idnits 2.17.1 draft-bormann-coap-misc-25.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 seems to lack a Security Considerations section. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 735 has weird spacing: '... code mid...' -- The document date (May 21, 2013) is 3993 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'RFCXXXX' is mentioned on line 1450, but not defined -- Looks like a reference, but probably isn't: '0' on line 1092 == Unused Reference: 'CoRE201' is defined on line 354, but no explicit reference was found in the text == Outdated reference: A later version (-18) exists of draft-ietf-core-coap-16 == Outdated reference: A later version (-16) exists of draft-ietf-core-observe-08 == Outdated reference: A later version (-26) exists of draft-ietf-httpbis-p1-messaging-22 == Outdated reference: A later version (-26) exists of draft-ietf-httpbis-p4-conditional-22 == Outdated reference: A later version (-26) exists of draft-ietf-httpbis-p6-cache-22 ** Obsolete normative reference: RFC 2616 (Obsoleted by RFC 7230, RFC 7231, RFC 7232, RFC 7233, RFC 7234, RFC 7235) == Outdated reference: A later version (-09) exists of draft-bormann-cbor-00 -- Obsolete informational reference (is this intentional?): RFC 5226 (Obsoleted by RFC 8126) Summary: 2 errors (**), 0 flaws (~~), 10 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 CoRE Working Group C. Bormann 3 Internet-Draft K. Hartke 4 Intended status: Informational Universitaet Bremen TZI 5 Expires: November 22, 2013 May 21, 2013 7 Miscellaneous additions to CoAP 8 draft-bormann-coap-misc-25 10 Abstract 12 This short I-D makes a number of partially interrelated proposals how 13 to solve certain problems in the CoRE WG's main protocol, the 14 Constrained Application Protocol (CoAP). The current version has 15 been resubmitted to keep information about these proposals available; 16 the proposals are not all fleshed out at this point in time. 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 November 22, 2013. 35 Copyright Notice 37 Copyright (c) 2013 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. Observing Resources in CoAP . . . . . . . . . . . . . . . . . 4 54 3. The Base-Uri Option . . . . . . . . . . . . . . . . . . . . . 6 55 4. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 7 56 5. References . . . . . . . . . . . . . . . . . . . . . . . . . 7 57 5.1. Normative References . . . . . . . . . . . . . . . . . . 7 58 5.2. Informative References . . . . . . . . . . . . . . . . . 8 59 Appendix A. The Nursery (Things that still need to ripen a bit) 9 60 A.1. Envelope Options . . . . . . . . . . . . . . . . . . . . 9 61 A.2. Payload-Length Option . . . . . . . . . . . . . . . . . . 10 62 A.3. URI Authorities with Binary Adresses . . . . . . . . . . 10 63 A.4. Length-aware number encoding (o256) . . . . . . . . . . . 11 64 A.5. SMS encoding . . . . . . . . . . . . . . . . . . . . . . 13 65 A.5.1. ASCII-optimized SMS encoding . . . . . . . . . . . . 14 66 A.6. CONNECT . . . . . . . . . . . . . . . . . . . . . . . . . 17 67 A.6.1. Requesting a Tunnel with CONNECT . . . . . . . . . . 17 68 A.6.2. Using a CONNECT Tunnel . . . . . . . . . . . . . . . 17 69 A.6.3. Closing down a CONNECT Tunnel . . . . . . . . . . . . 18 70 Appendix B. The Museum (Things we did, but maybe not exactly 71 this way) . . . . . . . . . . . . . . . . . . . . . 18 72 B.1. Getting rid of artificial limitations . . . . . . . . . . 18 73 B.1.1. Beyond 270 bytes in a single option . . . . . . . . . 19 74 B.1.2. Beyond 15 options . . . . . . . . . . . . . . . . . . 20 75 B.1.3. Implementing the option delimiter for 15 or more 76 options . . . . . . . . . . . . . . . . . . . . . . . 23 77 B.1.4. Option Length encoding beyond 270 bytes . . . . . . . 23 78 B.2. Registered Option . . . . . . . . . . . . . . . . . . . . 26 79 B.2.1. A Separate Suboption Number Space . . . . . . . . . . 26 80 B.2.2. Opening Up the Option Number Space . . . . . . . . . 27 81 B.3. Enabling Protocol Evolution . . . . . . . . . . . . . . . 31 82 B.3.1. Potential new option number allocation . . . . . . . 32 83 B.4. Patience, Leisure, and Pledge . . . . . . . . . . . . . . 34 84 B.4.1. Patience . . . . . . . . . . . . . . . . . . . . . . 34 85 B.4.2. Leisure . . . . . . . . . . . . . . . . . . . . . . . 35 86 B.4.3. Pledge . . . . . . . . . . . . . . . . . . . . . . . 35 87 B.4.4. Option Formats . . . . . . . . . . . . . . . . . . . 36 88 Appendix C. The Cemetery (Things we won't do) . . . . . . . . . 36 89 C.1. Example envelope option: solving #230 . . . . . . . . . . 36 90 C.2. Example envelope option: proxy-elective options . . . . . 37 91 C.3. Stateful URI compression . . . . . . . . . . . . . . . . 38 92 Appendix D. Experimental Options . . . . . . . . . . . . . . . . 39 93 D.1. Options indicating absolute time . . . . . . . . . . . . 39 94 D.2. Representing Durations . . . . . . . . . . . . . . . . . 41 95 D.3. Rationale . . . . . . . . . . . . . . . . . . . . . . . . 42 96 D.4. Pseudo-Floating Point . . . . . . . . . . . . . . . . . . 42 97 D.5. A Duration Type for CoAP . . . . . . . . . . . . . . . . 43 98 D.6. CONTOUR (CoAP Non-trivial Option Useful Representation) . 50 99 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 50 101 1. Introduction 103 The CoRE WG is tasked with standardizing an Application Protocol for 104 Constrained Networks/Nodes, CoAP [I-D.ietf-core-coap]. This protocol 105 is intended to provide RESTful [REST] services not unlike HTTP 106 [RFC2616], while reducing the complexity of implementation as well as 107 the size of packets exchanged in order to make these services useful 108 in a highly constrained network of themselves highly constrained 109 nodes. 111 This objective requires restraint in a number of sometimes 112 conflicting ways: 114 o reducing implementation complexity in order to minimize code size, 116 o reducing message sizes in order to minimize the number of 117 fragments needed for each message (in turn to maximize the 118 probability of delivery of the message), the amount of 119 transmission power needed and the loading of the limited-bandwidth 120 channel, 122 o reducing requirements on the environment such as stable storage, 123 good sources of randomness or user interaction capabilities. 125 This draft attempts to address a number of problems not yet 126 adequately solved in [I-D.ietf-core-coap]. The solutions proposed to 127 these problems are somewhat interrelated and are therefore presented 128 in one draft. As of the current version of the draft, the main body 129 is almost empty, since few significant problems remain with CoAP or 130 its satellite specifications. 132 The appendix contains the "CoAP cemetery" (Appendix C, possibly later 133 to move into its own draft), documenting roads that the WG decided 134 not to take, in order to spare readers from reinventing them in vain. 135 There is also a "CoAP museum" (Appendix B), which documents previous 136 forms of proposals part of which did make it into the main documents 137 in one form or another. Finally, the "CoAP nursery" (Appendix A) 138 contains half- to fully-baked proposals that might become interesting 139 as the basis for future extensions. 141 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 142 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 143 document are to be interpreted as described in [RFC2119]. 145 The term "byte" is used in its now customary sense as a synonym for 146 "octet". 148 2. Observing Resources in CoAP 150 (Co-Author for this section: Matthias Kovatsch) 152 There are two open issues related to -observe 153 [I-D.ietf-core-observe]: 155 o mixing freshness and observation lifetime, and 157 o non-cacheable resources. 159 To solve the first issue, we think that -observe should be clarified 160 as follows: 162 A server sends at least some notifications as confirmable messages. 163 Each confirmable notification is an opportunity for the server to 164 check if the client is still there. If the client acknowledges the 165 notification, it is assumed to be well and alive and still interested 166 in the resource. If it rejects the message with a reset message or 167 if it doesn't respond, it is assumed not longer to be interested and 168 is removed from the list of observers. So an observation 169 relationship can potentially go on forever, if the client 170 acknowledges each confirmable notification. If the server doesn't 171 send a notification for a while and wants to check if the client is 172 still there, it may send a confirmable notification with the current 173 resource state to check that. 175 So there is no mixing of freshness and lifetime going on. 177 The other issue is a bit less trivial to solve. The problem is that 178 normal CoAP and -observe actually have very different freshness 179 models: 181 Normally, when a client wants to know the current state of a 182 resource, it retrieves a representation, uses it and stores it in its 183 cache. Later, when it wants to know the current state again, it can 184 either use the stored representation provided that it's still fresh, 185 or retrieve a new representation, use it and store it in its cache. 187 If a server knows when the state of the resource will change the next 188 time, it can set the Max-Age of the representation to an accurate 189 time span. So the change of the resource state will coincide with 190 the expiration of the freshness of the representation stored in the 191 client's cache (ignoring network latency). 193 But if the resource changes its state unpredictably at any time, the 194 server can set the Max-Age only to an estimate. If the state then 195 actually changes before the freshness expires, the client wrongly 196 believes it has fresh information. Conversely, if the freshness 197 expires and the client wants to know the current state, the client 198 wrongly believes it has to make a new request although the 199 representation is actually still fresh (this is defused by ETag 200 validation). 202 -observe doesn't have these kinds of problems: the server does not 203 have to predict when the resource will change its state the next 204 time. It just sends a notification when it does. The new 205 representation invalidates the old representation stored in the 206 client's cache. So the client always has a fresh representation that 207 it can use when it wants to know the current resource state without 208 ever having to make a request. An explicit Max-Age is not needed for 209 determining freshness. 211 But -observe has a different set of problems: 213 The first problem is that the resource may change its state more 214 often than there is bandwidth available or the client can handle. 215 Thus, -observe cannot make any guarantee that a client will see every 216 state change. The solution is that -observe guarantees that the 217 client will eventually see the latest state change, and follows a 218 best effort approach to enable the client to see as many state 219 changes as possible. 221 The second problem is that, when a notification doesn't arrive for a 222 while, the client does not know if the resource did not change its 223 state or if the server lost its state and forgot that the client is 224 interested in the resource. We propose the following solution: With 225 each notification that the server sends, it makes a promise to send 226 another notification, and that it will send this next notification at 227 latest after a certain time span. This time span is included with 228 each notification. So when no notification arrives for a while and 229 the time span has not expired yet, the client assumes that the 230 resource did not change its state. If the time span has expired, no 231 notification has arrived and the client wants to know the current 232 state of the resource, it has to make a new request. 234 The third problem is that, when an intermediary is observing a 235 resource and wants to create a response from a representation stored 236 in its cache, it needs to specify a Max-Age. But the intermediary 237 cannot predict when it will receive the next notification, because 238 the next notification can arrive at any time. Unlike the origin 239 server, it also doesn't have the application-specific knowledge that 240 the origin server has. We propose the following solution: With each 241 notification a server sends, it includes a value that an intermediary 242 should use to calculate the Max-Age. 244 To summarize: 246 o A notification doesn't have a Max-Age; it's fresh until the next 247 notification arrives. A notification is the promise for another 248 notification that will arrive at latest after Next-Notification- 249 At-Latest. This value is included with every notification. The 250 promise includes that the server attempts to transmit a 251 notification to the client for the promised time span, even if the 252 client does not seem to respond, e.g., due to a temporary network 253 outage. 255 o A notification also contains another value, called Max-Age-Hint. 256 This value is used by a cache to calculate a Max-Age for the 257 representation if needed. In a cache, the Max-Age-Hint of a 258 representation is counted down like Max-Age. When it reaches 259 zero, however, the representation can be still used to satisfy 260 requests, but is non-cacheable (i.e., Max-Age is 0). The Max-Age- 261 Hint must be less than or equal to Next-Notification-At-Latest. 263 We see two possible ways to encode Next-Notification-At-Latest and 264 Max-Age-Hint in a message: 266 o The first way is to require the values of Next-Notification-At- 267 Latest and Max-Age-Hint to be the same, although they are 268 conceptually unrelated. Then, a single option in the message can 269 be used to hold both values. 271 o The second way is to include two options, one for Next- 272 Notification-At-Latest and one for Max-Age-Hint. Since Next- 273 Notification-At-Latest is less than or equal to Max-Age-Hint, the 274 first option should indicates Max-Age-Hint, and the second option 275 Next-Notification-At-Latest minus Max-Age-Hint with a default 276 value of 0. 278 3. The Base-Uri Option 280 A proxy that forwards a response with embedded URIs may need to 281 indicate a base URI relative to which the embedded URIs need to be 282 interpreted that is different from the original request URI. E.g., 283 when the proxy forwarded the request to a multicast address, it may 284 need to indicate which specific server sent the response. A similar 285 requirement is the need to provide a request URI relative to which 286 the Location-* options can be interpreted. 288 The Base-Uri Option can be used in a response to provide this 289 information. It is structured like the Proxy-Uri option, but it is 290 elective and safe to forward (whether it is a cache-key is 291 irrelevant, as it is a response option only). 293 +--------+----------+-----------+ 294 | Number | Name | Reference | 295 +--------+----------+-----------+ 296 | TBD | Base-Uri | [RFCXXXX] | 297 +--------+----------+-----------+ 299 4. Acknowledgements 301 This work was partially funded by the Klaus Tschira Foundation and by 302 Intel Corporation. 304 Of course, much of the content of this draft is the result of 305 discussions with the [I-D.ietf-core-coap] authors. 307 Patience and Leisure were influenced by a mailing list discussion 308 with Esko Dijk, Kepeng Li, and Salvatore Loreto - thanks! 310 5. References 312 5.1. Normative References 314 [I-D.ietf-core-coap] 315 Shelby, Z., Hartke, K., and C. Bormann, "Constrained 316 Application Protocol (CoAP)", draft-ietf-core-coap-16 317 (work in progress), May 2013. 319 [I-D.ietf-core-observe] 320 Hartke, K., "Observing Resources in CoAP", draft-ietf- 321 core-observe-08 (work in progress), February 2013. 323 [I-D.ietf-httpbis-p1-messaging] 324 Fielding, R. and J. Reschke, "Hypertext Transfer Protocol 325 (HTTP/1.1): Message Syntax and Routing", draft-ietf- 326 httpbis-p1-messaging-22 (work in progress), February 2013. 328 [I-D.ietf-httpbis-p4-conditional] 329 Fielding, R. and J. Reschke, "Hypertext Transfer Protocol 330 (HTTP/1.1): Conditional Requests", draft-ietf- 331 httpbis-p4-conditional-22 (work in progress), February 332 2013. 334 [I-D.ietf-httpbis-p6-cache] 335 Fielding, R., Nottingham, M., and J. Reschke, "Hypertext 336 Transfer Protocol (HTTP/1.1): Caching", draft-ietf- 337 httpbis-p6-cache-22 (work in progress), February 2013. 339 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 340 Requirement Levels", BCP 14, RFC 2119, March 1997. 342 [RFC2616] Fielding, R., Gettys, J., Mogul, J., Frystyk, H., 343 Masinter, L., Leach, P., and T. Berners-Lee, "Hypertext 344 Transfer Protocol -- HTTP/1.1", RFC 2616, June 1999. 346 [RFC4648] Josefsson, S., "The Base16, Base32, and Base64 Data 347 Encodings", RFC 4648, October 2006. 349 [RFC6256] Eddy, W. and E. Davies, "Using Self-Delimiting Numeric 350 Values in Protocols", RFC 6256, May 2011. 352 5.2. Informative References 354 [CoRE201] , "Clarify use of retransmission window for duplicate 355 detection", CoRE ticket #201, 2012, 356 . 358 [CoRE214] , "Adopt vendor-defined option into core-coap", CoRE 359 ticket #214, 2012, 360 . 362 [CoRE230] , "Multiple Location options need to be processed as a 363 unit", CoRE ticket #230, 2012, 364 . 366 [CoRE241] , "Proxy Safe & Cache Key indication for options", CoRE 367 ticket #241, 2012, 368 . 370 [I-D.bormann-cbor] 371 Bormann, C. and P. Hoffman, "Concise Binary Object 372 Representation (CBOR)", draft-bormann-cbor-00 (work in 373 progress), May 2013. 375 [REST] Fielding, R., "Architectural Styles and the Design of 376 Network-based Software Architectures", 2000. 378 [RFC1924] Elz, R., "A Compact Representation of IPv6 Addresses", RFC 379 1924, April 1996. 381 [RFC2817] Khare, R. and S. Lawrence, "Upgrading to TLS Within HTTP/ 382 1.1", RFC 2817, May 2000. 384 [RFC5198] Klensin, J. and M. Padlipsky, "Unicode Format for Network 385 Interchange", RFC 5198, March 2008. 387 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 388 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 389 May 2008. 391 [RFC6648] Saint-Andre, P., Crocker, D., and M. Nottingham, 392 "Deprecating the "X-" Prefix and Similar Constructs in 393 Application Protocols", BCP 178, RFC 6648, June 2012. 395 Appendix A. The Nursery (Things that still need to ripen a bit) 397 A.1. Envelope Options 399 As of [I-D.ietf-core-coap], options can take one of four types, two 400 of which are mostly identical: 402 o uint: A non-negative integer which is represented in network byte 403 order using a variable number of bytes (see [I-D.ietf-core-coap] 404 Appendix A); 406 o string: a sequence of bytes that is nominally a Net-Unicode string 407 [RFC5198]; 409 o opaque: a sequence of bytes. 411 o empty (not explicitly identified as a fourth type in 412 [I-D.ietf-core-coap]). 414 It turns out some options would benefit from some internal structure. 415 Also, it may be a good idea to be able to bundle multiple options 416 into one, in order to ensure consistency for a set of elective 417 options that need to be processed all or nothing (i.e., the option 418 becomes critical as soon as another option out of the set is 419 processed, too). 421 In this section, we introduce a fifth CoAP option type: Envelope 422 options. 424 An envelope option is a sequence of bytes that looks and is 425 interpreted exactly like a CoAP sequence of options. Instead of an 426 option count or an end-of-option marker, the sequence of options is 427 terminated by the end of the envelope option. 429 The nested options (options inside the envelope option) may come from 430 the same number space as the top-level CoAP options, or the envelope 431 option may define its own number space - this choice needs to be 432 defined for each envelope option. 434 If the top-level number space is used, the envelope option typically 435 will restrict the set of options that actually can be used in the 436 envelope. In particular, it is unlikely that an envelope option will 437 allow itself inside the envelope (this would be a recursive option). 439 Envelope options are a general, but simple mechanism. Some of its 440 potential uses are illustrated by two examples in the cemetery: 441 Appendix C.1 and Appendix C.2. (Each of these examples has its own 442 merits and demerits, which led us to decide not to pursue either of 443 them right now, but this should be discussed separately from the 444 concept of Envelope options employed in the examples.) 446 A.2. Payload-Length Option 448 Not all transport mappings may provide an unambiguous length of the 449 CoAP message. For UDP, it may also be desirable to pack more than 450 one CoAP message into one UDP payload (aggregation); in that case, 451 for all but the last message there needs to be a way to delimit the 452 payload of that message. 454 This can be solved using a new option, the Payload-Length option. If 455 this option is present, the value of this option is an unsigned 456 integer giving the length of the payload of the message (note that 457 this integer can be zero for a zero-length payload, which can in turn 458 be represented by a zero-length option value). (In the UDP 459 aggregation case, what would have been in the payload of this message 460 after "payload-length" bytes is then actually one or more additional 461 messages.) 463 A.3. URI Authorities with Binary Adresses 465 One problem with the way URI authorities are represented in the URI 466 syntax is that the authority part can be very bulky if it encodes an 467 IPv6 address in ASCII. 469 Proposal: Provide an option "Uri-Authority-Binary" that can be an 470 even number of bytes between 2 and 18 except 12 or 14. 472 o If the number of bytes is 2, the destination IP address of the 473 packet transporting the CoAP message is implied. 475 o If the number of bytes is 4 or 6, the first four bytes of the 476 option value are an IPv4 address in binary. 478 o If the number of bytes is 8 or 10, the first eight bytes are the 479 lower 64 bits of an IPv6 address; the upper eight bytes are 480 implied from the destination address of the packet transporting 481 the CoAP message. 483 o If the number of bytes is 16 or 18, the first 16 bytes are an IPv6 484 address. 486 o If two more bytes remain, this is a port number (as always in 487 network byte order). 489 The resulting authority is (conceptually translated into ASCII and) 490 used in place of an Uri-Authority option, or inserted into a Proxy- 491 Uri. Examples: 493 +--------------+------------------+---------+-----------------------+ 494 | Proxy-Uri | Uri-Authority- | Uri- | URI | 495 | | Binary | Path | | 496 +--------------+------------------+---------+-----------------------+ 497 | (none) | (none) | (none) | "/" | 498 | | | | | 499 | (none) | (none) | 'temp' | "/temp" | 500 | | | | | 501 | (none) | 2 bytes: 61616 | 'temp' | "coap://[DA]:61616/te | 502 | | | | mp" | 503 | | | | | 504 | (none) | 16 bytes: | temp | "coap://[2000::1]/tem | 505 | | 2000::1 | | p" | 506 | | | | | 507 | 'http://' | 10 bytes: | (none) | "http://[DA::123:45]: | 508 | | ::123:45 + 616 | | 616" | 509 | | | | | 510 | 'http:///tem | 18 bytes: | (none) | "http://[2000::1]:616 | 511 | p' | 2000::1 + 616 | | /temp" | 512 +--------------+------------------+---------+-----------------------+ 514 A.4. Length-aware number encoding (o256) 516 The number encoding defined in Appendix A of [I-D.ietf-core-coap] has 517 one significant flaw: Every number has an infinite number of 518 representations, which can be derived by adding leading zero bytes. 519 This runs against the principle of minimizing unnecessary choice. 520 The resulting uncertainty in encoding ultimately leads to unnecessary 521 interoperability failures. (It also wastes a small fraction of the 522 encoding space, i.e., it wastes bytes.) 523 We could solve the first, but not the second, by outlawing leading 524 zeroes, but then we have to cope with error cases caused by illegal 525 values, another source of interoperability problems. 527 The number encoding "o256" defined in this section avoids this flaw. 528 The suggestion is not to replace CoAP's "uint" encoding wholesale 529 (CoAP is already too widely implemented for such a change), but to 530 consider this format for new options. 532 The basic requirements for such an encoding are: 534 o numbers are encoded as a sequence of zero or more bytes 536 o each number has exactly one encoding 538 o for a < b, encoding-size(a) <= encoding-size(b) -- i.e., with 539 larger numbers, the encoding only gets larger, never smaller 540 again. 542 o within each encoding size (0 bytes, 1 byte, etc.), lexicographical 543 ordering of the bytes is the same as numeric ordering 545 Obviously, there is only one encoding that satisfies all these 546 requirements. As illustrated by Figure 1, this is unambiguously 547 derived by 549 1. enumerating all possible byte sequences, ordered by length and 550 within the same length in lexicographic ordering, and, 552 2. assigning sequential cardinals. 554 0x'' -> 0 555 0x'00' -> 1 556 0x'01' -> 2 557 0x'02' -> 3 558 ... 559 0x'fe' -> 255 560 0x'ff' -> 256 561 0x'0000' -> 257 562 0x'0001' -> 258 563 ... 564 0x'fefd' -> 65534 565 0x'fefe' -> 65535 566 0x'feff' -> 65536 567 ... 568 0x'ffff' -> 65792 569 0x'000000' -> 65793 570 0x'000001' -> 65794 572 Figure 1: Enumerating byte sequences by length and then lexicographic 573 order 575 This results in an exceedingly simple algorithm: each byte is 576 interpreted in the base-256 place-value system, but stands for a 577 number between 1 and 256 instead of 0 to 255. We therefore call this 578 encoding "o256" (one-to-256). 0 is always encoded in zero bytes; 1 579 to 256 is one byte, 257 (0x101) to 65792 (0x10100) is two bytes, 580 65793 (0x10101) to 16843008 (0x1010100) is three bytes, etc. 582 To further illustrate the algorithmic simplicity, pseudocode for 583 encoding and decoding is given in Figure 2 and Figure 3, respectively 584 (in the encoder, "prepend" stands for adding a byte at the _leading_ 585 edge, the requirement for which is a result of the network byte 586 order). Note that this differs only in a single subtraction/addition 587 (resp.) of one from the canonical algorithm for Appendix A uints. 589 while num > 0 590 num -= 1 591 prepend(num & 0xFF) 592 num >>= 8 593 end 595 Figure 2: o256 encoder (pseudocode) 597 num = 0 598 each_byte do |b| 599 num <<= 8 600 num += b + 1 601 end 603 Figure 3: o256 decoder (pseudocode) 605 On a more philosophical note, it can be observed that o256 solves the 606 inverse problem of Self-Delimiting Numeric Values (SDNV) [RFC6256]: 607 SDNV encodes variable-length numbers together with their length 608 (allowing decoding without knowing their length in advance, deriving 609 delimiting information from the number encoding). o256 encodes 610 variable-length numbers when there is a way to separately convey the 611 length (as in CoAP options), encoding (and later deriving) a small 612 part of the numeric value into/from that size information. 614 A.5. SMS encoding 616 For use in SMS applications, CoAP messages can be transferred using 617 SMS binary mode. However, there is operational experience showing 618 that some environments cannot successfully send a binary mode SMS. 620 For transferring SMS in character mode (7-bit characters), 621 base64-encoding [RFC4648] is an obvious choice. 3 bytes of message 622 (24 bits) turn into 4 characters, which cosume 28 bits. The overall 623 overhead is approximately 17 %; the maximum message size is 120 bytes 624 (160 SMS characters). 626 If a more compact encoding is desired, base85 encoding can be 627 employed (however, probably not the version defined in [RFC1924] 628 -- instead, the version used in tools such as btoa and PDF 629 should be chosen). However, this requires division operations. 630 Also, the base85 character set includes several characters that 631 cannot be transferred in a single 7-bit unit in SMS and/or are known 632 to cause operational problems. A modified base85 character set can 633 be defined to solve the latter problem. 4 bytes of message (32 bits) 634 turn into 5 characters, which consume 35 bits. The overall overhead 635 is approximately 9.3 %; the resulting maximum message size is 128 636 bytes (160 SMS characters). 638 Base64 and base85 do not make use of the fact that much CoAP data 639 will be ASCII-based. Therefore, we define the following experimental 640 SMS encoding. 642 A.5.1. ASCII-optimized SMS encoding 644 Not all 128 theoretically possible SMS characters are operationally 645 free of problems. We therefore define: 647 Shunned code characters: @ sign, as it maps to 0x00 649 LF and CR signs (0x0A, 0x0D) 651 uppercase C cedilla (0x09), as it is often mistranslated in 652 gateways 654 ESC (0x1B), as it is used in certain character combinations only 656 Some ASCII characters cannot be transferred in the base SMS character 657 set, as their code positions are taken by non-ASCII characters. 658 These are simply encoded with their ASCII code positions, e.g., an 659 underscore becomes a section mark (even though underscore has a 660 different code position in the SMS character set). 662 Equivalently translated input bytes: $, @, [, \, ], ^, _, `, {, |, 663 }, ~, DEL 665 In other words, bytes 0x20 to 0x7F are encoded into the same code 666 positions in the 7-bit character set. 668 Out of the remaining code characters, the following SMS characters 669 are available for encoding: 671 Non-equivalently translated (NET) code characters: 0x01 to 0x08, (8 672 characters) 674 0x0B, 0x0C, (2 characters) 676 0x0E to 0x1A, (13 characters) 678 0x1C to 0x1F, (4 characters) 680 Of the 27 NET code characters, 18 are taken as prefix characters (see 681 below), and 8 are defined as directly translated characters: 683 Directly translated bytes: Equivalently translated input bytes are 684 represented as themselves 686 0x00 to 0x07 are represented as 0x01 to 0x08 688 This leaves 0x08 to 0x1F and 0x80 to 0xFF. Of these, the bytes 0x80 689 to 0x87 and 0xA0 to 0xFF are represented as the bytes 0x00 to 0x07 690 (represented by characters 0x01 to 0x08) and 0x20 to 0x7F, with a 691 prefix of 1 (see below). The characters 0x08 to 0x1F are represented 692 as the characters 0x28 to 0x3F with a prefix of 2 (see below). The 693 characters 0x88 to 0x9F are represented as the characters 0x48 to 694 0x5F with a prefix of 2 (see below). (Characters 0x01 to 0x08, 0x20 695 to 0x27, 0x40 to 0x47, and 0x60 to 0x7f with a prefix of 2 are 696 reserved for future extensions, which could be used for some 697 backreferencing or run-length compression.) 699 Bytes that do not need a prefix (directly translated bytes) are sent 700 as is. Any byte that does need a prefix (i.e., 1 or 2) is preceded 701 by a prefix character, which provides a prefix for this and the 702 following two bytes as follows: 704 +------+-----+---+------+-----+ 705 | 0x0B | 100 | | 0x15 | 200 | 706 +------+-----+---+------+-----+ 707 | 0x0C | 101 | | 0x16 | 201 | 708 | | | | | | 709 | 0x0E | 102 | | 0x17 | 202 | 710 | | | | | | 711 | 0x0F | 110 | | 0x18 | 210 | 712 | | | | | | 713 | 0x10 | 111 | | 0x19 | 211 | 714 | | | | | | 715 | 0x11 | 112 | | 0x1A | 212 | 716 | | | | | | 717 | 0x12 | 120 | | 0x1C | 220 | 718 | | | | | | 719 | 0x13 | 121 | | 0x1D | 221 | 720 | | | | | | 721 | 0x14 | 122 | | 0x1E | 222 | 722 +------+-----+---+------+-----+ 724 (This leaves one non-shunned character, 0x1F, for future extension.) 726 The coding overhead of this encoding for random bytes is similar to 727 Base85, without the need for a division/multiplication. For bytes 728 that are mostly ASCII characters, the overhead can easily become 729 negative. (Conversely, for bytes that are more likely to be non- 730 ASCII than in a random sequence of bytes, the overhead becomes 731 greater.) 733 So, for instance, for the CoAP message in Figure 4: 735 ver tt code mid 736 1 ack 2.05 17033 737 content_type 40 738 token sometok 739 3c 2f 3e 3b 74 69 74 6c 65 3d 22 47 65 6e 65 72 |;title="Gener| 740 61 6c 20 49 6e 66 6f 22 3b 63 74 3d 30 2c 3c 2f |al Info";ct=0,;if="clock"| 742 3b 72 74 3d 22 54 69 63 6b 73 22 3b 74 69 74 6c |;rt="Ticks";titl| 743 65 3d 22 49 6e 74 65 72 6e 61 6c 20 43 6c 6f 63 |e="Internal Cloc| 744 6b 22 3b 63 74 3d 30 2c 3c 2f 61 73 79 6e 63 3e |k";ct=0,| 745 3b 63 74 3d 30 |;ct=0 | 747 Figure 4: CoAP response message as captured and decoded 749 The 116 byte unencoded message is shown as ASCII characters in Figure 750 5 (\xDD stands for the byte with the hex digits DD): 752 bEB\x89\x11(\xA7sometok;title="General Info";ct=0, 753 ;if="clock";rt="Ticks";title="Internal Clock";ct=0,;ct=0 755 Figure 5: CoAP response message shown as unencoded characters 757 The equivalent SMS encoding is shown as equivalent-coded SMS 758 characters in Figure 6 (7 bits per character, \x12 is a 220 prefix 759 and \x0B is a 100 prefix, the rest is shown in equivalent encoding), 760 adding two characters of prefix overhead, for a total length of 118 761 7-bit characters or 104 (103.25 plus padding) bytes: 763 bEB\x12I1(\x0B'sometok;title="General Info";ct=0, 764 ;if="clock";rt="Ticks";title="Internal Clock";ct=0,;ct=0 766 Figure 6: CoAP response message shown as SMS-encoded characters 768 A.6. CONNECT 770 [RFC2817] defines the HTTP CONNECT method to establish a TCP tunnel 771 through a proxy so that end-to-end TLS connections can be made 772 through the proxy. Recently, a requirement for similar functionality 773 has been discussed for CoAP. This section defines a straw-man 774 CONNECT method and related methods and response codes for CoAP. 776 (IANA considerations for this section TBD.) 778 A.6.1. Requesting a Tunnel with CONNECT 780 CONNECT is allocated as a new method code in the "CoAP Method Codes" 781 registry. When a client makes a CONNECT request to an intermediary, 782 the intermediary evaluates the Uri-Host, Uri-Port, and/or the 783 authority part of the Proxy-Uri Options in a way that is defined by 784 the security policy of the intermediary. If the security policy 785 allows the allocation of a tunnel based on these parameters, the 786 method returns an empty payload and a response code of 2.30 Tunnel 787 Established. Other possible response codes include 4.03 Forbidden. 789 It may be the case that the intermediary itself can only reach the 790 requested origin server through another intermediary. In this case, 791 the first intermediary SHOULD make a CONNECT request of that next 792 intermediary, requesting a tunnel to the authority. A proxy MUST NOT 793 respond with any 2.xx status code unless it has either a direct or 794 tunnel connection established to the authority. 796 An origin server which receives a CONNECT request for itself MAY 797 respond with a 2.xx status code to indicate that a tunnel is 798 established to itself. 800 Code 2.30 "Tunnel Established" is allocated as a new response code in 801 the "CoAP Response Codes" registry. 803 A.6.2. Using a CONNECT Tunnel 805 Any successful (2.xx) response to a CONNECT request indicates that 806 the intermediary has established a tunnel to the requested host and 807 port. The tunnel is bound to the requesting end-point and the Token 808 supplied in the request (as always, the default Token is admissible). 809 The tunnel can be used by the client by making a DATAGRAM request. 811 DATAGRAM is allocated as a new method code in the "CoAP Method Codes" 812 registry. When a client makes a DATAGRAM request to an intermediary, 813 the intermediary looks up the tunnel bound to the client end-point 814 and Token supplied in the DATAGRAM request (no other Options are 815 permitted). If a tunnel is found and the intermediary's security 816 policy permits, the intermediary forwards the payload of the DATAGRAM 817 request as the UDP payload towards the host and port established for 818 the tunnel. No response is defined for this request (note that the 819 request can be given as a CON or NON request; for CON, there will be 820 an ACK on the message layer if the tunnel exists). 822 The security policy on the intermediary may restrict the allowable 823 payloads based on its security policy, possibly considering host and 824 port. An inadmissible payload SHOULD cause a 4.03 Forbidden response 825 with a diagnostic message as payload. 827 The UDP payload of any datagram received from the tunnel and admitted 828 by the security policy is forwarded to the client as the payload of a 829 2.31 "Datagram Received" response. The response does not carry any 830 Option except for Token, which identifies the tunnel towards the 831 client. 833 Code 2.31 "Datagram Received" is allocated as a new response code in 834 the "CoAP Response Codes" registry. 836 An origin server that has established a tunnel to itself processes 837 the CoAP payloads of related DATAGRAM requests as it would process an 838 incoming UDP payload, and forwards what would be outgoing UDP 839 payloads in 2.31 "Datagram Received" responses. 841 A.6.3. Closing down a CONNECT Tunnel 843 A 2.31 "Datagram Received" response may be replied to with a RST, 844 which closes down the tunnel. Similarly, the Token used in the 845 tunnel may be reused by the client for a different purpose, which 846 also closes down the tunnel. 848 Appendix B. The Museum (Things we did, but maybe not exactly this way) 850 B.1. Getting rid of artificial limitations 852 _Artificial limitations_ are limitations of a protocol or system that 853 are not rooted in limitations of actual capabilities, but in 854 arbitrary design decisions. Proper system design tries to avoid 855 artificial limitations, as these tend to cause complexity in systems 856 that need to work with these limitations. 858 E.g., the original UNIX filesystem had an artificial limitation of 859 the length of a path name component to 14 bytes. This led to a 860 cascade of workarounds in programs that manipulate file names: E.g., 861 systematically replacing a ".el" extension in a filename with a 862 ".elc" for the compiled file might exceed the limit, so all ".el" 863 files were suddenly limited to 13-byte filenames. 865 Note that, today, there still is a limitation in most file system 866 implementations, typically at 255. This just happens to be high 867 enough to rarely be of real-world concern; we will refer to this case 868 as a "painless" artificial limitation. 870 CoAP-08 had two highly recognizable artificial limitations in its 871 protocol encoding 873 o The number of options in a single message is limited to 15 max. 875 o The length of an option is limited to 270 max. 877 It has been argued that the latter limitation causes few problems, 878 just as the 255-byte path name component limitation in filenames 879 today causes few problems. Appendix B.1.1 provided a design to 880 extend this; as a precaution to future extensions of this kind, the 881 current encoding for length 270 (eight ones in the extension byte) 882 could be marked as reserved today. Since, Matthias Kovatsch has 883 proposed a simpler scheme that seems to gain favor in the WG, see 884 Appendix B.1.4. 886 The former limitation has been solved in CoAP-09. A historical 887 discussion of other approaches for going beyond 15 options is in 888 Appendix B.1.2. Appendix B.1.3 discusses implementation. 890 B.1.1. Beyond 270 bytes in a single option 892 The authors would argue that 270 as the maximum length of an option 893 is already beyond the "painless" threshold. 895 If that is not the consensus of the WG, the scheme can easily be 896 extended as in Figure 7: 898 for 15..269: 899 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 900 | Option Delta | 1 1 1 1 | Length - 15 | 901 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 902 | Option Value ... 903 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 905 for 270..65805: 907 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 908 | Option Delta | 1 1 1 1 | 1 1 1 1 1 1 1 1 | 909 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 910 | Length - 270 (in network byte order) | 911 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 912 | Option Value ... 913 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 915 Figure 7: Ridiculously Long Option Header 917 The infinite number of obvious variations on this scheme are left as 918 an exercise to the reader. 920 Again, as a precaution to future extensions, the current encoding for 921 length 270 (eight ones in the extension byte) could be marked as 922 reserved today. 924 B.1.2. Beyond 15 options 926 (This section keeps discussion that is no longer needed as we have 927 agreed to do what is documented in Appendix B.1.3). 929 The limit of 15 options is motivated by the fixed four-bit field "OC" 930 that is used for indicating the number of options in the fixed-length 931 CoAP header (Figure 8). 933 0 1 2 3 934 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 935 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 936 |Ver| T | OC | Code | Message ID | 937 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 938 | Options (if any) ... 939 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 940 | Payload (if any) ... 941 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 943 Figure 8: Four-byte fixed header in a CoAP Message 945 Note that there is another fixed four-bit field in CoAP: the option 946 length (Figure 9 - note that this figure is not to the same scale as 947 the previous figure): 949 0 1 2 3 4 5 6 7 950 +---+---+---+---+---+---+---+---+ 951 | Option Delta | Length | for 0..14 952 +---+---+---+---+---+---+---+---+ 953 | Option Value ... 954 +---+---+---+---+---+---+---+---+ 955 Figure 9: Short Option Header 957 Since 15 is inacceptable for a maximum option length, the all-ones 958 value (15) was taken out of the set of allowable values for the short 959 header, and a long header was introduced that allows the insertion of 960 an extension byte (Figure 10): 962 for 15..270: 963 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 964 | Option Delta | 1 1 1 1 | Length - 15 | 965 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 966 | Option Value ... 967 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 969 Figure 10: Long Option Header 971 We might want to use the same technique for the CoAP header as well. 972 There are two obvious places where the extension byte could be 973 placed: 975 1. right after the byte carrying the OC field, so the structure is 976 the same as for the option header; 978 2. right after the fixed-size CoAP header. 980 Both solutions lose the fixed-size-ness of the CoAP header. 982 Solution 1 has the disadvantage that the CoAP header is also changing 983 in structure: The extension byte is wedged between the first and the 984 second byte of the CoAP header. This is unfortunate, as the number 985 of options only comes into play when the option processing begins, so 986 it is more natural to use solution 2 (Figure 11): 988 0 1 2 3 989 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 990 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 991 |Ver| T | 15 | Code | Message ID | 992 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 993 | OC - 15 | Options ... 994 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 995 | Payload (if any) ... 996 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 998 Figure 11: Extended header for CoAP Messages with 15+ options 1000 This would allow for up to 270 options in a CoAP message, which is 1001 very likely way beyond the "painless" threshold. 1003 B.1.2.1. Implementation considerations 1005 For a message decoder, this extension creates relatively little pain, 1006 as the number of options only becomes interesting when the encoding 1007 turns to the options part of the message, which is then simply lead 1008 in by the extension byte if the four-bit field is 15. 1010 For a message encoder, this extension is not so rosy. If the encoder 1011 is constructing the message serially, it may not know in advance 1012 whether the number of options will exceed 14. None of the following 1013 implementation strategies is particularly savory, but all of them do 1014 work: 1016 1. Encode the options serially under the assumption that the number 1017 of options will be 14 or less. When the 15th option needs to be 1018 encoded, abort the option encoding, and restart it from scratch 1019 one byte further to the left. 1021 2. Similar to 1, except that the bytes already encoded are all moved 1022 one byte to right, the extension byte is inserted, and the option 1023 encoding process is continued. 1025 3. The encoder always leaves space for the extension byte (at least 1026 if it can't prove the number will be less thatn 14). If the 1027 extension byte is not needed, an Option 0 with length 0 is 1028 encoded instead (i.e., one byte is wasted - this option is 1029 elective and will be ignored by the receiver). 1031 As a minimum, to enable strategy 3, the option 0 should be reserved 1032 at least for the case of length=0. 1034 B.1.2.2. What should we do now? 1036 As a minimum proposal for the next version of CoAP, the value 15 for 1037 OC should be marked as reserved today. 1039 B.1.2.3. Alternatives 1041 One alternative that has been discussed previously is to have an 1042 "Options" Option, which allows the carriage of multiple options in 1043 the belly of a single one. This could also be used to carry more 1044 than 15 options. However: 1046 o The conditional introduction of an Options option has 1047 implementation considerations that are likely to be more severe 1048 than the ones listed above; 1050 o since 270 bytes may not be enough for the encoding of _all_ 1051 options, the "Options" option would need to be repeatable. This 1052 creates many different ways to encode the same message, leading to 1053 combinatorial explosion in test cases for ensuring 1054 interoperability. 1056 B.1.2.4. Alternative: Going to a delimiter model 1058 Another alternative is to spend the additional byte not as an 1059 extended count, but as an option terminator. 1061 B.1.3. Implementing the option delimiter for 15 or more options 1063 Implementation note: As can be seen from the proof of concept code 1064 in Figure 12, the actual implementation cost for a decoder is 1065 around 4 lines of code (or about 8-10 machine code instructions). 1067 while numopt > 0 1068 nextbyte = ... get next byte 1070 if numopt == 15 # new 1071 break if nextbyte == 0xF0 # new 1072 else # new 1073 numopt -= 1 1074 end # new 1076 ... decode delta and length from nextbyte and handle them 1077 end 1079 Figure 12: Implementing the Option Terminator 1081 Similarly, creating the option terminator needs about four more lines 1082 (not marked "old" in the C code in Figure 13). 1084 b0 = 0x40 + (tt << 4); /* old */ 1085 buffer[0] = b0 + 15; /* guess first byte */ 1087 .... encode options .... /* old */ 1089 if (option_count >= 15 || first_fragment_already_shipped) 1090 buffer[pos++] = 0xF0; /* use delimiter */ 1091 else /* save a byte: */ 1092 buffer[0] = b0 + option_count; /* old: backpatch */ 1094 Figure 13: Creating the Option Terminator 1096 B.1.4. Option Length encoding beyond 270 bytes 1097 For option lengths beyond 270 bytes, we reserve the value 255 of an 1098 extension byte to mean "add 255, read another extension byte" Figure 1099 14. While this causes the length of the option header to grow 1100 linearly with the size of the option value, only 0.4 % of that size 1101 is used. With a focus on short options, this encoding is justified. 1103 for 15..269: 1104 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1105 | Option Delta | 1 1 1 1 | Length - 15 | 1106 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1107 | Option Value ... 1108 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1110 for 270..524: 1111 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1112 | Option Delta | 1 1 1 1 | 1 1 1 1 1 1 1 1 | 1113 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1114 | Length - 270 | Option Value ... 1115 +---+---+---+---+---+---+---+---+ 1116 | 1117 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1119 for 525..779: 1120 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1121 | Option Delta | 1 1 1 1 | 1 1 1 1 1 1 1 1 | 1122 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1123 | 1 1 1 1 1 1 1 1 | Length - 525 | 1124 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1125 | Option Value ... 1126 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1128 for 780..1034: 1129 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1130 | Option Delta | 1 1 1 1 | 1 1 1 1 1 1 1 1 | 1131 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1132 | 1 1 1 1 1 1 1 1 | 1 1 1 1 1 1 1 1 | 1133 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1134 | Length - 780 | Option Value ... 1135 +---+---+---+---+---+---+---+---+ 1136 | 1137 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ 1139 Figure 14: Options beyond 270 bytes 1141 Options that are longer than 1034 bytes MUST NOT be sent; an option 1142 that has 255 (all one bits) in the field called "Length - 780" MUST 1143 be rejected upon reception as an invalid option. 1145 In the process, the maximum length of all options that are currently 1146 set at 270 should now be set to a carefully chosen value. With the 1147 purely encoding-based limit gone, Uri-Proxy should now be restored to 1148 be a non-repeatable option. 1150 A first proposal for a new set of per-option length restrictions 1151 follows: 1153 +--------+---------------------+-----+------+--------+--------+ 1154 | number | name | min | max | type | repeat | 1155 +--------+---------------------+-----+------+--------+--------+ 1156 | 1 | content_type | 0 | 2 | uint | - | 1157 | | | | | | | 1158 | 2 | max_age | 0 | 4 | uint | - | 1159 | | | | | | | 1160 | 3 | proxy_uri | 1 | 1023 | string | - | 1161 | | | | | | | 1162 | 4 | etag | 1 | 8 | opaque | yes | 1163 | | | | | | | 1164 | 5 | uri_host | 1 | 255 | string | - | 1165 | | | | | | | 1166 | 6 | location_path | 0 | 255 | string | yes | 1167 | | | | | | | 1168 | 7 | uri_port | 0 | 2 | uint | - | 1169 | | | | | | | 1170 | 8 | location_query | 0 | 255 | string | yes | 1171 | | | | | | | 1172 | 9 | uri_path | 0 | 255 | string | yes | 1173 | | | | | | | 1174 | 10 | observe | 0 | 2 | uint | - | 1175 | | | | | | | 1176 | 11 | token | 1 | 8 | opaque | - | 1177 | | | | | | | 1178 | 12 | accept | 0 | 2 | uint | yes | 1179 | | | | | | | 1180 | 13 | if_match | 0 | 8 | opaque | yes | 1181 | | | | | | | 1182 | 14 | registered_elective | 1 | 1023 | opaque | yes | 1183 | | | | | | | 1184 | 15 | uri_query | 1 | 255 | string | yes | 1185 | | | | | | | 1186 | 17 | block2 | 0 | 3 | uint | - | 1187 | | | | | | | 1188 | 18 | size | 0 | 4 | uint | - | 1189 | | | | | | | 1190 | 19 | block1 | 0 | 3 | uint | - | 1191 | | | | | | | 1192 | 21 | if_none_match | 0 | 0 | empty | - | 1193 | | | | | | | 1194 | 25 | registered_critical | 1 | 1023 | opaque | yes | 1195 +--------+---------------------+-----+------+--------+--------+ 1197 (Option 14 with a length of 0 is a fencepost only.) 1199 B.2. Registered Option 1201 CoAP's option encoding is highly efficient, but works best with small 1202 option numbers that do not require much fenceposting. The CoAP 1203 Option Number Registry therefore has a relatively heavyweight 1204 registration requirement: "IETF Review" as described in [RFC5226]. 1206 However, there is also considerable benefit in a much looser registry 1207 policy, enabling a first-come-first-served policy for a relatively 1208 large option number space. 1210 Here, we discuss two solutions that enable such a registry. One is 1211 to define a separate mechanism for registered options, discussed in 1212 Appendix B.2.1. Alternatively, we could make it easier to use a 1213 larger main option number space, discussed in Appendix B.2.2. 1215 B.2.1. A Separate Suboption Number Space 1217 This alternative defines a separate space of suboption numbers, with 1218 an expert review [RFC5226] (or even first-come-first-served) 1219 registration policy. If expert review is selected for this registry, 1220 it would be with a relatively loose policy delegated to the expert. 1221 This draft proposes leaving the registered suboption numbers 0-127 to 1222 expert review with a policy that mainly focuses on the availability 1223 of a specification, and 128-16383 for first-come-first-served where 1224 essentially only a name is defined. 1226 The "registered" options are used in conjunction with this suboption 1227 number registry. They use two normal CoAP option numbers, one for 1228 options with elective semantics (Registered-Elective) and one for 1229 options with critical semantics (Registered-Critical). The suboption 1230 numbers are not separate, i.e. one registered suboption number might 1231 have some elective semantics and some other critical semantics (e.g., 1232 for the request and the response leg of an exchange). The option 1233 value starts with an SDNV [RFC6256] of the registered suboption 1234 number. (Note that there is no need for an implementation to 1235 understand SDNVs, it can treat the prefixes as opaque. One could 1236 consider the SDNVs as a suboption prefix allocation guideline for 1237 IANA as opposed to a number encoding.) 1238 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1239 |1 0 0 0 0 0 0 1|0 1 1 1 0 0 1 1| value... | 1240 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1241 \___SDNV of registered number___/ 1243 Figure 15: Example option value for registered option 1245 Note that a Registered Option cannot be empty, because there would be 1246 no space for the SDNV. Also, the empty option 14 is reserved for 1247 fenceposting ([I-D.ietf-core-coap], section 3.2). (Obviously, once a 1248 Registered-Elective Option is in use, there is never a need for a 1249 fence-post for option number 14.) 1251 The Registered-Elective and Registered-Critical Options are 1252 repeatable. 1254 +------+----------+---------------------+--------+--------+---------+ 1255 | No. | C/E | Name | Format | Length | Default | 1256 +------+----------+---------------------+--------+--------+---------+ 1257 | 14 | Elective | Registered-Elective | (see | 1-1023 | (none) | 1258 | | | | above) | B | | 1259 | | | | | | | 1260 | 25 | Critical | Registered-Critical | (see | 1-1023 | (none) | 1261 | | | | above) | B | | 1262 +------+----------+---------------------+--------+--------+---------+ 1264 This solves CoRE issue #214 [CoRE214]. (How many options we need 1265 will depend on the resolution of #241 [CoRE241].) 1267 B.2.2. Opening Up the Option Number Space 1269 The disadvantage of the registered-... options is that there is a 1270 significant syntactic difference between options making use of this 1271 space and the usual standard options. This creates a problem not 1272 unlike that decried in [RFC6648]. 1274 The alternative discussed in this section reduces the distance by 1275 opening up the main Option number space instead. 1277 There is still a significant incentive to use low-numbered Options. 1278 However, the proposal reduces the penalty for using a high-numbered 1279 Option to two or three bytes. More importantly, using a cluster of 1280 related high-numbered options only carries a total penalty of two or 1281 three bytes. 1283 The main reason high-numbered options are expensive to use and thus 1284 the total space is relatively limited is that the option delta 1285 mechanism only allows increasing the current option number by up to 1286 14 per one-byte fencepost. To use, e.g., Option number 1234 together 1287 with the usual set of low-numbered Options, one needs to insert 88 1288 fence-post bytes. This is prohibitive. 1290 Enabling first-come-first-served probably requires easily addressing 1291 a 16-bit option number space, with some potential increase later in 1292 the lifetime of the protocol (say, 10 to 15 years from now). 1294 To enable the use of large option numbers, one needs a way to advance 1295 the Option number in bigger steps than possible by the Option Delta. 1296 So we propose a new construct, the Long Jump construct, to move the 1297 Option number forward. 1299 B.2.2.1. Long Jump construct 1301 The following construct can occur in front of any Option: 1303 0 1 2 3 4 5 6 7 1304 +---+---+---+---+---+---+---+---+ 1305 | 1 1 1 1 | 0 0 0 1 | 0xf1 (Delta = 15) 1306 +---+---+---+---+---+---+---+---+ 1308 0 1 2 3 4 5 6 7 1309 +---+---+---+---+---+---+---+---+ 1310 | 1 1 1 1 | 0 0 1 0 | 0xf2 1311 +---+---+---+---+---+---+---+---+ 1312 | Long Jump Value | (Delta/8)-2 1313 +---+---+---+---+---+---+---+---+ 1315 0 1 2 3 4 5 6 7 1316 +---+---+---+---+---+---+---+---+ 1317 | 1 1 1 1 | 0 0 1 1 | 0xf3 1318 +---+---+---+---+---+---+---+---+ 1319 | Long Jump Value, MSB | 1320 +---+---+---+---+---+---+---+---+ (Delta/8)-258 1321 | Long Jump Value, LSB | 1322 +---+---+---+---+---+---+---+---+ 1324 Figure 16: Long Jump Format 1326 This construct is not by itself an Option. It can occur in front of 1327 any Option to increase the current Option number that then goes into 1328 its Option number calculation. The increase is done in multiples of 1329 eight. More specifically, the actual addition to the current Option 1330 number is computed as follows: 1332 Delta = ((Long Jump Value) + N) * 8 1334 where N is 2 for the one-byte version and N is 258 for the two-byte 1335 version. 1337 A Long Jump MUST be followed by an actual Option, i.e., it MUST NOT 1338 be followed by another Long Jump or an end-of-options indicator. A 1339 message violating this MUST be rejected as malformed. 1341 Long Jumps do NOT count as Options in the Option Count field of the 1342 header (i.e., they cannot by themselves end the Option sequence). 1344 B.2.2.2. Discussion 1346 Adding a mechanism at this late stage creates concerns of backwards 1347 compatibility. A message sender never needs to implement long-jumps 1348 unless it wants to make use of a high-numbered option. So this 1349 mechanism can be added once a high-numbered option is added. A 1350 message receiver, though, would more or less unconditionally have to 1351 implement the mechanism, leading to unconditional additional 1352 complexity. There are good reasons to minimize this, as follows: 1354 o The increase in multiples of eight allows looking at an option and 1355 finding out whether it is critical or not even if the Long Jump 1356 value has just been skipped (as opposed to having been processed 1357 fully). (It also allows accessing up to approximately 2048 1358 options with a two-byte Long Jump.) This allows a basic 1359 implementation that does not implement any high-numbered options 1360 to simply ignore long jumps and any elective options behind them, 1361 while still properly reacting to critical options. 1363 o There is probably a good reason to disallow long-jumps that lead 1364 to an option number of 42 and less, enabling simple receivers to 1365 do the above simplification. 1367 o It might seem obvious to remove the fenceposting mechanism 1368 altogether in favor of long jumps. This is not advisable: 1369 Fenceposting already has zero implementation effort at the 1370 receiver, and the overhead at the sender is very limited (it is 1371 just a third kind of jump, at one byte per jump). Beyond 42, 1372 senders can ignore the existence of fenceposts if they want 1373 (possibly obviating the need for more complex base-14 arithmetic). 1375 There is no need for a finer granularity than 8, as the Option 1376 construct following can also specify a Delta of 0..14. (A 1377 granularity of 16 will require additional fenceposting where an 1378 option delta of 15 would happen to be required otherwise, which we 1379 have reserved. It can be argued that 16 is still the better choice, 1380 as fenceposting is already in the code path.) 1381 The Long Jump construct takes 0xf1 and 0xf2 from the space available 1382 for initial bytes of Options. (Note that we previously took 0xf0 to 1383 indicate end-of-options for OC=15.) 1385 Varying N with the length as defined above makes it unambiguous 1386 whether a one- or two-byte Long Jump is to be used. Setting N=2 for 1387 the one-byte version makes it clear that a Delta of 8 is to be 1388 handled the usual way (i.e., by Option Delta itself and/or 1389 fenceposting). If the delta is not small and not 7 modulo 8, there 1390 is still a choice between using the smaller multiple of 8 and a 1391 larger Delta in the actual Option or v.v., this biases the choice 1392 towards a larger Long Jump and a smaller following Delta, which is 1393 also easier to implement as it reduces the number of choice points. 1395 B.2.2.3. Example 1397 The following sequence of bytes would encode a Uri-Path Option of 1398 "foo" followed by Options 1357 (value "bar") and 1360 (value "baz"): 1400 93 65 6f 6f Option 9 (0 + 9, "foo") 1401 f1 a6 Long Jump by 1344 1402 43 62 61 72 Option 1357 (9 + 1344 + 4, "bar") 1403 33 62 61 7a Option 1360 (1357 + 3, "baz") 1405 Figure 17: Example using a Long Jump construct 1407 where f1 a6 is the long jump forward by (0xa6+2)*8=1344 option 1408 numbers. The total option count (OC) for the CoAP header is 3. Note 1409 that even if f1 a6 is skipped, the 1357 (which then appears as an 1410 Option number 13) is clearly visible as Critical. 1412 B.2.2.4. IANA considerations 1414 With the scheme proposed above, we could have three tiers of Option 1415 Numbers, differing in their allocation policy [RFC5226]: 1417 +---------------+-------------------------+ 1418 | Option Number | Policy | 1419 +---------------+-------------------------+ 1420 | 0..255 | Standards Action | 1421 | | | 1422 | 256..2047 | Designated Expert | 1423 | | | 1424 | 2048..65535 | First Come First Served | 1425 +---------------+-------------------------+ 1427 For the inventor of a new option, this would provide a small 1428 incentive to go through the designated expert for some minimal cross- 1429 checking in order to be able to use the two-byte long-jump. 1431 This draft adds option numbers to Table 2 of [I-D.ietf-core-coap]: 1433 +--------+---------------------+-----------+ 1434 | Number | Name | Reference | 1435 +--------+---------------------+-----------+ 1436 | 14 | Registered-Elective | [RFCXXXX] | 1437 | | | | 1438 | 25 | Registered-Critical | [RFCXXXX] | 1439 +--------+---------------------+-----------+ 1441 Table 1: New CoAP Option Numbers 1443 This draft adds a suboption registry, initially empty. 1445 +------------+-----------------------------+-----------+ 1446 | Number | Name | Reference | 1447 +------------+-----------------------------+-----------+ 1448 | 0..127 | (allocate on export review) | [RFCXXXX] | 1449 | | | | 1450 | 128..16383 | (allocate fcfs) | [RFCXXXX] | 1451 +------------+-----------------------------+-----------+ 1453 Table 2: CoAP Suboption Numbers 1455 B.3. Enabling Protocol Evolution 1457 To enable a protocol to evolve, it is critical that new capabilities 1458 can be introduced without requiring changes in components that don't 1459 really care about the capability. One such probem is exhibited by 1460 CoAP options: If a proxy does not understand an elective option in a 1461 request, it will not be able to forward it to the origin server, 1462 rendering the new option ineffectual. Worse, if a proxy does not 1463 understand a critical option in a request, it will not be able to 1464 operate on the request, rendering the new option damaging. 1466 As a conclusion to the Ticket #230 discussion in the June 4th interim 1467 call, we decided to solve the identification of options that a proxy 1468 can safely forward even if not understood (previously called Proxy- 1469 Elective). 1471 The proposal is to encode this information in the option number, just 1472 like the way the information that an option is critical is encoded 1473 now. This leads to two bits with semantics: the lowest bit continues 1474 to be the critical bit, and the next higher bit is now the "unsafe" 1475 bit (i.e., this option is not safe to forward unless understood by 1476 the proxy). 1478 Another consideration (for options that are not unsafe to forward) is 1479 whether the option should serve as a cache key in a request. HTTP 1480 has a vary header that indicates in the response which header fields 1481 were considered by the origin server to be cache keys. In order to 1482 avoid this complexity, we should be able to indicate this information 1483 right in the option number. However, reserving another bit is 1484 wasteful, in particular as there are few safe-to-forward options that 1485 are not cache-keys. 1487 Therefore, we propose the following bit allocation in an option 1488 number: 1490 xxx nnn UC 1492 Figure 18 1494 (where xxx is a variable length prefix, as option numbers are not 1495 bounded upwards). UC is the unsafe and critical bits. For U=0 only, 1496 if nnn is equal to 111 binary, the option does not serve as a cache 1497 key (for U=1, the proxy has to know the option to act on it, so there 1498 is no point in indicating whether it is a cache key). There is no 1499 semantic meaning of xxx. 1501 Note that clients and servers are generally not interested in this 1502 information. A proxy may use an equivalent of the following C code 1503 to derive the characteristics of an option number "onum": 1505 Critical = (onum & 1); 1506 UnSafe = (onum & 2); 1507 NoCache = ((onum & 0x1e) == 0x1c); 1509 Figure 19 1511 Discussion: This requires a renumbering of all options. 1513 This renumbering may also be considered as an opportunity to make 1514 the numbering straight again shortly before nailing down the 1515 protocol 1517 In particular, Content-Type is now probably better considered to 1518 be elective. 1520 B.3.1. Potential new option number allocation 1521 We want to give one example for a revised allocation of option 1522 numbers. Option numbers are given as decimal numbers, one each for 1523 xxx, nnn, and UC, with the UC values as follows 1525 +-----------+------------+------------------------------------+ 1526 | UC binary | UC decimal | meaning | 1527 +-----------+------------+------------------------------------+ 1528 | 00 | 0 | (safe, elective, 111=no-cache-key) | 1529 | | | | 1530 | 01 | 1 | (safe, critical, 111=no-cache-key) | 1531 | | | | 1532 | 10 | 2 | (unsafe, elective) | 1533 | | | | 1534 | 11 | 3 | (unsafe, critical) | 1535 +-----------+------------+------------------------------------+ 1537 The table is: 1539 +-----+-------+---------+-----------------------+-------------------+ 1540 | New | xx | Old | Name | Comment | 1541 | | nnn | | | | 1542 | | UC | | | | 1543 +-----+-------+---------+-----------------------+-------------------+ 1544 | 4 | 0 1 0 | 1 | Content-Type | category change | 1545 | | | | | (elective) | 1546 | | | | | | 1547 | 8 | 0 2 0 | 4 | ETag | | 1548 | | | | | | 1549 | 12 | 0 3 0 | 12 | Accept | | 1550 | | | | | | 1551 | 16 | 0 4 0 | 6 | Location-Path | | 1552 | | | | | | 1553 | 20 | 0 5 0 | 8 | Location-Query | | 1554 | | | | | | 1555 | 24 | 0 6 0 | - | (unused) | | 1556 | | | | | | 1557 | 28 | 0 7 0 | 18 | Size | needs nnn=111 | 1558 | | | | | | 1559 | 32 | 1 0 0 | 20/22 | Patience | | 1560 | | | | | | 1561 | 64 | 2 x 0 | - | Location-reserved | (nnn = 0..3, 4 | 1562 | | | | | reserved numbers) | 1563 | | | | | | 1564 | 1 | 0 0 1 | 13 | If-Match | | 1565 | | | | | | 1566 | 5 | 0 1 1 | 21 | If-None-Match | | 1567 | | | | | | 1568 | 2 | 0 0 2 | 2 | Max-Age | | 1569 | | | | | | 1570 | 6 | 0 1 2 | 10 | Observe | | 1571 | | | | | | 1572 | 10 | 0 2 2 | xx | Observe-2 | | 1573 | | | | | | 1574 | 14 | 0 3 2 | xx | (unused) | was fencepost | 1575 | | | | | | 1576 | 3 | 0 0 3 | 3 | Proxy-Uri | | 1577 | | | | | | 1578 | 7 | 0 1 3 | 5 | Uri-Host | | 1579 | | | | | | 1580 | 11 | 0 2 3 | 7 | Uri-Port | | 1581 | | | | | | 1582 | 15 | 0 3 3 | 9 | Uri-Path | | 1583 | | | | | | 1584 | 19 | 0 4 3 | 15 | Uri-Query | | 1585 | | | | | | 1586 | 23 | 0 5 3 | 11 | Token | | 1587 | | | | | | 1588 | 27 | 0 6 3 | 17 | Block2 | | 1589 | | | | | | 1590 | 31 | 0 7 3 | 19 | Block1 | yes, we can use | 1591 | | | | | nnn=111 with U=1 | 1592 +-----+-------+---------+-----------------------+-------------------+ 1594 B.4. Patience, Leisure, and Pledge 1596 A number of options might be useful for controlling the timing of 1597 interactions. 1599 (This section also addresses core-coap ticket #177.) 1601 B.4.1. Patience 1603 A client may have a limited time period in which it can actually make 1604 use of the response for a request. Using the Patience option, it can 1605 provide an (elective) indication how much time it is willing to wait 1606 for the response from the server, giving the server license to ignore 1607 or reject the request if it cannot fulfill it in this period. 1609 If the server knows early that it cannot fulfill the request in the 1610 time requested, it MAY indicate this with a 5.04 "Timeout" response. 1611 For non-safe methods (such as PUT, POST, DELETE), the server SHOULD 1612 indicate whether it has fulfilled the request by either responding 1613 with 5.04 "Timeout" (and not further processing the request) or by 1614 processing the request normally. 1616 Note that the value of the Patience option should be chosen such that 1617 the client will be able to make use of the result even in the 1618 presence of the expected network delays for the request and the 1619 response. Similarly, when a proxy receives a request with a Patience 1620 option and cannot fulfill that request from its cache, it may want to 1621 adjust the value of the option before forwarding it to an upstream 1622 server. 1624 (TBD: The various cases that arise when combining Patience with 1625 Observe.) 1627 The Patience option is elective. Hence, a client MUST be prepared to 1628 receive a normal response even after the chosen Patience period (plus 1629 an allowance for network delays) has elapsed. 1631 B.4.2. Leisure 1633 Servers generally will compute an internal value that we will call 1634 Leisure, which indicates the period of time that will be used for 1635 responding to a request. A Patience option, if present, can be used 1636 as an upper bound for the Leisure. Leisure may be non-zero for 1637 congestion control reasons, in particular for responses to multicast 1638 requests. For these, the server should have a group size estimate G, 1639 a target rate R (which both should be chosen conservatively) and an 1640 estimated response size S; a rough lower bound for Leisure can then 1641 be computed as follows: 1643 lb_Leisure = S * G / R 1645 Figure 20: Computing a lower bound for the Leisure 1647 E.g., for a multicast request with link-local scope on an 2.4 GHz 1648 IEEE 802.15.4 (6LoWPAN) network, G could be (relatively 1649 conservatively) set to 100, S to 100 bytes, and the target rate to 8 1650 kbit/s = 1 kB/s. The resulting lower bound for the Leisure is 10 1651 seconds. 1653 To avoid response implosion, responses to multicast requests SHOULD 1654 be dithered within a Leisure period chosen by the server to fall 1655 between these two bounds. 1657 Currently, we don't foresee a need to signal a value for Leisure from 1658 client to server (beyond the signalling provided by Patience) or from 1659 server to client, but an appropriate Option might be added later. 1661 B.4.3. Pledge 1662 In a basic observation relationship [I-D.ietf-core-observe], the 1663 server makes a pledge to keep the client in the observation 1664 relationship for a resource at least until the max-age for the 1665 resource is reached. 1667 To save the client some effort in re-establishing observation 1668 relationships each time max-age is reached, the server MAY want to 1669 extend its pledge beyond the end of max-age by signalling in a 1670 response/notification an additional time period using the Pledge 1671 Option, in parallel to the Observe Option. 1673 The Pledge Option MUST NOT be used unless the server can make a 1674 reasonable promise not to lose the observation relationship in this 1675 time frame. 1677 Currently, we don't foresee a need to signal a value for Pledge from 1678 client to server, but an appropriate behavior might be added later 1679 for this option when sent in a request. 1681 B.4.4. Option Formats 1683 +-----+----------+----------+-----------------+--------+---------+ 1684 | No. | C/E | Name | Format | Length | Default | 1685 +-----+----------+----------+-----------------+--------+---------+ 1686 | 22 | Elective | Patience | Duration in mis | 1 B | (none) | 1687 | | | | | | | 1688 | 24 | Elective | Pledge | Duration in s | 1 B | 0 | 1689 +-----+----------+----------+-----------------+--------+---------+ 1691 All timing options use the Duration data type (see Appendix D.2), 1692 however Patience (and Leisure, if that ever becomes an option) uses a 1693 timebase of mibiseconds (mis = 1/1024 s) instead of seconds. (This 1694 reduces the range of the Duration from ~ 91 days to 128 minutes.) 1696 Implementation note: As there are no strong accuracy requirements on 1697 the clocks employed, making use of any existing time base of 1698 milliseconds is a valid implementation approach (2.4 % off). 1700 None of the options may be repeated. 1702 Appendix C. The Cemetery (Things we won't do) 1704 This annex documents roads that the WG decided not to take, in order 1705 to spare readers from reinventing them in vain. 1707 C.1. Example envelope option: solving #230 1708 Ticket #230 [CoRE230] points out a design flaw of 1709 [I-D.ietf-core-coap]: When we split the elective Location option of 1710 draft -01 into multiple elective options, we made it possible that an 1711 implementation might process some of these and ignore others, leading 1712 to an incorrect interpretation of the Location expressed by the 1713 server. 1715 There are several more or less savory solutions to #230. 1717 Each of the elective options that together make up the Location could 1718 be defined in such a way that it makes a requirement on the 1719 processing of the related option (essentially revoking their elective 1720 status once the option under consideration is actually processed). 1721 This falls flat as soon as another option is defined that would also 1722 become part of the Location: existing implementations would not know 1723 that the new option is also part of the cluster that is re- 1724 interpreted as critical. The potential future addition of Location- 1725 Host and Location-Port makes this a valid consideration. 1727 A better solution would be to define an elective Envelope Option 1728 called Location. Within a Location Option, the following top-level 1729 options might be allowed (now or in the future): 1731 o Uri-Host 1733 o Uri-Port 1735 o Uri-Path 1737 o Uri-Query 1739 This would unify the code for interpreting the top-level request 1740 options that indicate the request URI with the code that interprets 1741 the Location URI. 1743 The four options listed are all critical, while the envelope is 1744 elective. This gives exactly the desired semantics: If the envelope 1745 is processed at all (which is elective), the nested options are 1746 critical and all need to be processed. 1748 C.2. Example envelope option: proxy-elective options 1749 Another potential application of envelope options is motivated by the 1750 observation that new critical options might not be implemented by all 1751 proxies on the CoAP path to an origin server. So that this does not 1752 become an obstacle to introducing new critical options that are of 1753 interest only to client and origin server, the client might want to 1754 mark some critical options proxy-elective, i.e. elective for a proxy 1755 but still critical for the origin server. 1757 One way to do this would be an Envelope option, the Proxy-Elective 1758 Option. A client might bundle a number of critical options into a 1759 critical Proxy-Elective Option. A proxy that processes the message 1760 is obliged to process the envelope (or reject the message), where 1761 processing means passing on the nested options towards the origin 1762 server (preferably again within a Proxy-Elective option). It can 1763 pass on the nested options, even ones unknown to the proxy, knowing 1764 that the client is happy with proxies not processing all of them. 1766 (The assumption here is that the Proxy-Elective option becomes part 1767 of the base standard, so all but the most basic proxies would know 1768 how to handle it.) 1770 C.3. Stateful URI compression 1772 Is the approximately 25 % average saving achievable with Huffman- 1773 based URI compression schemes worth the complexity? Probably not, 1774 because much higher average savings can be achieved by introducing 1775 state. 1777 Henning Schulzrinne has proposed for a server to be able to supply a 1778 shortened URI once a resource has been requested using the full- 1779 length URI. Let's call such a shortened referent a _Temporary 1780 Resource Identifier_, _TeRI_ for short. This could be expressed by a 1781 response option as shown in Figure 21. 1783 0 1784 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 1785 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1786 | duration | TeRI... 1787 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1789 Figure 21: Option for offering a TeRI in a response 1791 The TeRI offer option indicates that the server promises to offer 1792 this resources under the TeRI given for at least the time given as 1793 the duration. Another TeRI offer can be made later to extend the 1794 duration. 1796 Once a TeRI for a URI is known (and still within its lifetime), the 1797 client can supply a TeRI instead of a URI in its requests. The same 1798 option format as an offer could be used to allow the client to 1799 indicate how long it believes the TeRI will still be valid (so that 1800 the server can decide when to update the lifetime duration). TeRIs 1801 in requests could be distinguished from URIs e.g. by using a 1802 different option number. 1804 Proposal: Add a TeRI option that can be used in CoAP requests and 1805 responses. 1807 Add a way to indicate a TeRI and its duration in a link-value. 1809 Do not add any form of stateless URI encoding. 1811 Benefits: Much higher reduction of message size than any stateless 1812 URI encoding could achieve. 1814 As the use of TeRIs is entirely optional, minimal complexity nodes 1815 can get by without implementing them. 1817 Drawbacks: Adds considerable state and complexity to the protocol. 1819 It turns out that real CoAP URIs are short enough that TeRIs are 1820 not needed. 1822 (Discuss the security implications of TeRIs.) 1824 Appendix D. Experimental Options 1826 This annex documents proposals that need significant additional 1827 discussion before they can become part of (or go back to) the main 1828 CoAP specification. They are not dead, but might die if there turns 1829 out to be no good way to solve the problem. 1831 D.1. Options indicating absolute time 1833 HTTP has a number of headers that may indicate absolute time: 1835 o "Date", defined in Section 14.18 in [RFC2616] (Section 9.3 in 1836 [I-D.ietf-httpbis-p1-messaging]), giving the absolute time a 1837 response was generated; 1839 o "Last-Modified", defined in Section 14.29 in [RFC2616], 1840 (Section 6.6 in [I-D.ietf-httpbis-p4-conditional], giving the 1841 absolute time of when the origin server believes the resource 1842 representation was last modified; 1844 o "If-Modified-Since", defined in Section 14.25 in [RFC2616], "If- 1845 Unmodified-Since", defined in Section 14.28 in [RFC2616], and "If- 1846 Range", defined in Section 14.27 in [RFC2616] can be used to 1847 supply absolute time to gate a conditional request; 1849 o "Expires", defined in Section 14.21 in [RFC2616] (Section 3.3 in 1850 [I-D.ietf-httpbis-p6-cache]), giving the absolute time after which 1851 a response is considered stale. 1853 o The more obscure headers "Retry-After", defined in Section 14.37 1854 in [RFC2616], and "Warning", defined in section 14.46 in 1855 [RFC2616], also may employ absolute time. 1857 [I-D.ietf-core-coap] defines a single "Date" option, which however 1858 "indicates the creation time and date of a given resource 1859 representation", i.e., is closer to a "Last-Modified" HTTP header. 1860 HTTP's caching rules [I-D.ietf-httpbis-p6-cache] make use of both 1861 "Date" and "Last-Modified", combined with "Expires". The specific 1862 semantics required for CoAP needs further consideration. 1864 In addition to the definition of the semantics, an encoding for 1865 absolute times needs to be specified. 1867 In UNIX-related systems, it is customary to indicate absolute time as 1868 an integer number of seconds, after midnight UTC, January 1, 1970. 1869 Unless negative numbers are employed, this time format cannot 1870 represent time values prior to January 1, 1970, which probably is not 1871 required for the uses ob absolute time in CoAP. 1873 If a 32-bit integer is used and allowance is made for a sign-bit in a 1874 local implementation, the latest UTC time value that can be 1875 represented by the resulting 31 bit integer value is 03:14:07 on 1876 January 19, 2038. If the 32-bit integer is used as an unsigned 1877 value, the last date is 2106-02-07, 06:28:15. 1879 The reach can be extended by: - moving the epoch forward, e.g. by 40 1880 years (= 1262304000 seconds) to 2010-01-01. This makes it impossible 1881 to represent Last-Modified times in that past (such as could be 1882 gatewayed in from HTTP). - extending the number of bits, e.g. by 1883 one more byte, either always or as one of two formats, keeping the 1884 32-bit variant as well. 1886 Also, the resolution can be extended by expressing time in 1887 milliseconds etc., requiring even more bits (e.g., a 48-bit unsigned 1888 integer of milliseconds would last well after year 9999.) 1890 For experiments, an experimental "Date" option is defined with the 1891 semantics of HTTP's "Last-Modified". It can carry an unsigned 1892 integer of 32, 40, or 48 bits; 32- and 40-bit integers indicate the 1893 absolute time in seconds since 1970-01-01 00:00 UTC, while 48-bit 1894 integers indicate the absolute time in milliseconds since 1970-01-01 1895 00:00 UTC. 1897 However, that option is not really that useful until there is a "If- 1898 Modified-Since" option as well. 1900 (Also: Discuss nodes without clocks.) 1902 D.2. Representing Durations 1904 Various message types used in CoAP need the representation of 1905 *durations*, i.e. of the length of a timespan. In SI units, these 1906 are measured in seconds. CoAP durations represent integer numbers of 1907 seconds, but instead of representing these numbers as integers, a 1908 more compact single-byte pseudo-floating-point (pseudo-FP) 1909 representation is used (Figure 22). 1911 0 1 2 3 4 5 6 7 1912 +---+---+---+---+---+---+---+---+ 1913 | 0... value | 1914 +---+---+---+---+---+---+---+---+ 1916 +---+---+---+---+---+---+---+---+ 1917 | 1... mantissa | exponent | 1918 +---+---+---+---+---+---+---+---+ 1920 Figure 22: Duration in (8,4) pseudo-FP representation 1922 If the high bit is clear, the entire n-bit value (including the high 1923 bit) is the decoded value. If the high bit is set, the mantissa 1924 (including the high bit, with the exponent field cleared out but 1925 still present) is shifted left by the exponent to yield the decoded 1926 value. 1928 The (n,e)-pseudo-FP format can be decoded with a single line of code 1929 (plus a couple of constant definitions), as demonstrated in Figure 1930 23. 1932 #define N 8 1933 #define E 4 1934 #define HIBIT (1 << (N - 1)) 1935 #define EMASK ((1 << E) - 1) 1936 #define MMASK ((1 << N) - 1 - EMASK) 1938 #define DECODE_8_4(r) (r < HIBIT ? r : (r & MMASK) << (r & EMASK)) 1939 Figure 23: Decoding an (8,4) pseudo-FP value 1941 Note that a pseudo-FP encoder needs to consider rounding; different 1942 applications of durations may favor rounding up or rounding down the 1943 value encoded in the message. 1945 The highest pseudo-FP value, represented by an all-ones byte (0xFF), 1946 is reserved to indicate an indefinite duration. The next lower value 1947 (0xEF) is thus the highest representable value and is decoded as 1948 7340032 seconds, a little more than 12 weeks. 1950 D.3. Rationale 1952 Where CPU power and memory is abundant, a duration can almost always 1953 be adequately represented by a non-negative floating-point number 1954 representing that number of seconds. Historically, many APIs have 1955 also used an integer representation, which limits both the resolution 1956 (e.g., if the integer represents the duration in seconds) and often 1957 the range (integer machine types have range limits that may become 1958 relevant). UNIX's "time_t" (which is used for both absolute time and 1959 durations) originally was a signed 32-bit value of seconds, but was 1960 later complemented by an additional integer to add microsecond 1961 ("struct timeval") and then later nanosecond ("struct timespec") 1962 resolution. 1964 Three decisions need to be made for each application of the concept 1965 of duration: 1967 o the *resolution*. What rounding error is acceptable? 1969 o the *range*. What is the maximum duration that needs to be 1970 represented? 1972 o the *number of bits* that can be expended. 1974 Obviously, these decisions are interrelated. Typically, a large 1975 range needs a large number of bits, unless resolution is traded. For 1976 most applications, the actual requirement for resolution are limited 1977 for longer durations, but can be more acute for shorter durations. 1979 D.4. Pseudo-Floating Point 1981 Constrained systems typically avoid the use of floating-point (FP) 1982 values, as 1984 o simple CPUs often don't have support for floating-point datatypes 1985 o software floating-point libraries are expensive in code size and 1986 slow. 1988 In addition, floating-point datatypes used to be a significant 1989 element of market differentiation in CPU design; it has taken the 1990 industry a long time to agree on a standard floating point 1991 representation. 1993 These issues have led to protocols that try to constrain themselves 1994 to integer representation even where the ability of a floating point 1995 representation to trade range for resolution would be beneficial. 1997 The idea of introducing _pseudo-FP_ is to obtain the increased range 1998 provided by embedding an exponent, without necessarily getting stuck 1999 with hardware datatypes or inefficient software floating-point 2000 libraries. 2002 For the purposes of this draft, we define an (n,e)-pseudo-FP as a 2003 fixed-length value of n bits, e of which may be used for an exponent. 2004 Figure 22 illustrates an (8,4)-pseudo-FP value. 2006 If the high bit is clear, the entire n-bit value (including the high 2007 bit) is the decoded value. If the high bit is set, the mantissa 2008 (including the high bit, but with the exponent field cleared out) is 2009 shifted left by the exponent to yield the decoded value. 2011 The (n,e)-pseudo-FP format can be decoded with a single line of code 2012 (plus a couple of constant definition), as demonstrated in Figure 23. 2014 Only non-negative numbers can be represented by this format. It is 2015 designed to provide full integer resolution for values from 0 to 2016 2^(n-1)-1, i.e., 0 to 127 in the (8,4) case, and a mantissa of n-e 2017 bits from 2^(n-1) to (2^n-2^e)*2^(2^e-1), i.e., 128 to 7864320 in the 2018 (8,4) case. By choosing e carefully, resolution can be traded 2019 against range. 2021 Note that a pseudo-FP encoder needs to consider rounding; different 2022 applications of durations may favor rounding up or rounding down the 2023 value encoded in the message. This requires a little more than a 2024 single line of code (which is left as an exercise to the reader, as 2025 the most efficient expression depends on hardware details). 2027 D.5. A Duration Type for CoAP 2028 CoAP needs durations in a number of places. In [I-D.ietf-core-coap], 2029 durations occur in the option "Subscription-lifetime" as well as in 2030 the option "Max-age". (Note that the option "Date" is not a 2031 duration, but a point in time.) Other durations of this kind may be 2032 added later. 2034 Most durations relevant to CoAP are best expressed with a minimum 2035 resolution of one second. More detailed resolutions are unlikely to 2036 provide much benefit. 2038 The range of lifetimes and caching ages are probably best kept below 2039 the order of magnitude of months. An (8,4)-pseudo-FP has the maximum 2040 value of 7864320, which is about 91 days; this appears to be adequate 2041 for a subscription lifetime and probably even for a maximum cache 2042 age. Figure 24 shows the values that can be expressed. (If a larger 2043 range for the latter is indeed desired, an (8,5)-pseudo-FP could be 2044 used; this would last 15 milleniums, at the cost of having only 3 2045 bits of accuracy for values larger than 127 seconds.) 2047 Proposal: A single duration type is used throughout CoAP, based on 2048 an (8,4)-pseudo-FP giving a duration in seconds. 2050 Benefits: Implementations can use a single piece of code for 2051 managing all CoAP-related durations. 2053 In addition, length information never needs to be managed for 2054 durations that are embedded in other data structures: All 2055 durations are expressed by a single byte. 2057 It might be worthwhile to reserve one duration value, e.g. 0xFF, for 2058 an indefinite duration. 2060 Duration Seconds Encoded 2061 ----------- ---------- ------- 2062 00:00:00 0x00000000 0x00 2063 00:00:01 0x00000001 0x01 2064 00:00:02 0x00000002 0x02 2065 00:00:03 0x00000003 0x03 2066 00:00:04 0x00000004 0x04 2067 00:00:05 0x00000005 0x05 2068 00:00:06 0x00000006 0x06 2069 00:00:07 0x00000007 0x07 2070 00:00:08 0x00000008 0x08 2071 00:00:09 0x00000009 0x09 2072 00:00:10 0x0000000a 0x0a 2073 00:00:11 0x0000000b 0x0b 2074 00:00:12 0x0000000c 0x0c 2075 00:00:13 0x0000000d 0x0d 2076 00:00:14 0x0000000e 0x0e 2077 00:00:15 0x0000000f 0x0f 2078 00:00:16 0x00000010 0x10 2079 00:00:17 0x00000011 0x11 2080 00:00:18 0x00000012 0x12 2081 00:00:19 0x00000013 0x13 2082 00:00:20 0x00000014 0x14 2083 00:00:21 0x00000015 0x15 2084 00:00:22 0x00000016 0x16 2085 00:00:23 0x00000017 0x17 2086 00:00:24 0x00000018 0x18 2087 00:00:25 0x00000019 0x19 2088 00:00:26 0x0000001a 0x1a 2089 00:00:27 0x0000001b 0x1b 2090 00:00:28 0x0000001c 0x1c 2091 00:00:29 0x0000001d 0x1d 2092 00:00:30 0x0000001e 0x1e 2093 00:00:31 0x0000001f 0x1f 2094 00:00:32 0x00000020 0x20 2095 00:00:33 0x00000021 0x21 2096 00:00:34 0x00000022 0x22 2097 00:00:35 0x00000023 0x23 2098 00:00:36 0x00000024 0x24 2099 00:00:37 0x00000025 0x25 2100 00:00:38 0x00000026 0x26 2101 00:00:39 0x00000027 0x27 2102 00:00:40 0x00000028 0x28 2103 00:00:41 0x00000029 0x29 2104 00:00:42 0x0000002a 0x2a 2105 00:00:43 0x0000002b 0x2b 2106 00:00:44 0x0000002c 0x2c 2107 00:00:45 0x0000002d 0x2d 2108 00:00:46 0x0000002e 0x2e 2109 00:00:47 0x0000002f 0x2f 2110 00:00:48 0x00000030 0x30 2111 00:00:49 0x00000031 0x31 2112 00:00:50 0x00000032 0x32 2113 00:00:51 0x00000033 0x33 2114 00:00:52 0x00000034 0x34 2115 00:00:53 0x00000035 0x35 2116 00:00:54 0x00000036 0x36 2117 00:00:55 0x00000037 0x37 2118 00:00:56 0x00000038 0x38 2119 00:00:57 0x00000039 0x39 2120 00:00:58 0x0000003a 0x3a 2121 00:00:59 0x0000003b 0x3b 2122 00:01:00 0x0000003c 0x3c 2123 00:01:01 0x0000003d 0x3d 2124 00:01:02 0x0000003e 0x3e 2125 00:01:03 0x0000003f 0x3f 2126 00:01:04 0x00000040 0x40 2127 00:01:05 0x00000041 0x41 2128 00:01:06 0x00000042 0x42 2129 00:01:07 0x00000043 0x43 2130 00:01:08 0x00000044 0x44 2131 00:01:09 0x00000045 0x45 2132 00:01:10 0x00000046 0x46 2133 00:01:11 0x00000047 0x47 2134 00:01:12 0x00000048 0x48 2135 00:01:13 0x00000049 0x49 2136 00:01:14 0x0000004a 0x4a 2137 00:01:15 0x0000004b 0x4b 2138 00:01:16 0x0000004c 0x4c 2139 00:01:17 0x0000004d 0x4d 2140 00:01:18 0x0000004e 0x4e 2141 00:01:19 0x0000004f 0x4f 2142 00:01:20 0x00000050 0x50 2143 00:01:21 0x00000051 0x51 2144 00:01:22 0x00000052 0x52 2145 00:01:23 0x00000053 0x53 2146 00:01:24 0x00000054 0x54 2147 00:01:25 0x00000055 0x55 2148 00:01:26 0x00000056 0x56 2149 00:01:27 0x00000057 0x57 2150 00:01:28 0x00000058 0x58 2151 00:01:29 0x00000059 0x59 2152 00:01:30 0x0000005a 0x5a 2153 00:01:31 0x0000005b 0x5b 2154 00:01:32 0x0000005c 0x5c 2155 00:01:33 0x0000005d 0x5d 2156 00:01:34 0x0000005e 0x5e 2157 00:01:35 0x0000005f 0x5f 2158 00:01:36 0x00000060 0x60 2159 00:01:37 0x00000061 0x61 2160 00:01:38 0x00000062 0x62 2161 00:01:39 0x00000063 0x63 2162 00:01:40 0x00000064 0x64 2163 00:01:41 0x00000065 0x65 2164 00:01:42 0x00000066 0x66 2165 00:01:43 0x00000067 0x67 2166 00:01:44 0x00000068 0x68 2167 00:01:45 0x00000069 0x69 2168 00:01:46 0x0000006a 0x6a 2169 00:01:47 0x0000006b 0x6b 2170 00:01:48 0x0000006c 0x6c 2171 00:01:49 0x0000006d 0x6d 2172 00:01:50 0x0000006e 0x6e 2173 00:01:51 0x0000006f 0x6f 2174 00:01:52 0x00000070 0x70 2175 00:01:53 0x00000071 0x71 2176 00:01:54 0x00000072 0x72 2177 00:01:55 0x00000073 0x73 2178 00:01:56 0x00000074 0x74 2179 00:01:57 0x00000075 0x75 2180 00:01:58 0x00000076 0x76 2181 00:01:59 0x00000077 0x77 2182 00:02:00 0x00000078 0x78 2183 00:02:01 0x00000079 0x79 2184 00:02:02 0x0000007a 0x7a 2185 00:02:03 0x0000007b 0x7b 2186 00:02:04 0x0000007c 0x7c 2187 00:02:05 0x0000007d 0x7d 2188 00:02:06 0x0000007e 0x7e 2189 00:02:07 0x0000007f 0x7f 2190 00:02:08 0x00000080 0x80 2191 00:02:24 0x00000090 0x90 2192 00:02:40 0x000000a0 0xa0 2193 00:02:56 0x000000b0 0xb0 2194 00:03:12 0x000000c0 0xc0 2195 00:03:28 0x000000d0 0xd0 2196 00:03:44 0x000000e0 0xe0 2197 00:04:00 0x000000f0 0xf0 2198 00:04:16 0x00000100 0x81 2199 00:04:48 0x00000120 0x91 2200 00:05:20 0x00000140 0xa1 2201 00:05:52 0x00000160 0xb1 2202 00:06:24 0x00000180 0xc1 2203 00:06:56 0x000001a0 0xd1 2204 00:07:28 0x000001c0 0xe1 2205 00:08:00 0x000001e0 0xf1 2206 00:08:32 0x00000200 0x82 2207 00:09:36 0x00000240 0x92 2208 00:10:40 0x00000280 0xa2 2209 00:11:44 0x000002c0 0xb2 2210 00:12:48 0x00000300 0xc2 2211 00:13:52 0x00000340 0xd2 2212 00:14:56 0x00000380 0xe2 2213 00:16:00 0x000003c0 0xf2 2214 00:17:04 0x00000400 0x83 2215 00:19:12 0x00000480 0x93 2216 00:21:20 0x00000500 0xa3 2217 00:23:28 0x00000580 0xb3 2218 00:25:36 0x00000600 0xc3 2219 00:27:44 0x00000680 0xd3 2220 00:29:52 0x00000700 0xe3 2221 00:32:00 0x00000780 0xf3 2222 00:34:08 0x00000800 0x84 2223 00:38:24 0x00000900 0x94 2224 00:42:40 0x00000a00 0xa4 2225 00:46:56 0x00000b00 0xb4 2226 00:51:12 0x00000c00 0xc4 2227 00:55:28 0x00000d00 0xd4 2228 00:59:44 0x00000e00 0xe4 2229 01:04:00 0x00000f00 0xf4 2230 01:08:16 0x00001000 0x85 2231 01:16:48 0x00001200 0x95 2232 01:25:20 0x00001400 0xa5 2233 01:33:52 0x00001600 0xb5 2234 01:42:24 0x00001800 0xc5 2235 01:50:56 0x00001a00 0xd5 2236 01:59:28 0x00001c00 0xe5 2237 02:08:00 0x00001e00 0xf5 2238 02:16:32 0x00002000 0x86 2239 02:33:36 0x00002400 0x96 2240 02:50:40 0x00002800 0xa6 2241 03:07:44 0x00002c00 0xb6 2242 03:24:48 0x00003000 0xc6 2243 03:41:52 0x00003400 0xd6 2244 03:58:56 0x00003800 0xe6 2245 04:16:00 0x00003c00 0xf6 2246 04:33:04 0x00004000 0x87 2247 05:07:12 0x00004800 0x97 2248 05:41:20 0x00005000 0xa7 2249 06:15:28 0x00005800 0xb7 2250 06:49:36 0x00006000 0xc7 2251 07:23:44 0x00006800 0xd7 2252 07:57:52 0x00007000 0xe7 2253 08:32:00 0x00007800 0xf7 2254 09:06:08 0x00008000 0x88 2255 10:14:24 0x00009000 0x98 2256 11:22:40 0x0000a000 0xa8 2257 12:30:56 0x0000b000 0xb8 2258 13:39:12 0x0000c000 0xc8 2259 14:47:28 0x0000d000 0xd8 2260 15:55:44 0x0000e000 0xe8 2261 17:04:00 0x0000f000 0xf8 2262 18:12:16 0x00010000 0x89 2263 20:28:48 0x00012000 0x99 2264 22:45:20 0x00014000 0xa9 2265 1d 01:01:52 0x00016000 0xb9 2266 1d 03:18:24 0x00018000 0xc9 2267 1d 05:34:56 0x0001a000 0xd9 2268 1d 07:51:28 0x0001c000 0xe9 2269 1d 10:08:00 0x0001e000 0xf9 2270 1d 12:24:32 0x00020000 0x8a 2271 1d 16:57:36 0x00024000 0x9a 2272 1d 21:30:40 0x00028000 0xaa 2273 2d 02:03:44 0x0002c000 0xba 2274 2d 06:36:48 0x00030000 0xca 2275 2d 11:09:52 0x00034000 0xda 2276 2d 15:42:56 0x00038000 0xea 2277 2d 20:16:00 0x0003c000 0xfa 2278 3d 00:49:04 0x00040000 0x8b 2279 3d 09:55:12 0x00048000 0x9b 2280 3d 19:01:20 0x00050000 0xab 2281 4d 04:07:28 0x00058000 0xbb 2282 4d 13:13:36 0x00060000 0xcb 2283 4d 22:19:44 0x00068000 0xdb 2284 5d 07:25:52 0x00070000 0xeb 2285 5d 16:32:00 0x00078000 0xfb 2286 6d 01:38:08 0x00080000 0x8c 2287 6d 19:50:24 0x00090000 0x9c 2288 7d 14:02:40 0x000a0000 0xac 2289 8d 08:14:56 0x000b0000 0xbc 2290 9d 02:27:12 0x000c0000 0xcc 2291 9d 20:39:28 0x000d0000 0xdc 2292 10d 14:51:44 0x000e0000 0xec 2293 11d 09:04:00 0x000f0000 0xfc 2294 12d 03:16:16 0x00100000 0x8d 2295 13d 15:40:48 0x00120000 0x9d 2296 15d 04:05:20 0x00140000 0xad 2297 16d 16:29:52 0x00160000 0xbd 2298 18d 04:54:24 0x00180000 0xcd 2299 19d 17:18:56 0x001a0000 0xdd 2300 21d 05:43:28 0x001c0000 0xed 2301 22d 18:08:00 0x001e0000 0xfd 2302 24d 06:32:32 0x00200000 0x8e 2303 27d 07:21:36 0x00240000 0x9e 2304 30d 08:10:40 0x00280000 0xae 2305 33d 08:59:44 0x002c0000 0xbe 2306 36d 09:48:48 0x00300000 0xce 2307 39d 10:37:52 0x00340000 0xde 2308 42d 11:26:56 0x00380000 0xee 2309 45d 12:16:00 0x003c0000 0xfe 2310 48d 13:05:04 0x00400000 0x8f 2311 54d 14:43:12 0x00480000 0x9f 2312 60d 16:21:20 0x00500000 0xaf 2313 66d 17:59:28 0x00580000 0xbf 2314 72d 19:37:36 0x00600000 0xcf 2315 78d 21:15:44 0x00680000 0xdf 2316 84d 22:53:52 0x00700000 0xef 2317 91d 00:32:00 0x00780000 0xff (reserved) 2319 Figure 24 2321 D.6. CONTOUR (CoAP Non-trivial Option Useful Representation) 2323 So far, CoAP options have been simple. For non-trivial options of 2324 the future, e.g. for interaction with cellular systems, it is useful 2325 to have an option value encoding that can represent tree-structured 2326 data. Several such data structures are good candidates. If none of 2327 them work out, the CBOR encoding defined in [I-D.bormann-cbor] may be 2328 useful. 2330 Authors' Addresses 2332 Carsten Bormann 2333 Universitaet Bremen TZI 2334 Postfach 330440 2335 Bremen D-28359 2336 Germany 2338 Phone: +49-421-218-63921 2339 Email: cabo@tzi.org 2341 Klaus Hartke 2342 Universitaet Bremen TZI 2343 Postfach 330440 2344 Bremen D-28359 2345 Germany 2347 Phone: +49-421-218-63905 2348 Email: hartke@tzi.org