idnits 2.17.1 draft-knodel-e2ee-definition-00.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 22, 2021) is 1158 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Model-T M. Knodel 3 Internet-Draft CDT 4 Intended status: Informational F. Baker 5 Expires: August 26, 2021 6 O. Kolkman 7 ISOC 8 S. Celi 9 Cloudflare 10 G. Grover 11 Centre for Internet and Society 12 February 22, 2021 14 Definition of End-to-end Encryption 15 draft-knodel-e2ee-definition-00 17 Abstract 19 End-to-end encryption (E2EE) is an application of cryptography in 20 communications systems between endpoints. E2EE systems are unique in 21 providing features of confidentiality, integrity and authenticity for 22 users. Improvements to E2EE strive to maximise the system's security 23 while balancing usability and availability. Users of E2EE 24 communications expect trustworthy providers of secure implementations 25 to respect and protect their right to whisper. 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at https://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on August 26, 2021. 44 Copyright Notice 46 Copyright (c) 2021 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (https://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 62 2. Formal definition of end-to-end encryption . . . . . . . . . 3 63 2.1. End-to-end principle . . . . . . . . . . . . . . . . . . 3 64 2.2. Encryption . . . . . . . . . . . . . . . . . . . . . . . 4 65 3. End-to-end encrypted systems design . . . . . . . . . . . . . 5 66 3.1. Features . . . . . . . . . . . . . . . . . . . . . . . . 5 67 3.1.1. Necessary features . . . . . . . . . . . . . . . . . 5 68 3.1.2. Optional/desirable features . . . . . . . . . . . . . 6 69 3.2. Challenges . . . . . . . . . . . . . . . . . . . . . . . 6 70 4. End-user expectations . . . . . . . . . . . . . . . . . . . . 8 71 4.1. A conversation is confidential . . . . . . . . . . . . . 8 72 4.2. Providers are trustworthy . . . . . . . . . . . . . . . . 8 73 4.3. Access by a third-party is impossible . . . . . . . . . . 9 74 4.4. Pattern inference is minimised . . . . . . . . . . . . . 9 75 4.5. The e2ee system is not compromised . . . . . . . . . . . 9 76 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 10 77 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 78 7. Security Considerations . . . . . . . . . . . . . . . . . . . 10 79 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 80 9. Informative References . . . . . . . . . . . . . . . . . . . 10 81 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 11 83 1. Introduction 85 This document defines end-to-end encryption (E2EE) using three 86 different dimensions that together comprise a full definition of 87 E2EE, which can be applied in a variety of contexts. 89 The first is a formal definition that draws on the basic 90 understanding of end points and cryptography. The second looks at 91 E2EE systems from a design perspective, both its fundamental features 92 and the direction of travel towards improving those features. Lastly 93 we consider the expectations of the user of E2EE systems. 95 These dimensions taken as a whole comprise a generally comprehensible 96 picture of consensus at the IETF as to what is end-to-end encryption, 97 irrespective of application, from messaging to video conferencing, 98 and between any number of end points. 100 2. Formal definition of end-to-end encryption 102 An end-to-end encrypted communications system, irrespective of the 103 content or the specific methods employed, relies on two important and 104 rigorous technical concepts: The end-to-end principle, defined in the 105 IETF because of its importance to internet protocols; and encryption, 106 an application of cryptography and the primary means employed by the 107 IETF to secure internet protocols. 109 2.1. End-to-end principle 111 An important question in any review of the End-to-End Principle 112 [RFC3724] is "what constitutes an end?" Intuitively, an "end" either 113 sends messages or receives them, usually both; other systems on the 114 path are just that - other systems. 116 In 1984 the "end-to-end argument" was introduced [saltzer] as a 117 design principle that helps guide placement of functions among the 118 modules of a distributed computer system. It suggests that functions 119 placed at low levels of a system may be redundant or of little value 120 when compared with the cost of providing them at that low level. 121 It's used to design around questions about which parts of the system 122 should make which decisions, and as such the identity of the actual 123 "speaker" or "end" may be less obvious than it appears. The 124 communication described by Saltzer is between communicating 125 processes, which may or may not be on the same physical machine, and 126 may be implemented in various ways. For example, a BGP speaker is 127 often implemented as a process that manages the Routing Information 128 Base (RIB) and communicates with other BGP speakers using an 129 operating system service that implements TCP. The RIB manager might 130 find itself searching the RIB for prefixes that should be advertised 131 to a peer, and performing "writes" to TCP for each one. TCP in this 132 context often implements a variant of the algorithm described in RFC 133 868 (the "Nagle algorithm"), which accumulates writes in a buffer 134 until there is no data in flight between the communicants, and then 135 sends it - which might happen several times during a single search by 136 the RIB manager. In that sense, the RIB manager might be thought of 137 as the "end", because it decides what should be communicated, or TCP 138 might be the "end", because it actually sends the TCP Segment, 139 detects errors if they occur, retransmits it if necessary, and 140 ultimately decides that the segment has been successfully 141 transferred. 143 However despite the nuance for engineers, it is now widely accepted 144 that the communication system itself begins and ends with the user 146 [RFC8890]. We imagine people (through an application's user 147 interface) as components in a subsystem's design. An important 148 exception to this in E2EE systems might be the use of public key 149 infrastructure where a third party is often used in the 150 authentication phase to enhance the larger system's trust model. 152 Another important question is "what statement exactly summarizes the 153 end-to-end principle?". Saltzer answered this in two ways, the first 154 of which is that the service implementing the transaction is most 155 correct if it implements the intent of the application that sent it, 156 which would be to move the message toward the destination address in 157 the relevant IP header. Salzer's more thorough treatment, however, 158 deals with end cases that come up in implementation: "Examples 159 discussed in the paper", according to the abstract, "include bit 160 error recovery, security using encryption, duplicate message 161 suppression, recovery from system crashes, and delivery 162 acknowledgement." It also notes that there is occasionally a 163 rationale for ignoring the end-to-end arguments for the purposes of 164 optimization. There may be other user expectations or design 165 features, some explained below, which need to be balanced with the 166 end-to-end argument. 168 2.2. Encryption 170 From draft-dkg-hrpc-glossary-00, encryption is fundamental to the 171 end-to-end principle. "End-to-End : The principal of extending 172 characteristics of a protocol or system as far as possible within the 173 system. For example, end-to-end instant message encryption would 174 conceal communication content from one user's instant messaging 175 application through any intermediate devices and servers all the way 176 to the recipient's instant messaging application. If the message was 177 decrypted at any intermediate point-for example at a service 178 provider-then the property of end-to-end encryption would not be 179 present."[dkg] Note that this only talks about the contents of the 180 communication and not the metadata generated from it. 182 The way to achieve a truly end-to-end communications system is indeed 183 to encrypt the content of the data exchanged between the endpoints, 184 eg sender(s) and receiver(s). The more common end-to-end technique 185 for encrypting uses a double-ratchet algorithm with an authenticated 186 encryption scheme, present in many modern messenger applications such 187 as those considered in the IETF Messaging Layer Security working 188 group, whose charter is to create a document that satisfies the need 189 for several Internet applications for group key establishment and 190 message protection protocols.[mls] OpenPGP, mostly used for email, 191 uses a different technique to achieve encryption. It is also 192 chartered in the IETF to create an specification that covers object 193 encryption, object signing, and identity certification.[openpgp] Both 194 protocols rely on the use of asymmetric and symmetric encryption, and 195 have to exchange public keys with amongst end points. 197 There are dozens of documents in the RFC Series that fundamentally 198 and technically define encryption schemes. Perhaps interesting work 199 to be done would be to survey all existing documents of this kind to 200 define, in aggregate, their common features. The point is, the IETF 201 has clear mandate and demonstrated expertise in defining the 202 specifics of encrypted communications of the internet. 204 3. End-to-end encrypted systems design 206 When looking at E2EE systems from a design perspective, the first 207 consideration is the list of fundamental features that distinguish an 208 E2EE system from one that does not employ E2EE. Secondly one must 209 consider the direction of travel for improving the features of E2EE 210 systems. In other words, what challenges are the designers, 211 developers and implementers of E2EE systems facing? 213 The features and challenges listed below are framed holistically 214 rather than from the perspective of their design, development, 215 implementation or use. 217 3.1. Features 219 Defining a technology can also be done by inspecting what it does, or 220 is meant to do, in the form of features. The features of end-to-end 221 encryption from an implementation perspective can be inspected across 222 several important categories: 1) the necessary features of E2EE of 223 authenticity, confidentiality and integrity, whereas features of 2) 224 availability, deniability, forward secrecy and post-compromise 225 security are enhancements to E2EE systems. 227 3.1.1. Necessary features 229 Authenticity A system provides message authenticity if the recipient 230 is certain who sent the message and the sender is certain who 231 received it. 233 Confidentiality A system provides message confidentiality if only 234 the sender and intended recipient(s) can read the message 235 plaintext, i.e. messages are encrypted by the sender such that 236 only the intended recipient(s) can decrypt them. 238 Integrity A system provides message integrity when it guarantees 239 that messages has not been modified in transit, i.e. a recipient 240 is assured that the message they have received is exactly what the 241 sender intented to sent. 243 3.1.2. Optional/desirable features 245 Availability A system provides high availability if the user is able 246 to get to the message when they so desire and potentially from 247 more than one device, i.e. a message arrives to a recipient even 248 if they have been offline for a long time. 250 Deniability Deniability ensures that anyone with a record of the 251 transcript, including message recipients, cannot cryptographically 252 prove to others that a particular participant of a communication 253 authored the message. As demonstrated by the Signal and OTR 254 protocols, this property has to exist in conjunction with message 255 authenticity, i.e. participants in a communication have to be 256 assured that they are communicating with the intended parties but 257 this assurance cannot be proof to any other parties. 259 Forward secrecy Forward secrecy is a security property that prevents 260 attackers from decrypting all encrypted data they have previously 261 captured over a communication channel, even if they have 262 compromised one of the endpoints. Forward secrecy is usually 263 achieved by updating the encryption/decryption keys, and older 264 ones are deleted periodically. 266 Post-compromise security Post-compromise security is a security 267 property that seeks to guarantee a way to recover from an end- 268 point compromise (and consequently that communication sent post- 269 compromise is protected with the same security properties that 270 existed before the compromise). It is usually achieved by adding 271 ephemeral key exchanges to the derivation of encryption/decryption 272 keys. 274 3.2. Challenges 276 Earlier we defined end-to-end encryption using formal definitions 277 assumed by internet protocol implementations. And because "the IETF 278 is a place for state-of-the-art producing high quality, relevant 279 technical documents that influence the way people design, use, and 280 manage the Internet" we can be confident that current deployments of 281 end-to-end encrypted technologies in the IETF indicate the cutting 282 edge of their developments, yet another way to define what is, or 283 ideally should be, how a technology is defined. 285 Below is an exhaustive, yet vaguely summarised, list of the 286 challenges currently faced by protocol designers of end-to-end 287 encrypted systems. In other words, in order to realise the goals of 288 end-to-end encrypted systems, both for users and implementers (see 289 previous section), these problems must be tackled. Problems that 290 fall outside of this list are likely 1) unnecessary feature requests 291 that negligibly, or do nothing to, achieve the aims of end-to-end 292 encrypted systems or 2) in some way antithetical to the goals of end- 293 to-end encrypted systems. 295 Public key verification is very difficult for users to manage. 296 Authentication of the two ends is required for confidential 297 conversations. Therefore solving the problem of verification of 298 public keys is a major concern for any end-to-end encrypted system 299 design. Some applications bind together the account identity and the 300 key, and leaves users to establish a trust relationship between them, 301 assisted by public key fingerprint information. 303 Users want to smoothly switch application use between devices, but 304 this comes at a cost to the security of user data. Thus there is a 305 problem of availability in end-to-end encrypted systems because the 306 account identity's private key is generated by and stored on the end- 307 user's original device and to move the private key to another device 308 compromises the security of one of the end-points of the system. 310 Existing protocols are vulnerable to meta-data analysis, even though 311 meta-data is often much more sensitive than content. Meta-data is 312 plaintext information that travels across the wire and includes 313 delivery-relevant details that central servers need such as the 314 account identity of end-points, timestamps, message size. Meta-data 315 is difficult to obfuscate efficiently. 317 Users need to communicate in groups, but this presents major problems 318 of scale for end-to-end encryption systems that rely on public key 319 cryptography. 321 The whole of a user's data should remain secure if only one message 322 is compromised. However, for encrypted communication, you must 323 currently choose between forward secrecy or the ability to 324 communicate asynchronously. This presents a problem for application 325 design that uses end-to-end encryption for asynchronous messaging 326 over email, RCS, etc. 328 Users of E2EE systems should be able to communicate with any medium 329 of their choice, from plain text to large files, however there is 330 often a resource problem because there are no open protocols to allow 331 users to securely share the same resource in an end-to-end encrypted 332 system. Client-side, eg end-point, activities like URL unfurling 333 scanning. 335 Usability considerations are sometimes in conflict with security 336 considerations, such as message read status, typing indicators, URL/ 337 link previews. 339 Deployment is notoriously challenging for any software application 340 where maintenance and updates can be particularly disastrous for 341 obsolete cryptographic libraries. 343 4. End-user expectations 345 While the formal definition and properties of an E2EE system relate 346 to communication security, they do not draw from a comprehensive 347 threat model or speak to what users expect from E2EE communication. 348 It is in this context that some E2EE designs and architectures may 349 ultimately run contrary to user expectations of E2EE systems. 350 [GEC-EU] Although some system designs do not directly violate "the 351 math" of encryption algorithms, they do so by implicating and 352 weakening other important aspects of an E2EE _system_. 354 4.1. A conversation is confidential 356 Users talking to one another in an E2EE system should be the only 357 ones that know what they are talking about [RFC7624]. People have 358 the right to privacy as defined in international human rights law and 359 within the right to free expression and to hold opinions is inferred 360 the right to whisper, whether or not they are using digital 361 communications or walking through a field. 363 4.2. Providers are trustworthy 365 While trustworth can be rigourously defined from an engineering 366 perspective, for the purposes of this document we choose a definition 367 of Trustworthy inspired by an internal workshop by Internet Society 368 staff: 370 Trustworthy A system is completely trustworthy if and only if it is 371 completely resilient, reliable, accountable, and secure in a way 372 that consistently meets users' expectations. The opposite of 373 trustworthy is untrustworthy. 375 This definition is complete in its positive and negative aspects: 376 what it is, eg "Worthy of confidence" and what it isn't, eg in RFC 377 7258: "behavior that subverts the intent of communicating 378 partieswithout the agreement of those parties." [RFC7258] 380 Therefore, a trustworthy end-to-end encrypted communication system is 381 the set of functions needed by two or more parties to communicate 382 among each other in a confidential and authenticated fashion without 383 any third party having access to the content of that communication 384 where the functions that offer the confidentiallity and authenticity 385 are trustworthy. 387 4.3. Access by a third-party is impossible 389 No matter the specifics, any methods used to access to the content of 390 the messages by a third party would violate a user's expectations of 391 E2EE messaging. "[T]hese access methods scan message contents on the 392 user's [device]", which are then "scanned for matches against a 393 database of prohibited content before, and sometimes after, the 394 message is sent to the recipient." [GEC-EU] 396 If a method makes private communication, intended to be sent over an 397 encrypted channel between end points, available to parties other than 398 the recipient, without formally interfering with channel 399 confidentiality, that method violates the understood expectation of 400 that security property. 402 4.4. Pattern inference is minimised 404 Analyses such as traffic fingerprinting or homomorphic encryption 405 computations should be considred outside the scope of an E2EE 406 system's goals of providing secure communications to end users. 408 Such methods of analyses, outside of or as part of E2EE system 409 design, allows third parties to draw inferences from communication 410 that was intended to be confidential. "By allowing private user data 411 to be scanned via direct access by servers and their providers," the 412 use of these methods should be considered an affront to "the privacy 413 expectations of users of end-to-end encrypted communication systems." 414 [GEC-EU] 416 Not only should an E2EE system value user privacy by not allowing 417 pattern inference, it should actively be attempting to solve issues 418 of metadata and traceability (enhanced metadata) through further 419 innovation that stays ahead of advances in these techniques. 421 4.5. The e2ee system is not compromised 423 RFC 3552 talks about the Internet Threat model such as the assumption 424 that the user can expect any communications systems, but perhaps 425 especially E2EE systems, not being compromised.[RFC3552] Compromises 426 to E2EE systems are often referred to as "backdoors" but are often 427 presented as additional design features like "key escrow." Users of 428 E2EE systems would not expect a front, back or side door entrance 429 into their confidential conversations and would expect a provider to 430 actively resist- technically and legally- compromise through these 431 means. 433 5. Conclusions 435 From messaging to video conferencing, there are many competing 436 features in an E2EE system that is secure and usable. The most well 437 designed system cannot meet the expectations of every user, nor does 438 an ideal system exist from any dimension. E2EE is a technology that 439 is constantly improving to achieve the ideal as defined in this 440 document. 442 Features and functionalities of e2ee systems should be developed and 443 improved in service of end user expectations for privacy preserving 444 communications. 446 6. Acknowledgements 448 Fred Baker, Stephen Farrell, Richard Barnes, Olaf Kolkman all 449 contributed to the early strategic thinking of this document and 450 whether it would be useful to the IETF community. 452 The folks at Riseup and the LEAP Encryption Access Project have 453 articulated brilliantly the hardest parts of end-to-end encryption 454 systems that serve the end users' right to whisper. 456 Ryan Polk at the Internet Society has energy to spare when it comes 457 to organising meaningful contributions, like this one, for the 458 technical advisors of the Global Encryption Coalition. 460 7. Security Considerations 462 As this draft concerns an informational document, there are no 463 security considerations. 465 8. IANA Considerations 467 This document has no actions for IANA. 469 9. Informative References 471 [dkg] Gillmor, D., "Human Rights Protocol Considerations 472 Glossary", 2015, 473 . 475 [GEC-EU] Global Encryption Coaliation, ., "Breaking encryption 476 myths: What the European Commission's leaked report got 477 wrong about online security", 2020, 478 . 481 [mls] IETF, ., "Messaging Layer Security", 2018, 482 . 484 [openpgp] IETF, ., "Open Specification for Pretty Good Privacy", 485 2020, 486 . 488 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 489 Text on Security Considerations", BCP 72, RFC 3552, 490 DOI 10.17487/RFC3552, July 2003, 491 . 493 [RFC3724] Kempf, J., Ed., Austein, R., Ed., and IAB, "The Rise of 494 the Middle and the Future of End-to-End: Reflections on 495 the Evolution of the Internet Architecture", RFC 3724, 496 DOI 10.17487/RFC3724, March 2004, 497 . 499 [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an 500 Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 501 2014, . 503 [RFC7624] Barnes, R., Schneier, B., Jennings, C., Hardie, T., 504 Trammell, B., Huitema, C., and D. Borkmann, 505 "Confidentiality in the Face of Pervasive Surveillance: A 506 Threat Model and Problem Statement", RFC 7624, 507 DOI 10.17487/RFC7624, August 2015, 508 . 510 [RFC8890] Nottingham, M., "The Internet is for End Users", RFC 8890, 511 DOI 10.17487/RFC8890, August 2020, 512 . 514 [saltzer] Saltzer, et al, J., "End-to-end arguments in system 515 design", 1984, 516 . 519 Authors' Addresses 521 Mallory Knodel 522 CDT 524 Email: mknodel@cdt.org 525 Fred Baker 527 Email: fredbaker.IETF@gmail.com 529 Olaf Kolkman 530 ISOC 532 Email: kolkman@isoc.org 534 Sofia Celi 535 Cloudflare 537 Email: cherenkov@riseup.net 539 Gurshabad Grover 540 Centre for Internet and Society 542 Email: gurshabad@cis-india.org