. This process can be repeated for additional offer/answer 141 exchanges. 143 Regarding ICE, JSEP decouples the ICE state machine from the overall 144 signaling state machine, as the ICE state machine must remain in the 145 browser, because only the browser has the necessary knowledge of 146 candidates and other transport info. Performing this separation also 147 provides additional flexibility; in protocols that decouple session 148 descriptions from transport, such as Jingle, the transport 149 information can be sent separately; in protocols that don't, such as 150 SIP, the information can be used in the aggregated form. Sending 151 transport information separately can allow for faster ICE and DTLS 152 startup, since the necessary roundtrips can occur while waiting for 153 the remote side to accept the session. 155 The JSEP approach does come with a minor downside. As the 156 application now is responsible for driving the signaling state 157 machine, slightly more application code is necessary to perform call 158 setup; the application must call the right APIs at the right times, 159 and convert the session descriptions and ICE information into the 160 defined messages of its chosen signaling protocol, instead of simply 161 forwarding the messages emitted from the browser. 163 One way to mitigate this is to provide a Javascript library that 164 hides this complexity from the developer, which would implement the 165 state machine and serialization of the desired signaling protocol. 166 For example, this library could convert easily adapt the JSEP API 167 into the exact ROAP API [I-D.jennings-rtcweb-signaling], thereby 168 implementing the ROAP signaling protocol. Such a library could of 169 course also implement other popular signaling protocols, including 170 SIP or Jingle. In this fashion we can enable greater control for the 171 experienced developer without forcing any additional complexity on 172 the novice developer. 174 2. Other Approaches Considered 176 Another approach that was considered for JSEP was to move the 177 mechanism for generating offers and answers out of the browser as 178 well. Instead of providing createOffer/createAnswer methods within 179 the browser, this approach would instead expose a getCapabilities API 180 which would provide the application with the information it needed in 181 order to generate its own session descriptions. This increases the 182 amount of work that the application needs to do; it needs to know how 183 to generate session descriptions from capabilities, and especially 184 how to generate the correct answer from an arbitrary offer and the 185 supported capabilities. While this could certainly be addressed by 186 using a library like the one mentioned above, it basically forces the 187 use of said library even for a simple example. Exposing createOffer/ 188 createAnswer avoids that problem, but still allows applications to 189 generate their own offers/answers if they choose, using the 190 description generated by createOffer as an indication of the 191 browser's capabilities. 193 Note also that while JSEP transfers more control to Javascript, it is 194 not intended to be an example of a "low-level" API. The general 195 argument against a low-level API is that there are too many necessary 196 API points, and they can be called in any order, leading to something 197 that is hard to specify and test. In the approach proposed here, 198 control is performed via session descriptions; this requires only a 199 few APIs to handle these descriptions, and they are evaluated in a 200 specific fashion, which reduces the number of possible states and 201 interactions. 203 3. Terminology 205 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 206 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 207 document are to be interpreted as described in RFC 2119 [RFC2119]. 209 4. Semantics and Syntax 211 4.1. Signaling Model 213 JSEP does not specify a particular signaling model or state machine, 214 other than the generic need to exchange RFC 3264 offers and answers 215 in order for both sides of the session to know how to conduct the 216 session. JSEP provides mechanisms to create offers and answers, as 217 well as to apply them to a session. However, the actual mechanism by 218 which these offers and answers are communicated to the remote side, 219 including addressing, retransmission, forking, and glare handling, is 220 left entirely up to the application. 222 +-----------+ +-----------+ 223 | Web App |<--- App-Specific Signaling -->| Web App | 224 +-----------+ +-----------+ 225 ^ ^ 226 | SDP | SDP 227 V V 228 +-----------+ +-----------+ 229 | Browser |<----------- Media ------------>| Browser | 230 +-----------+ +-----------+ 232 Figure 1: JSEP Signaling Model 234 4.2. Session Descriptions and State Machine 236 In order to establish the media plane, the user agent needs specific 237 parameters to indicate what to transmit to the remote side, as well 238 as how to handle the media that is received. These parameters are 239 determined by the exchange of session descriptions in offers and 240 answers, and there are certain details to this process that must be 241 handled in the JSEP APIs. 243 Whether a session description was sent or received affects the 244 meaning of that description. For example, the list of codecs sent to 245 a remote party indicates what the local side is willing to decode, 246 and what the remote party should send. Not all parameters follow 247 this rule; for example, the SRTP parameters [RFC4568] sent to a 248 remote party indicate what the local side will use to encrypt, and 249 thereby how the remote party should expect to receive. 251 In addition, various RFCs put different conditions on the format of 252 offers versus answers. For example, a offer may propose multiple 253 SRTP configurations, but an answer may only contain a single SRTP 254 configuration. 256 Lastly, while the exact media parameters are only known only after a 257 offer and an answer have been exchanged, it is possible for the 258 offerer to receive media after they have sent an offer and before 259 they have received an answer. To properly process incoming media in 260 this case, the offerer's media handler must be aware of the details 261 of the offerer before the answer arrives. 263 Therefore, in order to handle session descriptions properly, the user 264 agent needs: 266 1. To know if a session description pertains to the local or remote 267 side. 269 2. To know if a session description is an offer or an answer. 271 3. To allow the offer to be specified independently of the answer. 273 JSEP addresses this by adding both a setLocalDescription and a 274 setRemoteDescription method and having session description objects 275 contain a type field indicating the type of session description being 276 supplied. This satisfies the requirements listed above for both the 277 offerer, who first calls setLocalDescription(sdp [offer]) and then 278 later setRemoteDescription(sdp [answer]), as well as for the 279 answerer, who first calls setRemoteDescription(sdp [offer]) and then 280 later setLocalDescription(sdp [answer]). While it could be possible 281 to implicitly determine the value of the offer/answer argument, 282 requiring it to be specified explicitly is more robust, allowing 283 invalid combinations (i.e. an answer before an offer) to generate an 284 appropriate error. 286 JSEP also allows for an answer to be treated as provisional by the 287 application. Provisional answers provide a way for an answerer to 288 communicate initial session parameters back to the offerer, in order 289 to allow the session to begin, while allowing a final answer to be 290 specified later. This concept of a final answer is important to the 291 offer/answer model; when such an answer is received, any extra 292 resources allocated by the caller can be released, now that the exact 293 session configuration is known. These "resources" can include things 294 like extra ICE components, TURN candidates, or video decoders. 295 Provisional answers, on the other hand, do no such deallocation 296 results; as a result, multiple dissimilar provisional answers can be 297 received and applied during call setup. 299 In [RFC3264], the constraints at the signaling level is that only one 300 offer can be outstanding for a given session but from the media stack 301 level, a new offer can be generated at any point. For example, when 302 using SIP for signaling, if one offer is sent, then cancelled using a 303 SIP CANCEL, another offer can be generated even though no answer was 304 received for the first offer. To support this, the JSEP media layer 305 can provide an offer whenever the Javascript application needs one 306 for the signaling. The answerer can send back zero or more 307 provisional answers, and finally end the offer-answer exchange by 308 sending a final answer. The state machine for this is as follows: 310 +-----------+ 311 | | 312 | | 313 | Stable |<---------------\ 314 | | | 315 | | | 316 +-----------+ | 317 ^ | | 318 | | OFFER | 319 ANSWER | | | ANSWER 320 | V | 321 +-----------+ +-----------+ 322 | | | | 323 | | PRANSWER | | 324 | Offer |-------- >| Pranswer | 325 | | | | 326 | |----\ | |----\ 327 +-----------+ | +-----------+ | 328 ^ | ^ | 329 | | | | 330 \-----/ \-----/ 331 OFFER PRANSWER 333 Figure 2: JSEP State Machine 335 Aside from these state transitions, there is no other difference 336 between the handling of provisional ("pranswer") and final ("answer") 337 answers. 339 4.3. Session Description Format 341 In the WebRTC specification, session descriptions are formatted as 342 SDP messages. While this format is not optimal for manipulation from 343 Javascript, it is widely accepted, and frequently updated with new 344 features. Any alternate encoding of session descriptions would have 345 to keep pace with the changes to SDP, at least until the time that 346 this new encoding eclipsed SDP in popularity. As a result, JSEP 347 continues to use SDP as the internal representation for its session 348 descriptions. 350 However, to simplify Javascript processing, and provide for future 351 flexibility, the SDP syntax is encapsulated within a 352 SessionDescription object, which can be constructed from SDP, and be 353 serialized out to SDP. If future specifications agree on a JSON 354 format for session descriptions, we could easily enable this object 355 to generate and consume that JSON. 357 Other methods may be added to SessionDescription in the future to 358 simplify handling of SessionDescriptions from Javascript. Though it 359 is unclear exactly what manipulations developer will commonly want to 360 do to SDP, it would be simple to write a Javascript library to 361 perform these manipulations. 363 4.4. ICE 365 When a new ICE candidate is available, the ICE Agent will notify the 366 application via a callback; these candidates will automatically be 367 added to the local session description. When all candidates have 368 been gathered, the callback will also be invoked to signal that the 369 gathering process is complete. 371 4.4.1. ICE Candidate Trickling 373 Candidate trickling is a technique through which a caller may 374 incrementally provide candidates to the callee after the initial 375 offer has been dispatched; the semantics of "Trickle ICE" are defined 376 in [I-D.rescorla-mmusic-ice-trickle]. This process allows the callee 377 to begin acting upon the call and setting up the ICE (and perhaps 378 DTLS) connections immediately, without having to wait for the caller 379 to gather all possible candidates. This results in faster call 380 startup in cases where gathering is not performed prior to initating 381 the call. 383 JSEP supports optional candidate trickling by providing APIs that 384 provide control and feedback on the ICE candidate gathering process. 385 Applications that support candidate trickling can send the initial 386 offer immediately and send individual candidates when they get the 387 notified of a new candidate; applications that do not support this 388 feature can simply wait for the indication that gathering is 389 complete, and then create and send their offer, with all the 390 candidates, at this time. 392 Upon receipt of trickled candidates, the receiving application will 393 supply them to its ICE Agent. This triggers the ICE Agent to start 394 using the new remote candidates for connectivity checks. 396 4.4.1.1. ICE Candidate Format 398 As with session descriptions, the syntax of the IceCandidate object 399 provides some abstraction, but can be easily converted to and from 400 the SDP a=candidate lines. 402 The a=candidate lines are the only SDP information that is contained 403 within IceCandidate, as they represent the only information needed 404 that is not present in the initial offer (i.e. for trickle 405 candidates). This information is carried with the same syntax as the 406 "a=candidate" line in SDP. For example: 408 a=candidate:1 1 UDP 1694498815 192.0.2.33 10000 typ host 410 The IceCandidate object also contains fields to indicate which m= 411 line it should be associated with. The m line can be identified in 412 one of two ways; either by a m-line index, or a MID. The m-line 413 index is a zero-based index, referring to the Nth m-line in the SDP. 414 The MID uses the "media stream identification", as defined in [RFC 415 3388], to identify the m-line. WebRTC implementations creating an 416 ICE Candidate object MUST populate both of these fields. 417 Implementations receiving an ICE Candidate object SHOULD use the MID 418 if they implement that functionality, or the m-line index, if not. 420 4.5. Interactions With Forking 422 Some call signaling systems allow various types of forking where an 423 SDP Offer may be provided to more than one device. For example, SIP 424 RFC 3261 defines both a "Parallel Search" and "Sequential Search". 425 Although these are primarily signaling level issues that are outside 426 the scope of JSEP, they do have some impact on the configuration of 427 the media plane, which is relevant. When forking is happening at the 428 signaling layer, the Javascript application responsible for the 429 signaling needs to make the decisions about what media should be sent 430 or received at any point of time and which remote endpoint it should 431 communicate with. JSEP is used to make sure the media engine can 432 make the RTP and media perform as required by the application. The 433 basic operations that the applications can have the media engine do 434 are: 436 Start exchanging media to a given remote peer but keep all the 437 resources reserved in the offer. 439 Start exchanging media with a given remote peer and free any 440 resources in the offer that are not being used. 442 4.5.1. Sequential Forking 444 Sequential forking involves a call being dispatched to multiple 445 remote callees, where each callee can accept the call, but only one 446 active session ever exists at a time; no mixing of received media is 447 performed. 449 JSEP handles serial forking well, allowing the application to easily 450 control the policy for selecting the desired remote endpoint. When 451 an answer arrives from one of the callees, the application can choose 452 to apply it either as a provisional answer, leaving open the 453 possibility of using a different answer in the future, or apply it as 454 a final answer, ending the setup flow. 456 In a "first-one-wins" situation, the first answer will be applied as 457 a final answer, and the application will reject any subsequent 458 answers. In SIP parlance, this would be ACK + BYE. 460 In a "last-one-wins" situation, all answers would be applied as 461 provisional answers, and any previous call leg will be terminated. 462 At some point, the application will end the setup process, perhaps 463 with a timer; at this point, the application could reapply the 464 existing remote description as a final answer. 466 4.5.2. Parallel Forking 468 Parallel forking involves a call being dispatched to multiple remote 469 callees, where each callee can accept the call, and multiple 470 simultaneous active signaling sessions can be established as a 471 result. If multiple callees send media at the same time, the 472 possibilities for handling this are described in Section 3.1 of RFC 473 3960. Most SIP devices today only support exchanging media with a 474 single device at a time, and do not try to mix multiple early media 475 audio sources, as that could result in a confusing situation. For 476 example. consider having a European ringback tone mixed together with 477 the North American ringback tone - the resulting sound would not be 478 like either tone, and would confuse the user. If the signaling 479 application wishes to only exchange media with one of the remote 480 endpoints at a time, then from a media engine point of view, this is 481 exactly like the sequential forking case. 483 In the parallel forking case where the Javascript application wishes 484 to simultaneously exchange media with multiple peers, the flow is 485 slightly more complex, but the Javascript application can follow the 486 strategy that RFC 3960 describes using UPDATE. (It is worth noting 487 that use cases where this is the desired behavior are very unusual.) 488 The UPDATE approach allows the signaling to set up a separate media 489 flow for each peer that it wishes to exchange media with. In JSEP, 490 this offer used in the UPDATE would be formed by simply creating a 491 new PeerConnection and making sure that the same local media streams 492 have been added into this new PeerConnection. Then the new 493 PeerConnection object would produce a SDP offer that could be used by 494 the signaling to perform the UPDATE strategy discussed in RFC 3690. 496 As a result of sharing the media streams, the application will end up 497 with N parallel PeerConnection sessions, each with a local and remote 498 description and their own local and remote addresses. The media flow 499 from these sessions can be managed by specifying SDP direction 500 attributes in the descriptions, or the application can choose to play 501 out the media from all sessions mixed together. Of course, if the 502 application wants to only keep a single session, it can simply 503 terminate the sessions that it no longer needs. 505 4.6. Session Rehydration 507 In the event that the local application state is reinitialized, 508 either due to a user reload of the page, or a decision within the 509 application to reload itself (perhaps to update to a new version), it 510 is possible to keep an existing session alive via a process called 511 "rehydration". 513 With rehydration, the current signaling state is persisted somewhere 514 outside of the page, perhaps on the application server, or in browser 515 local storage. The page is then reloaded, and a new session object 516 is created in Javascript. The saved signaling state is now 517 retrieved, and a new PeerConnection object is created for the 518 session. At this point a new offer can be generated by the new 519 PeerConnection, with new ICE and SDES credentials. This can then be 520 used to re-initiate the session with the existing remote endpoint, 521 who simply sees the new offer as an in-call renegotiation, and will 522 reply with an answer that can be supplied to setRemoteDescription. 523 ICE processing proceeds as usual, and as soon as connectivity is 524 established, the session will be back up and running again. 526 Open Issue: EKR proposed an alternative rehydration approach where 527 the actual internal PeerConnection object in the browser was kept 528 alive for some time after the web page was killed and provided some 529 way for a new page to acquire the old PeerConnection object. 531 5. Interface 533 This section details the basic operations that must be present to 534 implement JSEP functionality. The actual API exposed in the W3C API 535 may have somewhat different syntax, but should map easily to these 536 concepts. 538 5.1. SDP Requirements 540 Note: The text in this section may not represent working group 541 consensus and is put here so that the working group can discuss it 542 and find out how to change it such that it does have consensus. 544 When generating SDP blobs, either for offers or answers, the 545 generated SDP needs to conform to the following specifications. 546 Similarly, in order to properly process received SDP blobs, 547 implementations need to implement the functionality described in the 548 following specifications. This list is derived from 549 [I-D.ietf-rtcweb-rtp-usage]. 551 RFC4566 is the base SDP specification and MUST be implemented. 553 RFC5124 MUST be supported for signaling RTP/SAVPF RTP profile. 555 RFC5104 MUST be implemented to signal RTCP based feedback. 557 RFC5761 MUST be implemented to signal multiplexing of RTP and 558 RTCP. 560 RFC5245 MUST be implemented for signaling the ICE candidate lines 561 corresponding to each media stream. 563 RFC3264 MUST be implemented to signal information about media 564 direction. 566 The RFC5888 grouping framework MUST be implemented for signaling 567 the grouping information. 569 RFC5506 MAY be implemented to signal Reduced-Size RTCP messages. 571 RFC5576 MAY be implemented to signal RTP SSRC values. 573 RFC3556 with bandwidth modifiers MAY be supported for specifying 574 RTCP bandwidth as a fraction of the media bandwidth, RTCP fraction 575 allocated to the senders and setting maximum media bit-rate 576 boundaries. 578 As required by RFC 4566 Section 5.13 JSEP implementations MUST ignore 579 unknown attributes (a=) lines. 581 Example SDP for RTCWeb call flows can be found in 582 [I-D.nandakumar-rtcweb-sdp]. 584 5.2. Methods 586 5.2.1. createOffer 588 The createOffer method generates a blob of SDP that contains a RFC 589 3264 offer with the supported configurations for the session, 590 including descriptions of the local MediaStreams attached to this 591 PeerConnection, the codec/RTP/RTCP options supported by this 592 implementation, and any candidates that have been gathered by the ICE 593 Agent. A constraints parameters may be supplied to provide 594 additional control over the generated offer, e.g. to get a full set 595 of session capabilities, or to request a new set of ICE credentials. 597 In the initial offer, the generated SDP will contain all desired 598 functionality for the session (certain parts that are supported but 599 not desired by default may be omitted); for each SDP line, the 600 generation of the SDP must follow the appropriate process for 601 generating an offer. In the event createOffer is called after the 602 session is established, createOffer will generate an offer that is 603 compatible with the current session, incorporating any changes that 604 have been made to the session since the last complete offer-answer 605 exchange, such as addition or removal of streams. If no changes have 606 been made, the offer will be identical to the current local 607 description. 609 Session descriptions generated by createOffer must be immediately 610 usable by setLocalDescription; if a system has limited resources 611 (e.g. a finite number of decoders), createOffer should return an 612 offer that reflects the current state of the system, so that 613 setLocalDescription will succeed when it attempts to acquire those 614 resources. Because this method may need to inspect the system state 615 to determine the currently available resources, it may be implemented 616 as an async operation. 618 Calling this method may do things such as generate new ICE 619 credentials, but does not change media state. 621 5.2.2. createAnswer 623 The createAnswer method generates a blob of SDP that contains a RFC 624 3264 SDP answer with the supported configuration for the session that 625 is compatible with the parameters supplied in the offer. Like 626 createOffer, the returned blob contains descriptions of the local 627 MediaStreams attached to this PeerConnection, the codec/RTP/RTCP 628 options negotiated for this session, and any candidates that have 629 been gathered by the ICE Agent. A constraints parameter may be 630 supplied to provide additional control over the generated answer. 632 As an answer, the generated SDP will contain a specific configuration 633 that specifies how the media plane should be established. 635 Session descriptions generated by createAnswer must be immediately 636 usable by setLocalDescription; like createOffer, the returned 637 description should reflect the current state of the system. Because 638 this method may need to inspect the system state to determine the 639 currently available resources, it may need to be implemented as an 640 async operation. 642 Calling this method may do things such as generate new ICE 643 credentials, but does not change media state. 645 5.2.3. SessionDescriptionType 647 Session description objects (RTCSessionDescription) may be of type 648 "offer", "pranswer", and "answer". These types provide information 649 as to how the description parameter should be parsed, and how the 650 media state should be changed. 652 "offer" indicates that a description should be parsed as an offer; 653 said description may include many possible media configurations. A 654 description used as an "offer" may be applied anytime the 655 PeerConnection is in a stable state, or as an update to a previously 656 sent but unanswered "offer". 658 "pranswer" indicates that a description should be parsed as an 659 answer, but not a final answer, and so should not result in the 660 freeing of allocated resources. It may result in the start of media 661 transmission, if the answer does not specify an inactive media 662 direction. A description used as a "pranswer" may be applied as a 663 response to an "offer", or an update to a previously sent "answer". 665 "answer" indicates that a description should be parsed as an answer, 666 the offer-answer exchange should be considered complete, and any 667 resources (decoders, candidates) that are no longer needed can be 668 released. A description used as an "answer" may be applied as a 669 response to a "offer", or an update to a previously sent "pranswer". 671 The application can use some discretion on whether an answer should 672 be applied as provisional or final. For example, in a serial forking 673 scenario, an application may receive multiple "final" answers, one 674 from each remote endpoint. The application could accept the initial 675 answers as provisional answers, and only apply an answer as final 676 when it receives one that meets its criteria (e.g. a live user 677 instead of voicemail). 679 5.2.3.1. Creating Answers 681 Most web applications will not need to create answers using the 682 "pranswer" type. The general recommendation for a web application 683 would be to create an answer more or less immediately after receiving 684 the offer, instead of waiting for a human user to provide input. 685 Later when the human input is received, the applications can create a 686 new offer to update the previous offer/answer pair. Some 687 applications may not be able to do this, particularly ones that Some 688 application may not be able to do this, particular ones that are 689 attempting to gateway to other signaling protocols. 691 Consider a typical web application that will set up a data channel, 692 an audio channel, and a video channel. When an endpoint receives an 693 offer with these channels, it could send an answer accepting the data 694 channel for two-way data, and accepting the audio and video tracks as 695 receive-only. It could then ask the user if they wanted to transmit 696 audio and video to the far end, acquire the local media streams, and 697 send a new offer to the remote side moving the audio and video to be 698 two-way media. By the time the human has authorized sending media, 699 it is likely that the ICE and DTLS handshaking with the remote side 700 will already be set up. 702 5.2.4. setLocalDescription 704 The setLocalDescription method instructs the PeerConnection to apply 705 the supplied SDP blob as its local configuration. The type field 706 indicates whether the blob should be processed as an offer, 707 provisional answer, or final answer; offers and answers are checked 708 differently, using the various rules that exist for each SDP line. 710 This API changes the local media state; among other things, it sets 711 up local resources for receiving and decoding media. In order to 712 successfully handle scenarios where the application wants to offer to 713 change from one media format to a different, incompatible format, the 714 PeerConnection must be able to simultaneously support use of both the 715 old and new local descriptions (e.g. support codecs that exist in 716 both descriptions) until a final answer is received, at which point 717 the PeerConnection can fully adopt the new local description, or roll 718 back to the old description if the remote side denied the change. 720 If setRemoteDescription was previous called with an offer, and 721 setLocalDescription is called with an answer (provisional or final), 722 and the media directions are compatible, this will result in the 723 starting of media transmission. 725 5.2.5. setRemoteDescription 727 The setRemoteDescription method instructs the PeerConnection to apply 728 the supplied SDP blob as the desired remote configuration. As in 729 setLocalDescription, the type field of the indicates how the blob 730 should be processed. 732 This API changes the local media state; among other things, it sets 733 up local resources for sending and encoding media. 735 If setRemoteDescription was previous called with an offer, and 736 setLocalDescription is called with an answer (provisional or final), 737 and the media directions are compatible, this will result in the 738 starting of media transmission. 740 5.2.6. localDescription 742 The localDescription method returns a copy of the current local 743 configuration, i.e. what was most recently passed to 744 setLocalDescription, plus any local candidates that have been 745 generated by the ICE Agent. 747 A null object will be returned if the local description has not yet 748 been established. 750 5.2.7. remoteDescription 752 The remoteDescription method returns a copy of the current remote 753 configuration, i.e. what was most recently passed to 754 setRemoteDescription, plus any remote candidates that have been 755 supplied via processIceMessage. 757 A null object will be returned if the remote description has not yet 758 been established. 760 5.2.8. updateIce 762 The updateIce method allows the configuration of the ICE Agent to be 763 changed during the session, primarily for changing which types of 764 local candidates are provided to the application and used for 765 connectivity checks. A callee may initially configure the ICE Agent 766 to use only relay candidates, to avoid leaking location information, 767 but update this configuration to use all candidates once the call is 768 accepted. 770 Regardless of the configuration, the gathering process collects all 771 available candidates, but excluded candidates will not be surfaced in 772 onicecallback or used for connectivity checks. 774 This call may result in a change to the state of the ICE Agent, and 775 may result in a change to media state if it results in connectivity 776 being established. 778 5.2.9. addIceCandidate 780 The addIceCandidate method provides a remote candidate to the ICE 781 Agent, which will be added to the remote description. Connectivity 782 checks will be sent to the new candidate. 784 This call will result in a change to the state of the ICE Agent, and 785 may result in a change to media state if it results in connectivity 786 being established. 788 6. Configurable SDP Parameters 790 Note: This section is still very early and is likely to 791 significantly change as we get a better understanding of the a) the 792 use cases for this b) the implications at the protocol level c) 793 feedback from implementors on what they can do. 795 The following is a partial list of SDP parameters that an application 796 may want to control, in either local or remote descriptions, using 797 this API. 799 o remove or reorder codecs (m=) 801 o change codec attributes (a=fmtp; ptime) 803 o enable/disable BUNDLE (a=group) 805 o enable/disable RTCP mux (a=rtcp-mux) 807 o change send resolution or framerate (TBD) 809 o change desired recv resolution or framerate (TBD) 811 o change total bandwidth (b=) 813 o remove desired AVPF mechanisms (a=rtcp-fb) 815 o remove RTP header extensions (a=rtphdr-ext) 817 o add/change SSRC grouping (e.g. FID, RTX, etc) (a=ssrc-group) 819 o add SSRC attributes (a=ssrc) 821 o change media send/recv state (a=sendonly/recvonly/inactive) 823 For example, an application could implement call hold by adding an 824 a=inactive attribute to its local description, and then applying and 825 signaling that description. 827 7. Security Considerations 829 TODO 831 8. IANA Considerations 833 This document requires no actions from IANA. 835 9. Acknowledgements 837 Harald Alvestrand, Dan Burnett, Neil Stratford, Eric Rescorla, Anant 838 Narayanan, and Adam Bergkvist all provided valuable feedback on this 839 proposal. Suhas Nandakumar provided text and input for SDP 840 requirements. Matthew Kaufman provided the observation that keeping 841 state out of the browser allows a call to continue even if the page 842 is reloaded. 844 10. References 846 10.1. Normative References 848 [I-D.rescorla-mmusic-ice-trickle] 849 Rescorla, E., Uberti, J., and E. Ivov, "Trickle ICE: 850 Incremental Provisioning of Candidates for the Interactive 851 Connectivity Establishment (ICE) Protocol", 852 draft-rescorla-mmusic-ice-trickle-00 (work in progress), 853 October 2012. 855 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 856 Requirement Levels", BCP 14, RFC 2119, March 1997. 858 [RFC3264] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model 859 with Session Description Protocol (SDP)", RFC 3264, 860 June 2002. 862 [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 863 Description Protocol", RFC 4566, July 2006. 865 10.2. Informative References 867 [I-D.ietf-rtcweb-rtp-usage] 868 Perkins, C., Westerlund, M., and J. Ott, "Web Real-Time 869 Communication (WebRTC): Media Transport and Use of RTP", 870 draft-ietf-rtcweb-rtp-usage-04 (work in progress), 871 July 2012. 873 [I-D.jennings-rtcweb-signaling] 874 Jennings, C., Rosenberg, J., and R. Jesup, "RTCWeb Offer/ 875 Answer Protocol (ROAP)", 876 draft-jennings-rtcweb-signaling-01 (work in progress), 877 October 2011. 879 [I-D.nandakumar-rtcweb-sdp] 880 Nandakumar, S. and C. Jennings, "SDP for the WebRTC", 881 draft-nandakumar-rtcweb-sdp-00 (work in progress), 882 October 2012. 884 [RFC4568] Andreasen, F., Baugher, M., and D. Wing, "Session 885 Description Protocol (SDP) Security Descriptions for Media 886 Streams", RFC 4568, July 2006. 888 [RFC5245] Rosenberg, J., "Interactive Connectivity Establishment 889 (ICE): A Protocol for Network Address Translator (NAT) 890 Traversal for Offer/Answer Protocols", RFC 5245, 891 April 2010. 893 [W3C.WD-webrtc-20111027] 894 Bergkvist, A., Burnett, D., Narayanan, A., and C. 895 Jennings, "WebRTC 1.0: Real-time Communication Between 896 Browsers", World Wide Web Consortium WD WD-webrtc- 897 20111027, October 2011, 898 . 900 Appendix A. JSEP Implementation Examples 902 A.1. Example API Flows 904 Below are several sample flows for the new PeerConnection and library 905 APIs, demonstrating when the various APIs are called in different 906 situations and with various transport protocols. For clarity and 907 simplicity, the createOffer/createAnswer calls are assumed to be 908 synchronous in these examples, whereas the actual APIs are async. 910 A.1.1. Call using ROAP 912 This example demonstrates a ROAP call, without the use of trickle 913 candidates. 915 // Call is initiated toward Answerer 916 OffererJS->OffererUA: pc = new PeerConnection(); 917 OffererJS->OffererUA: pc.addStream(localStream, null); 918 OffererUA->OffererJS: iceCallback(candidate); 919 OffererJS->OffererUA: offer = pc.createOffer(null); 920 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 921 OffererJS->AnswererJS: {"type":"OFFER", "sdp":offer } 923 // OFFER arrives at Answerer 924 AnswererJS->AnswererUA: pc = new PeerConnection(); 925 AnswererJS->AnswererUA: pc.setRemoteDescription("offer", msg.sdp); 926 AnswererUA->AnswererJS: onaddstream(remoteStream); 927 AnswererUA->OffererUA: iceCallback(candidate); 929 // Answerer accepts call 930 AnswererJS->AnswererUA: peer.addStream(localStream, null); 931 AnswererJS->AnswererUA: answer = peer.createAnswer(msg.sdp, null); 932 AnswererJS->AnswererUA: peer.setLocalDescription("answer", answer); 933 AnswererJS->OffererJS: {"type":"ANSWER","sdp":answer } 935 // ANSWER arrives at Offerer 936 OffererJS->OffererUA: peer.setRemoteDescription("answer", answer); 937 OffererUA->OffererJS: onaddstream(remoteStream); 939 // ICE Completes (at Answerer) 940 AnswererUA->AnswererJS: onopen(); 941 AnswererUA->OffererUA: Media 943 // ICE Completes (at Offerer) 944 OffererUA->OffererJS: onopen(); 945 OffererJS->AnswererJS: {"type":"OK" } 946 OffererUA->AnswererUA: Media 948 A.1.2. Call using XMPP 950 This example demonstrates an XMPP call, making use of trickle 951 candidates. 953 // Call is initiated toward Answerer 954 OffererJS->OffererUA: pc = new PeerConnection(); 955 OffererJS->OffererUA: pc.addStream(localStream, null); 956 OffererJS->OffererUA: offer = pc.createOffer(null); 957 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 958 OffererJS: xmpp = createSessionInitiate(offer); 959 OffererJS->AnswererJS: 961 OffererJS->OffererUA: pc.startIce(); 962 OffererUA->OffererJS: onicecandidate(cand); 963 OffererJS: createTransportInfo(cand); 964 OffererJS->AnswererJS: 966 // session-initiate arrives at Answerer 967 AnswererJS->AnswererUA: pc = new PeerConnection(); 968 AnswererJS: offer = parseSessionInitiate(xmpp); 969 AnswererJS->AnswererUA: pc.setRemoteDescription("offer", offer); 970 AnswererUA->AnswererJS: onaddstream(remoteStream); 972 // transport-infos arrive at Answerer 973 AnswererJS->AnswererUA: candidate = parseTransportInfo(xmpp); 974 AnswererJS->AnswererUA: pc.addIceCandidate(candidate); 975 AnswererUA->AnswererJS: onicecandidate(cand) 976 AnswererJS: createTransportInfo(cand); 977 AnswererJS->OffererJS: 979 // transport-infos arrive at Offerer 980 OffererJS->OffererUA: candidates = parseTransportInfo(xmpp); 981 OffererJS->OffererUA: pc.addIceCandidate(candidates); 983 // Answerer accepts call 984 AnswererJS->AnswererUA: peer.addStream(localStream, null); 985 AnswererJS->AnswererUA: answer = peer.createAnswer(offer, null); 986 AnswererJS: xmpp = createSessionAccept(answer); 987 AnswererJS->AnswererUA: pc.setLocalDescription("answer", answer); 988 AnswererJS->OffererJS: 990 // session-accept arrives at Offerer 991 OffererJS: answer = parseSessionAccept(xmpp); 992 OffererJS->OffererUA: peer.setRemoteDescription("answer", answer); 993 OffererUA->OffererJS: onaddstream(remoteStream); 995 // ICE Completes (at Answerer) 996 AnswererUA->AnswererJS: onopen(); 997 AnswererUA->OffererUA: Media 999 // ICE Completes (at Offerer) 1000 OffererUA->OffererJS: onopen(); 1001 OffererUA->AnswererUA: Media 1003 A.1.3. Adding video to a call, using XMPP 1005 This example demonstrates an XMPP call, where the XMPP content-add 1006 mechanism is used to add video media to an existing session. For 1007 simplicity, candidate exchange is not shown. 1009 Note that the offerer for the change to the session may be different 1010 than the original call offerer. 1012 // Offerer adds video stream 1013 OffererJS->OffererUA: pc.addStream(videoStream) 1014 OffererJS->OffererUA: offer = pc.createOffer(null); 1015 OffererJS: xmpp = createContentAdd(offer); 1016 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 1017 OffererJS->AnswererJS: 1019 // content-add arrives at Answerer 1020 AnswererJS: offer = parseContentAdd(xmpp); 1021 AnswererJS->AnswererUA: pc.setRemoteDescription("offer", offer); 1022 AnswererJS->AnswererUA: answer = pc.createAnswer(offer, null); 1023 AnswererJS->AnswererUA: pc.setLocalDescription("answer", answer); 1024 AnswererJS: xmpp = createContentAccept(answer); 1025 AnswererJS->OffererJS: 1027 // content-accept arrives at Offerer 1028 OffererJS: answer = parseContentAccept(xmpp); 1029 OffererJS->OffererUA: pc.setRemoteDescription("answer", answer); 1031 A.1.4. Simultaneous add of video streams, using XMPP 1033 This example demonstrates an XMPP call, where new video sources are 1034 added at the same time to a call that already has video; since adding 1035 these sources only affects one side of the call, there is no 1036 conflict. The XMPP description-info mechanism is used to indicate 1037 the new sources to the remote side. 1039 // Offerer and "Answerer" add video streams at the same time 1040 OffererJS->OffererUA: pc.addStream(offererVideoStream2) 1041 OffererJS->OffererUA: offer = pc.createOffer(null); 1042 OffererJS: xmpp = createDescriptionInfo(offer); 1043 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 1044 OffererJS->AnswererJS: 1046 AnswererJS->AnswererUA: pc.addStream(answererVideoStream2) 1047 AnswererJS->AnswererUA: offer = pc.createOffer(null); 1048 AnswererJS: xmpp = createDescriptionInfo(offer); 1049 AnswererJS->AnswererUA: pc.setLocalDescription("offer", offer); 1050 AnswererJS->OffererJS: 1052 // description-info arrives at "Answerer", and is acked 1053 AnswererJS: offer = parseDescriptionInfo(xmpp); 1054 AnswererJS->OffererJS: // ack 1056 // description-info arrives at Offerer, and is acked 1057 OffererJS: offer = parseDescriptionInfo(xmpp); 1058 OffererJS->AnswererJS: // ack 1060 // ack arrives at Offerer; remote offer is used as an answer 1061 OffererJS->OffererUA: pc.setRemoteDescription("answer", offer); 1063 // ack arrives at "Answerer"; remote offer is used as an answer 1064 AnswererJS->AnswererUA: pc.setRemoteDescription("answer", offer); 1066 A.1.5. Call using SIP 1068 This example demonstrates a simple SIP call (e.g. where the client 1069 talks to a SIP proxy over WebSockets). 1071 // Call is initiated toward Answerer 1072 OffererJS->OffererUA: pc = new PeerConnection(); 1073 OffererJS->OffererUA: pc.addStream(localStream, null); 1074 OffererUA->OffererJS: onicecandidate(candidate); 1075 OffererJS->OffererUA: offer = pc.createOffer(null); 1076 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 1077 OffererJS: sip = createInvite(offer); 1078 OffererJS->AnswererJS: SIP INVITE w/ SDP 1080 // INVITE arrives at Answerer 1081 AnswererJS->AnswererUA: pc = new PeerConnection(); 1082 AnswererJS: offer = parseInvite(sip); 1083 AnswererJS->AnswererUA: pc.setRemoteDescription("offer", offer); 1084 AnswererUA->AnswererJS: onaddstream(remoteStream); 1085 AnswererUA->OffererUA: onicecandidate(candidate); 1087 // Answerer accepts call 1088 AnswererJS->AnswererUA: peer.addStream(localStream, null); 1089 AnswererJS->AnswererUA: answer = peer.createAnswer(offer, null); 1090 AnswererJS: sip = createResponse(200, answer); 1091 AnswererJS->AnswererUA: peer.setLocalDescription("answer", answer); 1092 AnswererJS->OffererJS: 200 OK w/ SDP 1094 // 200 OK arrives at Offerer 1095 OffererJS: answer = parseResponse(sip); 1096 OffererJS->OffererUA: peer.setRemoteDescription("answer", answer); 1097 OffererUA->OffererJS: onaddstream(remoteStream); 1098 OffererJS->AnswererJS: ACK 1100 // ICE Completes (at Answerer) 1101 AnswererUA->AnswererJS: onopen(); 1102 AnswererUA->OffererUA: Media 1104 // ICE Completes (at Offerer) 1105 OffererUA->OffererJS: onopen(); 1106 OffererUA->AnswererUA: Media 1108 A.1.6. Handling early media (e.g. 1-800-GO FEDEX), using SIP 1110 This example demonstrates how early media could be handled; for 1111 simplicity, only the offerer side of the call is shown. 1113 // Call is initiated toward Answerer 1114 OffererJS->OffererUA: pc = new PeerConnection(); 1115 OffererJS->OffererUA: pc.addStream(localStream, null); 1116 OffererUA->OffererJS: onicecandidate(candidate); 1117 OffererJS->OffererUA: offer = pc.createOffer(null); 1118 OffererJS->OffererUA: pc.setLocalDescription("offer", offer); 1119 OffererJS: sip = createInvite(offer); 1120 OffererJS->AnswererJS: SIP INVITE w/ SDP 1122 // 180 Ringing is received by offerer, w/ SDP 1123 OffererJS: answer = parseResponse(sip); 1124 OffererJS->OffererUA: pc.setRemoteDescription("pranswer", answer); 1125 OffererUA->OffererJS: onaddstream(remoteStream); 1127 // ICE Completes (at Offerer) 1128 OffererUA->OffererJS: onopen(); 1129 OffererUA->AnswererUA: Media 1131 // 200 OK arrives at Offerer 1132 OffererJS: answer = parseResponse(sip); 1133 OffererJS->OffererUA: pc.setRemoteDescription("answer", answer); 1134 OffererJS->AnswererJS: ACK 1136 Appendix B. Change log 1138 Changes in draft -02: 1140 o Converted from nroff 1142 o Removed comparisons to old approaches abandoned by the working 1143 group 1145 o Removed stuff that has moved to W3C specificaiton 1147 o Align SDP handling with W3C draft 1149 o Clarified section on forking. 1151 Changes in draft -01: 1153 o Added diagrams for architecture and state machine. 1155 o Added sections on forking and rehydration. 1157 o Clarified meaning of "pranswer" and "answer". 1159 o Reworked how ICE restarts and media directions are controlled. 1161 o Added list of parameters that can be changed in a description. 1163 o Updated suggested API and examples to match latest thinking. 1165 o Suggested API and examples have been moved to an appendix. 1167 Changes in draft -00: 1169 o Migrated from draft-uberti-rtcweb-jsep-02. 1171 Authors' Addresses 1173 Justin Uberti 1174 Google 1175 747 6th Ave S 1176 Kirkland, WA 98033 1177 USA 1179 Email: justin@uberti.name 1181 Cullen Jennings 1182 Cisco 1183 170 West Tasman Drive 1184 San Jose, CA 95134 1185 USA 1187 Email: fluffy@iii.ca