idnits 2.17.1 draft-irtf-hrpc-research-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 : ---------------------------------------------------------------------------- ** The abstract seems to contain references ([HRC2012], [RFC6973], [NETmundial], [Brown], [UNGA2013], [ICCPR], [UDHR], [ICESCR]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (September 19, 2016) is 2776 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 3303 -- Looks like a reference, but probably isn't: '2' on line 3305 -- Looks like a reference, but probably isn't: '3' on line 3307 == Unused Reference: 'RFC4033' is defined on line 2974, but no explicit reference was found in the text -- Obsolete informational reference (is this intentional?): RFC 226 (Obsoleted by RFC 247) -- Obsolete informational reference (is this intentional?): RFC 760 (Obsoleted by RFC 791) -- Obsolete informational reference (is this intentional?): RFC 793 (Obsoleted by RFC 9293) -- Obsolete informational reference (is this intentional?): RFC 1631 (Obsoleted by RFC 3022) -- Obsolete informational reference (is this intentional?): RFC 1766 (Obsoleted by RFC 3066, RFC 3282) -- Obsolete informational reference (is this intentional?): RFC 1866 (Obsoleted by RFC 2854) -- Obsolete informational reference (is this intentional?): RFC 2460 (Obsoleted by RFC 8200) -- Obsolete informational reference (is this intentional?): RFC 3536 (Obsoleted by RFC 6365) -- Obsolete informational reference (is this intentional?): RFC 3979 (Obsoleted by RFC 8179) -- Obsolete informational reference (is this intentional?): RFC 7540 (Obsoleted by RFC 9113) -- Obsolete informational reference (is this intentional?): RFC 7626 (Obsoleted by RFC 9076) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 15 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Human Rights Protocol Considerations Research Group N. ten Oever 3 Internet-Draft Article19 4 Intended status: Informational C. Cath 5 Expires: March 23, 2017 Oxford Internet Institute 6 September 19, 2016 8 Research into Human Rights Protocol Considerations 9 draft-irtf-hrpc-research-00 11 Abstract 13 The evergrowing interconnectedness of Internet and society increases 14 the impact of the Internet on the lives of individuals. Because of 15 this, the design and development of the architecture of the Internet 16 also has a growing impact on society. This has led to an broad 17 recognition that human rights [UDHR] [ICCPR] [ICESCR] have a role in 18 the development and management of the Internet [HRC2012] [UNGA2013] 19 [NETmundial]. It has also been argued that the Internet should be 20 strengthened as a human rights enabling environment [Brown]. 22 This document provides a proposal for a vocabulary to discuss the 23 relation between human rights and Internet protocols, an overview of 24 the discussion in technical and academic literature and communities, 25 a proposal for the mapping of the relation between human rights and 26 technical concepts, and a proposal for guidelines for human rights 27 considerations, similar to the work done on the guidelines for 28 privacy considerations [RFC6973]. 30 This document is not an Internet Standards Track specification; it is 31 published for informational purposes. 33 This document is a product of the Internet Research Task Force 34 (IRTF). The IRTF publishes the results of Internet-related research 35 and development activities. This documents aims to be a consensus 36 document of the Human Rights Protocol Consideration Research Group of 37 the Internet Research Task Force (IRTF). 39 Discussion of this draft at: hrpc@irtf.org // 40 https://www.irtf.org/mailman/listinfo/hrpc 42 Status of This Memo 44 This Internet-Draft is submitted in full conformance with the 45 provisions of BCP 78 and BCP 79. 47 Internet-Drafts are working documents of the Internet Engineering 48 Task Force (IETF). Note that other groups may also distribute 49 working documents as Internet-Drafts. The list of current Internet- 50 Drafts is at http://datatracker.ietf.org/drafts/current/. 52 Internet-Drafts are draft documents valid for a maximum of six months 53 and may be updated, replaced, or obsoleted by other documents at any 54 time. It is inappropriate to use Internet-Drafts as reference 55 material or to cite them other than as "work in progress." 57 This Internet-Draft will expire on March 23, 2017. 59 Copyright Notice 61 Copyright (c) 2016 IETF Trust and the persons identified as the 62 document authors. All rights reserved. 64 This document is subject to BCP 78 and the IETF Trust's Legal 65 Provisions Relating to IETF Documents 66 (http://trustee.ietf.org/license-info) in effect on the date of 67 publication of this document. Please review these documents 68 carefully, as they describe your rights and restrictions with respect 69 to this document. Code Components extracted from this document must 70 include Simplified BSD License text as described in Section 4.e of 71 the Trust Legal Provisions and are provided without warranty as 72 described in the Simplified BSD License. 74 Table of Contents 76 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 77 2. Vocabulary used . . . . . . . . . . . . . . . . . . . . . . . 5 78 3. Research Questions . . . . . . . . . . . . . . . . . . . . . 11 79 4. Literature and Discussion Review . . . . . . . . . . . . . . 11 80 5. Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 14 81 5.1. Data Sources . . . . . . . . . . . . . . . . . . . . . . 15 82 5.1.1. Discourse analysis of RFCs . . . . . . . . . . . . . 15 83 5.1.2. Interviews with members of the IETF community . . . . 15 84 5.1.3. Participant observation in Working Groups . . . . . . 16 85 5.2. Data analysis strategies . . . . . . . . . . . . . . . . 16 86 5.2.1. Identifying qualities of technical concepts that 87 relate to human rights . . . . . . . . . . . . . . . 16 88 5.2.2. Translating human rights to technical terms . . . . . 18 89 5.2.3. IPv4 . . . . . . . . . . . . . . . . . . . . . . . . 20 90 5.2.4. DNS . . . . . . . . . . . . . . . . . . . . . . . . . 22 91 5.2.5. HTTP . . . . . . . . . . . . . . . . . . . . . . . . 25 92 5.2.6. XMPP . . . . . . . . . . . . . . . . . . . . . . . . 29 93 5.2.7. Peer to Peer . . . . . . . . . . . . . . . . . . . . 31 94 5.2.8. Virtual Private Network . . . . . . . . . . . . . . . 33 95 5.2.9. HTTP Status Code 451 . . . . . . . . . . . . . . . . 35 96 5.2.10. Middleboxes . . . . . . . . . . . . . . . . . . . . . 37 97 5.2.11. DDOS attacks . . . . . . . . . . . . . . . . . . . . 38 98 5.3. Model for developing human rights protocol considerations 40 99 5.3.1. Human rights threats . . . . . . . . . . . . . . . . 40 100 5.3.2. Guidelines for human rights considerations . . . . . 41 101 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 55 102 7. Security Considerations . . . . . . . . . . . . . . . . . . . 56 103 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 56 104 9. Research Group Information . . . . . . . . . . . . . . . . . 56 105 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 56 106 10.1. Informative References . . . . . . . . . . . . . . . . . 56 107 10.2. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 71 109 1. Introduction 111 "There's a freedom about the Internet: As long as we accept the 112 rules of sending packets around, we can send packets containing 113 anything to anywhere." 115 [Berners-Lee] 117 "The Internet isn't value-neutral, and neither is the IETF." 119 [RFC3935] 121 This document aims to expose the relation between protocols and human 122 rights, propose possible guidelines to protect the Internet as a 123 human-rights-enabling environment in future protocol development, in 124 a manner similar to the work done for Privacy Considerations in 125 [RFC6973], and to increase the awareness in both the human rights 126 community and the technical community on the importance of the 127 technical workings of the Internet and its impact on human rights. 129 Open, secure and reliable connectivity is necessary (although not 130 sufficient) to excercise the human rights such as freedom of 131 expression and freedom of association, as defined in the Universal 132 Declaration of Human Rights [UDHR]. The Internet aims to be a global 133 network of networks that provides unfettered connectivity to all 134 users at all times and for any content [RFC1958]. This objective of 135 stimulating global connectivity contributes to the Internet's role as 136 an enabler of human rights. Next to that, the strong commitment to 137 security [RFC1984] [RFC3365] and privacy [RFC6973] [RFC7258] in the 138 Internet's architectural design contribute to the strengthening of 139 the Internet as a human rights enabling environment. One could even 140 argue that the Internet is not only an enabler of human rights, but 141 that human rights lie at the basis of, and are ingrained in, the 142 architecture of the network. Internet connectivity increases the 143 capacity for individuals to exercise their rights, the core of the 144 Internet, its architectural design is therefore closely intertwined 145 with the human rights framework [CathFloridi]. The quintessential 146 link between the Internet's architecture and human rights has been 147 argued by many. [Bless] for instance argues that, 'to a certain 148 extent, the Internet and its protocols have already facilitated the 149 realization of human rights, e.g., the freedom of assembly and 150 expression. In contrast, measures of censorship and pervasive 151 surveillance violate fundamental human rights.' [Denardis15] argues 152 that 'Since the first hints of Internet commercialization and 153 internationalization, the IETF has supported strong security in 154 protocol design and has sometimes served as a force resisting 155 protocol-enabled surveillance features.' By doing so, the IETF 156 enabled the manifestation of the right to privacy, through the 157 Internet's architecture. Additionally, access to information gives 158 people access to knowledge that enables them to help satisfy other 159 human rights, as such the Internet increasingly becoming a pre- 160 condition for human rights rather than a supplement. 162 Openness of communications of the technical design fostered freedom 163 of communication as a core value, however as the scale and the 164 commercialization of the Internet grew, topics like access, rights 165 and connectivity are forced to compete with other values. Therefore, 166 important human rights enabling characteristics of the Internet might 167 be degraded if they're not properly defined, described and protected 168 as such. And, the other way around, not protecting human right 169 enabling characteristics could also result in (partial) loss of 170 functionality and connectivity, and other inherent parts of the 171 Internet's architecture. New protocols, particularly those that 172 upgrade the core infrastructure of the Net, should be designed to 173 continue to enable fundamental human rights. 175 The IETF has produced guidelines and procedures to ensure and 176 galvanize the privacy and security of the network in protocol 177 development. This document aims to explore the possibility of the 178 development of similar procedures for guidelines for human rights 179 considerations to ensure that protocols developed in the IETF do not 180 have an adverse impact on the realization of human rights on the 181 Internet. By carefully considering the answers to the questions 182 posed in the final part of this document, document authors should be 183 able to produce a comprehensive analysis that can serve as the basis 184 for discussion on whether the protocol adequately protects against 185 human rights threats. 187 This document has been developed within the framework of the Human 188 Rights Protocols Considerations Research Group, based on discussions 189 on the hrpc mailinglist and during hrpc sessions, where this document 190 also has been extensively discussed. The draft in its current form 191 and iteration has received five in-depth reviews on list, and 192 received many comments from inside and outside the IRTF and IETF 193 community. The authors believe that the issues that have been raised 194 by the reviewers have been addressed. 196 2. Vocabulary used 198 In the discussion of human rights and Internet architecture concepts 199 developed in computer science, networking, law, policy-making and 200 advocacy are coming together [Dutton],[Kaye], [Franklin]. The same 201 concepts might have a very different meaning and implications in 202 other areas of expertise. In order to foster a constructive 203 interdisciplinary debate, and minimize differences in interpretation, 204 the following glossary is provided, building as much as possible on 205 existing definitions, and where these were not available definitions 206 have been developed. 208 Accessibility Full Internet Connectivity as described in [RFC4084] 209 to provide unfettered access to the Internet 211 The design of protocols, services or implementation that provide 212 an enabling environment for people with disabilities. 214 The ability to receive information available on the Internet 216 Anonymity The condition of an identity being unknown or concealed. 217 [RFC4949] 219 Anonymous A state of an individual in which an observer or attacker 220 cannot identify the individual within a set of other individuals 221 (the anonymity set). [RFC6973] 223 Authenticity The fact that the data does indeed come from the source 224 it claims to come from. (It is strongly linked with Integrity, 225 see below). 227 Censorship resistance Methods and measures to prevent Internet 228 censorship. 230 Confidentiality The non-disclosure of information to any unintended 231 person or host or party. 233 Connectivity The extent to which a device or network is able to 234 reach other devices or networks to exchange data. The Internet is 235 the tool for providing global connectivity [RFC1958]. 237 Content-agnosticism Treating network traffic identically regardless 238 of content. 240 Debugging Debugging is a methodical process of finding and reducing 241 the number of bugs, or defects, or malfunctions in a protocol or 242 its implementation, thus making it behave as expected. It also 243 includes analyzing the consequences that might have emanate from 244 the error. Debugging tends to be harder when various subsystems 245 are tightly coupled, as changes in one may cause bugs to emerge in 246 another. [WP-Debugging] 248 The process through which people troubleshoot a technical issue, 249 which may include inspection of program source code or device 250 configurations. Can also include tracing or monitoring packet 251 flow. 253 Decentralized Opportunity for implementation or deployment of 254 standards, protocols or systems without one single point of 255 control. 257 End-to-End The principal of extending characteristics of a protocol 258 or system as far as possible within the system. technical this 259 means that intermediaries should not modify messages but simply 260 route them to their desired end-points as capabilities should be 261 given by the end-points, that the network then interconnects 262 rather than controls. For example, end-to-end instant message 263 encryption would conceal communications from one user's instant 264 messaging application through any intermediate devices and servers 265 all the way to the recipient's instant messaging application. If 266 the message was decrypted at any intermediate point-for example at 267 a service provider-then the property of end-to-end encryption 268 would not be present. 270 One of the key architectural guidelines of the Internet is the 271 end-to-end principle in the papers by Saltzer, Reed, and Clark 272 [Saltzer] [Clark]. The end-to-end principle was originally 273 articulated as a question of where best not to put functions in a 274 communication system. Yet, in the ensuing years, it has evolved 275 to address concerns of maintaining openness, increasing 276 reliability and robustness, and preserving the properties of user 277 choice and ease of new service development as discussed by 278 Blumenthal and Clark in [Blumenthal]; concerns that were not part 279 of the original articulation of the end-to-end principle. 280 [RFC3724] 282 Federation The possibility of connecting autonomous and possibly 283 centralized systems into single system without a central 284 authority. 286 Heterogenity The Internet is characterized by heterogeneity on many 287 levels: devices and nodes, router scheduling algorithms and queue 288 management mechanisms, routing protocols, levels of multiplexing, 289 protocol versions and implementations, underlying link layers 290 (e.g., point-to-point, multi-access links, wireless, FDDI, etc.), 291 in the traffic mix and in the levels of congestion at different 292 times and places. Moreover, as the Internet is composed of 293 autonomous organizations and Internet service providers, each with 294 their own separate policy concerns,there is a large heterogeneity 295 of administrative domains and pricing structures. As a result, 296 the heterogeneity principle proposed in [RFC1958] needs to be 297 supported by design. [FIArch] 299 Integrity Maintenance and assurance of the accuracy and consistency 300 of data to ensure it has not been (intentionally or 301 unintentionally) altered. 303 Internet censorship Internet censorship is the intentional 304 suppression of information originating, flowing or stored on 305 systems connected to the Internet where that information is 306 relevant for decision making to some entity. [Elahi] 308 Inter-operable A property of a documented standard or protocol which 309 allows different independent implementations to work with each 310 other without any restricted access or functionality. 312 Internet Standards as an Arena for Conflict Pursuant to the 313 principle of constant change, since the function and scope of the 314 Internet evolves, so does the role of the IETF in developing 315 standards. Internet standards are adopted on the basis of a 316 series of criteria, including high technical quality, support by 317 community consensus, and their overall benefit to the Internet. 318 The latter calls for an assessment of the interests of all 319 affected parties and the specifications' impact on the Internet's 320 users. In this respect, the effective exercise of the human 321 rights of the Internet users is a relevant consideration that 322 needs to be appreciated in the standardization process insofar as 323 it is directly linked to the reliability and core values of the 324 Internet. [RFC1958] [RFC0226] [RFC3724] 326 Internationalization (i18n) The practice of making protocols, 327 standards, and implementations usable in different languages and 328 scripts. (see Localization) 330 (cf [RFC6365]) In the IETF, "internationalization" means to add or 331 improve the handling of non-ASCII text in a protocol. [RFC6365] 332 A different perspective, more appropriate to protocols that are 333 designed for global use from the beginning, is the definition used 334 by W3C: 336 "Internationalization is the design and development of a product, 337 application or document content that enables easy localization for 338 target audiences that vary in culture, region, or language." 339 [W3Ci18nDef] 341 Many protocols that handle text only handle one charset (US- 342 ASCII), or leave the question of encoding up to local guesswork 343 (which leads, of course, to interoperability problems) [RFC3536]. 344 If multiple charsets are permitted, they must be explicitly 345 identified [RFC2277]. Adding non-ASCII text to a protocol allows 346 the protocol to handle more scripts, hopefully all of the ones 347 useful in the world. In today's world, that is normally best 348 accomplished by allowing Unicode encoded in UTF-8 only, thereby 349 shifting conversion issues away from individual choices. 351 Localization (l10n) The practice of translating an implementation to 352 make it functional in a specific language or for users in a 353 specific locale (see Internationalization). 355 (cf [RFC6365]): The process of adapting an internationalized 356 application platform or application to a specific cultural 357 environment. In localization, the same semantics are preserved 358 while the syntax may be changed. [FRAMEWORK] 360 Localization is the act of tailoring an application for a 361 different language or script or culture. Some internationalized 362 applications can handle a wide variety of languages. Typical 363 users only understand a small number of languages, so the program 364 must be tailored to interact with users in just the languages they 365 know. The major work of localization is translating the user 366 interface and documentation. Localization involves not only 367 changing the language interaction, but also other relevant changes 368 such as display of numbers, dates, currency, and so on. The 369 better internationalized an application is, the easier it is to 370 localize it for a particular language and character encoding 371 scheme. 373 Open standards Conform [RFC2606]: Various national and international 374 standards bodies, such as ANSI, ISO, IEEE, and ITU-T, develop a 375 variety of protocol and service specifications that are similar to 376 Technical Specifications defined here. National and international 377 groups also publish "implementors' agreements" that are analogous 378 to Applicability Statements, capturing a body of implementation- 379 specific detail concerned with the practical application of their 380 standards. All of these are considered to be "open external 381 standards" for the purposes of the Internet Standards Process. 383 Openness The quality of the unfiltered Internet that allows for free 384 access to other hosts. 386 Absence of centralized points of control - a feature that is 387 assumed to make it easy for new users to join and new uses to 388 unfold [Brown]. 390 Permissionless innovation The freedom and ability to freely create 391 and deploy new protocols on top of the communications constructs 392 that currently exist. 394 Privacy The right of an entity (normally a person), acting in its 395 own behalf, to determine the degree to which it will interact with 396 its environment, including the degree to which the entity is 397 willing to share its personal information with others. [RFC4949] 399 The right of individuals to control or influence what information 400 related to them may be collected and stored and by whom and to 401 whom that information may be disclosed. 403 Privacy is a broad concept relating to the protection of 404 individual or group autonomy and the relationship between an 405 individual or group and society, including government, companies 406 and private individuals. It is often summarized as "the right to 407 be left alone" but it encompasses a wide range of rights including 408 protections from intrusions into family and home life, control of 409 sexual and reproductive rights, and communications secrecy. It is 410 commonly recognized as a core right that underpins human dignity 411 and other values such as freedom of association and freedom of 412 speech. 414 The right to privacy is also recognized in nearly every national 415 constitution and in most international human rights treaties. It 416 has been adjudicated upon both by international and regional 417 bodies. The right to privacy is also legally protected at the 418 national level through provisions in civil and/or criminal codes. 420 Reliable Reliability ensures that a protocol will execute its 421 function consistently and error resistant as described and 422 function without unexpected result. A system that is reliable 423 degenerates gracefully and will have a documented way to announce 424 degradation. It also has mechanisms to recover from failure 425 gracefully, and if applicable, allow for partial healing. 427 Resilience The maintaining of dependability and performance in the 428 face of unanticipated changes and circumstances. 430 Robustness The resistance of protocols and their implementations to 431 errors, and to involuntary, legal or malicious attempts to disrupt 432 its mode of operations. [RFC0760] [RFC0791] [RFC0793] [RFC1122]. 433 Or framed more positively, a system can provide functionality 434 consistently and without errors despite involuntary, legal or 435 malicious attempts to disrupt its mode of operations. 437 Scalable The ability to handle increased or decreased workloads 438 predictably within defined expectations. There should be a clear 439 definition of its scope and applicability. The limits of a 440 systems scalability should be defined. 442 Stateless / stateful In computing, a stateless protocol is a 443 communications protocol that treats each request as an independent 444 transaction that is unrelated to any previous request so that the 445 communication consists of independent pairs of request and 446 response. A stateless protocol does not require the server to 447 retain session information or status about each communications 448 partner for the duration of multiple requests. In contrast, a 449 protocol which requires keeping of the internal state on the 450 server is known as a stateful protocol. [WP-Stateless] 452 Strong encryption / cryptography Used to describe a cryptographic 453 algorithm that would require a large amount of computational power 454 to defeat it. [RFC4949] 456 Transparent "transparency" refers to the original Internet concept 457 of a single universal logical addressing scheme, and the 458 mechanisms by which packets may flow from source to destination 459 essentially unaltered. [RFC2775] 461 The combination of reliability, confidentiality, integrity, 462 anonymity, and authenticity is what makes up security on the 463 Internet. 465 ( Reliability ) 466 ( Confidentiality ) 467 ( Integrity ) = communication and information security 468 ( Authenticity ) 469 ( Anonymity ) 471 The combination of the end-to-end principle, interoperability, 472 resilience, reliability and robustness are the enableing factors that 473 result in on the Internet. 475 ( End-to-End ) 476 ( Interoperability ) 477 ( Resilience ) 478 ( Reliability ) = connectivity 479 ( Robustness ) 480 ( Autonomy ) 481 ( Simplicity ) 483 3. Research Questions 485 The Human Rights Protocol Considerations Research Group (hrpc) in the 486 Internet Research Taskforce (IRTF) embarked on its mission to answer 487 the following two questions which are also the main two questions 488 which this documents seeks to answer: 490 1. How can Internet protocols and standards impact human rights, 491 either by enabling them or by creating a restrictive environment? 493 2. Can guidelines be developed to improve informed and transparent 494 decision making about potential human rights impact of protocols? 496 4. Literature and Discussion Review 498 Protocols and standards are regularly seen as merely performing 499 technical functions. However, these protocols and standards do not 500 exist outside of their technical context nor outside of their 501 political, historical, economic, legal or cultural context. This is 502 best exemplified by the way in which protocols have become part and 503 parcel of political processes and public policies: one only has to 504 look at the IANA transition, the RFC on pervasive monitoring or 505 global innovation policy for concrete examples [Denardis15]. To 506 quote [Abbate]: "protocols are politics by other means". Since the 507 late 1990's a burgeoning group of academics and practitioners 508 researched questions surrounding the societal impact of protocols. 509 These studies vary in focus and scope: some focus on specific 510 standards [Davidsonetal] [Musiani], others look into the political, 511 legal, commercial or social impact of protocols [BrownMarsden] 512 [Lessig], [Mueller] and yet others look at how the engineers' 513 personal set of values get translated into technology 514 [Abbate],[CathFloridi] [Denardis15] [WynsbergheMoura]. 516 Commercial and political influences on the management of the 517 Internet's architecture are well-documented in the academic 518 literature and will thus not be discussed here [Benkler] [Brownetal] 519 [Denardis15] [Lessig] [Mueller] [Zittrain]. It is sufficient to 520 say that the IETF community consistently tries to push back against 521 the standardization of surveillance and certain other issues that 522 negatively influence end-users' experience of and trust in the 523 Internet [Denardis14]. The role human rights play in engineering, 524 architecture and protocol design is much less clear. 526 It is very important to understand how protocols and standards impact 527 human rights. In particular because Standard Developing 528 Organizations (SDOs) are increasingly becoming venues where social 529 values (like human rights) are discussed, although often from a 530 technological point of view. These SDOs are becoming a new focal 531 point for discussions about values-by-design, and the role of 532 technical engineers in protecting or enabling human rights 533 [Brownetal] [Clarketal] [Denardis14] [CathFloridi] [Lessig] 534 [Rachovitsa]. 536 In the academic literature five clear positions can be discerned, in 537 relation to the role of human rights in protocol design and how to 538 account for these human rights in protocol development: Clark et al. 539 argue that there is a need to 'design for variation in outcome, so 540 that the outcome can be different in different places, and the tussle 541 takes place within the design (...) [as] Rigid designs will be 542 broken; designs that permit variation will flex under pressure and 543 survive [Clarketal].' They hold that human rights should not be 544 hard-coded into protocols because of four reasons: first, the rights 545 in the UDHR are not absolute. Second, technology is not the only 546 tool in the tussle over human rights. Third, there are inherent 547 dangers to blunting the tools of enforcement and last but not least, 548 it is dangerous to make promises that can't be kept. The open nature 549 of the Internet will never, they argue, be enough to fully protect 550 individuals' human rights. 552 Conversely, Brown et al. [Brownetal] state that 'some key, universal 553 values - of which the UDHR is the most legitimate expression - should 554 be baked into the architecture at design time.' They argue that 555 design choices have offline consequences, and are able shape the 556 power positions of groups or individuals in society. As such, the 557 individuals making these technical decisions have a moral obligation 558 to take into account the impact of their decisions on society, and by 559 extension human rights. Brown et al recognise that values and the 560 implementation of human rights vary across the globe. Yet they argue 561 that all members of the United Nations have found 'common agreement 562 on the values proclaimed in the Universal Declaration of Human 563 Rights. In looking for the most legitimate set of global values to 564 embed in the future Internet architecture, the UDHR has the 565 democratic assent of a significant fraction of the planet's 566 population, through their elected representatives." 568 The main disagreement between these two academic positions lies 569 mostly in the question on whether a particular value system should be 570 embedded into the Internet's architecture or whether the architecture 571 needs to account for a varying set of values. 573 A third position that is similar to that of Brown et al., is taken by 574 [Broeders] who argues that 'we must find ways to continue 575 guaranteeing the overall integrity and functionality of the public 576 core of the Internet.' He argues that the best way to do this is by 577 declaring the backbone of the Internet - which includes the TCP/IP 578 protocol suite, numerous standards, the Domain Name System (DNS), and 579 routing protocols - a common public good. This is a different 580 approach than that of [Clarketal] and [Brownetal] because Broeders 581 does not suggest that social values should (or should not) be 582 explicitly coded into the Internet's architecture, but rather that 583 the existing architecture should be seen as an entity of public 584 value. 586 Bless and Orwat [Bless] represent a fourth position. They argue that 587 it is too early to make any definitive claims, but that there is a 588 need for more careful analysis of the impact of protocol design 589 choices on human rights. They also argue that it is important to 590 search for solutions that 'create awareness in the technical 591 community about impact of design choices on social values. And work 592 towards a methodology for co-design of technical and institutional 593 systems.' 595 Berners-Lee and Halpin argue that the Internet could lead to even new 596 capacities, and these capacities may over time be viewed as new kinds 597 of rights. For example, Internet access may be viewed as a human 598 right in of itself if it is taken to be a pre-condition for other 599 rights, even if it could not have been predicted at the declaration 600 of the UNHDR after the end of World War 2.[BernersLeeHalpin]. This 601 last position is interesting to keep in mind, but beyond the remit of 602 this document. 604 It is important to give some background to the academic discussion on 605 this issue. As it stems from the issues as they arise in the field 606 of technical engineering. They also are important to document as 607 they inform the position of the authors of this document. Our 608 position is that hard-coding human rights into protocols is very 609 complicated as each situation is dependent on its context. At this 610 point is difficult to say whether hard-coding human rights into 611 protocols is wise (or feasible). It is however important to make 612 conscious and explicit design decisions that take into account the 613 human rights protocol considerations guidelines developed below. 614 This will ensure that the impact protocols can have on human rights 615 is clear and explicit, both for developers and for users. In 616 addition, it ensures that the impact of specific protocol on human 617 rights is carefully considered and that concrete design decisions are 618 documented in the protocol. 620 This document details the steps taken in the research into human 621 rights protocol considerations by the hrpc research group to clarify 622 the relation between technical concepts used in the IETF and human 623 rights. This document sets out some preliminary steps and 624 considerations for engineers to take into account when developing 625 standards and protocols. 627 5. Methodology 629 Mapping the relation between human rights, protocols and 630 architectures is a new research challenge, which requires a good 631 amount of interdisciplinary and cross organizational cooperation to 632 develop a consistent methodology. 634 The methodological choices made in this document are based on the 635 political science-based method of discourse analysis and ethnographic 636 research methods [Cath]. This work departs from the assumption that 637 language reflects the understanding of concepts. Or as [Jabri] 638 holds, policy documents are 'social relations represented in texts 639 where language is used to construct meaning and representation'. 640 This process happens in 'the social space of society' [Schroeder] and 641 manifests itself in institutions and organizations [King], exposed 642 using the ethnographic methods of semi-structured interviews and 643 participant observation. Or in non-academic language, the way the 644 language in IETF/IRTF documents describes and approaches the issues 645 they are trying to address is an indicator for the underlying social 646 assumptions and relations of the engineers to their engineering. By 647 reading and analyzing these documents, as well as interviewing 648 engineers and participating in the IETF/IRTF working groups, it is 649 possible to distill the relation between human rights, protocols and 650 the Internet's architecture. 652 The discourse analysis was operationalized using qualitative and 653 quantitative means. The first step taken by the research group was 654 reading RFCs and other official IETF documents. The second step was 655 the use of a python-based analyzer, using the tool Big Bang, adapted 656 by Nick Doty [Doty] to scan for the concepts that were identified as 657 important architectural principles (distilled on the initial reading 658 and supplemented by the interviews and participant observation). 659 Such a quantitative method is very precise and speeds up the research 660 process [Richie]. But this tool is unable to understand 'latent 661 meaning' [Denzin]. In order to mitigate these issues of automated 662 word-frequency based approaches, and to get a sense of the 'thick 663 meaning' [Geertz] of the data, a second qualitative analysis of the 664 data set was performed. These various rounds of discourse analysis 665 were used to inform the interviews and further data analysis. As 666 such the initial rounds of quantitative discourse analysis were used 667 to inform the second rounds of qualitative analysis.The results from 668 the qualitative interviews were again used to feed new concepts into 669 the quantitative discourse analysis. As such the two methods 670 continued to support and enrich each other. 672 The ethnographic methods of the data collection and processing 673 allowed the research group to acquire the data necessary to 'provide 674 a holistic understanding of research participants' views and actions' 675 [Denzin] that highlighted ongoing issues and case studies where 676 protocols impact human rights. The interview participants were 677 selected through purposive sampling [Babbie], as the research group 678 was interested in getting a wide variety of opinions on the role of 679 human rights in guiding protocol development. This sampling method 680 also ensured that individuals with extensive experience working at 681 the IETF in various roles were targeted. The interviewees included 682 individuals in leadership positions (Working Group (WG) chairs, Area 683 Directors (ADs)), 'regular participants', individuals working for 684 specific entities (corporate, civil society, political, academic) and 685 represented various backgrounds, nationalities and genders. 687 5.1. Data Sources 689 In order to map the potential relation between human rights and 690 protocols, the HRPC research group gathered data from three specific 691 sources: 693 5.1.1. Discourse analysis of RFCs 695 To start addressing the issue, a mapping exercise analyzing Internet 696 architecture and protocols features, vis-a-vis their possible impact 697 on human rights was undertaken. Therefore, research on the language 698 used in current and historic RFCs and mailing list discussions was 699 undertaken to expose core architectural principles, language and 700 deliberations on human rights of those affected by the network. 702 5.1.2. Interviews with members of the IETF community 704 Interviews with the current and past members of the Internet 705 Architecture Board (IAB), current and past members of the Internet 706 Engineering Steering Group (IESG) and chairs of selected working 707 groups and RFC authors was done at the IETF92 Dallas meeting in March 708 2015. To get an insider understanding of how they view the 709 relationship (if any) between human rights and protocols to play out 710 in their work. 712 5.1.3. Participant observation in Working Groups 714 By participating in various working groups, in person at IETF 715 meetings and on mailinglists, information was gathered about the 716 IETFs day-to-day workings. From which general themes, technical 717 concepts, and use-cases about human rights and protocols were 718 extracted. 720 5.2. Data analysis strategies 722 The data above was processed using three consecutive strategies: 723 mapping protocols related to human rights, extracting concepts from 724 these protocols, and creation of a common glossary (detailed under 725 "2.vocabulary used"). Before going over these strategies some 726 elaboration on the process of identifying technical concepts as they 727 relate to human rights needs to be given: 729 5.2.1. Identifying qualities of technical concepts that relate to human 730 rights 732 5.2.1.1. Mapping protocols and standards related to human rights 734 By combining data from the three data sources named above, an 735 extensive list of protocols and standards that potentially enable the 736 Internet as a tool for freedom of expression and association was 737 assembly. In order to determine the enabling (or inhibiting) 738 features we relied on direct references of such impact in the RFCs, 739 as well as input from the community. On the basis of this analysis a 740 list of RFCs that describe standards and protocols that are 741 potentially closely related to human rights was compiled. 743 5.2.1.2. Extracting concepts from mapped RFCs 745 Mapping the protocols and standards that are related to human rights 746 and create a human rights enabeling environment was the first step. 747 For that we needed to focus on specific technical concepts that 748 underlie these protocols and standards. On the basis of this list a 749 number of technical concepts that appeared frequently was extracted, 750 and used to create a second list of technical terms that, when 751 combined, create an enabling environment for excercising human rights 752 on the Internet. 754 5.2.1.3. Building a common vocabulary of technical concepts that impact 755 human rights 757 While interviewing experts, mapping RFCs and compiling technical 758 definitions several concepts of convergence and divergence were 759 identified. To ensure that the discussion was based on a common 760 understanding of terms and vocabulary, a list of definitions was 761 created. The definitions are based on the wording found in various 762 IETF documents, and if these were unavailable definitions were taken 763 from definitions from other Standards Developing Organizations or 764 academic literature, as indicated in the vocabulary section. 766 5.2.1.4. Translating Human Rights Concepts into Technical Definitions 768 The previous steps allowed for the clarification of relation between 769 human rights and technical concepts. The steps taken show how the 770 research process zoomed in, from compiling a broad lists of protocols 771 and standards that relate to human rights to extracting the precise 772 technical concepts that make up these protocols and standards, in 773 order to understand the relationship between the two. This sub- 774 section presents the next step: translating human rights to technical 775 concepts by matching the individuals components of the rights to the 776 accompanying technical concepts, allowing for the creation of a list 777 of technical concepts that when combined create an enabling 778 environment for human rights. 780 5.2.1.5. List technical terms that combined create enabling environment 781 for human rights 783 On the basis of the prior steps the following list of technical 784 terms, that when combined create an enabling environment for human 785 rights, such a freedom of expression and freedom of association, was 786 drafted. 788 Architectural principles Enabling features 789 and characteristics for user rights 791 /------------------------------------------------\ 792 | | 793 +=================|=============================+ | 794 = | = | 795 = | End to end = | 796 = | Reliability = | 797 = | Resilience = Access as | 798 = | Interoperability = Human Right | 799 = Good enough | Transparency = | 800 = principle | Data minimization = | 801 = | Permissionless innovation = | 802 = Simplicity | Graceful degradation = | 803 = | Connectivity = | 804 = | Heterogeneity = | 805 = | = | 806 = | = | 807 = \------------------------------------------------/ 808 = = 809 +===============================================+ 811 5.2.2. Translating human rights to technical terms 813 The combination of the technical concepts that have been gathered the 814 steps above have been grouped according to their impact on specific 815 rights as they have been mentioned in the interviews done at IETF92 816 as well as study of literature (see literature and discussion review 817 above). 819 This analysis aims to assist protocol developers by better 820 understanding the roles specific technical concepts with regards to 821 the possibility to exercise human rights on the Internet. 823 This analysis does not claim to be an complete or exhaustive mapping 824 of all possible ways in which a protocols could potentially impact 825 human rights, but it presents an initial concept mapping based on 826 interviews and literature and discussion review. 828 ( Connectivity ) 829 ( Privacy ) 830 ( Security ) = Right to freedom of expression 831 ( Content agnosticism ) 832 ( Internationalization ) 833 ( Censorship resistance ) 834 ( Open Standards ) 835 ( Heterogeneity support ) 837 ( Anonymity ) 838 ( Privacy ) = Right to non-discrimination 839 ( Pseudonymity ) 840 ( Content agnosticism ) 841 ( Accessibility ) 843 ( Content Agnosticism ) 844 ( Security ) = Right to equal protection 846 ( Accessibility ) 847 ( Internationalization ) = Right to political participation 848 ( Censorship resistance ) 849 ( Accessibility ) 851 ( Open standards ) 852 ( Localization ) = Right to participate in cultural life, 853 ( Internationalization ) arts and science & 854 ( Censorship resistance ) Right to education 855 ( Accessibility ) 857 ( Connectivity ) 858 ( Decentralization ) 859 ( Censorship resistance ) = Right to freedom of assembly 860 ( Pseudonymity ) and association 861 ( Anonymity ) 862 ( Security ) 864 ( Reliability ) 865 ( Confidentiality ) 866 ( Integrity ) = Right to security 867 ( Authenticity ) 868 ( Anonymity ) 870 5.2.2.1. Map cases of protocols that adversely impact human rights or 871 are enablers thereof 873 Given the information above, the following list of cases of protocols 874 that adversely impact or enable human rights was formed. 876 It is important to note that the assessment here is not a general 877 judgment on these protocols. When they were conceived, there were 878 many criteria to take into account. For instance, relying on an 879 external server can be bad for freedom of speech (it creates one more 880 control point, where censorship could be applied) but it may be a 881 necessity if the endpoints are not connected and reachable 882 permanently. So, when we say "protocol X has feature Y, which may 883 endanger the freedom of speech", it does not mean that protocol X is 884 bad and even less that its authors were evil. The goal here is to 885 show, with actual examples, that the design of protocols have 886 practical consequences for some human rights and these consequences 887 have to be considered at design time. 889 5.2.3. IPv4 891 The Internet Protocol version 4 (IPv4), also known as 'layer 3' of 892 the Internet, and specified as a common encapsulation and protocol 893 header, is defined in [RFC0791]. The evolution of Internet 894 communications led to continued development in this area, 895 encapsulated in the development of version 6 (IPv6) of the protocol 896 in [RFC2460]. In spite of this updated protocol, we find that 25 897 years after the specification of version 6 of the protocol, the older 898 v4 standard continues to account for a sizeable majority of Internet 899 traffic, and most of the issues discussed here (with the big 900 exception of NAT, see Address Translation) are valid for IPv4 as well 901 as IPv6. 903 The Internet was designed as a platform for free and open 904 communication, most notably encoded in the end-to-end principle, and 905 that philosophy is also present in the technical implementation of 906 the Internet Protocol. [RFC3724] While the protocol was designed to 907 exist in an environment where intelligence is at the end hosts, it 908 has proven to provide sufficient information that a more intelligent 909 network core can make policy decisions and enforce policy shaping and 910 restricting the communications of end hosts. These capabilities for 911 network control and limitations of the freedom of expression by end 912 hosts can be traced back to the IPv4 design, helping us understand 913 which technical protocol decisions have led to harm of this human 914 rights. A feature that can harm freedom of expression as well as the 915 right to privacy through misuse of the Internet Protocol is the 916 exploitation of the public visibility of the host pairs for all 917 communications, and the corresponding ability to discriminate and 918 block traffic as a result of that metadata. 920 5.2.3.1. Network visibility of Source and Destination 922 The IPv4 protocol header contains fixed location fields for both the 923 source and destination IP addresses [RFC0791]. These addresses 924 identify both the host sending and receiving each message, and allow 925 the core network to understand who is talking to whom, and to 926 practically limit communication selectively between pairs of hosts. 927 Blocking of communication based on the pair of source and destination 928 is one of the most common limitations on the ability for hosts to 929 communicate today, [caida] and can be seen as a restriction of the 930 ability for those hosts to assemble or to consensually express 931 themselves. 933 Inclusion of an Internet-wide identified source in the IP header is 934 not the only possible design, especially since the protocol is most 935 commonly implemented over Ethernet networks exposing only link-local 936 identifiers. [RFC0894] A variety of alternative designs including 937 source routing, which would allow for the sender to choose a per 938 defined (safe) route, and spoofing of the source IP address are 939 technically supported by the protocol, but neither are considered 940 good practice on the Internet. While projects like [torproject] 941 provide an alternative implementation of anonymity in connections, 942 they have been developed in spite of the IPv4 protocol design. 944 5.2.3.2. Protocols 946 The other major feature of the IP protocol header is that it 947 specifies the protocol encapsulated in each message in an easily 948 observable form, and does not encourage a design where the 949 encapsulated protocol is not available to a network observer. This 950 design has resulted in a proliferation of routers which inspect the 951 inner protocol, and also led to a stagnation where only the TCP and 952 UDP protocols are widely supported across the Internet. While the IP 953 protocol was designed as the entire set of metadata needed for 954 routing, subsequent enhanced routers have found value on making 955 policy decisions based on the contents of TCP and UDP headers as 956 well, and are encoded with the assumption that only these protocols 957 will be used for data transfer. [spdy] [RFC4303] defines an encrypted 958 encapsulation of additional protocols, but lacks widespread 959 deployment and faces the same challenge as any other protocol of 960 providing sufficient metadata with each message for routers to make 961 positive policy decisions. Protocols like [RFC4906] have seen 962 limited wide-area uptake, and these alternate designs are frequently 963 re-implemented on top of UDP. [quic] 965 5.2.3.3. Address Translation and Mobility 967 A major structural shift in the Internet which undermined the 968 protocol design of IPv4, and significantly reduced the freedom of end 969 users to communicate and assemble is the introduction of network 970 address translation. [RFC1631] Network address translation is a 971 process whereby organizations and autonomous systems connect two 972 networks by translating the IPv4 source and destination addresses 973 between the two. This process puts the router performing the 974 translation into a privileged position, where it can decide which 975 subset of communications are worthy of translation, and whether an 976 unknown request for communication will be correctly forwarded to a 977 host on the other network. 979 This process of translation has widespread adoption despite promoting 980 a process that goes against the stated end-to-end process of the 981 underlying protocol [natusage]. In contrast, the proposed mechanism 982 to provide support for mobility and forwarding to clients which may 983 move, encoded instead as an option in the IP protocol in [RFC5944], 984 has failed to gain traction. In this situation the compromise made 985 in the design of the protocol resulted in a technology that is not 986 coherent with the end-to-end principles and thus creates and extra 987 possible hurdle for freedom of expression in its design, even though 988 a viable alternative that would do this exists. There is a 989 particular problem surrounding NATs and VPN (as well as other 990 connections used for privacy purposes) as they sometimes cause these 991 not to work. 993 5.2.4. DNS 995 The Domain Name System (DNS) [RFC1035], provides service discovery 996 capabilities, and provides a mechanism to associate human readable 997 names with services. The DNS system is organized around a set of 998 independently operated 'Root Servers' run by organizations around the 999 web which enact ICANN's policy by answering queries for which 1000 organizations have been delegated to manage registration under each 1001 Top Level Domain (TLD). The DNS is organized as a rooted tree, and 1002 this brings up political and social concerns over control. Top Level 1003 domains are maintained and determined by ICANN. These namespaces 1004 encompass several classes of services. The initial name spaces 1005 including '.Com' and '.Net', provide common spaces for expression of 1006 ideas, though their policies are enacted through US based companies. 1007 Other name spaces are delegated to specific nationalities, and may 1008 impose limits designed to focus speech in those forums both to 1009 promote speech from that nationality, and to comply with local limits 1010 on expression and social norms. Finally, the system has recently 1011 been expanded with additional generic and sponsored name spaces, for 1012 instance '.travel' and '.ninja', which are operated by a range of 1013 organizations which may independently determine their registration 1014 policies. This new development has both positive and negative 1015 implications in terms of enabling human rights. Some individuals 1016 argue that it undermines the right to freedom of expression because 1017 some of these new gtlds have restricted policies on registration and 1018 particular rules on hate speech content. Others argue that precisely 1019 these properties are positive because they enable certain (mostly 1020 minority) communities to build safer spaces for association, thereby 1021 enabling their right to freedom of association. An often mentioned 1022 example is an application like .gay. 1024 DNS has significant privacy issues per [RFC7626]. Most notable the 1025 lack of encryption to limit the visibility of requests for domain 1026 resolution from intermediary parties, and a limited deployment of 1027 DNSSEC to provide authentication, allowing the client to know that 1028 they received a correct, "authoritative", answer to a query. In 1029 response to the privacy issues, the IETF DNS PRIVate Exchange 1030 (DPRIVE) Working Group is developing mechanisms to provide 1031 confidentiality to DNS transactions, to address concerns surrounding 1032 pervasive monitoring [RFC7258]. 1034 Authentication through DNSSEC creates a validation path for records. 1035 This authentication protects against forged or manipulated DNS data. 1036 As such DNSSEC protects the directory look-up and makes hijacking of 1037 a session harder. This is important because currently interference 1038 with the operation of the DNS is becoming one of the central 1039 mechanisms used to block access to websites. This interference 1040 limits both the freedom of expression of the publisher to offer their 1041 content, and the freedom of assembly for clients to congregate in a 1042 shared virtual space. Even though DNSSEC doesn't prevent censorship, 1043 it makes it clear that the returned information is not the 1044 information that was requested, which contributes to the right to 1045 security and increases trust in the network. It is however important 1046 to note that DNSSEC is currently not widely supported or deployed by 1047 domain name registrars, making it difficult to authenticate and use 1048 correctly. 1050 5.2.4.1. Removal of records 1052 There have been a number of cases where the records for a domain are 1053 removed from the name system due to real-world events. Examples of 1054 this removal includes the 'seizure' of wikileaks [bbc-wikileaks] and 1055 the names of illegally operating gambling operations by the United 1056 States Immigrations and Customs Enforcement unit, which compelled the 1057 US-based registry in charge of the .com TLD to hand ownership of 1058 those domains over to the US government. The same technique has been 1059 used in Libya to remove sites in violation of "our Country's Law and 1060 Morality (which) do not allow any kind of pornography or its 1061 promotion." [techyum] 1063 At a protocol level, there is no technical auditing for name 1064 ownership, as in alternate systems like [namecoin]. As a result, 1065 there is no ability for users to differentiate seizure from the 1066 legitimate transfer of name ownership, which is purely a policy 1067 decision of registrars. While DNSSEC addresses network distortion 1068 events described below, it does not tackle this problem. 1070 (While mentioning alternative techniques, this is not a complete 1071 comparison of DNS with Namecoin: the latter has its own problems and 1072 limitations. The idea here is to show that there are several 1073 possible choices, and they have consequences for human rights.) 1075 5.2.4.2. Distortion of records 1077 The most common mechanism by which the DNS system is abused to limit 1078 freedom of expression is through manipulation of protocol messages by 1079 the network. One form occurs at an organizational level, where 1080 client computers are instructed to use a local DNS resolver 1081 controlled by the organization. The DNS resolver will then 1082 selectively distort responses rather than request the authoritative 1083 lookup from the upstream system. The second form occurs through the 1084 use of deep packet inspection, where all DNS protocol messages are 1085 inspected by the network, and objectionable content is distorted, as 1086 can be observed in Chinese network. 1088 A notable instance of distortion occurred in Greece [ververis], where 1089 a study found evidence of both of deep packet inspection to distort 1090 DNS replies, and overblocking of content. ISPs prevented clients 1091 from resolving the names of domains which they were instructed to do 1092 through a governmental order, prompting this particular blocking 1093 systems there. 1095 At a protocol level, the effectiveness of these attacks is made 1096 possible by a lack of authentication in the DNS protocol. DNSSEC 1097 provides the ability to determine authenticity of responses when 1098 used, but it is not regularly checked by resolvers. DNSSEC is not 1099 effective when the local resolver for a network is complicit in the 1100 distortion, for instance when the resolver assigned for use by an ISP 1101 is the source of injection. Selective distortion of records is also 1102 been made possible by the predictable structure of DNS messages, 1103 which make it computationally easy for a network device to watch all 1104 passing messages even at high speeds, and the lack of encryption, 1105 which allows the network to distort only an objectionable subset of 1106 protocol messages. Specific distortion mechanisms are discussed 1107 further in [hall]. 1109 5.2.4.3. Injection of records 1111 Responding incorrectly to requests for name lookups is the most 1112 common mechanism that in-network devices use to limit the ability of 1113 end users to discover services. A deviation, which accomplishes a 1114 similar objective may be seen as different from a freedom of 1115 expression perspective, is the injection of incorrect responses to 1116 queries. The most prominent example of this behavior occurs in 1117 China, where requests for lookups of sites deemed inappropriate will 1118 trigger the network to respond with a false response, causing the 1119 client to ignore the real response when it subsequently arrives. 1120 [greatfirewall] Unlike the other forms of discussion mentioned above, 1121 injection does not stifle the ability of a server to announce it's 1122 name, it instead provides another voice which answers sooner. This 1123 is effective because without DNSSEC, the protocol will respond to 1124 whichever answer is received first, without listening for subsequent 1125 answers. 1127 5.2.5. HTTP 1129 The Hypertext Transfer Protocol (HTTP), described in its version 1.1 1130 in RFC 7230 to 7237, is a request-response application protocol 1131 developed throughout the 1990s, and factually contributed to the 1132 exponential growth of the Internet and the inter-connection of 1133 populations around the world. Because of its simple design, HTTP has 1134 become the foundation of most modern Internet platforms and 1135 communication systems, from websites, to chat systems, and computer- 1136 to-computer applications. In its manifestation with the World Wide 1137 Web, HTTP radically revolutionized the course of technological 1138 development and the ways people interact with online content and with 1139 each other. 1141 However, HTTP is also a fundamentally insecure protocol, that doesn't 1142 natively provide encryption properties. While the definition of the 1143 Secure Sockets Layer (SSL), and later of Transport Layer Security 1144 (TLS), also happened during the 1990s, the fact that HTTP doesn't 1145 mandate the use of such encryption layers to developers and service 1146 providers, caused a very late adoption of encryption. Only in the 1147 middle of the 2000s did we observed big Internet service providers, 1148 such as Google, starting to provide encrypted access to their web 1149 services. 1151 The lack of sensitivity and understanding of the critical importance 1152 of securing web traffic incentivized malicious and offensive actors 1153 to develop, deploy and utilize at large interception systems and 1154 later active injection attacks, in order to swipe large amounts of 1155 data, compromise Internet-enabled devices. The commercial 1156 availability of systems and tools to perform these types of attacks 1157 also led to a number of human rights abuses that have been discovered 1158 and reported over the years. 1160 Generally we can identify in Traffic Interception and Traffic 1161 Manipulation the two most problematic attacks that can be performed 1162 against applications employing a clear-text HTTP transport layer. 1163 That being said, the IETF and especially the General Area Review Team 1164 (Gen-ART), is taking steady steps to move to the encrypted version of 1165 HTTP, HTTPSecure (HTTPS). 1167 5.2.5.1. Traffic Interception 1169 While we are seeing an increasing trend in the last couple of years 1170 to employ SSL/TLS as a secure traffic layer for HTTP-based 1171 applications, we are still far from seeing an ubiquitous use of 1172 encryption on the World Wide Web. It is important to consider that 1173 the adoption of SSL/TLS is also a relatively recent phenomena. 1174 E-mail providers such as riseup.net provided SSL on by default as on 1175 of the first. Google introduced an option for its GMail users to 1176 navigate with SSL only in 2008 [Rideout], and turned SSL on by 1177 default later in 2010 [Schillace]. It took an increasing amount of 1178 security breaches and revelations on global surveillance from Edward 1179 Snowden to have other Internet service providers follow Google's 1180 lead. For example, Yahoo enabled SSL/TLS by default on its webmail 1181 services only towards the end of 2013 [Peterson]. 1183 TLS itself has been subject to many attacks and bugs which can be 1184 attributed to some fundamental design weaknesses such as lack of a 1185 state machine, which opens a vulnerability for a Triple Handshake 1186 Attack, and flaws caused by early U.S. government restrictions on 1187 cryptography, leading to cipher-suite downgrade attacks (Logjam 1188 attack). These vulnerabilities have been corrected in TLS1.3. 1189 [Bhargavan] [Adrian] 1191 HTTP upgrading to HTTPS is also vulnerable to having an attacker 1192 remove the "S" in any links to HTTPS URIs from a web-page transferred 1193 in cleartext over HTTP, an attack called "SSL Stripping" [sslstrip]. 1194 Thus, for high security use of HTTPS IETF standards such as HSTS 1195 [RFC6797], certificate pinning [RFC7469] and/or DANE [RFC6698] should 1196 be used. 1198 As we learned through the Snowden's revelations, intelligence 1199 agencies have been intercepting and collecting unencrypted traffic at 1200 large for many years. There are documented examples of such mass 1201 surveillance programs with GCHQ's TEMPORA and NSA's XKEYSCORE. 1202 Through these programs NSA/GCHQ have been able to swipe large amounts 1203 of data including email and instant messaging communications which 1204 have been transported by the respective providers in clear for years, 1205 unsuspecting of the pervasiveness and scale of governments' efforts 1206 and investment into global mass surveillance capabilities. 1208 However, similar mass interception of unencrypted HTTP communications 1209 is also often employed at a nation-level by less democratic countries 1210 by exercising control over state-owned Internet Service Providers 1211 (ISP) and through the use of commercially available monitoring, 1212 collection, and censorship equipment. Over the last few years a lot 1213 of information has come to public attention on the role and scale of 1214 a surveillance industry dedicated to develop interception gear of 1215 different types, making use of known and unknown weaknesses in 1216 existing protocols [RFC7258]. We have several records of such 1217 equipment being sold and utilized by oppressive regimes in order to 1218 monitor entire segments of population especially at times of social 1219 and political distress, uncovering massive human rights abuses. For 1220 example, in 2013 the group Telecomix revealed that the Syrian regime 1221 was making use of BlueCoat products in order to intercept clear-text 1222 traffic as well as to enforce censorship of unwanted content [RSF]. 1223 Similarly in 2012 it was found that the French Amesys provided the 1224 Gaddafi's government with equipment able to intercept emails, 1225 Facebook traffic, and chat messages at a country level. The use of 1226 such systems, especially in the context of the Arab Spring and of 1227 civil uprisings against the dictatorships, has caused serious 1228 concerns of significant human rights abuses in Libya. 1230 5.2.5.2. Traffic Manipulation 1232 The lack of a secure transport layer under HTTP connections not only 1233 exposes the users to interception of the content of their 1234 communications, but is more and more commonly abused as a vehicle for 1235 actively comprosing computers and mobile devices. If an HTTP session 1236 travels in the clear over the network, any node positioned at any 1237 point in the network is able to perform man-in-the-middle attacks and 1238 observe, manipulate, and hijack the session and modify the content of 1239 the communication in order to trigger unexpected behavior by the 1240 application generating the traffic. For example, in the case of a 1241 browser the attacker would be able to inject malicious code in order 1242 to exploit vulnerabilities in the browser or any of its plugins. 1243 Similarly, the attacker would be able to intercept, add malware, and 1244 repackage binary software updates that are very commonly downloaded 1245 in clear by applications such as word processors and media players. 1246 If the HTTP session would be encrypted, the tampering of the content 1247 would not be possible, and these network injection attacks would not 1248 be successful. 1250 While traffic manipulation attacks have been long known, documented, 1251 and prototyped especially in the context of WiFi and LAN networks, in 1252 the last few years we observed an increasing investment into the 1253 production and sale of network injection equipment both available 1254 commercially as well as deployed at scale by intelligence agencies. 1256 For example we learned from some of the documents provided by Edward 1257 Snowden to the press, that the NSA has constructed a global network 1258 injection infrastructure, called QUANTUM, able to leverage mass 1259 surveillance in order to identify targets of interests and 1260 subsequently task man-on-the-side attacks to ultimately compromise a 1261 selected device. Among other attacks, NSA makes use of an attack 1262 called QUANTUMINSERT [Haagsma] which intercepts and hijacks an 1263 unencrypted HTTP communication and forces the requesting browser to 1264 redirect to a host controlled by NSA instead of the intended website. 1265 Normally, the new destination would be an exploitation service, 1266 referred in Snowden documents as FOXACID, which would attempt at 1267 executing malicious code in the context of the target's browser. The 1268 Guardian reported in 2013 that NSA has for example been using these 1269 techniques to target users of the popular anonymity service Tor 1270 [Schneier]. The German NDR reported in 2014 that NSA has also been 1271 using its mass surveillance capabilities to identify Tor users at 1272 large [Appelbaum]. 1274 Recently similar capabilities of Chinese authorities have been 1275 reported as well in what has been informally called the "Great 1276 Cannon" [Marcak], which raised numerous concerns on the potential 1277 curb on human rights and freedom of speech due to the increasing 1278 tighter control of Chinese Internet communications and access to 1279 information. 1281 Network injection attacks are also made widely available to state 1282 actors around the world through the commercialization of similar, 1283 smaller scale equipment that can be easily acquired and deployed at a 1284 country-wide level. Companies like FinFisher and HackingTeam are 1285 known to have network injection gear within their products portfolio, 1286 respectively called FinFly ISP and RCS Network Injector 1287 [Marquis-Boire]. The technology devised and produced by HackingTeam 1288 to perform network traffic manipulation attacks on HTTP 1289 communications is even the subject of a patent application in the 1290 United States [Googlepatent]. Access to offensive technologies 1291 available on the commercial lawful interception market has been 1292 largely documented to have lead to human rights abuses and 1293 illegitimate surveillance of journalists, human rights defenders, and 1294 political activists in many countries around the world. Companies 1295 like FinFisher and HackingTeam have been found selling their products 1296 to oppressive regimes with little concern for bad human rights 1297 records [Collins]. While network injection attacks haven't been the 1298 subject of much attention, they do enable even unskilled attackers to 1299 perform silent and very resilient compromises, and unencrypted HTTP 1300 remains one of the main vehicles. 1302 There is a new version of HTTP, called HTTP/2, which was published as 1303 [RFC7540] and which aimed to be largely backwards compatible but also 1304 offer new option such as data compression of HTTP headers and 1305 pipelining of request and multiplexing multiple requests over a 1306 single TCP connection. In addition to decreasing latency to improve 1307 page loading speeds it also facilitates more efficient use of 1308 connectivity in low-bandwith environments, which is an enabler for 1309 freedom of expression, the right to assembly, right to political 1310 participation and the right to participate in cultural life, art and 1311 science. [RFC7540] does not mandate Transport Layer Security or any 1312 other form of encryption, is also does not support opportunistic 1313 encryption, so the vulnerabilities listed above for HTTP/1 are also 1314 valid for HTTP/2 as defined in [RFC7540]. 1316 5.2.6. XMPP 1318 The Extensible Messaging and Presence Protocol (XMPP), specified in 1319 [RFC6120], provides a standard for interactive chat messaging, and 1320 has evolved to encompass interoperable text, voice, and video chat. 1321 The protocol is structured as a federated network of servers, similar 1322 to email, where users register with a local server which acts one 1323 their behalf to cache and relay messages. This protocol design has 1324 many advantages, allowing servers to shield clients from denial of 1325 service and other forms of retribution for their expression, and 1326 designed to avoid central entities which could control the ability to 1327 communicate or assemble using the protocol. 1329 None-the-less, there are plenty of aspects of the protocol design of 1330 XMPP which shape the ability for users to communicate freely, and to 1331 assembly through the protocol. The protocol also has facets that may 1332 stifle speech as users self-censor for fear of surveillance, or find 1333 themselves unable to express themselves freely. 1335 5.2.6.1. User Identification 1337 The XMPP specification dictates that clients are identified with a 1338 resource (node@domain/home [1] / node@domain/work [2]) to distinguish 1339 the conversations to specific devices. While the protocol does not 1340 specify that the resource must be exposed by the client's server to 1341 remote users, in practice this has become the default behavior. In 1342 doing so, users can be tracked by remote friends and their servers, 1343 who are able to monitor presence not just of the user, but of each 1344 individual device the user logs in with. This has proven to be 1345 misleading to many users, [pidgin] since many clients only expose 1346 user level rather than device level presence. Likewise, user 1347 invisibility so that communication can occur while users don't notify 1348 all buddies and other servers of their availability is not part of 1349 the formal protocol, and has only been added as an extension within 1350 the XML stream rather than enforced by the protocol. 1352 5.2.6.2. Surveillance of Communication 1354 The XMPP protocol specifies the standard by which communication of 1355 channels may be encrypted, but it does not provide visibility to 1356 clients of whether their communications are encrypted on each link. 1357 In particular, even when both clients ensure that they have an 1358 encrypted connection to their XMPP server to ensure that their local 1359 network is unable to read or disrupt the messages they send, the 1360 protocol does not provide visibility into the encryption status 1361 between the two servers. As such, clients may be subject to 1362 selective disruption of communications by an intermediate network 1363 which disrupts communications based on keywords found through Deep 1364 Packet Inspection. While many operators have commited to only 1365 establishing encrypted links from their servers in recognition of 1366 this vulnerability, it remains impossible for users to audit this 1367 behavior and encrypted connections are not required by the protocol 1368 itself [xmppmanifesto]. 1370 In particular, section 13.14 of the protocol specification [RFC6120] 1371 explicitly acknowledges the existence of a downgrade attack where an 1372 adversary controlling an intermediate network can force the inter 1373 domain federation between servers to revert to a non-encrypted 1374 protocol were selective messages can then be disrupted. 1376 5.2.6.3. Group Chat Limitations 1378 Group chat in the XMPP protocol is defined as an extension within the 1379 XML specification of the XMPP protocol (https://xmpp.org/extensions/ 1380 xep-0045.html). However, it is not encoded or required at a protocol 1381 level, and not uniformly implemented by clients. 1383 The design of multi-user chat in the XMPP protocol suffers from 1384 extending a protocol that was not designed with assembly of many 1385 users in mind. In particular, in the federated protocol provided by 1386 XMPP, multi-user communities are implemented with a distinguished 1387 'owner', who is granted control over the participants and structure 1388 of the conversation. 1390 Multi-user chat rooms are identified by a name specified on a 1391 specific server, so that while the overall protocol may be federated, 1392 the ability for users to assemble in a given community is moderated 1393 by a single server. That server may block the room and prevent 1394 assembly unilaterally, even between two users neither of whom trust 1395 or use that server directly. 1397 5.2.7. Peer to Peer 1399 Peer-to-Peer (P2P) is a network architecture in which all the 1400 participant nodes can be responsible for the storage and 1401 dissemination of information from any other node (defined in 1402 [RFC7574], an IETF standard that used a P2P architecture). A P2P 1403 network is a logical overlay that lives on top of the physical 1404 network, and allows nodes (or "peers") participating to it to 1405 establish contact and exchange information directly from one to each 1406 other. The implementation of a P2P network may very widely: it may 1407 be structured or unstructured, and it may implement stronger or 1408 weaker cryptographic and anonymity properties. While its most common 1409 application has traditionally been file-sharing (and other types of 1410 content delivery systems), P2P is increasingly becoming a popular 1411 architecture for networks and applications that require (or 1412 encourage) decentralization. A prime example is Bitcoin (and similar 1413 cryptocurrencies), as well as Skype, Spotify and other proprietary 1414 multimedia applications. 1416 In a time of heavily centralized online services, peer-to-peer is 1417 often seen as an alternative, more democratic, and resistant 1418 architecture that displaces structures of control over data and 1419 communications and delegates all peers equally to be responsible for 1420 the functioning, integrity, and security of the data. While in 1421 principle peer-to-peer remains critical to the design and development 1422 of future content distribution, messaging, and publishing systems, it 1423 poses numerous security and privacy challenges which are mostly 1424 delegated to individual developers to recognize, analyze, and solve 1425 in each implementation of a given P2P network. 1427 5.2.7.1. Network Poisoning 1429 Since content, and in some occasions peer lists, are safeguarded and 1430 distributed by its members, P2P networks are prone to what are 1431 generally defined as "poisoning attacks". Poisoning attacks might be 1432 directed directly at the data that is being distributed, for example 1433 by intentionally corrupting it, or at the index tables used to 1434 instruct the peers where to fetch the data, or at routing tables, 1435 with the attempt of providing connecting peers with lists of rogue or 1436 non-existing peers, with the intention to effectively cause a Denial 1437 of Service on the network. 1439 5.2.7.2. Throttling 1441 Peer-to-Peer traffic (and BitTorrent in particular) represents a high 1442 percentage of global Internet traffic and it has become increasingly 1443 popular for Internet Service Providers to perform throttling of 1444 customers lines in order to limit bandwidth usage [torrentfreak1] and 1445 sometimes probably as an effect of the ongoing conflict between 1446 copyright holders and file-sharing communities [wikileaks]. Such 1447 throtteling undermines the end-to-end principle. 1449 Throttling the peer-to-peer traffic makes some uses of P2P networks 1450 ineffective and it might be coupled with stricter inspection of 1451 users' Internet traffic through Deep Packet Inspection techniques 1452 which might pose additional security and privacy risks. 1454 5.2.7.3. Tracking and Identification 1456 One of the fundamental and most problematic issues with traditional 1457 peer-to-peer networks is a complete lack of anonymization of its 1458 users. For example, in the case of BitTorrent, all peers' IP 1459 addresses are openly available to the other peers. This has lead to 1460 an ever-increasing tracking of peer-to-peer and file-sharing users 1461 [ars]. As the geographical location of the user is directly exposed, 1462 and so could be his identity, the user might become target of 1463 additional harassment and attacks, being of physical or legal nature. 1464 For example, it is known that in Germany law firms have made 1465 extensive use of peer-to-peer and file-sharing tracking systems in 1466 order to identify downloaders and initiate legal actions looking for 1467 compensations [torrentfreak2]. 1469 It is worth noting that there are varieties of P2P networks that 1470 implement cryptographic practices and that introduce anonymization of 1471 its users. Such implementations may be proved to be successful in 1472 resisting censorship of content, and tracking of the network peers. 1473 A primary example is FreeNet [freenet1], a free software application 1474 designed to significantly increase the difficulty of users and 1475 content identification, and dedicated to foster freedom of speech 1476 online [freenet2]. 1478 5.2.7.4. Sybil Attacks 1480 In open-membership P2P networks, a single attacker can pretend to be 1481 many participants, typically by creating multiple fake identities of 1482 whatever kind the P2P network uses [Douceur]. Attackers can use 1483 Sybil attacks to bias choices the P2P network makes collectively 1484 toward the attacker's advantage, e.g., by making it more likely that 1485 a particular data item (or some threshold of the replicas or shares 1486 of a data item) are assigned to attacker-controlled participants. If 1487 the P2P network implements any voting, moderation, or peer review- 1488 like functionality, Sybil attacks may be used to "stuff the ballots" 1489 toward the attacker's benefit. Companies and governments can use 1490 Sybil attacks on discussion-oriented P2P systems for "astroturfing" 1491 or creating the appearance of mass grassroots support for some 1492 position where there is none in reality. It is important to know 1493 that there are no known complete, environmentally sustainable, and 1494 fully distributed solutions to Sybil attacks, and routing via 1495 'friends' allows users to be de-anonymized via their social graph. 1497 5.2.7.5. Conclusions 1499 Encrypted P2P and Anonymous P2P networks already emerged and provided 1500 viable platforms for sharing material [tribler], publish content 1501 anonymously, and communicate securely [bitmessage].These platforms 1502 are not perfect, and more research needs to be done. If adopted at 1503 large, well-designed and resistant P2P networks might represent a 1504 critical component of a future secure and distributed Internet, 1505 enabling freedom of speech and freedom of information at scale. 1507 5.2.8. Virtual Private Network 1509 5.2.8.1. Introduction 1511 A Virtual Private Network (VPN) is a point-to-point connection that 1512 enables two computers to communicate over an encrypted tunnel. There 1513 are multiple implementations and protocols used in provisioning a 1514 VPN, and they generally diversify by encryption protocol or 1515 particular requirements, most commonly in proprietary and enterprise 1516 solutions. VPNs are used commonly either to enable some devices to 1517 communicate through peculiar network configurations, or in order to 1518 use some privacy and security properties in order to protect the 1519 traffic generated by the end user; or both. VPNs have also become a 1520 very popular technology among human rights defenders, dissidents, and 1521 journalists worldwide to avoid local illegitimate wiretapping and 1522 eventually also to circumvent censorship. Among human rights 1523 defenders VPNs are often debated as a potential alternative to Tor or 1524 other anonymous networks. Such comparison is misleading, as some of 1525 the privacy and security properties of VPNs are often misunderstood 1526 by less tech-savvy users, which could ultimately lead to unintended 1527 problems. 1529 As VPNs increased in popularity, commercial VPN providers have 1530 started growing in business and are very commonly picked by human 1531 rights defenders and people at risk, as they are normally provided 1532 with an easy-to-use service and sometimes even custom applications to 1533 establish the VPN tunnel. Not being able to control the 1534 configuration of the network, and even less so the security of the 1535 application, assessing the general privacy and security state of 1536 common VPNs is very hard. Often such services have been discovered 1537 leaking information, and their custom applications have been found 1538 flawed. While Tor and similar networks receive a lot of scrutiny 1539 from the public and the academic community, commercial or non- 1540 commercial VPN networks are way less analyzed and understood, and it 1541 might be valuable to establish some standards to guarantee a minimal 1542 level of privacy and security to those who need them the most. 1544 5.2.8.2. No anonymity against VPN provider 1546 One of the common misconception among users of VPNs is the level of 1547 anonymity VPN can provide. This sense of anonymity can be betrayed 1548 by a number of attacks or misconfigurations of the VPN provider. It 1549 is important to remember that, contrarily to Tor and similar systems, 1550 VPN was not designed to provide anonymity properties. From a 1551 technical point of view, the VPN might leak identifiable information, 1552 or might be subject of correlation attacks that could expose the 1553 originating address of the connecting user. Most importantly, it is 1554 vital to understand that commercial and non-commercial VPN providers 1555 are bound by the law of the jurisdiction they reside in or in which 1556 their infrastructure is located, and they might be legally forced to 1557 turn over data of specific users if legal investigations or 1558 intelligence requirements dictate so. In such cases, if the VPN 1559 providers retain logs, it is possible that the information of the 1560 user is provided to the user's adversary and leads to his or her 1561 identification. 1563 5.2.8.3. Logging 1565 With VPN being point-to-point connections, the service providers are 1566 in fact able to observe the original location of the connecting users 1567 and they are able to track at what time they started their session 1568 and eventually also to which destinations they're trying to connect 1569 to. If the VPN providers retain logs for long enough, they might be 1570 forced to turn over the relevant data or they might be otherwise 1571 compromised, leading to the same data getting exposed. A clear log 1572 retaining policy could be enforced, but considering that countries 1573 enforce very different levels of data retention policies, VPN 1574 providers should at least be transparent on what information do they 1575 store and for how long is being kept. 1577 5.2.8.4. 3rd Party Hosting 1579 VPN providers very commonly rely on 3rd parties to provision the 1580 infrastructure that is later going to be used to run VPN endpoints. 1581 For example, they might rely on external dedicated server hosting 1582 providers, or on uplink providers. In those cases, even if the VPN 1583 provider itself isn't retaining any significant logs, the information 1584 on the connecting users might be retained by those 3rd parties 1585 instead, introducing an additional collection point for the 1586 adversary. 1588 5.2.8.5. IPv6 Leakage 1590 Some studies proved that several commercial VPN providers and 1591 applications suffer of critical leakage of information through IPv6 1592 due to improper support and configuration [PETS2015VPN]. This is 1593 generally caused by a lack of proper configuration of the client's 1594 IPv6 routing tables. Considering that most popular browsers and 1595 similar applications have been supporting IPv6 by default, if the 1596 host is provided with a functional IPv6 configuration, the traffic 1597 that is generated might be leaked if the VPN application isn't 1598 designed to manipulate such traffic properly. 1600 5.2.8.6. DNS Leakage 1602 Similarly, VPN services that aren't handling DNS requests and are not 1603 running DNS servers of their own, might be prone to DNS leaking which 1604 might not only expose sensitive information on the activity of the 1605 user, but could also potentially lead to DNS hijacking attacks and 1606 following compromises. 1608 5.2.8.7. Traffic Correlation 1610 As revelations of mass surveillance have been growing in the press, 1611 additional details on attacks on secure Internet communications have 1612 come to the public's attention. Among these, VPN appeared to be a 1613 very interesting target for attacks and collection efforts. Some 1614 implementations of VPN appear to be particularly vulnerable to 1615 identification and collection of key exchanges which, some Snowden 1616 documents revealed, are systematically collected and stored for 1617 future reference. The ability of an adversary to monitor network 1618 connections at many different points over the Internet, can allow 1619 them to perform traffic correlation attacks and identify the origin 1620 of certain VPN traffic by cross referencing the connection time of 1621 the user to the endpoint and the connection time of the endpoint to 1622 the final destination. These types of attacks, although very 1623 expensive and normally only performed by very resourceful 1624 adversaries, have been documented [spiegel] to be already in practice 1625 and could completely vanify the use of a VPN and ultimately expose 1626 the activity and the identity of a user at risk. 1628 5.2.9. HTTP Status Code 451 1630 Every Internet user has run into the '404 Not Found' Hypertext 1631 Transfer Protocol (HTTP) status code when trying, and failing, to 1632 access a particular website [Cath]. It is a response status that the 1633 server sends to the browser, when the server cannot locate the URL. 1634 '403 Forbidden' is another example of this class of code signals that 1635 gives users information about what is going on. In the '403' case 1636 the server can be reached, but is blocking the request because the 1637 user is trying to access content forbidden to them. This can be 1638 because the specific user is not allowed access to the content (like 1639 a government employee trying to access pornography on a work- 1640 computer) or because access is restricted to all users (like social 1641 network sites in certain countries). As surveillance and censorship 1642 of the Internet is becoming more commonplace, voices were raised at 1643 the IETF to introduce a new status code that indicates when something 1644 is not available for 'legal reasons' (like censorship): 1646 The 451 status code would allow server operators to operate with 1647 greater transparency in circumstances where issues of law or public 1648 policy affect their operation. This transparency may be beneficial 1649 both to these operators and to end-users [Bray]. 1651 The status code is named '451', a reference to Bradbury's famous 1652 novel on censorship, and the temperature (in Fahrenheit) at which 1653 bookpaper autoignites. 1655 During the IETF92 meeting in Dallas, there was discussion about the 1656 usefulness of '451'. The main tension revolved around the lack of an 1657 apparent machine-readable technical use of the information. The 1658 extent to which '451' is just 'political theatre' or whether it has a 1659 concrete technical use was heatedly debated. Some argued that 'the 1660 451 status code is just a status code with a response body' others 1661 said it was problematic because 'it brings law into the picture'. 1662 Again others argued that it would be useful for individuals, or 1663 organizations like the 'Chilling Effects' project, crawling the web 1664 to get an indication of censorship (IETF discussion on '451' - 1665 author's field notes March 2015). There was no outright objection 1666 during the Dallas meeting against moving forward on status code 1667 '451', and on December 18, 2015 the Internet Engineering Steering 1668 Group approved publication of 'An HTTP Status Code to Report Legal 1669 Obstacles'. It is now an IETF approved HTTP status code to signal 1670 when resource access is denied as a consequence of legal demands 1671 [RFC7725]. 1673 What is interesting about this particular case is that not only 1674 technical arguments but also the status code's outright potential 1675 political use for civil society played a substantial role in shaping 1676 the discussion, and the decision to move forward with this 1677 technology. 1679 It is nonetheless important to note that HTTP status code 451 is not 1680 a solution to detect all occasions of censorship. A large swath of 1681 Internet filtering occurs in the network rather than the server 1682 itself. For these forms of censorship 451 plays a limited role, as 1683 the servers will not be able to send the code, because they haven't 1684 received the requests (as is the case with servers with resources 1685 blocked by the Chinese Golden shield). Such filtering regimes are 1686 unlikely to voluntarily inject a 451 status code. The use of 451 is 1687 most likely to apply in the case of cooperative, legal versions of 1688 content removal resulting from requests to providers. One can think 1689 of content that is removed or blocked for legal reasons, like 1690 copyright infringement, gambling laws, child abuse, et cetera. Large 1691 Internet companies and search engines are constantly asked to censor 1692 content in various jurisdictions. 451 allows this to be easily 1693 discovered, for instance by initiatives like the Lumen Database. 1695 Overall, the strength of 451 lies in its ability to provide 1696 transparency by giving the reason for blocking, and giving the end- 1697 user the ability to file a complaint. It allows organizations to 1698 easily measure censorship in an automated way, and prompts the user 1699 to access the content via another path (e.g. TOR, VPNs) when (s)he 1700 encounters the 451 status code. 1702 Status code 451 impact human rights by making censorship more 1703 transparent and measurable. The status code increases transparency 1704 both by signaling the existence of censorship (instead of a much more 1705 broad HTTP error message like HTTP status code 404) as well as 1706 providing details of the legal restriction, which legal authority is 1707 imposing it, and what class of resources it applies to. This 1708 empowers the user to seek redress. 1710 5.2.10. Middleboxes 1712 On the current Internet, transparency on how packets reach a 1713 destination is no longer a given. This is due to the increased 1714 presence of firewalls, spam filters, and network address translators 1715 networks (NATs) - or middleboxes as these hosts are often called - 1716 that make use of higher-layer fields to function [Walfish]. This 1717 development is contentious. The debate also unfolded at the IETF, 1718 specifically at the Session Protocol Underneath Datagrams (SPUD) 1719 Birds of a Feather (BOF) meeting held at the IETF conference in March 1720 2015. The discussion at the BOF focused on questions about adding 1721 meta-data, or other information to traffic flows, to enable the 1722 sharing of information with middleboxes in that flow. During the 1723 sessions two competing arguments were distilled. On the one hand 1724 adding additional data would allow for network optimization, and 1725 hence improve traffic carriage. On the other hand, there are risks 1726 of information leakage and other privacy and security concerns. 1728 Middleboxes, and the protocols guiding them, influence individuals' 1729 ability to communicate online freely and privately. Repeatedly 1730 mentioned in the discussion was the danger of censorship that comes 1731 with middleboxes, and the IETF's role to prevent such censorship from 1732 happening. Middleboxes essentially undermine the end-to-end 1733 principle by inserting themselves in the network, and acting as 1734 intermediaries. Although there are many advantages, such as 1735 increased security and network performance, to having middleboxes 1736 they also have downsides. They are known to limit the choice of 1737 transport protocols and drop packets that don't conform. As such, 1738 limiting both freedom of expression online and undermining the end- 1739 to-end principle. 1741 Middleboxes are becoming a proxy for the debate on the extent to 1742 which commercial interests are a valid reason to undermine the end- 1743 to-end principle. The potential for abuse and censoring, and thus 1744 ultimately the impact of middleboxes on the Internet as a place of 1745 unfiltered, unmonitored freedom of speech, is real. It is impossible 1746 to make any definitive statements about the direction the debate on 1747 middleboxes will take at the IETF. The opinions expressed in the 1748 SPUD BOF and by the various interviewees indicate that a majority of 1749 engineers are trying to mitigate the negative effects of middleboxes 1750 on freedom of speech, but their ability to act is limited by their 1751 larger commercial context that is expanding the use of middleboxes. 1753 5.2.11. DDOS attacks 1755 Many individuals, not excluding IETF engineers, have argued that DDoS 1756 attacks are fundamentally against freedom of speech. Technically 1757 DDoS attacks are when one or multiple host overload the bandwidth or 1758 resources of another host by flooding it with traffic, causing it to 1759 temporarily stop being available to users. One can roughly 1760 differentiate three types of DDoS attacks: Volume Based Attacked 1761 (This attack aims to make the host unreachable by using up all it's 1762 bandwith, often used techniques are: UDP floods and ICMP floods), 1763 Protocol Attacks (This attacks aims to use up actual server 1764 resources, often used techniques are SYN floods, fragmented packet 1765 attacks, and Ping of Death [RFC4949]) and Application Layer Attacks 1766 (this attack aims to bring down a server, such as the webserver). 1768 In their 2010 report Zuckerman et al argue that DDoS attacks are a 1769 malicious tool because they are increasingly used by governments to 1770 attack and silence critics. Their research demonstrates that in many 1771 countries independent media outlets and human rights organizations 1772 are the victim of DDoS attacks, which are directly or indirectly 1773 linked to their governments. These types of attacks are particularly 1774 complicated because attribution is difficult, creating a situation in 1775 which governments can effectively censor content, while being able to 1776 deny involvement in the attacks [Zuckerman]. DDoS attacks can thus 1777 stifle freedom of expression, complicate the ability of independent 1778 media and human rights organizations to exercise their right to 1779 (online) freedom of association, while facilitating the ability of 1780 governments to censor dissent. When it comes to comparing DDoS 1781 attacks to protests in offline life, it is important to remember that 1782 only a limited number of DDoS attacks involved solely willing 1783 participants. In most cases, the clients are hacked computers of 1784 unrelated parties that have not consented to being part of a DDoS 1785 (for exceptions see Operation Abibil [Abibil] or the Iranian Green 1786 Movement DDoS [GreenMovement]). 1788 In addition, DDoS attacks are increasingly used as an extortion 1789 tactic, with criminals flooding a website - rendering it inaccessible 1790 - until the owner pays them a certain amount of money to stop the 1791 attack. The costs of mitigating such attacks, either by improving 1792 security to prevent them or paying off the attackers, ends up being 1793 paid by the consumer. 1795 Some people may say that DDoS attacks are the only mean to be heard, 1796 in the current Internet. It is true that Internet has no "public 1797 space", a place where everyone can go and protest and raise issues in 1798 front of everyone. The Internet, and specially the Web, is a more a 1799 series of commercial malls, hold by private entities and where free 1800 speech is limited. This is a complicated issue for human rights, and 1801 specially freedom of speech, but it does not mean that DDoS are a 1802 solution to this real problem. 1804 All of these issues seem to suggest that the IETF should try to 1805 ensure that their protocols cannot be used for DDoS attacks. 1806 Decreasing the number of vulnerabilities in protocols and (outside of 1807 IETF) the number of bugs in the network stacks of routers or 1808 computers could address this issue. The IETF can clearly play a role 1809 in bringing about some of these changes. it is important to consider 1810 that DDoS attacks are sometimes seen as a method for exercising 1811 freedom of speech [Sauter]. 1813 Eventhough it is important to consider that DDoS attacks are 1814 sometimes seen as a method for exercising freedom of speech [Sauter]. 1815 There is a need for the IETF to be consistent in the face of attacks 1816 (an attack is an attack is an attack) to maintain the viability of 1817 the network. Arguing that some DDoS attacks should be allowed, based 1818 on the motivation of the attackers complicates the work of the IETF. 1819 Because it approaches PM regardless of the motivation of the 1820 attackers (see [RFC7258]) for reasoning), taking the motivation of 1821 the attackers into account for DDoS would indirectly undermine the 1822 ability of the IETF to protect the right to privacy because it 1823 introduces an element of inconsistency into how the IETF deals with 1824 attacks. 1826 David Clark recently published a paper warning that the future of the 1827 Internet is in danger. He argues that the private sector control 1828 over the Internet is too strong, limiting the myriad of ways in which 1829 it can be used [Daedalus], including for freedom of speech. But just 1830 because freedom of speech, dissent, and protest are human rights, and 1831 DDoS is a potential expression of those rights, doesn't mean that 1832 DDoS in and of itself is a right. To widen the analogy, just because 1833 the Internet is a medium through which the right to freedom of 1834 expression can be exercised does not make access to the Internet or 1835 specific ICTs or NCTs a human right. Uses of DDoS might or might not 1836 be legitimate for political reasons, but the IETF has no means or 1837 methods to assess this, and in general enabling DDoS would mean a 1838 deterioration of the network and thus freedom of expression. 1840 In summation, the IETF cannot be expected to take a moral stance on 1841 DDoS attacks, or create protocols to enable some attacks and inhibit 1842 others. But what it can do is critically reflect on its role in 1843 creating a commercialized Internet without a defacto public space or 1844 inherent protections for freedom of speech. 1846 5.3. Model for developing human rights protocol considerations 1848 Having established how human rights relate to standards and 1849 protocols, a common vocabulary of technical concepts that impact 1850 human rights and how these technical concept can be combined to 1851 ensure that the Internet remains an enabling environment for human 1852 rights means the contours of a model for developing human rights 1853 protocol considerations has taken shape. This subsection provides 1854 the last step by detailing how the technical concepts identified 1855 above relate to human rights, and what questions engineers should ask 1856 themselves when developing or improving protocols. In short, it 1857 presents a set of human rights protocol considerations. 1859 5.3.1. Human rights threats 1861 Human rights threats on the Internet come in a myriad of forms. 1862 Protocols and standards can harm or enable the right to freedom of 1863 expression, right to non-discrimination, right to equal protection, 1864 right to participate in cultural life, arts and science, right to 1865 freedom of assembly and association, and the right to security. An 1866 end-user who is denied access to certain services, data or websites 1867 may be unable to disclose vital information about the malpractices of 1868 a government or other authority. A person whose communications are 1869 monitored may be prevented from exercising their right to freedom of 1870 association or participate in political processes [Penney]. In a 1871 worst-case scenario, protocols that leak information can lead to 1872 physical danger. A realistic example to consider is when opposition 1873 group members (or those identified as such) in totalitarian regimes 1874 are subjected to torture on the basis of information gathered by the 1875 regime through information leakage in protocols. 1877 This sections details several 'common' threats to human rights, 1878 indicating how each of these can lead to human rights violations/ 1879 harms and present several examples of how these threats to human 1880 rights materialize on the Internet. This threat modeling is inspired 1881 by [RFC6973] Privacy Considerations for Internet Protocols, which is 1882 based on the security threat analysis. This method is by no means a 1883 perfect solution for assessing human rights risks in Internet 1884 protocols and systems; it is however the best approach currently 1885 available. Certain specific human rights threats are indirectly 1886 considered in Internet protocols as part of the security 1887 considerations [RFC3552], but privacy guidelines [RFC6973] or 1888 reviews, let alone human rights impact assessments of protocols are 1889 not standardized or implemented. 1891 Many threats, enablers and risks are linked to different rights. 1892 This is not unsurprising if one takes into account that human rights 1893 are interrelated, interdependent and indivisible. Here however we're 1894 not discussing all human rights because not all human rights are 1895 relevant to ICTs in general and protocols and standards in particular 1896 [Bless]. This is by no means an attempt to cherry picks rights, if 1897 other rights seem relevant, please contact the authors and/or the 1898 hrpc mailinglist. 1900 5.3.2. Guidelines for human rights considerations 1902 This section provides guidance for document authors in the form of a 1903 questionnaire about protocols and their (potential) impact. The 1904 questionnaire may be useful at any point in the design process, 1905 particularly after document authors have developed a high-level 1906 protocol model as described in [RFC4101]. 1908 Protocols and Internet Standard might benefit from a documented 1909 discussion of potential human rights risks arising from potential 1910 misapplications of the protocol or technology described in the RFC. 1911 This might be coupled with an Applicability Statement for that RFC. 1913 Note that the guidance provided in this section does not recommend 1914 specific practices. The range of protocols developed in the IETF is 1915 too broad to make recommendations about particular uses of data or 1916 how human rights might be balanced against other design goals. 1917 However, by carefully considering the answers to the following 1918 questions, document authors should be able to produce a comprehensive 1919 analysis that can serve as the basis for discussion on whether the 1920 protocol adequately protects against specific human rights threats. 1921 This guidance is meant to help the thought process of a human rights 1922 analysis; it does not provide specific directions for how to write a 1923 human rights protocol considerations section (following the example 1924 set in [RFC6973]), and the addition of a human rights protocol 1925 considerations section has also not yet been proposed. 1927 5.3.2.1. Technical concepts as they relate to human rights 1929 5.3.2.1.1. Connectivity 1931 Question(s): Does your protocol add application-specific functions to 1932 intermediary nodes? Could this functionality be added to end nodes 1933 instead of intermediary nodes? 1935 Explanation: The end-to-end principle [Saltzer] which aims to extend 1936 characteristics of a protocol or system as far as possible within the 1937 system, or in other words 'the intelligence is end to end rather than 1938 hidden in the network' [RFC1958]. Middleboxes (which can be Content 1939 Delivery Networks, Firewalls, NATs or other intermediary nodes that 1940 provide other 'services' than routing), and the protocols guiding 1941 them, influence individuals' ability to communicate online freely and 1942 privately. The potential for abuse and intentional and unintentional 1943 censoring and limiting permissionless innovation, and thus ultimately 1944 the impact of middleboxes on the Internet as a place of unfiltered, 1945 unmonitored freedom of speech, is real. 1947 Example: End-to-end instant message encryption would conceal 1948 communications from one user's instant messaging application through 1949 any intermediate devices and servers all the way to the recipient's 1950 instant messaging application. If the message was decrypted at any 1951 intermediate point-for example at a service provider-then the 1952 property of end-to-end encryption would not be present. 1954 Impacts: 1956 - Right to freedom of expression 1958 - Right to freedom of assembly and association 1960 5.3.2.1.2. Privacy 1962 Question(s): Did you have a look at the Guidelines in the Privacy 1963 Considerations for Internet Protocols [RFC6973] section 7? Could 1964 your protocol in any way impact the confidentiality of protocol 1965 metadata? Could your protocol counter traffic analysis, or data 1966 minimization? 1968 Explanation: Privacy refers to the right of an entity (normally a 1969 person), acting in its own behalf, to determine the degree to which 1970 it will interact with its environment, including the degree to which 1971 the entity is willing to share its personal information with others. 1972 [RFC4949]. 1974 Example: See [RFC6973] 1976 Impacts: 1978 - Right to freedom of expression 1980 - Right to non-discrimination 1982 5.3.2.1.3. Content agnosticism 1984 Question(s): If your protocol impacts packet handling, does it look 1985 at the packet content? Is it making decisions based on the content 1986 of the packet? Is the protocol transparent about its decisions? 1987 Does your protocol prioritize certain content or services over 1988 others? 1990 Explanation: Content agnosticism refers to the notion that network 1991 traffic is treated identically regardless of content. 1993 Example: Content agnosticism prevents content-based discrimination 1994 against packets. This is important because changes to this principle 1995 can lead to a two-tiered Internet, where certain packets are 1996 prioritized over others on the basis of their content. Effectively 1997 this would mean that although all users are entitled to receive their 1998 packets at a certain speed, some users become more equal than others. 2000 Impacts: 2002 - Right to freedom of expression 2004 - Right to non-discrimination 2006 - Right to equal protection 2008 5.3.2.1.4. Security 2010 Question(s): Did you have a look at Guidelines for Writing RFC Text 2011 on Security Considerations [RFC3552]? Have you found any attacks 2012 that are out of scope for your protocol? Would these attacks be 2013 pertinent to the human rights enabling features of the Internet (as 2014 descibred throughout this document)? 2016 Explanation: Most people speak of security as if it were a single 2017 monolithic property of a protocol or system, however, upon 2018 reflection; one realizes that it is clearly not true. Rather, 2019 security is a series of related but somewhat independent properties. 2020 Not all of these properties are required for every application. We 2021 can loosely divide security goals into those related to protecting 2022 communications (COMMUNICATION SECURITY, also known as COMSEC) and 2023 those relating to protecting systems (ADMINISTRATIVE SECURITY or 2024 SYSTEM SECURITY). Since communications are carried out by systems 2025 and access to systems is through communications channels, these goals 2026 obviously interlock, but they can also be independently provided 2027 [RFC3552]. Security needs to be also be approached in terms of 2028 advesaries, and passive global adversaries whose attack is pervasive 2029 surveillance now need to be taken into consideration when designing 2030 new protocols. 2032 Example: See [RFC3552]. 2034 Impacts: 2036 - Right to freedom of expression 2038 - Right to freedom of assembly and association 2040 - Right to non discrimination 2042 5.3.2.1.5. Internationalization 2044 Question(s): Does your protocol have text strings that are readable 2045 or entered by humans? Does your protocol allow Unicode encoded in 2046 UTF-8 only,If other character sets /encodings are allowed, does your 2047 protocol mandates a proper tagging of the charset? Did you have a 2048 look at [RFC6365]? 2050 Explanation: Internationalization refers to the practice of making 2051 protocols, standards, and implementations usable in different 2052 languages and scripts. (see Localization). In the IETF, 2053 internationalization means to add or improve the handling of non- 2054 ASCII text in a protocol. [RFC6365] A different perspective, more 2055 appropriate to protocols that are designed for global use from the 2056 beginning, is the definition used by W3C: 2058 "Internationalization is the design and development of a 2059 product, application or document content that enables easy 2060 localization for target audiences that vary in culture, region, 2061 or language." {{W3Ci18nDef}} 2063 Many protocols that handle text only handle one charset (US-ASCII), 2064 or leave the question of what CCS and encoding are used up to local 2065 guesswork (which leads, of course, to interoperability problems). If 2066 multiple charsets are permitted, they must be explicitly identified 2068 [RFC2277]. Adding non-ASCII text to a protocol allows the protocol 2069 to handle more scripts, hopefully representing users across the 2070 world. In today's world, that is normally best accomplished by 2071 allowing Unicode encoded in UTF-8 only. 2073 Example: See localization Impacts: 2075 - Right to freedom of expression 2077 - Right to political participation 2079 - Right to participate in cultural life, arts and science 2081 - Right to political participation 2083 5.3.2.1.6. Censorship resistance 2085 Question(s): Does this protocol introduce new identifiers that might 2086 be associated with persons or content? Does your protocol make it 2087 apparent or transparent when filtering happens? Can your protocol 2088 contribute to filtering in a way it could be implemented to censor 2089 data or services? Could this be designed to ensure this doesn't 2090 happen? 2092 Explanation: Censorship resistance refers to the methods and measures 2093 to prevent Internet censorship. 2095 Example: Identifiers of content exposed within a protocol might be 2096 used to facilitate censorship, as in the case of IP based censorship, 2097 which affects protocols like HTTP. Filtering can be made apparent by 2098 the use of status code 451 - which allows server operators to operate 2099 with greater transparency in circumstances where issues of law or 2100 public policy affect their operation [Bray]. 2102 Impacts: 2104 - Right to freedom of expression 2106 - Right to political participation 2108 - Right to participate in cultural life, arts and science 2110 - Right to freedom of assembly and association 2112 5.3.2.1.7. Open Standards 2114 Question(s): Is your protocol fully documented in a way that it could 2115 be easily implemented, improved, build upon and/or further developed? 2116 Do you depend on proprietary code for the implementation, running or 2117 further development of your protocol? Does your protocol favor a 2118 particular proprietary specification over technically equivalent and 2119 competing specification(s), for instance by making any incorporated 2120 vendor specification "required" or "recommended" [RFC2026]? Do you 2121 normatively reference another standard that is not available without 2122 cost? Are you aware of any patents that would prevent your standard 2123 from being fully implemented [RFC3979] [RFC6701]? 2125 Explanation: The Internet was able to developed into the global 2126 network of networks because of the existence of open, non-proprietary 2127 standards [Zittrain]. They are crucial for enabling 2128 interoperability. Yet, open standards are not explicitly defined 2129 within the IETF. On the subject, [RFC2606] states: Various national 2130 and international standards bodies, such as ANSI, ISO, IEEE, and ITU- 2131 T, develop a variety of protocol and service specifications that are 2132 similar to Technical Specifications defined at the IETF. National 2133 and international groups also publish "implementors' agreements" that 2134 are analogous to Applicability Statements, capturing a body of 2135 implementation-specific detail concerned with the practical 2136 application of their standards. All of these are considered to be 2137 "open external standards" for the purposes of the Internet Standards 2138 Process. Similarly, [RFC3935] does not define open standards but 2139 does emphasize the importance of 'open process': any interested 2140 person can participate in the work, know what is being decided, and 2141 make his or her voice heard on the issue. Part of this principle is 2142 the IETF's commitment to making its documents, WG mailing lists, 2143 attendance lists, and meeting minutes publicly available on the 2144 Internet. 2146 Open standards are important as they allow for permissionless 2147 innovation, which is important to maintain the freedom and ability to 2148 freely create and deploy new protocols on top of the communications 2149 constructs that currently exist. It is at the heart of the Internet 2150 as we know it, and to maintain its fundamentally open nature, we need 2151 to be mindful of the need for developing open standards. 2153 All standards that need to be normatively implemented should be 2154 freely available and with reasonable protection for patent 2155 infringement claims, so it can also be implemented in open source or 2156 free software. Patents have often held back open standardization or 2157 been used against those deploying open stadards, particularly in the 2158 domain of cryptography [newegg]. Patents in open standards or in 2159 normative references to other standards should have a patent 2160 disclosure [notewell], royalty-free licensing [patentpolicy], or some 2161 other form of reasonable protection. Reasonable patent protection 2162 should includes but is not limited to cryptographic primitives. 2164 Example: [RFC6108] describes a system for providing critical end-user 2165 notifications to web browsers, which has been deployed by Comcast, an 2166 Internet Service Provider (ISP). Such a notification system is being 2167 used to provide near-immediate notifications to customers, such as to 2168 warn them that their traffic exhibits patterns that are indicative of 2169 malware or virus infection. There are other proprietary systems that 2170 can perform such notifications, but those systems utilize Deep Packet 2171 Inspection (DPI) technology. In contrast to DPI, this document 2172 describes a system that does not rely upon DPI, and is instead based 2173 in open IETF standards and open source applications. 2175 Impacts: 2177 - Right to freedom of expression 2179 - Right to participate in cultural life, arts and science 2181 5.3.2.1.8. Heterogeneity Support 2183 Question(s): Does your protocol support heterogeneity by design? 2184 Does your protocol allow for multiple types of hardware? Does your 2185 protocol allow for multiple types of application protocols? Is your 2186 protocol liberal in what it receives and handles? Will it remain 2187 usable and open if the context changes? Does your protocol allow 2188 there to be well-defined extension points? Do these extension points 2189 to allow open innovation possibly have security and privacy 2190 ramifications, and if so,how can these be dealt with? 2192 Explanation: The Internet is characterized by heterogeneity on many 2193 levels: devices and nodes, router scheduling algorithms and queue 2194 management mechanisms, routing protocols, levels of multiplexing, 2195 protocol versions and implementations, underlying link layers (e.g., 2196 point-to-point, multi-access links, wireless, FDDI, etc.), in the 2197 traffic mix and in the levels of congestion at different times and 2198 places. Moreover, as the Internet is composed of autonomous 2199 organizations and Internet service providers, each with their own 2200 separate policy concerns, there is a large heterogeneity of 2201 administrative domains and pricing structures. As a result, the 2202 heterogeneity principle proposed in [RFC1958] needs to be supported 2203 by design [FIArch]. 2205 Example: Heterogeneity is inevitable and needs be supported by 2206 design. Multiple types of hardware must be allowed for, e.g. 2207 transmission speeds differing by at least 7 orders of magnitude, 2208 various computer word lengths, and hosts ranging from memory-starved 2209 microprocessors up to massively parallel supercomputers. Multiple 2210 types of application protocol must be allowed for, ranging from the 2211 simplest such as remote login up to the most complex such as 2212 distributed databases [RFC1958]. 2214 Impacts: - Right to freedom of expression - Right to security 2216 5.3.2.1.9. Anonymity 2218 Question(s): Did you have a look at the Privacy Considerations for 2219 Internet Protocols [RFC6973], especially section 6.1.1 ? 2221 Explanation: Anonymity refers to the condition of an identity being 2222 unknown or concealed [RFC4949]. It is an important feature for many 2223 end-users, as it allows them different degrees of privacy online. 2225 Example: Often standards expose private information, it is important 2226 to consider ways to mitigate the obvious privacy impacts. For 2227 instance, a feature which uses deep packet inspection or geolocation 2228 data could refuse to open this data to third parties, that might be 2229 able to connect the data to a physical person. 2231 Impacts: 2233 - Right to non-discrimination 2235 - Right to political participation 2237 - Right to freedom of assembly and association 2239 - Right to security 2241 5.3.2.1.10. Pseudonymity 2243 Question(s): Have you considered the Privacy Considerations for 2244 Internet Protocols [RFC6973], especially section 6.1.2 ? Does this 2245 specification collect personally derived data? Does the standard 2246 utilize data that is personally-derived, i.e. derived from the 2247 interaction of a single person, or their device or address? Does 2248 this specification generate personally derived data, and if so how 2249 will that data be handled? 2251 Explanation: Pseudonymity - the ability to disguise one's identity 2252 online - is an important feature for many end-users, as it allows 2253 them different degrees of disguised identity and privacy online. 2255 Example: Designing a standard that exposes private information, it is 2256 important to consider ways to mitigate the obvious impacts. For 2257 instance, a feature which uses deep packet inspection or geolocation 2258 data could refuse to open this data to third parties, that might be 2259 able to connect the data to a physical person. 2261 Impacts: 2263 - Right to non-discrimination 2265 - Right to freedom of assembly and association 2267 5.3.2.1.11. Accessibility 2269 Question(s): Is your protocol designed to provide an enabling 2270 environment for people who are not able-bodied? Have you looked at 2271 the W3C Web Accessibility Initiative for examples and guidance? Is 2272 your protocol optimized for low bandwidth and high latency 2273 connections? Could your protocol also be developed in a stateless 2274 manner? 2276 Explanation: The Internet is fundamentally designed to work for all 2277 people, whatever their hardware, software, language, culture, 2278 location, or physical or mental ability. When the Internet meets 2279 this goal, it is accessible to people with a diverse range of 2280 hearing, movement, sight, and cognitive ability [W3CAccessibility]. 2281 Sometimes in the design of protocols, websites, web technologies, or 2282 web tools, barriers are created that exclude people from using the 2283 Web. 2285 Example: The HTML protocol as defined in [RFC1866] specifically 2286 requires that every image must have an alt attribute (with a few 2287 exceptions for HTML5) to ensure images are accessible for people that 2288 cannot themselves decipher non-text content in web pages. 2290 Impacts: 2292 - Right to non-discrimination 2294 - Right to freedom of assembly and association 2296 - Right to education 2298 - Right to political participation 2300 5.3.2.1.12. Localization 2302 Question(s): Does your protocol uphold the standards of 2303 internationalization? Have made any concrete steps towards 2304 localizing your protocol for relevant audiences? 2306 Explanation: Localization refers to the adaptation of a product, 2307 application or document content to meet the language, cultural and 2308 other requirements of a specific target market (a locale) 2309 [W3Ci18nDef]. It is also described as the practice of translating an 2310 implementation to make it functional in a specific language or for 2311 users in a specific locale (see Internationalization). 2313 Example: The Internet is a global medium, but many of its protocols 2314 and products are developed with a certain audience in mind, that 2315 often share particular characteristics like knowing how to read and 2316 write in ASCII and knowing English. This limits the ability of a 2317 large part of the world's online population from using the Internet 2318 in a way that is culturally and linguistically accessible. An 2319 example of a protocol that has taken into account the view that 2320 individuals like to have access to data in their native language can 2321 be found in [RFC1766]. This protocol labels the information content 2322 with an identifier for the language in which it is written. And this 2323 allows information to be presented in more than one language. 2325 Impacts: 2327 - Right to non-discrimination 2329 - Right to participate in cultural life, arts and science 2331 - Right to Freedom of Expression 2333 5.3.2.1.13. Decentralization 2335 Question(s): Can your protocol be implemented without one single 2336 point of control? If applicable, can your protocol be deployed in a 2337 federated manner? What is the potential for discrimination against 2338 users of your protocol? How can use of your protocol be used to 2339 implicate users? Does your protocol create additional centralized 2340 points of control? 2342 Explanation: Decentralization is one of the central technical 2343 concepts of the architecture, and embraced as such by the IETF 2344 [RFC3935]. It refers to the absence or minimization of centralized 2345 points of control - a feature that is assumed to make it easy for new 2346 users to join and new uses to unfold {{Brown}. It also reduces issues 2347 surrounding single points of failure, and distributes the network 2348 such that it continues to function if one or several nodes are 2349 disabled. With the commercialization of the Internet in the early 2350 1990's there has been a slow move to move away from decentralization, 2351 to the detriment of the technical benefits of having a decentralized 2352 Internet. 2354 Example: The bits traveling the Internet are increasingly susceptible 2355 to monitoring and censorship, from both governments and Internet 2356 service providers, as well as third (malicious) parties. The ability 2357 to monitor and censor is further enabled by the increased 2358 centralization of the network that creates central infrastructure 2359 points that can be tapped in to. The creation of peer-to-peer 2360 networks and the development of voice-over-IP protocols using peer- 2361 to-peer technology in combination with distributed hash table (DHT) 2362 for scalability are examples of how protocols can preserve 2363 decentralization [Pouwelse]. 2365 Impacts: 2367 - Right to freedom of assembly and association 2369 5.3.2.1.14. Reliability 2371 Question(s): Is your protocol fault tolerant? Does it degrade 2372 gracefully? Do you have a documented way to announce degradation? 2373 Do you have measures in place for recovery or partial healing from 2374 failure? Can your protocol maintain dependability and performance in 2375 the face of unanticipated changes or circumstances? 2377 Explanation: Reliability ensures that a protocol will execute its 2378 function consistently and error resistant as described, and function 2379 without unexpected result. A system that is reliable degenerates 2380 gracefully and will have a documented way to announce degradation. 2381 It also has mechanisms to recover from failure gracefully, and if 2382 applicable, allow for partial healing. As with confidentiality, the 2383 growth of the Internet and fostering innovation in services depends 2384 on users having confidence and trust [RFC3724] in the network. For 2385 reliability it is necessary that services notify the users if a 2386 delivery fails. In the case of real-time systems in addition to the 2387 reliable delivery the protocol needs to safeguard timeliness. 2389 Example: In the modern IP stack structure, a reliable transport layer 2390 requires an indication that transport processing has successfully 2391 completed, such as given by TCP's ACK message [RFC0793], and not 2392 simply an indication from the IP layer that the packet arrived. 2393 Similarly, an application layer protocol may require an application- 2394 specific acknowledgement that contains, among other things, a status 2395 code indicating the disposition of the request (See [RFC3724]). 2397 Impacts: 2399 - Right to security 2401 5.3.2.1.15. Confidentiality 2403 Question(s): Does this protocol expose information related to 2404 identifiers or data? If so, does it do so to each other protocol 2405 entity (i.e., recipients, intermediaries, and enablers) [RFC6973]? 2406 What options exist for protocol implementers to choose to limit the 2407 information shared with each entity? What operational controls are 2408 available to limit the information shared with each entity? 2410 What controls or consent mechanisms does the protocol define or 2411 require before personal data or identifiers are shared or exposed via 2412 the protocol? If no such mechanisms or controls are specified, is it 2413 expected that control and consent will be handled outside of the 2414 protocol? 2416 Does the protocol provide ways for initiators to share different 2417 pieces of information with different recipients? If not, are there 2418 mechanisms that exist outside of the protocol to provide initiators 2419 with such control? 2421 Does the protocol provide ways for initiators to limit which 2422 information is shared with intermediaries? If not, are there 2423 mechanisms that exist outside of the protocol to provide users with 2424 such control? Is it expected that users will have relationships that 2425 govern the use of the information (contractual or otherwise) with 2426 those who operate these intermediaries? Does the protocol prefer 2427 encryption over clear text operation? 2429 Does the protocol provide ways for initiators to express individuals' 2430 preferences to recipients or intermediaries with regard to the 2431 collection, use, or disclosure of their personal data? 2433 Explanation: Confidentiality refers to keeping your data secret from 2434 unintended listeners [RFC3552]. The growth of the Internet depends 2435 on users having confidence that the network protects their private 2436 information [RFC1984]. 2438 Example: Protocols that do not encrypt their payload make the entire 2439 content of the communication available to the idealized attacker 2440 along their path. Following the advice in [RFC3365], most such 2441 protocols have a secure variant that encrypts the payload for 2442 confidentiality, and these secure variants are seeing ever-wider 2443 deployment. A noteworthy exception is DNS [RFC1035], as DNSSEC 2444 [RFC4033]does not have confidentiality as a requirement. This 2445 implies that, in the absence of changes to the protocol as presently 2446 under development in the IETF's DNS Private Exchange (DPRIVE) working 2447 group, all DNS queries and answers generated by the activities of any 2448 protocol are available to the attacker. When store-and-forward 2449 protocols are used (e.g., SMTP [RFC5321]), intermediaries leave this 2450 data subject to observation by an attacker that has compromised these 2451 intermediaries, unless the data is encrypted end-to-end by the 2452 application-layer protocol or the implementation uses an encrypted 2453 store for this data [RFC7624]. 2455 Impacts: 2457 - Right to security 2459 5.3.2.1.16. Integrity 2461 Question(s): Does your protocol maintain and assure the accuracy of 2462 data? Does your protocol maintain and assure the consistency of 2463 data? Does your protocol in any way allow for the data to be 2464 (intentionally or unintentionally) altered? 2466 Explanation: Integrity refers to the maintenance and assurance of the 2467 accuracy and consistency of data to ensure it has not been 2468 (intentionally or unintentionally) altered. 2470 Example: See authenticity 2472 Impacts: 2474 - Right to security 2476 5.3.2.1.17. Authenticity 2478 Question(s): Do you have sufficient measures to confirm the truth of 2479 an attribute of a single piece of data or entity? Can the attributes 2480 get garbled along the way (see security)? If relevant have you 2481 implemented IPsec, DNSsec, HTTPS and other Standard Security Best 2482 Practices? 2484 Explanation: Authenticity ensures that data does indeed come from the 2485 source it claims to come from. This is important to prevent attacks 2486 or unauthorized access and use of data. 2488 Example: Authentication of data is important to prevent 2489 vulnerabilities and attacks, like man-in-the-middle-attacks. These 2490 attacks happen when a third party (often for malicious reasons) 2491 intercepts a communication between two parties, inserting themselves 2492 in the middle and posing as both parties. In practice this looks as 2493 follows: 2495 Alice wants to communicate with Bob. 2496 Alice sends data to Bob. 2497 Corinne intercepts the data sent to Bob. 2498 Corinne reads and alters the message to Bob. 2499 Bob cannot see the data did not come from Alice but from Corinne. 2500 Corinne intercepts and alters the communication as it is sent between 2501 Alice and Bob. 2502 Corinne knows all. 2504 Impacts: 2506 - Right to security 2508 5.3.2.1.18. Acceptability 2510 Question(s): Do your protocols follow the principle of non- 2511 discrimination? Do your protocols follow the principle of content 2512 agnosticism? Does your protocol take into account the needs of 2513 special needs (Internet) groups, like the audio-visually impaired? 2514 Also see availability. 2516 Explanation: The Internet is a global medium. Yet, there continue to 2517 be issues surrounding acceptability - the extent to which standards 2518 are non-discriminatory and relevant to the widest range of end-users 2519 - that need to be resolved. Many standards are not suitable for end- 2520 users who are not-ablebodied, or otherwise restricted in their 2521 ability to access the Internet in its current form (text, data and 2522 English heavy). Development of new standards should consider the 2523 ways in which they exclude or include non-traditional user 2524 communities. 2526 Example: Designing a feature that could make access to websites for 2527 non-able bodied people more difficult. 2529 Impacts: 2531 - Right to education 2533 - Right to freedom of expression 2535 - Right to freedom of assembly and association 2537 5.3.2.1.19. Adaptability 2539 Question(s): Is your protocol written in such a way that is would be 2540 easy for other protocols to be developed on top of it, or to interact 2541 with it? Does your protocol impact permissionless innovation? See 2542 'Connectivity' above. 2544 Explanation: Adaptability is closely interrelated permissionless 2545 innovation, both maintain the freedom and ability to freely create 2546 and deploy new protocols on top of the communications constructs that 2547 currently exist. It is at the heart of the Internet as we know it, 2548 and to maintain its fundamentally open nature, we need to be mindful 2549 of the impact of protocols on maintaining or reducing permissionless 2550 innovation to ensure the Internet can continue to develop. 2552 Example: WebRTC generates audio and/or video data. In order to 2553 ensure that WebRTC can be used in different locations by different 2554 parties it is important that standard Javascript APIs are developed 2555 to support applications from different voice service providers. 2556 Multiple parties will have similar capabilities, in order to ensure 2557 that all parties can build upon existing standards these need to be 2558 adaptable, and allow for permissionless innovation. 2560 Impacts: 2562 - Right to education 2564 - Freedom of expression 2566 - Freedom of assembly and association 2568 6. Acknowledgements 2570 A special thanks to all members of the hrpc RG who contributed to 2571 this draft. The following deserve a special mention: 2573 - Joana Varon for helping draft the first iteration of the 2574 methodology, previous drafts and the direction of the film Net of 2575 Rights and working on the interviews at IETF92 in Dallas. 2577 - Daniel Kahn Gillmor (dkg) for helping with the first iteration of 2578 the glossary as well as a lot of technical guidance, support and 2579 language suggestions. 2581 - Claudio Guarnieri for writing the first iterations of the case 2582 studies on VPN, HTTP, and Peer to Peer. 2584 - Will Scott for writing the first iterations of the case studies on 2585 DNS, IP, XMPP. 2587 - Avri Doria for proposing writing a glossary in the first place, 2588 help writing the initial proposals and Internet Drafts and 2589 contributing to the glossary. 2591 and Stephane Bortzmeyer, John Curran, Barry Shein, Joe Hall, Joss 2592 Wright, Harry Halpin, and Tim Sammut who made a lot of excellent 2593 suggestions, many of which found their way directly into the text. 2594 We want to thank Stephane Bortzemeyer, Shane Kerr, Giovane Moura, 2595 James Gannon, and Scott Craig for their reviews and testing the HRPC 2596 guidelines in the wild. We would also like to thank Molly Sauter, 2597 Arturo Filasto, Nathalie Marechal, Eleanor Saitta and all others who 2598 provided input on the draft or the conceptualization of the idea. 2600 7. Security Considerations 2602 As this document concerns a research document, there are no security 2603 considerations. 2605 8. IANA Considerations 2607 This document has no actions for IANA. 2609 9. Research Group Information 2611 The discussion list for the IRTF Human Rights Protocol Considerations 2612 proposed working group is located at the e-mail address hrpc@ietf.org 2613 [3]. Information on the group and information on how to subscribe to 2614 the list is at https://www.irtf.org/mailman/listinfo/hrpc 2616 Archives of the list can be found at: https://www.irtf.org/mail- 2617 archive/web/hrpc/current/index.html 2619 10. References 2621 10.1. Informative References 2623 [Abbate] Abbate, J., "Inventing the Internet", MIT Press , 2000, 2624 . 2626 [Abibil] Danchev, D., "Dissecting 'Operation Ababil' - an OSINT 2627 Analysis", 2012, . 2630 [Adrian] Adrian, D., Bhargavan, K., Durumeric, Z., Gaudry, P., 2631 Green, M., Halderman, J., Heninger, N., Springall, D., 2632 Thome, E., Valenta, L., VanderSloot, B., Wustrow, E., 2633 Zanella Beguelin, S., and P. Zimmermann, "Imperfect 2634 Forward Secrecy: How Diffie-Hellman Fails in Practice", 2635 ACM Conference on Computer and Communications Security 2636 2015: 5-17 , 2015. 2638 [Appelbaum] 2639 Appelbaum, J., Gibson, A., Kabish, V., Kampf, L., and L. 2640 Ryge, "NSA targets the privacy-conscious", 2015, 2641 . 2644 [Babbie] Babbie, E., "The Basics of Social Research", Belmont CA 2645 Cengage , 2010. 2647 [Benkler] Benkler, Y., "The wealth of Networks - How social 2648 production transforms markets and freedom", New Haven and 2649 London - Yale University Press , 2006, 2650 . 2652 [Berners-Lee] 2653 Berners-Lee, T. and M. Fischetti, "Weaving the Web,", 2654 HarperCollins p 208, 1999. 2656 [BernersLeeHalpin] 2657 Berners-Lee, T. and H. Halpin, "Defend the Web", 2012, 2658 . 2661 [Bhargavan] 2662 Bhargavan, K., Delignat-Lavaud, A., Fournet, C., Pironti, 2663 A., and P. Strub, "Triple Handshakes and Cookie Cutters: 2664 Breaking and Fixing Authentication over TLS", IEEE 2665 Symposium on Security and Privacy 2014: 98-113 , 2014. 2667 [Bless] Bless, R. and C. Orwat, "Values and Networks", 2015. 2669 [Blumenthal] 2670 Blumenthal, M. and D. Clark, "Rethinking the design of the 2671 Internet: The end-to-end arguments vs. the brave new 2672 world", ACM Transactions on Internet Technology, Vol. 1, 2673 No. 1, August 2001, pp 70-109. , 2001. 2675 [Bray] Bray, T., "A New HTTP Status Code for Legally-restricted 2676 Resources", 2016, . 2679 [Broeders] 2680 Broeders, D., "The public core of the Internet", WRR , 2681 2015, 2682 . 2685 [Brown] Brown, I. and M. Ziewitz, "A Prehistory of Internet 2686 Governance", Research Handbook on Governance of the 2687 Internet. Cheltenham, Edward Elgar. , 2013. 2689 [BrownMarsden] 2690 Brown, I. and C. Marsden, "Regulating code", MIT Press , 2691 2013, . 2693 [Brownetal] 2694 Brown, I., Clark, D., and D. Trossen, "Should specific 2695 values be embedded in the Internet Architecture?", Sigcomm 2696 , 2010, . 2699 [Cath] Cath, C., "A Case Study of Coding Rights: Should Freedom 2700 of Speech Be Instantiated in the Protocols and Standards 2701 Designed by the Internet Engineering Task Force?", 2015, 2702 . 2705 [CathFloridi] 2706 Cath, C. and L. Floridi, "The Design of the Internet's 2707 Architecture by the Internet Engineering Task Force (IETF) 2708 and Human Rights", September 2016. 2710 [Clark] Clark, D., "The Design Philosophy of the DARPA Internet 2711 Protocols", Proc SIGCOMM 88, ACM CCR Vol 18, Number 4, 2712 August 1988, pp. 106-114. , 1988. 2714 [Clarketal] 2715 Clark, D., Wroclawski, J., Sollins, K., and R. Braden, 2716 "Tussle in cyberspace - defining tomorrow's Internet", ACM 2717 Digital Library , 2005, . 2720 [Collins] Collins, K., "Hacking Team's oppressive regimes customer 2721 list revealed in hack", 2015, 2722 . 2725 [Daedalus] 2726 Clark, D., "The Contingent Internet", Daedalus Winter 2727 2016, Vol. 145, No. 1. p. 9-17 , 2016, 2728 . 2730 [Davidsonetal] 2731 Davidson, A., Morris, J., and R. Courtney, "Strangers in a 2732 strange land", Telecommunications Policy Research 2733 Conference , 2002, 2734 . 2736 [Denardis14] 2737 Denardis, L., "The Global War for Internet Governance", 2738 Yale University Press , 2014, 2739 . 2741 [Denardis15] 2742 Denardis, L., "The Internet Design Tension between 2743 Surveillance and Security", IEEE Annals of the History of 2744 Computing (volume 37-2) , 2015, . 2746 [Denzin] Denzin, N. and Y. Lincoln, "Handbook of Qualitative 2747 Research", Thousand Oaks CA Sage , 2000, 2748 . 2751 [Doty] Doty, N., "Automated text analysis of Requests for Comment 2752 (RFCs)", 2014, . 2754 [Douceur] Douceur, J., "The Sybil Attack", 2002, 2755 . 2758 [Dutton] Dutton, W., "Freedom of Connection, Freedom of Expression: 2759 the Changing legal and regulatory Ecology Shaping the 2760 Internet.", 2011, . 2763 [Elahi] Elahi, T. and I. Goldberg, "CORDON - A taxonomy of 2764 Internet Censorship Resistance Strategies", 2012, 2765 . 2768 [FIArch] "Future Internet Design Principles", January 2012, 2769 . 2772 [FRAMEWORK] 2773 ISO/IEC, ., "Information technology - Framework for 2774 internationalization, prepared by ISO/IEC JTC 1/SC 22/WG 2775 20 ISO/IEC TR 11017", 1997. 2777 [Franklin] 2778 Franklin, U., "The Real World of Technology", 1999, 2779 . 2782 [Geertz] Clifford, G., "Kinship in Bali", Chicago University of 2783 Chicago Press. , 1975, 2784 . 2787 [Googlepatent] 2788 Google, ., "Method and device for network traffic 2789 manipulation", 2012, . 2792 [GreenMovement] 2793 Villeneuve, N., "Iran DDoS", 2009, 2794 . 2796 [HRC2012] United Nations Human Rights Council, "UN General Assembly 2797 Resolution "The right to privacy in the digital age" 2798 (A/C.3/68/L.45)", 2011, 2799 . 2801 [Haagsma] Haagsma, L., "Deep dive into QUANTUM INSERT", 2015, 2802 . 2805 [ICCPR] United Nations General Assembly, "International Covenant 2806 on Civil and Political Rights", 1976, 2807 . 2810 [ICESCR] United Nations General Assembly, "International Covenant 2811 on Economic, Social and Cultural Rights", 1966, 2812 . 2815 [Jabri] Jabri, V., "Discourses on Violence - conflict analysis 2816 reconsidered", Manchester University Press , 1996. 2818 [Kaye] Kaye, D., "Report of the Special Rapporteur on the 2819 promotion and protection of the right to freedom of 2820 opinion and expression", 2016, 2821 . 2824 [King] King, C., "Power, Social Violence and Civil Wars", 2825 Washington D.C. United States Institute of Peace Press , 2826 2007. 2828 [Lessig] Lessig, L., "Code - And Other Laws of Cyberspace, Version 2829 2.0.", New York Basic Books , 2006, . 2831 [Marcak] Marcak, B., Weaver, N., Dalek, J., Ensafi, R., Fifield, 2832 D., McKune, S., Rey, A., Scott-Railton, J., Deibert, R., 2833 and V. Paxson, "China's Great Fire Cannon", 2015, 2834 . 2836 [Marquis-Boire] 2837 Marquis-Boire, M., "Schrodinger's Cat Video and the Death 2838 of Clear-Text", 2014, . 2841 [Mueller] Mueller, M., "Networks and States", MIT Press , 2010, 2842 . 2844 [Musiani] Musiani, F., "Giants, Dwarfs and Decentralized 2845 Alternatives to Internet-based Services - An Issue of 2846 Internet Governance", Westminister Papers in Communication 2847 and Culture , 2015, . 2849 [NETmundial] 2850 NETmundial, "NETmundial Multistakeholder Statement", 2014, 2851 . 2854 [PETS2015VPN] 2855 Pera, V., Barbera, M., Tyson, G., Haddadi, H., and A. Mei, 2856 "A Glance through the VPN Looking Glass", 2015, 2857 . 2860 [Penney] Penney, J., "Chilling Effects: Online Surveillance and 2861 Wikipedia Use", 2016, . 2864 [Peterson] 2865 Peterson, A., Gellman, B., and A. Soltani, "Yahoo to make 2866 SSL encryption the default for Webmail users. Finally.", 2867 2013, . 2870 [Pouwelse] 2871 Pouwelse, Ed, J., "Media without censorship", 2012, 2872 . 2875 [RFC0226] Karp, P., "Standardization of host mnemonics", RFC 226, 2876 DOI 10.17487/RFC0226, September 1971, 2877 . 2879 [RFC0760] Postel, J., "DoD standard Internet Protocol", RFC 760, DOI 2880 10.17487/RFC0760, January 1980, 2881 . 2883 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, DOI 2884 10.17487/RFC0791, September 1981, 2885 . 2887 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, RFC 2888 793, DOI 10.17487/RFC0793, September 1981, 2889 . 2891 [RFC0894] Hornig, C., "A Standard for the Transmission of IP 2892 Datagrams over Ethernet Networks", STD 41, RFC 894, DOI 2893 10.17487/RFC0894, April 1984, 2894 . 2896 [RFC1035] Mockapetris, P., "Domain names - implementation and 2897 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 2898 November 1987, . 2900 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 2901 Communication Layers", STD 3, RFC 1122, DOI 10.17487/ 2902 RFC1122, October 1989, 2903 . 2905 [RFC1631] Egevang, K. and P. Francis, "The IP Network Address 2906 Translator (NAT)", RFC 1631, DOI 10.17487/RFC1631, May 2907 1994, . 2909 [RFC1766] Alvestrand, H., "Tags for the Identification of 2910 Languages", RFC 1766, DOI 10.17487/RFC1766, March 1995, 2911 . 2913 [RFC1866] Berners-Lee, T. and D. Connolly, "Hypertext Markup 2914 Language - 2.0", RFC 1866, DOI 10.17487/RFC1866, November 2915 1995, . 2917 [RFC1958] Carpenter, B., Ed., "Architectural Principles of the 2918 Internet", RFC 1958, DOI 10.17487/RFC1958, June 1996, 2919 . 2921 [RFC1984] IAB and IESG, "IAB and IESG Statement on Cryptographic 2922 Technology and the Internet", BCP 200, RFC 1984, DOI 2923 10.17487/RFC1984, August 1996, 2924 . 2926 [RFC2026] Bradner, S., "The Internet Standards Process -- Revision 2927 3", BCP 9, RFC 2026, DOI 10.17487/RFC2026, October 1996, 2928 . 2930 [RFC2277] Alvestrand, H., "IETF Policy on Character Sets and 2931 Languages", BCP 18, RFC 2277, DOI 10.17487/RFC2277, 2932 January 1998, . 2934 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2935 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 2936 December 1998, . 2938 [RFC2606] Eastlake 3rd, D. and A. Panitz, "Reserved Top Level DNS 2939 Names", BCP 32, RFC 2606, DOI 10.17487/RFC2606, June 1999, 2940 . 2942 [RFC2775] Carpenter, B., "Internet Transparency", RFC 2775, DOI 2943 10.17487/RFC2775, February 2000, 2944 . 2946 [RFC3365] Schiller, J., "Strong Security Requirements for Internet 2947 Engineering Task Force Standard Protocols", BCP 61, RFC 2948 3365, DOI 10.17487/RFC3365, August 2002, 2949 . 2951 [RFC3536] Hoffman, P., "Terminology Used in Internationalization in 2952 the IETF", RFC 3536, DOI 10.17487/RFC3536, May 2003, 2953 . 2955 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 2956 Text on Security Considerations", BCP 72, RFC 3552, DOI 2957 10.17487/RFC3552, July 2003, 2958 . 2960 [RFC3724] Kempf, J., Ed., Austein, R., Ed., and IAB, "The Rise of 2961 the Middle and the Future of End-to-End: Reflections on 2962 the Evolution of the Internet Architecture", RFC 3724, DOI 2963 10.17487/RFC3724, March 2004, 2964 . 2966 [RFC3935] Alvestrand, H., "A Mission Statement for the IETF", BCP 2967 95, RFC 3935, DOI 10.17487/RFC3935, October 2004, 2968 . 2970 [RFC3979] Bradner, S., Ed., "Intellectual Property Rights in IETF 2971 Technology", BCP 79, RFC 3979, DOI 10.17487/RFC3979, March 2972 2005, . 2974 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 2975 Rose, "DNS Security Introduction and Requirements", RFC 2976 4033, DOI 10.17487/RFC4033, March 2005, 2977 . 2979 [RFC4084] Klensin, J., "Terminology for Describing Internet 2980 Connectivity", BCP 104, RFC 4084, DOI 10.17487/RFC4084, 2981 May 2005, . 2983 [RFC4101] Rescorla, E. and IAB, "Writing Protocol Models", RFC 4101, 2984 DOI 10.17487/RFC4101, June 2005, 2985 . 2987 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", RFC 2988 4303, DOI 10.17487/RFC4303, December 2005, 2989 . 2991 [RFC4906] Martini, L., Ed., Rosen, E., Ed., and N. El-Aawar, Ed., 2992 "Transport of Layer 2 Frames Over MPLS", RFC 4906, DOI 2993 10.17487/RFC4906, June 2007, 2994 . 2996 [RFC4949] Shirey, R., "Internet Security Glossary, Version 2", FYI 2997 36, RFC 4949, DOI 10.17487/RFC4949, August 2007, 2998 . 3000 [RFC5321] Klensin, J., "Simple Mail Transfer Protocol", RFC 5321, 3001 DOI 10.17487/RFC5321, October 2008, 3002 . 3004 [RFC5944] Perkins, C., Ed., "IP Mobility Support for IPv4, Revised", 3005 RFC 5944, DOI 10.17487/RFC5944, November 2010, 3006 . 3008 [RFC6108] Chung, C., Kasyanov, A., Livingood, J., Mody, N., and B. 3009 Van Lieu, "Comcast's Web Notification System Design", RFC 3010 6108, DOI 10.17487/RFC6108, February 2011, 3011 . 3013 [RFC6120] Saint-Andre, P., "Extensible Messaging and Presence 3014 Protocol (XMPP): Core", RFC 6120, DOI 10.17487/RFC6120, 3015 March 2011, . 3017 [RFC6365] Hoffman, P. and J. Klensin, "Terminology Used in 3018 Internationalization in the IETF", BCP 166, RFC 6365, DOI 3019 10.17487/RFC6365, September 2011, 3020 . 3022 [RFC6698] Hoffman, P. and J. Schlyter, "The DNS-Based Authentication 3023 of Named Entities (DANE) Transport Layer Security (TLS) 3024 Protocol: TLSA", RFC 6698, DOI 10.17487/RFC6698, August 3025 2012, . 3027 [RFC6701] Farrel, A. and P. Resnick, "Sanctions Available for 3028 Application to Violators of IETF IPR Policy", RFC 6701, 3029 DOI 10.17487/RFC6701, August 2012, 3030 . 3032 [RFC6797] Hodges, J., Jackson, C., and A. Barth, "HTTP Strict 3033 Transport Security (HSTS)", RFC 6797, DOI 10.17487/ 3034 RFC6797, November 2012, 3035 . 3037 [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 3038 Morris, J., Hansen, M., and R. Smith, "Privacy 3039 Considerations for Internet Protocols", RFC 6973, DOI 3040 10.17487/RFC6973, July 2013, 3041 . 3043 [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an 3044 Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 3045 2014, . 3047 [RFC7469] Evans, C., Palmer, C., and R. Sleevi, "Public Key Pinning 3048 Extension for HTTP", RFC 7469, DOI 10.17487/RFC7469, April 3049 2015, . 3051 [RFC7540] Belshe, M., Peon, R., and M. Thomson, Ed., "Hypertext 3052 Transfer Protocol Version 2 (HTTP/2)", RFC 7540, DOI 3053 10.17487/RFC7540, May 2015, 3054 . 3056 [RFC7574] Bakker, A., Petrocco, R., and V. Grishchenko, "Peer-to- 3057 Peer Streaming Peer Protocol (PPSPP)", RFC 7574, DOI 3058 10.17487/RFC7574, July 2015, 3059 . 3061 [RFC7624] Barnes, R., Schneier, B., Jennings, C., Hardie, T., 3062 Trammell, B., Huitema, C., and D. Borkmann, 3063 "Confidentiality in the Face of Pervasive Surveillance: A 3064 Threat Model and Problem Statement", RFC 7624, DOI 3065 10.17487/RFC7624, August 2015, 3066 . 3068 [RFC7626] Bortzmeyer, S., "DNS Privacy Considerations", RFC 7626, 3069 DOI 10.17487/RFC7626, August 2015, 3070 . 3072 [RFC7725] Bray, T., "An HTTP Status Code to Report Legal Obstacles", 3073 RFC 7725, DOI 10.17487/RFC7725, February 2016, 3074 . 3076 [RSF] RSF, "Syria using 34 Blue Coat Servers to spy on Internet 3077 users", 2013, . 3080 [Rachovitsa] 3081 Rachovitsa, A., "Engineering 'Privacy by Design' in the 3082 Internet Protocols - Understanding Online Privacy both as 3083 a Technical and a Human Rights Issue in the Face of 3084 Pervasive Monitoring", International Journal of Law and 3085 Information Technology , 2015, . 3088 [Richie] Richie, J. and J. Lewis, "Qualitative Research Practice - 3089 A Guide for Social Science Students and Researchers", 3090 London Sage , 2003, . 3094 [Rideout] Rideout, A., "Making security easier", 2008, 3095 . 3098 [Saltzer] Saltzer, J., Reed, D., and D. Clark, "End-to-End Arguments 3099 in System Design", ACM TOCS, Vol 2, Number 4, November 3100 1984, pp 277-288. , 1984. 3102 [Sauter] Sauter, M., "The Coming Swarm", Bloomsbury, London , 2014. 3104 [Schillace] 3105 Schillace, S., "Default https access for Gmail", 2010, 3106 . 3109 [Schneier] 3110 Schneier, B., "Attacking Tor - how the NSA targets users' 3111 online anonymity", 2013, 3112 . 3115 [Schroeder] 3116 Schroeder, I. and B. Schmidt, "Introduction - Violent 3117 Imaginaries and Violent Practice", London and New York 3118 Routledge , 2001, . 3122 [UDHR] United Nations General Assembly, "The Universal 3123 Declaration of Human Rights", 1948, 3124 . 3126 [UNGA2013] 3127 United Nations General Assembly, "UN General Assembly 3128 Resolution "The right to privacy in the digital age" 3129 (A/C.3/68/L.45)", 2013, 3130 . 3132 [W3CAccessibility] 3133 W3C, "Accessibility", 2015, 3134 . 3136 [W3Ci18nDef] 3137 W3C, "Localization vs. Internationalization", 2010, 3138 . 3140 [WP-Debugging] 3141 "Debugging", n.d., . 3144 [WP-Stateless] 3145 "Stateless protocol", n.d., 3146 . 3148 [Walfish] Walfish, M., Stribling, J., Krohn, M., Balakrishnan, H., 3149 Morris, R., and S. Shenker, "Middleboxes No Longer 3150 Considered Harmful", 2004, . 3152 [WynsbergheMoura] 3153 Wynsberghe, A. and G. Moura, "The concept of embedded 3154 values and the example of internet security", 2013, 3155 . 3157 [Zittrain] 3158 Zittrain, J., "The Future of the Internet - And How to 3159 Stop It", Yale University Press , 2008, 3160 . 3163 [Zuckerman] 3164 Zuckerman, E., Roberts, H., McGrady, R., York, J., and J. 3165 Palfrey, "Report on Distributed Denial of Service (DDoS) 3166 Attacks", The Berkman Center for Internet and Society at 3167 Harvard University , 2010, 3168 . 3172 [ars] Anderson, N., "P2P researchers - use a blocklist or you 3173 will be tracked... 100% of the time", 2007, 3174 . 3178 [bbc-wikileaks] 3179 BBC, "Whistle-blower site taken offline", 2008, 3180 . 3182 [bitmessage] 3183 Bitmessage, "Bitmessage Wiki?", 2014, 3184 . 3186 [caida] Dainotti, A., Squarcella, C., Aben, E., Claffy, K., 3187 Chiesa, M., Russo, M., and A. Pescape, "Analysis of 3188 Country-wide Internet Outages Caused by Censorship", 2013, 3189 . 3192 [freenet1] 3193 Freenet, "What is Freenet?", n.d., 3194 . 3196 [freenet2] 3197 Ian Clarke, ., "The Philosphy behind Freenet?", n.d., 3198 . 3200 [greatfirewall] 3201 Anonymous, ., "Towards a Comprehensive Picture of the 3202 Great Firewall's DNS Censorship", 2014, 3203 . 3206 [hall] Hall, J., Aaron, M., and B. Jones, "A Survey of Worldwide 3207 Censorship Techniques", 2015, 3208 . 3211 [namecoin] 3212 Namecoin, "Namecoin - Decentralized secure names", 2015, 3213 . 3215 [natusage] 3216 Maier, G., Schneider, F., and A. Feldmann, "NAT usage in 3217 Residential Broadband networks", 2011, 3218 . 3221 [newegg] Mullin, J., "Newegg on trial: Mystery company TQP rewrites 3222 the history of encryption", 2013, . 3226 [notewell] 3227 IETF, "Note Well", 2015, . 3230 [patentpolicy] 3231 W3C, "W3C Patent Policy", 2004, 3232 . 3234 [pidgin] js, . and Pidgin Developers, "-XMPP- Invisible mode 3235 violating standard", July 2015, 3236 . 3238 [quic] The Chromium Project, "QUIC, a multiplexed stream 3239 transport over UDP", 2014, . 3242 [spdy] The Chromium Project, "SPDY - An experimental protocol for 3243 a faster web", 2009, . 3246 [spiegel] SPIEGEL, "Prying Eyes - Inside the NSA's War on Internet 3247 Security", 2014, 3248 . 3251 [sslstrip] 3252 Marlinspike, M., "Software >> sslstrip", 2011, 3253 . 3255 [techyum] Violet, ., "Official - vb.ly Link Shortener Seized by 3256 Libyan Government", 2010, . 3260 [torproject] 3261 The Tor Project, ., "Tor Project - Anonymity Online", 3262 2007, . 3264 [torrentfreak1] 3265 Van der Sar, E., "Proposal for research on human rights 3266 protocol considerations", 2015, . 3270 [torrentfreak2] 3271 Andy, ., "LAWYERS SENT 109,000 PIRACY THREATS IN GERMANY 3272 DURING 2013", 2014, . 3276 [tribler] Delft University of Technology, Department EWI/PDS/ 3277 Tribler, "About Tribler", 2013, . 3280 [ververis] 3281 Vasilis, V., Kargiotakis, G., Filasto, A., Fabian, B., and 3282 A. Alexandros, "Understanding Internet Censorship Policy - 3283 The Case of Greece", 2015, 3284 . 3287 [wikileaks] 3288 Sladek, T. and E. Broese, "Market Survey : Detection & 3289 Filtering Solutions to Identify File Transfer of Copyright 3290 Protected Content for Warner Bros. and movielabs", 2011, 3291 . 3294 [xmppmanifesto] 3295 Saint-Andre, P. and . XMPP Operators, "A Public Statement 3296 Regarding Ubiquitous Encryption on the XMPP Network", 3297 2014, 3298 . 3301 10.2. URIs 3303 [1] mailto:node@domain/home 3305 [2] mailto:node@domain/work 3307 [3] mailto:hrpc@ietf.org 3309 Authors' Addresses 3311 Niels ten Oever 3312 Article19 3314 EMail: niels@article19.org 3316 Corinne Cath 3317 Oxford Internet Institute 3319 EMail: corinnecath@gmail.com