idnits 2.17.1 draft-tenoever-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 : ---------------------------------------------------------------------------- ** There is 1 instance of too long lines in the document, the longest one being 14 characters in excess of 72. ** 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 == Line 173 has weird spacing: '...f other indiv...' == Line 274 has weird spacing: '...s to be appre...' == Line 295 has weird spacing: '...rse, to inter...' == Line 371 has weird spacing: '...itution and i...' == Line 559 has weird spacing: '...lved in both ...' == (4 more instances...) -- The document date (March 17, 2016) is 2955 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 2653 -- Looks like a reference, but probably isn't: '2' on line 2655 -- Looks like a reference, but probably isn't: '3' on line 2657 -- 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 2460 (Obsoleted by RFC 8200) -- Obsolete informational reference (is this intentional?): RFC 7626 (Obsoleted by RFC 9076) Summary: 2 errors (**), 0 flaws (~~), 7 warnings (==), 10 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: September 18, 2016 Oxford Internet Institute 6 March 17, 2016 8 Research into Human Rights Protocol Considerations 9 draft-tenoever-hrpc-research-00 11 Abstract 13 The increased intertwinement of Internet and society increases the 14 impact of the Internet on the lives of individuals. Because of this, 15 the design and development of the architecture of the Internet also 16 has an increasing impact on society. This has led to an increasing 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 strenghtened as a human rights enabeling environment [Brown]. 22 This document provides a proposal for a glossary to discuss the 23 relation between human rights and Internet protocols, an overview of 24 the discussion, a proposal for the mapping of the relation between 25 human rights and technical concepts, and a proposal for guidelines 26 for human rights considerations, similar to the work done on the 27 guidelines for privacy considerations [RFC6973]. 29 Discussion of this draft at: hrpc@irtf.org // 30 https://www.irtf.org/mailman/listinfo/hrpc 32 Status of This Memo 34 This Internet-Draft is submitted in full conformance with the 35 provisions of BCP 78 and BCP 79. 37 Internet-Drafts are working documents of the Internet Engineering 38 Task Force (IETF). Note that other groups may also distribute 39 working documents as Internet-Drafts. The list of current Internet- 40 Drafts is at http://datatracker.ietf.org/drafts/current/. 42 Internet-Drafts are draft documents valid for a maximum of six months 43 and may be updated, replaced, or obsoleted by other documents at any 44 time. It is inappropriate to use Internet-Drafts as reference 45 material or to cite them other than as "work in progress." 47 This Internet-Draft will expire on September 18, 2016. 49 Copyright Notice 51 Copyright (c) 2016 IETF Trust and the persons identified as the 52 document authors. All rights reserved. 54 This document is subject to BCP 78 and the IETF Trust's Legal 55 Provisions Relating to IETF Documents 56 (http://trustee.ietf.org/license-info) in effect on the date of 57 publication of this document. Please review these documents 58 carefully, as they describe your rights and restrictions with respect 59 to this document. Code Components extracted from this document must 60 include Simplified BSD License text as described in Section 4.e of 61 the Trust Legal Provisions and are provided without warranty as 62 described in the Simplified BSD License. 64 Table of Contents 66 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 67 2. Vocabulary used . . . . . . . . . . . . . . . . . . . . . . . 4 68 3. Research Questions . . . . . . . . . . . . . . . . . . . . . 10 69 4. Literature and Discussion Review . . . . . . . . . . . . . . 10 70 5. Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 12 71 5.1. Data Sources . . . . . . . . . . . . . . . . . . . . . . 13 72 5.1.1. Discourse analysis of RFCs . . . . . . . . . . . . . 13 73 5.1.2. Interviews with members of the IETF community during 74 IETF92 in Dallas . . . . . . . . . . . . . . . . . . 14 75 5.1.3. Participant observation in Working Groups . . . . . . 14 76 5.2. Data analysis strategies . . . . . . . . . . . . . . . . 14 77 5.2.1. Identifying qualities of technical concepts that 78 relate to human rights . . . . . . . . . . . . . . . 14 79 5.2.2. Translation human rights to technical terms . . . . . 16 80 5.2.3. IP . . . . . . . . . . . . . . . . . . . . . . . . . 17 81 5.2.4. DNS . . . . . . . . . . . . . . . . . . . . . . . . . 20 82 5.2.5. HTTP . . . . . . . . . . . . . . . . . . . . . . . . 22 83 5.2.6. XMPP . . . . . . . . . . . . . . . . . . . . . . . . 25 84 5.2.7. Peer to Peer . . . . . . . . . . . . . . . . . . . . 27 85 5.2.8. Virtual Private Network . . . . . . . . . . . . . . . 29 86 5.2.9. HTTP Status Code 451 . . . . . . . . . . . . . . . . 32 87 5.2.10. Middleboxes . . . . . . . . . . . . . . . . . . . . . 33 88 5.2.11. DDOS attacks . . . . . . . . . . . . . . . . . . . . 34 89 5.3. Model for developing human rights protocol considerations 37 90 5.3.1. Human rights threats . . . . . . . . . . . . . . . . 37 91 5.3.2. Guidelines for human rights considerations . . . . . 38 92 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 44 93 7. Security Considerations . . . . . . . . . . . . . . . . . . . 45 94 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 45 95 9. Research Group Information . . . . . . . . . . . . . . . . . 45 96 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 45 97 10.1. Normative References . . . . . . . . . . . . . . . . . . 45 98 10.2. Informative References . . . . . . . . . . . . . . . . . 46 99 10.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 56 101 1. Introduction 103 "There's a freedom about the Internet: As long as we accept the 104 rules of sending packets around, we can send packets containing 105 anything to anywhere." 107 [Berners-Lee] 109 This document aims to expose the relation between protocols and human 110 rights, propose possible guidelines to protect the Internet as a 111 human-rights-enabling environment in future protocol development, in 112 a manner similar to the work done for Privacy Considerations in 113 [RFC6973], and to to increase the awareness in both the human rights 114 community and the technical community on the importance of the 115 technical workings of the Internet and its impact on human rights. 117 Open, secure and reliable connectivity is necessary (although not 118 sufficient) to excercise the human rights such as freedom of 119 expression and freedom of association, as defined in the Universal 120 Declaration of Human Rights [UDHR]. The Internet aims to be a global 121 network of networks that provides unfettered connectivity to all 122 users at all times and for any content [RFC1958]. This objective of 123 stimulating global connectivity contributes to the Internets role as 124 an enabler of human rights. Next to that, the strong commitment to 125 security [RFC1984][RFC3365] and privacy [RFC6973] [RFC7258] in the 126 Internets architectural design equally strongly contributes to the 127 strenghtening of the Internet as a human rights enabeling 128 environment. One could even argue that the Internet is not only an 129 enabler of human rights, but that human rights lie at the basis of, 130 and are ingrained in, the architecture of the network. 132 While the Internet was designed with freedom and openness of 133 communications as core values, as the scale and the commercialization 134 of the Internet has grown greatly, topics like access, rights and 135 connectivity are forced to compete with other values. Therefore, 136 decisive and human rights enabling characteristics of the Internet 137 might be degraded if they're not properly defined, described and 138 protected as such. And, the other way around, not protecting human 139 right enabeling characteristics could also result in (partial) loss 140 of functionality and connectivity, and other inherent parts of the 141 Internets architecture. 143 The IETF has produced guidelines and procedures to ensure and 144 galvanize the privacy and security of the network in protocol 145 development. This document aims to explore the possibility of the 146 development of similar procedures for guidelines for human rights 147 considerations to ensure that protocols developed in the IETF do not 148 have an adverse impact on the enjoyment of human rights on the 149 Internet. 151 2. Vocabulary used 153 In the discussion of human rights and Internet architecture concepts 154 that have been developed and computer science, networking, law, 155 policy-making and advocacy are coming together. The same concepts 156 might have a very different meaning and implications in another area 157 of expertise. In order to foster a constructive interdisciplinary 158 debate, and minimize differences in interpretation, the following 159 glossary is provided. 161 Accessibility Full Internet Connectivity as described in [RFC4084] 162 to provide unfettered access to the Internet 164 The design of protocols, services or implementation that provide 165 an enabling environment for people with disabilities. 167 The ability to receive information available on the Internet 169 Anonymity The condition of an identity being unknown or concealed. 170 [RFC4949] 172 Anonymous A state of an individual in which an observer or attacker 173 cannot identify the individual within a set of other individuals 174 (the anonymity set). [RFC6973] 176 Authenticity The fact that the data does indeed come from the source 177 it claims to come from. (It is strongly linked with Integrity, 178 see below). 180 Censorship resistance Methods and measures to prevent Internet 181 censorship. 183 Confidentiality The non-disclosure of information to any unintended 184 person or host or party 186 Connectivity The extent to which a device or network is able to 187 reach other devices or networks to exchange data. The Internet is 188 the tool for providing global connectivity [RFC1958]. 190 Content-agnosticism Treating network traffic identically regardless 191 of content. 193 Debugging Debugging is a methodical process of finding and reducing 194 the number of bugs, or defects, or malfunctions in a protocol or 195 its implementation, thus making it behave as expected and analyse 196 the consequences that might have emanated from the error. 197 Debugging tends to be harder when various subsystems are tightly 198 coupled, as changes in one may cause bugs to emerge in another. 199 [WP-Debugging] 201 The process through which people troubleshoot a technical issue, 202 which may include inspection of program source code or device 203 configurations. Can also include tracing or monitoring packet 204 flow. 206 Decentralized Opportunity for implementation or deployment of 207 standards, protocols or systems without one single point of 208 control. 210 End-to-End The principal of extending characteristics of a protocol 211 or system as far as possible within the system. For example, end- 212 to-end instant message encryption would conceal communications 213 from one user's instant messaging application through any 214 intermediate devices and servers all the way to the recipient's 215 instant messaging application. If the message was decrypted at 216 any intermediate point-for example at a service provider-then the 217 property of end-to-end encryption would not be present. 219 One of the key architectural guidelines of the Internet is the 220 end-to-end principle in the papers by Saltzer, Reed, and Clark 221 [Saltzer] [Clark]. The end-to-end principle was originally 222 articulated as a question of where best not to put functions in a 223 communication system. Yet, in the ensuing years, it has evolved 224 to address concerns of maintaining openness, increasing 225 reliability and robustness, and preserving the properties of user 226 choice and ease of new service development as discussed by 227 Blumenthal and Clark in [Blumenthal]; concerns that were not part 228 of the original articulation of the end-to-end principle. 229 [RFC3724] 231 communication that takes place between communication end-points of 232 the same physical or logical functional level 234 Federation The possibility of connecting autonomous systems into a 235 single distributed system. 237 Heterogenity : The Internet is characterized by heterogeneity on 238 many levels: devices and nodes, router scheduling algorithms and 239 queue management mechanisms, routing protocols, levels of 240 multiplexing, protocol versions and implementations, underlying link 241 layers (e.g., point-to-point, multi-access links, wireless, FDDI, 242 etc.), in the traffic mix and in the levels of congestion at 243 different times and places. Moreover, as the Internet is composed of 244 autonomous organizations and internet service providers, each with 245 their own separate policy concerns, there is a large heterogeneity of 246 administrative domains and pricing structures. As a result, 247 heterogeneity principle is proposed in [RFC1958] to be supported by 248 design. [FIArch] 250 Integrity Maintenance and assurance of the accuracy and consistency 251 of data to ensure it has not been (intentionally or 252 unintentionally) altered 254 Internet censorship : Internet censorship is the intentional 255 suppression of information originating, flowing or stored on systems 256 connected to the Internet where that information is relevant for 257 decision making to some entity. [Elahi] 259 Inter-operable A property of a documented standard or protocol which 260 allows different independent implementations to work with each 261 other without any restricted negotiation, access or 262 functionality. 264 Internet Standards as an Arena for Conflict Pursuant to the 265 principle of constant change, since the function and scope of the 266 Internet evolves, so does the role of the IETF in developing 267 standards. Internet standards are adopted on the basis of a 268 series of criteria, including high technical quality, support by 269 community consensus, and their overall benefit to the Internet. 270 The latter calls for an assessment of the interests of all 271 affected parties and the specifications' impact on the Internet's 272 users. In this respect, the effective exercise of the human 273 rights of the Internet users is a relevant consideration that 274 needs to be appreciated in the standardization process insofar as 275 it is directly linked to the reliability and core values of the 276 Internet. [RFC1958] [RFC0226] [RFC3724] 278 Internationalization (i18n) The practice of making protocols, 279 standards, and implementations usable in different languages and 280 scripts. (see Localization) 282 (cf [RFC6365]) In the IETF, "internationalization" means to add or 283 improve the handling of non-ASCII text in a protocol. [RFC6365] 284 A different perspective, more appropriate to protocols that are 285 designed for global use from the beginning, is the definition used 286 by W3C: 288 "Internationalization is the design and development of a 289 product, application or document content that enables easy 290 localization for target audiences that vary in culture, region, 291 or language." [W3Ci18nDef] 293 Many protocols that handle text only handle one charset (US-ASCII), 294 or leave the question of what CCS and encoding are used up to local 295 guesswork (which leads, of course, to interoperability problems). 296 If multiple charsets are permitted, they must be explicitly 297 identified [RFC2277]. Adding non-ASCII text to a protocol allows the 298 protocol to handle more scripts, hopefully all of the ones useful in 299 the world. In today's world, that is normally best accomplished by 300 allowing Unicode encoded in UTF-8 only, thereby shifting conversion 301 issues away from individual choices. 303 Localization (l10n) The practice of translating an implementation to 304 make it functional in a specific language or for users in a 305 specific locale (see Internationalization) 307 (cf [RFC6365] The process of adapting an internationalized 308 application platform or application to a specific cultural 309 environment. In localization, the same semantics are preserved 310 while the syntax may be changed. [FRAMEWORK] 312 Localization is the act of tailoring an application for a different 313 language or script or culture. Some internationalized applications 314 can handle a wide variety of languages. Typical users only 315 understand a small number of languages, so the program must be 316 tailored to interact with users in just the languages they know. 318 The major work of localization is translating the user interface and 319 documentation. Localization involves not only changing the language 320 interaction, but also other relevant changes such as display of 321 numbers, dates, currency, and so on. The better internationalized an 322 application is, the easier it is to localize it for a particular 323 language and character encoding scheme. 325 Localization is rarely an IETF matter, and protocols that are merely 326 localized, even if they are serially localized for several locations, 327 are generally considered unsatisfactory for the global Internet. 329 Open standards Conform [RFC2606]: Various national and 330 international standards bodies, such as ANSI, ISO, IEEE, and 331 ITU-T, develop a variety of protocol and service 332 specifications that are similar to Technical Specifications 333 defined here. National and international groups also publish 334 "implementors' agreements" that are analogous to Applicability 335 Statements, capturing a body of implementation-specific detail 336 concerned with the practical application of their standards. All 337 of these are considered to be "open external standards" for 338 the purposes of the Internet Standards Process. 340 Openness The quality of the unfiltered Internet that allows for free 341 access to other hosts 343 Absence of centralised points of control - a feature that is 344 assumed to make it easy for new users to join and new uses to 345 unfold [Brown] 347 Permissionless innovation The freedom and ability of to freely 348 create and deploy new protocols on top of the communications 349 constructs that currently exist Privacy 351 The right of an entity (normally a person), acting in its own 352 behalf, to determine the degree to which it will interact with its 353 environment, including the degree to which the entity is willing 354 to share its personal information with others. [RFC4949] 356 The right of individuals to control or influence what information 357 related to them may be collected and stored and by whom and to 358 whom that information may be disclosed. 360 Privacy is a broad concept relating to the protection of 361 individual autonomy and the relationship between an individual and 362 society, including government, companies and private individuals. 363 It is often summarized as "the right to be left alone" but it 364 encompasses a wide range of rights including protections from 365 intrusions into family and home life, control of sexual and 366 reproductive rights, and communications secrecy. It is commonly 367 recognized as a core right that underpins human dignity and other 368 values such as freedom of association and freedom of speech. 370 The right to privacy is also recognized in nearly every national 371 constitution and in most international human rights treaties. It 372 has been adjudicated upon both by international and regional bodies. 373 The right to privacy is also legally protected at the national level 374 through provisions in civil and/or criminal codes. 376 Reliable Reliability ensures that a protocol will execute its 377 function consistently and error resistant as described and 378 function without unexpected result. A system that is reliable 379 degenerates gracefully and will have a documented way to announce 380 degradation. It also has mechanisms to recover from failure 381 gracefully, and if applicable, allow for partial healing. 383 Resilience The maintaining of dependability and performance in the 384 face of unanticipated changes and circumstances. 386 Robustness The resistance of protocols and their implementations to 387 errors, and to involuntary, legal or malicious attempts to disrupt 388 its mode of operations. [RFC0760] [RFC0791] [RFC0793] [RFC1122] 390 Scalable The ability to handle increased or decreased workloads 391 predictably within defined expectations. There should be a clear 392 definition of its scope and applicability. The limits of a 393 systems scalability should be defined. 395 Stateless / stateful In computing, a stateless protocol is a 396 communications protocol that treats each request as an independent 397 transaction that is unrelated to any previous request so that the 398 communication consists of independent pairs of request and 399 response. A stateless protocol does not require the server to 400 retain session information or status about each communications 401 partner for the duration of multiple requests. In contrast, a 402 protocol which requires keeping of the internal state on the 403 server is known as a stateful protocol. [WP-Stateless] 405 Strong encryption / cryptography Used to describe a cryptographic 406 algorithm that would require a large amount of computational power 407 to defeat it. [RFC4949] 409 Transparent "transparency" refers to the original Internet concept 410 of a single universal logical addressing scheme, and the 411 mechanisms by which packets may flow from source to destination 412 essentially unaltered. [RFC2775] 414 The combination of reliability, confidentiality, integrity, 415 anonymity, and authenticity is what makes up security on the 416 Internet. 418 ( Reliability ) 419 ( Confidentiality ) 420 ( Integrity ) = communication and information security (technical) 421 ( Authenticity ) 422 ( Anonymity ) 424 The combination of End-to-End, Interoperability, resilience, 425 reliability and robustness is what makes us connectivity on the 426 Internet 427 ( End-to-End ) 428 connectivity = ( Interoperability ) 429 ( Resilience ) 430 ( Reliability ) 431 ( Robustness ) 432 ( Autonomy ) 433 ( Simplicity ) 435 3. Research Questions 437 The Human Rights Protocol Considerations Research Group (hrpc) in the 438 Internet Research Taskforce (IRTF) embarked on its mission to answer 439 the following two questions which are also the main two questions 440 which this documents seeks to answer: 442 1. How can Internet protocols and standards impact human rights, 443 either by enabling them or by creating a restrictive environment? 445 2. Can guidelines be developed to improve informed and transparent 446 decision making about potential human rights impact of protocols? 448 4. Literature and Discussion Review 450 Protocols and standards are regularly seen as merely performing 451 technical functions. However, these protocols and standards do not 452 exist outside of their technical context nor outside of their 453 political,historical, economic, legal or cultural context. This is 454 best exemplified by the way in which protocols have become part and 455 parcel of political processes and public policies: one only has to 456 look at the IANA transition, the RFC on pervasive monitoring or 457 global innovation policy for concrete examples [Denardis15]. To 458 quote [Abbate]: "protocols are politics by other means". Since the 459 late 1990's a burgeoning group of academics and practitioners 460 researched questions surrounding the societal impact of protocols. 461 These studies vary in focus and scope: some focus on specific 462 standards [Davidsonetal] [Musiani], others look into the political, 463 legal, commercial or social impact of protocols [BrownMarsden] 464 [Lessig], [Mueller]. Commercial and political influences on the 465 management of the Internet's architecture are well-documented in the 466 academic literature and will thus not be discussed here [Benkler] 467 [Brownetal] [Denardis15] [Lessig] [Mueller] [Zittrain]. It is 468 enough to say that the IETF consistently tries to push back against 469 the standardization of surveillance and certain other issues that 470 negatively influence end-users' experience of the Internet 471 [Denardis14]. The role human rights play in technical engineering is 472 much less clear. Understanding how protocols and standards impact 473 human rights, especially the right to freedom of expression and 474 freedom of association and assembly is crucial. Questions at the 475 intersection of human rights and Internet architecture management are 476 particularly important as Internet Standard Developing Organizations 477 (SDOs) are the arenas for contention over human rights and the role 478 of technical engineers to protect human rights by design [Brownetal] 479 [Clarketal] [Denardis14] [Lessig] [Rachovitsa]. In the academic 480 literature four clear positions can be discerned, in relation to the 481 role of human rights in protocol design and how to account for these 482 human rights in protocol development: Clark et al. argue that there 483 is a need to 'design for variation in outcome, so that the outcome 484 can be different in different places, and the tussle takes place 485 within the design (...) [as] Rigid designs will be broken; designs 486 that permit variation will flex under pressure and survive 487 [Clarketal].' They hold that human rights should not be hard-coded 488 into protocols because of four reasons: first, the rights in the UDHR 489 are not absolute. Second, technology is not the only tool in the 490 tussle over human rights. Third, there are inherent dangers to 491 blunting the tools of enforcement and last but not least, it is 492 dangerous to make promises that can't be kept. The open nature of 493 the Internet will never, they argue, be enough to fully protect 494 individuals' human rights. 496 Conversely, Brown et al. [Brownetal] state that 'some key, universal 497 values - of which the UDHR is the most legitimate expression - should 498 be baked into the architecture at design time.' They argue that 499 design choices have offline consequences, and are able shape the 500 power positions of groups or individuals in society. As such, the 501 individuals making these technical decisions have a moral obligation 502 to take into account the impact of their decisions on society, and by 503 extension human rights. Brown et al recognise that values and the 504 implementation of human rights vary across the globe. Yet they argue 505 that all members of the United Nations have found 'common agreement 506 on the values proclaimed in the Universal Declaration of Human 507 Rights. In looking for the most legitimate set of global values to 508 embed in the future Internet architecture, the UDHR has the 509 democratic assent of a significant fraction of the planet's 510 population, through their elected representatives." 512 The main disagreement between these two positions lies mostly in the 513 question on whether a particular value system should be build into 514 the Internet's architecture or whether the architecture needs to 515 account for a varying set of values. A third position that is 516 similar to that of Brown et al., is taken by [Broeders] who argues 517 that 'we must find ways to continue guaranteeing the overall 518 integrity and functionality of the public core of the Internet.' He 519 argues that the best way to do this is by declaring the backbone of 520 the Internet - which includes the tcp/ip ProtocolSuite, numerous 521 standards, the Domain Name System (dns), and routing protocols- a 522 common public good. This is a different approach then that of 524 [Clarketal] and [Brownetal] because he does not suggest that social 525 values should (or should not) be explicitly coded into the Internet's 526 architecture, but rather that the existing architecture should be 527 seen as an entity of public value. Bless and Orwat [Bless] 528 represents a fourth position. They argue that 'pure technical 529 solutions for enabling, enforcing or restricting rights/values are 530 often costly, insufficient, inflexible, may have unintended 531 consequences or create stakeholders with too much power'. They argue 532 that it is important to search for solutions that 'create awareness 533 in the technical community about impact of design choices on social 534 values. And work towards a methodology for co-design of technical 535 and institutional systems.' Our position is that hard-coding human 536 rights into protocols in addition to being undesirable is also 537 impossible, because each situation is dependent on its context. It 538 is however important to make consicious design decisions that take 539 into account the human rights protocol considerations guidelines 540 developed below. This will ensure that the impact protocols can have 541 on human rights is clear and explicit, both for developers and for 542 users. In addition, it ensures that the impact of specific protocol 543 on human rights is carefully considered and that concrete design 544 decisions are documented in the protocol. 546 This document details the steps taken in the research into human 547 rights protocol considerations by the HRPC group to clarify the 548 relation between technical concepts used in the IETF and human 549 rights. And sets out some preliminary of steps and considerations 550 for engineers to take into account when developing standards and 551 protocols. 553 5. Methodology 555 Mapping the relation between human rights and protocols and 556 architectures is a new research challenge, which requires a good 557 amount of interdisciplinary and cross organizational cooperation to 558 develop a consistent methodology. While the authors of this first 559 draft are involved in both human rights advocacy and research on 560 Internet technologies - we believe that bringing this work into the 561 IRTF facilitates and improves this work by bringing human rights 562 experts together with the community of researchers and developers of 563 Internet standards and technologies. 565 The methodological choices made in this document are based on the 566 political science-based method of discourse analysis and ethnographic 567 research methods. This work departs from the assumption that 568 language reflects the understanding of concepts. Or as [Jabri] 569 holds, policy documents are 'social relations represented in texts 570 where language is used to construct meaning and representation'. 571 This process happens in 'the social space of society' [Schroeder] and 572 manifests itself in institutions and organizations [King] which were 573 exposed using the ethnographic methods of semi-structured interviews 574 and participant observation. 576 The discourse analysis was operationalized using qualitative and 577 quantitative means. The first step was for the research group to 578 read RFCs and other official IETF documents. The second step was the 579 use of a pyhon-based analyzer, using the tool Big Bang, adapted by 580 Nick Doty [Doty] to scan for the concepts that were identified as 581 important architetural principels (distilled on the initial reading 582 and supplemented by the interviews and participant observation). 583 Such a quantitative method is very precise and speeds up the research 584 process [Richie]. But this tool is unable to understand 'latent 585 meaning' [Denzin]. In order to mitigate these issues of automated 586 word-frequency based approaches, and to get a sense of the 'thick 587 meaning' [Geertz] of the data, a second qualitative analysis of the 588 data set was performed. These various rounds of discourse analysis 589 were used to inform the interviews and further data analysis. The 590 ethnographic methods of the data collection process allowed the 591 research group to acquire the data necessary to 'provide a holistic 592 understanding of research participants' views and actions' [Denzin] 593 that highlighted ongoing issues and case studies where protocols 594 impact human rights. The interview participants were selected 595 through purposive sampling [Babbie], as the research group was 596 interested in getting a wide variety of opinions on the role of human 597 rights in guiding protocol development. This sampling method also 598 ensured that the individuals with extensive experience of working at 599 the IETF in various roles were targeted. The interviewees included 600 individuals in leadership positions (Working Group (WG) chairs, Area 601 Directors (ADs)), 'regular participants', individuals working for 602 specific entities (corporate, civil society, political, academic) and 603 represented various backgrounds, nationalities and genders. 605 5.1. Data Sources 607 In order to map the potential relation between human rights and 608 protocols, so far, the HRPC research group gathered data from three 609 specific sources:
 611 5.1.1. Discourse analysis of RFCs 613 To start addressing the issue, a mapping exercise analyzing Internet 614 architecture and protocols features, vis-a-vis possible impact on 615 human rights is being undertaken. Therefore, research on the 616 language used in current and historic RFCs and mailing list 617 discussions is underway to expose core architectural principles, 618 language and deliberations on human rights of those affected by the 619 network. 621 5.1.2. Interviews with members of the IETF community during IETF92 in 622 Dallas 624 Interviews with the current and past members of the Internet 625 Architecture Board (IAB), current and past members of the Internet 626 Engineering Steering Group(IESG) and chairs of selected working 627 groups and RFC authors. To get an insider understanding of how they 628 view the relationship (if any) between human rights and protocols to 629 play out in their work. 631 5.1.3. Participant observation in Working Groups 633 By participating in various working groups, in person at IETF 634 meetings and on mailinglists, information was gathered about the 635 IETFs day-to-day workings. From which which general themes, 636 technical concepts, and use-cases about human rights and protocols 637 were extracted. 639 5.2. Data analysis strategies 641 The data above was processed using three consecutive strategies: 642 mapping protocols related to human rights, extracting concepts from 643 these protocols, and creation of a common glossary (detailed under 2. 644 vocabulary used). Before going into these strategies some 645 elaboration on the process of identifying technical concepts as they 646 related to human rights needs to be given. 648 5.2.1. Identifying qualities of technical concepts that relate to human 649 rights 651 5.2.1.1. Mapping protocols and standards related to human rights 653 By combining data from the three data sources named above, an 654 extensive list of protocols and standards that potentially enable 655 the internet as a tool for freedom of expression and association was 656 assembly. In order to determine this enabling (or inhibiting) 657 featured we relied on direct references of such impact in the RFCs, 658 as well as input from the community. On the basis of this analysis a 659 list of RFCs that describe standards and protocols that are 660 potentially more closely related to human rights were compiled. 662 5.2.1.2. Extracting concepts from mapped RFCs 664 Mapping the protocols and standards that are related to human rights 665 and creating an human rights enabeling environment was the first step 666 to focus on specific technical concepts that underlie these protocols 667 and standards. On the basis of this list number of technical 668 concepts that appeared frequently was extracted, and used to create a 669 list of technical terms that combined create the enabling environment 670 for excercising human rights on the Internet. 672 5.2.1.3. Building a common vocabulary of technical concepts that impact 673 human rights 675 While interviewing experts and mapping RFCs and compiling technical 676 definitions several concepts of convergence and divergence were 677 identified. To ensure that the discussion was based on a common 678 understanding of terms a list of terms was created. The definitions 679 are based on definitions from other IETF documents, and if these were 680 unavailable definitions were taken from definitions from other 681 Standards Developing Organizations or academic literature. 683 5.2.1.4. Translating Human Rights Concept into Technical Definitions 685 The previous steps allowed for the clarification of relation between 686 human rights and technical concepts. The steps taken show how the 687 research process zoomed in, from compiling a broad lists of protocols 688 and standards that relate to human rights to extracting the precies 689 technical concepts that make up these protocols and standards in 690 order to understand the relationship between the two. This sub- 691 section presents the next step: translating human rights to technical 692 concepts by matching the individuals components of the rights to the 693 accompanying technical concepts, allowing for the creation of a list 694 of technical concepts that combined create the enabling environment 695 for human rights. 697 5.2.1.5. List technical terms that combined create enabling environment 698 for human rights 700 On the basis of the prior steps the following list of technical terms 701 that combined create the enabling environment for human rights, such 702 a freedom of expression and freedom of association was drafted. 704 Architectural principles Enabling features 705 and characteristics for user rights 707 /------------------------------------------------\ 708 | | 709 +=================|=============================+ | 710 = | = | 711 = | End to end = | 712 = | Reliability = | 713 = | Resilience = Access as | 714 = | Interoperability = Human Right | 715 = Good enough | Transparency = | 716 = principle | Data minimization = | 717 = | Permissionless innovation = | 718 = Simplicity | Graceful degradation = | 719 = | Connectivity = | 720 = | Heterogeneity = | 721 = | = | 722 = | = | 723 = \------------------------------------------------/ 724 = = 725 +===============================================+ 727 5.2.2. Translation human rights to technical terms 729 This analysis aims to translate human rights concepts that impact or 730 are impacted by the Internet as follows: 732 The combination of content agnosticism, connectivity, security, 733 privacy (as defined in [RFC6973] ), and open standards are the 734 technical principles that underlay freedom of expression on the 735 Internet. 737 ( Connectivity ) 738 ( Privacy ) 739 ( Security ) = Right to freedom of expression 740 ( Content agnosticism ) 741 ( Internationalization ) 742 ( Censorship resistance ) 743 ( Open Standards ) 744 ( Heterogeneity support ) 746 ( Anonymity ) 747 ( Privacy ) = Right to non-discrimination 748 ( Pseudonymity ) 749 ( Content agnosticism ) 750 ( Accessibility ) 752 ( Content Agnosticism ) 753 ( Security ) = Right to equal protection 755 ( Anonymity ) 756 ( Privacy ) = Right to be presumed innocent 757 ( Security ) 759 ( Accessibility ) 760 ( Internationalization ) = Right to political participation 761 ( Censorship resistance ) 763 ( Open standards ) 764 ( Localization ) = Right to participate in cultural life, 765 ( Internationalization ) arts and science 766 ( Censorship resistance ) 768 ( Connectivity ) 769 ( Decentralization ) 770 ( Censorship resistance ) = Right to freedom of assembly 771 ( Pseudonymity ) and association 772 ( Anonymity ) 773 ( Security ) 775 ( Reliability ) 776 ( Confidentiality ) 777 ( Integrity ) = Right to security 778 ( Authenticity ) 779 ( Anonymity ) 781 5.2.2.1. Map cases of protocols that are adversely impact human rights 782 or are enablers thereof 784 Taken this information above, the following list of cases of 785 protocols that adversely impact or enable human rights was formed. 787 5.2.3. IP 789 The Internet Protocol version 4, known as 'layer 3' of the internet, 790 and specified as a common encapsulation and protocol header, is 791 defined by [RFC0791]. The evolution of Internet communications have 792 led to continued development in this area, encapsulated in the 793 development of version 6 of the protocol in [RFC2460]. In spite of 794 this updated protocol, we find that 25 years after the specification 795 of version 6 of the protocol, the older v4 standard continues to 796 account for a sizeable majority of internet traffic. 798 The internet was designed as a platform for free and open 799 communication, most notably encoded in the end-to-end principle, and 800 that philosophy is also present in the technical implementation of 801 the Internet Protocol. [RFC3724] While the protocol was designed to 802 exist in an environment where intelligence is at the end hosts, it 803 has proven to provide sufficient information that a more intelligent 804 network core can make policy decisions and enforce policy shaping and 805 restricting the communications of end hosts. These capabilities for 806 network control and limitations of the freedom of expression by end 807 hosts can be traced back to the IPv4 design, helping us understand 808 which technical protocol decisions have led to harm of these human 809 rights. 811 Two major shifts have occurred to harm freedom of expression through 812 misuse of the Internet Protocol. The first is the network's 813 exploitation of the public visibility of the host pairs for all 814 communications, and the corresponding ability to discriminate and 815 block traffic as a result of that metadata. The second is the 816 selective development of IP options. Protocol extensions including 817 Mobility and Multicasting have proposed alternate communication modes 818 and suggest that different forms of assembly could be supported by an 819 a robust IP layer. Instead, the protocol has limited the 820 deployability of such extensions by not providing a mechanism for 821 appropriate fallback behavior when unrecognized extensions are 822 encountered. 824 5.2.3.1. Network visibility of Source and Destination 826 The IPv4 protocol header contains fixed location fields for both the 827 source and destination IP addresses [RFC0791]. These addresses 828 identify both the host sending and receiving each message, and allow 829 the core network to understand who is talking to whom, and to 830 practically limit communication selectively between pairs of hosts. 831 Blocking of communication based on the pair of source and destination 832 is one of the most common limitations on the ability for hosts to 833 communicate today, [caida] and can be seen as a restriction of the 834 ability for those hosts to assemble or to consensually express 835 themselves. 837 Inclusion of an Internet-wide identified source in the IP header is 838 not the only possible design, especially since the protocol is most 839 commonly implemented over Ethernet networks exposing only link-local 840 identifiers. [RFC0894] A variety of alternative designs including 841 source routing, and spoofing of the source IP address are technically 842 supported by the protocol, but neither are regularly allowed on the 843 Internet. While projects like [torproject] provide an alternative 844 implementation of anonymity in connections, they have been developed 845 in spite of the IPv4 protocol design. 847 5.2.3.2. Protocols 849 The other major feature of the IP protocol header is that it 850 specifies the protocol encapsulated in each message in an easily 851 observable form, and does not encourage a design where the 852 encapsulated protocol is not available to a network observer. This 853 design has resulted in a proliferation of routers which inspect the 854 inner protocol, and has resulted in a stagnation where only the TCP 855 and UDP protocols are widely supported across the Internet. While 856 the IP protocol was designed as the entire set of metadata needed for 857 routing, subsequent enhanced routers have found value on making 858 policy decisions based on the contents of TCP and UDP headers as 859 well, and are encoded with the assumption that only these protocols 860 will be used for data transfer. [spdy] [RFC4303] defines an encrypted 861 encapsulation of additional protocols, but lacks widespread 862 deployment and faces the same challenge as any other protocol of 863 providing sufficient metadata with each message for routers to make 864 positive policy decisions. Protocols like [RFC4906] have seen 865 limited wide-area uptake, and these alternate designs are frequently 866 re-implemented on top of UDP. [quic] 868 5.2.3.3. Address Translation and Mobility 870 A major structural shift in the Internet which has undermined the 871 protocol design of IPv4, and has significantly reduced the freedom of 872 end users to communicate and assemble in the introduction network 873 address translation. [RFC1631] Network address translation is a 874 process whereby organizations and autonomous systems to connect two 875 networks by translating the IPv4 source and destination addresses 876 between the two. This process puts the router performing the 877 translation into a privileged position, where it can decide which 878 subset of communications are worthy of translation, and whether an 879 unknown request for communication will be correctly forwarded to a 880 host on the other network. 882 This process of translation has widespread adoption despite promoting 883 a process that goes against the stated end-to-end process of the 884 underlying protocol [natusage]. In contrast, the proposed mechanism 885 to provide support for mobility and forwarding to clients which may 886 move, encoded instead as an option in the IP protocol in [RFC5944], 887 has failed to gain traction. This situation again suggests that the 888 compromise made in design of the protocol has resulted in a 889 technology which failed to technical encode the freedom of expression 890 goals it was designed to promote. 892 5.2.4. DNS 894 The Domain Name System (DNS) [RFC1035], provides service discovery 895 capabilities, and provides a mechanism to associate human readable 896 names with services. The DNS system is organized around a set of 897 independently operated 'Root Servers' run by organizations around the 898 web which enact ICANN's policy by answering queries for which 899 organizations have been delegated to manage registration under each 900 Top Level Domain (TLD). Top Level domains are maintained and 901 determined by ICANN. These namespaces encompass several classes of 902 services. The initial name spaces including '.Com' and '.Net', 903 provide common spaces for expression of ideas, though their policies 904 are enacted through US based companies. Other name spaces are 905 delegated to specific nationalities, and may impose limits designed 906 to focus speech in those forums both to promote speech from that 907 nationality, and to comply with local limits on expression and social 908 norms. Finally, the system has been recently expanded with 909 additional generic and sponsored name spaces, for instance '.travel' 910 and '.ninja', which are operated by a range of organizations which 911 may independently determine their registration policies. 913 DNS has significant privacy issues per [RFC7626]. Most notable are 914 the lack of encryption to limit the visibility of requests for domain 915 resolution from intermediary parties, and a limited deployment of 916 DNSSEC to provide authentication, allowing the client to know that 917 they have received a correct, "authoritative", answer to a query. 918 Together, this situation results in ongoing harm to freedom of 919 expression as interference with the operation of DNS has become one 920 of the central mechanisms used to block access to websites. This 921 interference limits both the freedom of expression of the publisher 922 to offer their content, and the freedom of assembly for clients to 923 congregate in a shared virtual space. 925 There have been several mechanisms used impose these limitations 926 based on the technical design of the DNS protocol. These have led to 927 a number of situations where limits on expression have been imposed 928 through subversion of the DNS protocol. Each of these situations has 929 accompanying aspects of protocol design enabling those limitations. 931 5.2.4.1. Removal of records 933 There have been a number of cases where the records for a domain are 934 removed from the name system due to real-world events. Examples of 935 this removal includes the 'seizure' of wikileaks [bbc-wikileaks] and 936 the names of illegally operating gambling operations by the United 937 States ICE unit, which compelled the US-based registry in charge of 938 the .com TLD to hand ownership of those domains over to the 939 government. The same technique has been notably used by Libya to 940 remove sites in violation of "our Country's Law and Morality (which) 941 do not allow any kind of pornography or its promotion." [techyum] 943 At a protocol level, there is no technical auditing for name 944 ownership, as in alternate systems like [namecoin]. As a result, 945 there is no ability for users to differentiate seizure from the 946 legitimate transfer of name ownership, which is purely a policy 947 decision of registrars. While DNSSEC addresses network distortion 948 events described below, it does not tackle this problem, which has 949 the cooperation of (or compelled action by) the registry. 951 5.2.4.2. Distortion of records 953 The most common mechanism by which the DNS system is abused to limit 954 freedom of expression is through manipulation of protocol messages by 955 the network. One form occurs at an organizational level, where 956 client computers are instructed to use a local DNS resolver 957 controlled by the organization. The DNS resolver will then 958 selectively distort responses rather than request the authoritative 959 lookup from the upstream system. The second form occurs through the 960 use of deep packet inspection, where all DNS protocol messages are 961 inspected by the network, and objectionable content is distorted, as 962 in [turkey]. 964 A notable instance of distortion has occurred in Greece [ververis], 965 where a study found evidence of both of deep packet inspection to 966 distort DNS replies, and overblocking of content, where ISPs 967 prevented clients from resolving the names of domains which they were 968 not instructed to do through the governmental order prompting the 969 blocking systems there. 971 At a protocol level, the effectiveness of these attacks is made 972 possible by a lack of authentication in the DNS protocol. DNSSEC 973 provides the ability to determine authenticity of responses when 974 used, but it is not regularly checked by resolvers. DNSSEC is not 975 effective when the local resolver for a network is complicit in the 976 distortion, for instance when the resolver assigned for use by an ISP 977 is the source of injection. Selective distortion of records has also 978 been made possible by the predictable structure of DNS messages, 979 which make it computationally easy for a network device to watch all 980 passing messages even at high speeds, and the lack of encryption, 981 which allows the network to distort only an objectionable subset of 982 protocol messages. Specific distortion mechanisms are discussed 983 further in [draft-hall-censorship-tech-01]. 985 5.2.4.3. Injection of records 987 Responding incorrectly to requests for name lookups is the most 988 common mechanism that in-network devices use to limit the ability of 989 end users to discover services. A deviation which accomplishes a 990 similar objective, though may be seen as different from a freedom of 991 expression perspective, is the injection of incorrect responses to 992 queries. The most prominent example of this behavior occurs in 993 China, where requests for lookups of sites which have been deemed 994 inappropriate will trigger the network to respond with a bogus 995 response, causing the client to ignore the real response when it 996 subsequently arrives. [greatfirewall] Unlike the other forms of 997 discussion discussed above, injection does not stifle the ability of 998 a server to announce it's name, it instead provides another voice 999 which answers sooner. This is effective because without DNSSEC, the 1000 protocol will respond to whichever answer is received first, without 1001 listening for subsequent answers. 1003 5.2.5. HTTP 1005 The Hypertext Transfer Protocol (HTTP), described in its version 1.1 1006 in RFC 7230 to 7237, is a request-response application protocol 1007 developed throughout the 1990s, and factually contributed to the 1008 exponential growth of the Internet and the inter-connection of 1009 populations around the world. Because of its simple design, HTTP has 1010 become the foundation of most modern Internet platforms and 1011 communication systems, from websites, to chat systems, and computer- 1012 to-computer applications. In its manifestation with the World Wide 1013 Web, HTTP has radically revolutionized the course of technological 1014 development and the ways people interact with online content and with 1015 each other. 1017 However, HTTP is also a fundamentally insecure protocol, that doesn't 1018 natively provide encryption properties. While the definition of the 1019 Secure Sockets Layer (SSL), and later of Transport Layer Security 1020 (TLS), also happened during the 1990s, the fact that HTTP doesn't 1021 mandate the use of such encryption layers to developers and service 1022 providers, caused a very late adoption. Only in the middle of the 1023 2000s we observed big Internet service providers, such as Google, 1024 starting to provide encrypted access to their web services. 1026 The lack of sensitivity and understanding of the critical importance 1027 of securing web traffic incentivized malicious and offensive actors 1028 to develop, deploy and utilize at large interception systems and 1029 later active injection attacks, in order to swipe large amounts of 1030 data, compromise Internet-enabled devices. The commercial 1031 availability of systems and tools to perform these types of attacks 1032 also led to a number of human rights abuses that have been discovered 1033 and reported over the years and that painted a dark picture on the 1034 current state of control over the Internet. 1036 Generally we can identify in Traffic Interception and Traffic 1037 Manipulation the two most problematic attacks that can be performed 1038 against applications employing a clear-text HTTP transport layer. 1040 5.2.5.1. Traffic Interception 1042 While we are seeing an increasing trend in the last couple of years 1043 to employ SSL/TLS as a secure traffic layer for HTTP-based 1044 applications, we are still far from seeing an ubiquitous use of 1045 encryption on the World Wide Web. It is important to consider that 1046 the adoption of SSL/TLS is also a relatively recent phenomena. 1047 Google introduced an option for its GMail users to navigate with SSL 1048 only in 2008 [Rideout], and turned SSL on by default later in 2010 1049 [Schillace]. It took an increasing amount of scandalous security 1050 breaches and revelations on global surveillance from Edward Snowden 1051 to have other Internet service providers to follow Google's lead. 1052 For example, Yahoo enabled SSL/TLS by default on its webmail services 1053 only towards the end of 2013 [Peterson]. 1055 As we learned through the Snowden's revelations, intelligence 1056 agencies have been intercepting and collecting unencrypted traffic at 1057 large for many years. There are documented examples of such mass 1058 surveillance programs with GCHQ's TEMPORA and NSA's XKEYSCORE. 1059 Through these programs NSA/GCHQ have been able to swipe large amounts 1060 of data including email and instant messaging communications which 1061 have been transported by the respective providers in clear for years, 1062 unsuspecting of the pervasiveness and scale of governments' efforts 1063 and investment into global mass surveillance capabilities. 1065 However, similar mass interception of unencrypted HTTP communications 1066 is also often employed at a nation-level by less democratic countries 1067 by exercising control over state-owned Internet Service Providers 1068 (ISP) and through the use of commercially available monitoring, 1069 collection, and censorship equipment. Over the last few years a lot 1070 of information has come to public attention on the role and scale of 1071 a surveillance industry dedicated to develop interception gear of 1072 different types. We have several records of such equipment being 1073 sold and utilized by oppressive regimes in order to monitor entire 1074 segments of population especially at times of social and political 1075 distress, uncovering massive human rights abuses. For example, in 1076 2013 the group Telecomix revealed that the Syrian regime was making 1077 use of BlueCoat products in order to intercept clear-text traffic as 1078 well as to enforce censorship of unwanted content [RSF]. Similarly 1079 in 2012 it was found that the French Amesys provided the Gaddafi's 1080 government with equipment able to intercept emails, Facebook traffic, 1081 and chat messages ad a country level. The use of such systems, 1082 especially in the context of the Arab Spring and of civil uprisings 1083 against the dictatorships, has caused serious concerns of significant 1084 human rights abuses in Libya. 1086 5.2.5.2. Traffic Manipulation 1088 The lack of a secure transport layer over HTTP connections not only 1089 exposes the users to interception of the content of their 1090 communications, but is more and more commonly abused as a vehicle for 1091 active compromises of computers and mobile devices. If an HTTP 1092 session travels in clear over the network, any node positioned at any 1093 point in the network is able to perform man-in-the-middle attacks and 1094 observe, manipulate, and hijack the session and modify the content of 1095 the communication in order to trigger unexpected behavior by the 1096 application generating the traffic. For example, in the case of a 1097 browser the attacker would be able to inject malicious code in order 1098 to exploit vulnerabilities in the browser or any of its plugins. 1099 Similarly, the attacker would be able to intercept, trojanize, and 1100 repackage binary software updates that are very commonly downloaded 1101 in clear by applications such as word processors and media players. 1102 If the HTTP session would be encrypted, the tampering of the content 1103 would not be possible, and these network injection attacks would not 1104 be successful. 1106 While traffic manipulation attacks have been long known, documented, 1107 and prototyped especially in the context of WiFi and LAN networks, in 1108 the last few years we observed an increasing investment into the 1109 production and sale of network injection equipment both available 1110 commercially as well as deployed at scale by intelligence agencies. 1112 For example we learned from some of the documents provided by Edward 1113 Snowden to the press, that the NSA has constructed a global network 1114 injection infrastructure, called QUANTUM, able to leverage mass 1115 surveillance in order to identify targets of interests and 1116 subsequently task man-on-the-side attacks to ultimately compromise a 1117 selected device. Among other attacks, NSA makes use of an attack 1118 called QUANTUMINSERT [Haagsma] which intercepts and hijacks an 1119 unencrypted HTTP communication and forces the requesting browser to 1120 redirect to a host controlled by NSA instead of the intended website. 1121 Normally, the new destination would be an exploitation service, 1122 referred in Snowden documents as FOXACID, which would attempt at 1123 executing malicious code in the context of the target's browser. The 1124 Guardian reported in 2013 that NSA has for example been using these 1125 techniques to target users of the popular anonymity service Tor 1126 [Schneier]. The German NDR reported in 2014 that NSA has also been 1127 using its mass surveillance capabilities to identify Tor users at 1128 large [Appelbaum]. 1130 Recently similar capabilities of Chinese authorities have been 1131 reported as well in what has been informally called the "Great 1132 Cannon" [Marcak], which raised numerous concerns on the potential 1133 curb on human rights and freedom of speech due to the increasing 1134 tighter control of Chinese Internet communications and access to 1135 information. 1137 Network injection attacks are also made widely available to state 1138 actors around the world through the commercialization of similar, 1139 smaller scale equipment that can be easily acquired and deployed at a 1140 country-wide level. Companies like FinFisher and HackingTeam are 1141 known to have network injection gear within their products portfolio, 1142 respectively called FinFly ISP and RCS Network Injector 1143 [Marquis-Boire]. The technology devised and produced by HackingTeam 1144 to perform network traffic manipulation attacks on HTTP 1145 communications is even the subject of a patent application in the 1146 United States [Googlepatent]. Access to offensive technologies 1147 available on the commercial lawful interception market has been 1148 largely documented to have lead to human rights abuses and 1149 illegitimate surveillance of journalists, human rights defenders, and 1150 political activists in many countries around the world. Companies 1151 like FinFisher and HackingTeam have been found selling their products 1152 to oppressive regimes with little concern for bad human rights 1153 records [Collins]. While network injection attacks haven't been the 1154 subject of much attention, they do enable even unskilled attackers to 1155 perform silent and very resilient compromises, and unencrypted HTTP 1156 remains one of the main vehicles. 1158 5.2.6. XMPP 1160 The Extensible Messaging and Presence Protocol (XMPP), specified in 1161 [RFC6120], provides a standard for interactive chat messaging, and 1162 has evolved to encompass interoperable text, voice, and video chat. 1163 The protocol is structured as a federated network of servers, similar 1164 to email, where users register with a local server which acts one 1165 their behalf to cache and relay messages. This protocol design has 1166 many advantages, allowing servers to shield clients from denial of 1167 service and other forms of retribution for their expression, and 1168 designed to avoid central entities which could control the ability to 1169 communicate or assemble using the protocol. 1171 None-the-less, there are plenty of aspects of the protocol design of 1172 XMPP which shape the ability for users to communicate freely, and to 1173 assembly through the protocol. The protocol also has facets that may 1174 stifle speech as users self-censor for fear of surveillance, or find 1175 themselves unable to express themselves naturally. 1177 5.2.6.1. User Identification 1179 The XMPP specification dictates that clients are identified with a 1180 resource (node@domain/home [1] / node@domain/work [2]) to distinguish 1181 the conversations to specific devices. While the protocol does not 1182 specify that the resource must be exposed by the client's server to 1183 remote users, in practice this has become the default behavior. In 1184 doing so, users can be tracked by remote friends and their servers, 1185 who are able to monitor presence not just of the user, but of each 1186 individual device the user logs in with. This has proven to be 1187 misleading to many users, [pidgin] since many clients only expose 1188 user level rather than device level presence. Likewise, user 1189 invisibility so that communication can occur while users don't notify 1190 all buddies and other servers of their availability is not part of 1191 the formal protocol, and has only been added as an extension within 1192 the XML stream rather than enforced by the protocol. 1194 5.2.6.2. Surveillance of Communication 1196 The XMPP protocol specifies the standard by which communication of 1197 channels may be encrypted, but it does not provide visibility to 1198 clients of whether their communications are encrypted on each link. 1199 In particular, even when both clients ensure that they have an 1200 encrypted connection to their XMPP server to ensure that their local 1201 network is unable to read or disrupt the messages they send, the 1202 protocol does not provide visibility into the encryption status 1203 between the two servers. As such, clients may be subject to 1204 selective disruption of communications by an intermediate network 1205 which disrupts communications based on keywords found through Deep 1206 Packet Inspection. While many operators have commited to only 1207 establishing encrypted links from their servers in recognition of 1208 this vulnerability, it remains impossible for users to audit this 1209 behavior and encrypted connections are not required by the protocol 1210 itself [xmppmanifesto]. 1212 In particular, section 13.14 of the protocol specification [RFC6120] 1213 explicitly acknowledges the existence of a downgrade attack where an 1214 adversary controlling an intermediate network can force the inter 1215 domain federation between servers to revert to a non-encrypted 1216 protocol were selective messages can then be disrupted. 1218 5.2.6.3. Group Chat Limitations 1220 Group chat in the XMPP protocol is defined as an extension within the 1221 XML specification of the XMPP protocol (https://xmpp.org/extensions/ 1222 xep-0045.html). However, it is not encoded or required at a protocol 1223 level, and not uniformly implemented by clients. 1225 The design of multi-user chat in the XMPP protocol suffers from 1226 extending a protocol that was not designed with assembly of many 1227 users in mind. In particular, in the federated protocol provided by 1228 XMPP, multi-user communities are implemented with a distinguished 1229 'owner', who is granted control over the participants and structure 1230 of the conversation. 1232 Multi-user chat rooms are identified by a name specified on a 1233 specific server, so that while the overall protocol may be federated, 1234 the ability for users to assemble in a given community is moderated 1235 by a single server. That server may block the room and prevent 1236 assembly unilaterally, even between two users neither of whom trust 1237 or use that server directly. 1239 5.2.7. Peer to Peer 1241 Peer-to-Peer (P2P) is a network architecture (defined in [RFC7574]) 1242 in which all the participant nodes are equally responsible engaged 1243 into the storage and dissemination of information. A P2P network is 1244 a logical overlay that lives on top of the physical network, and 1245 allows nodes (or "peers") participating to it to establish contact 1246 and exchange information directly from one to each other. The 1247 implementation of a P2P network may very widely: it may be structured 1248 or unstructured, and it may implement stronger or weaker 1249 cryptographic and anonymity properties. While its most common 1250 application has traditionally been file-sharing (and other types of 1251 content delivery systems), P2P is increasingly becoming a popular 1252 architecture for networks and applications that require (or 1253 encourage) decentralization. A prime example is Bitcoin (and similar 1254 cryptocurrencies), as well as Skype, Spotify and other proprietary 1255 multimedia applications. 1257 In a time of heavily centralized online services, peer-to-peer is 1258 often seen as an alternative, more democratic, and resistant 1259 architecture that displaces structures of control over data and 1260 communications and delegates all peers equally to be responsible for 1261 the functioning, integrity, and security of the data. While in 1262 principle peer-to-peer remains critical to the design and development 1263 of future content distribution, messaging, and publishing systems, it 1264 poses numerous security and privacy challenges which are mostly 1265 delegated to individual developers to recognize, analyze, and solve 1266 in each implementation of a given P2P network. 1268 5.2.7.1. Network Poisoning 1270 Since content, and in some occasions peer lists, are safeguarded and 1271 distributed by its members, P2P networks are prone to what are 1272 generally defined as "poisoning attacks". Poisoning attacks might be 1273 directed directly at the data that is being distributed, for example 1274 by intentionally corrupting it, or at the index tables used to 1275 instruct the peers where to fetch the data, or at routing tables, 1276 with the attempt of providing connecting peers with lists of rogue or 1277 non-existing peers, with the intention to effectively cause a Denial 1278 of Service on the network. 1280 5.2.7.2. Throttling 1282 Peer-to-Peer traffic (and BitTorrent in particular) represents a high 1283 percentage of global Internet traffic and it has become increasingly 1284 popular for Internet Service Providers to perform throttling of 1285 customers lines in order to limit bandwidth usage [torrentfreak1] and 1286 sometimes probably as an effect of the ongoing conflict between 1287 copyright holders and file-sharing communities [wikileaks]. 1289 Throttling the peer-to-peer traffic makes some uses of P2P networks 1290 ineffective and it might be coupled with stricter inspection of 1291 users' Internet traffic through Deep Packet Inspection techniques 1292 which might pose additional security and privacy risks. 1294 5.2.7.3. Tracking and Identification 1296 One of the fundamental and most problematic issues with traditional 1297 peer-to-peer networks is a complete lack of anonymization of its 1298 users. For example, in the case of BitTorrent, all peers' IP 1299 addresses are openly available to the other peers. This has lead to 1300 an ever-increasing tracking of peer-to-peer and file-sharing users 1301 [ars]. As the geographical location of the user is directly exposed, 1302 and so could be his identity, the user might become target of 1303 additional harassment and attacks, being of physical or legal nature. 1304 For example, it is known that in Germany law firms have made 1305 extensive use of peer-to-peer and file-sharing tracking systems in 1306 order to identify downloaders and initiate legal actions looking for 1307 compensations [torrentfreak2]. 1309 It is worth nothing that there are varieties of P2P networks that 1310 implement cryptographic practices and that introduce anonymization of 1311 its users. Such implementations proved to be successful in resisting 1312 censorship of content, and tracking of the network peers. A primary 1313 example is FreeNet [freenet1], a free software application designed 1314 to significantly increase the difficulty of users and content 1315 identification, and dedicated to foster freedom of speech online 1316 [freenet2]. 1318 5.2.7.4. Sybil Attacks 1320 In open-membership P2P networks, a single attacker can pretend to be 1321 many participants, typically by creating multiple fake identities of 1322 whatever kind the P2P network uses [Douceur]. Attackers can use 1323 Sybil attacks to bias choices the P2P network makes collectively 1324 toward the attacker's advantage, e.g., by making it more likely that 1325 a particular data item (or some threshold of the replicas or shares 1326 of a data item) are assigned to attacker-controlled participants. If 1327 the P2P network implements any voting, moderation, or peer review- 1328 like functionality, Sybil attacks may be used to "stuff the ballots" 1329 toward the attacker's benefit. Companies and governments can use 1330 Sybil attacks on discussion-oriented P2P systems for "astroturfing" 1331 or creating the appearance of mass grassroots support for some 1332 position where there is none in reality. 1334 5.2.7.5. Conclusions 1336 Encrypted P2P and Anonymous P2P networks already emerged and provided 1337 viable platforms for sharing material, publish content anonymously, 1338 and communicate securely [bitmessage]. If adopted at large, well- 1339 designed and resistant P2P networks might represent a critical 1340 component of a future secure and distributed Internet, enabling 1341 freedom of speech and freedom of information at scale. 1343 5.2.8. Virtual Private Network 1345 5.2.8.1. Introduction 1347 A Virtual Private Network (VPN) is a point-to-point connection that 1348 enables two computers to communicate over an encrypted tunnel. There 1349 are multiple implementations and protocols used in provisioning a 1350 VPN, and they generally diversify by encryption protocol or 1351 particular requirements, most commonly in proprietary and enterprise 1352 solutions. VPNs are used commonly either to enable some devices to 1353 communicate through peculiar network configurations, or in order to 1354 use some privacy and security properties in order to protect the 1355 traffic generated by the end user; or both. VPNs have also become a 1356 very popular technology among human rights defenders, dissidents, and 1357 journalists worldwide to avoid local illegitimate wiretapping and 1358 eventually also to circumvent censorship. Among human rights 1359 defenders VPNs are often debated as a potential alternative to Tor or 1360 other anonymous networks. Such comparison is misleading, as some of 1361 the privacy and security properties of VPNs are often misunderstood 1362 by less tech-savvy users, which could ultimately lead to unintended 1363 problems. 1365 As VPNs increased in popularity, commercial VPN providers have 1366 started growing in business and are very commonly picked by human 1367 rights defenders and people at risk, as they are normally provided 1368 with an easy-to-use service and sometimes even custom applications to 1369 establish the VPN tunnel. Not being able to control the 1370 configuration of the network, and even less so the security of the 1371 application, assessing the general privacy and security state of 1372 common VPNs is very hard. Often such services have been discovered 1373 leaking information, and their custom applications have been found 1374 flawed. While Tor and similar networks receive a lot of scrutiny 1375 from the public and the academic community, commercial or non- 1376 commercial VPN networks are way less analyzed and understood, and it 1377 might be valuable to establish some standards to guarantee a minimal 1378 level of privacy and security to those who need them the most. 1380 5.2.8.2. False sense of Anonymity 1382 One of the common misconception among users of VPNs is the level of 1383 anonymity VPN can provide. This sense of anonymity can be betrayed 1384 by a number of attacks or misconfigurations of the VPN provider. It 1385 is important to remember that, contrarily to Tor and similar systems, 1386 VPN was not designed to provide anonymity properties. From a 1387 technical point of view, the VPN might leak identifiable information, 1388 or might be subject of correlation attacks that could expose the 1389 originating address of the connecting user. Most importantly, it is 1390 vital to understand that commercial and non-commercial VPN providers 1391 are bound by the law of the jurisdiction they reside in or in which 1392 their infrastructure is located, and they might be legally forced to 1393 turn over data of specific users if legal investigations or 1394 intelligence requirements dictate so. In such cases, if the VPN 1395 providers retain logs, it is possible that the information of the 1396 user is provided to the user's adversary and leads to his or her 1397 identification. 1399 5.2.8.3. Logging 1401 With VPN being point-to-point connections, the service providers are 1402 in fact able to observe the original location of the connecting users 1403 and they are able to track at what time they started their session 1404 and eventually also to which destinations they're trying to connect 1405 to. If the VPN providers retain logs for long enough, they might be 1406 forced to turn over the relevant data or they might be otherwise 1407 compromised, leading to the same data getting exposed. A clear log 1408 retaining policy could be enforced, but considering that countries 1409 enforce very different levels of data retention policies, VPN 1410 providers should at least be transparent on what information do they 1411 store and for how long is being kept. 1413 5.2.8.4. 3rd Party Hosting 1415 VPN providers very commonly rely on 3rd parties to provision the 1416 infrastructure that is later going to be used to run VPN endpoints. 1417 For example, they might rely on external dedicated server hosting 1418 providers, or on uplink providers. In those cases, even if the VPN 1419 provider itself isn't retaining any significant logs, the information 1420 on the connecting users might be retained by those 3rd parties 1421 instead, introducing an additional collection point for the 1422 adversary. 1424 5.2.8.5. IPv6 Leakage 1426 Some studies proved that several commercial VPN providers and 1427 applications suffer of critical leakage of information through IPv6 1428 due to improper support and configuration [PETS2015VPN]. This is 1429 generally caused by a lack of proper configuration of the client's 1430 IPv6 routing tables. Considering that most popular browsers and 1431 similar applications have been supporting IPv6 by default, if the 1432 host is provided with a functional IPv6 configuration, the traffic 1433 that is generated might be leaked if the VPN application isn't 1434 designed to manipulate such traffic properly. 1436 5.2.8.6. DNS Leakage 1438 Similarly, VPN services that aren't handling DNS requests and are not 1439 running DNS servers of their own, might be prone to DNS leaking which 1440 might not only expose sensitive information on the activity of the 1441 user, but could also potentially lead to DNS hijacking attacks and 1442 following compromises. 1444 5.2.8.7. Traffic Correlation 1446 As revelations of mass surveillance have been growing in the press, 1447 additional details on attacks on secure Internet communications have 1448 come to the public's attention. Among these, VPN appeared to be a 1449 very interesting target for attacks and collection efforts. Some 1450 implementations of VPN appear to be particularly vulnerable to 1451 identification and collection of key exchanges which, some Snowden 1452 documents revealed, are systematically collected and stored for 1453 future reference. The ability of an adversary to monitor network 1454 connections at many different points over the Internet, can allow 1455 them to perform traffic correlation attacks and identify the origin 1456 of certain VPN traffic by cross referencing the connection time of 1457 the user to the endpoint and the connection time of the endpoint to 1458 the final destination. These types of attacks, although very 1459 expensive and normally only performed by very resourceful 1460 adversaries, have been documented [spiegel] to be already in practice 1461 and could completely vanify the use of a VPN and ultimately expose 1462 the activity and the identity of a user at risk. 1464 5.2.9. HTTP Status Code 451 1466 Every Internet user has run into the '404 Not Found' Hypertext 1467 Transfer Protocol (HTTP) status code when trying, and failing, to 1468 access a particular website. It is a response status that the server 1469 sends to the browser, when the server cannot locate the URL. '403 1470 Forbidden' is another example of this class of code signals that 1471 gives users information about what is going on. In the '403' case 1472 the server can be reached, but is blocking the request because the 1473 user is trying to access content forbidden to them. This can be 1474 because the specific user is not allowed access to the content (like 1475 a government employee trying to access pornography on a work- 1476 computer) or because access is restricted to all users (like social 1477 network sites in certain countries). As surveillance and censorship 1478 of the Internet is becoming more commonplace, voices were raised at 1479 the IETF to introduce a new status code that indicates when something 1480 is not available for 'legal reasons' (like censorship): 1482 The 451 status code would allow server operators to operate with 1483 greater transparency in circumstances where issues of law or public 1484 policy affect their operation. This transparency may be beneficial 1485 both to these operators and to end-users [Bray]. 1487 The status code would be named '451', a reference to Bradbury's 1488 famous novel on censorship 1490 During the IETF meeting in Dallas, there was discussion about the 1491 usefulness of '451'. The main tension revolved around the lack of an 1492 apparent machine-readable technical use of the information. The 1493 extent to which '451' is just 'political theatre' or whether it has a 1494 concrete technical use was heatedly debated. Some argued that 'the 1495 451 status code is just a status code with a response body' others 1496 said it was problematic because 'it brings law into the picture'. 1497 Again others argued that it would be useful for individuals, or 1498 organizations like the 'Chilling Effects' project, crawling the web 1499 to get an indication of censorship (IETF discussion on '451' - 1500 author's field notes March 2015). There was no outright objection 1501 during the Dallas meeting against moving forward on status code 1502 '451', and on December 18, 2015 the Internet Engineering Steering 1503 Group approved publication of 'An HTTP Status Code to Report Legal 1504 Obstacles'. It is still in the process of becoming an RFC, but could 1505 effectively be used from the day of approval. 1507 What is interesting about this particular case is that not only 1508 technical arguments but also the status code's outright potential 1509 political use for civil society played a substantial role in shaping 1510 the discussion, and the decision to move forward with this 1511 technology. 1513 It is however important to note that 451 is not a solution to detect 1514 all occasions of censorship. A large swath of Internet filtering 1515 occurs in the network rather than the server itself. For these forms 1516 of censorship 451 plays a limited role, as the servers will not be 1517 able to send the code, because they haven't received the requests (as 1518 is the case with servers with resources blocked by the Chinese Golden 1519 shield). Such filtering regimes are unlikely to voluntarily inject a 1520 451 status code. The use of 451 is most likely to apply in the case 1521 of cooperative, legal versions of content removal resulting from 1522 requests to providers. One can think of content that is removed or 1523 blocked for legal reasons, like copyright infringement, gambling 1524 laws, child abuse, et cetera. The major use case is thus clearly on 1525 the Web server itself, not the network. Large Internet companies and 1526 search engines are constantly asked to censor content in various 1527 jurisdictions. 451 allows this to be easily discovered, for instance 1528 by initiatives like the Lumen Database. In the case of adversarial 1529 blocking done by a filtering entity on the network 451 is less 1530 useful. 1532 Overall, the strength of 451 lies in its ability to provide 1533 transparency by giving the reason for blocking, and giving the end- 1534 user the ability to file a complaint. It allows organizations to 1535 easily measure censorship in an automated way, and prompts the user 1536 to access the content via another path (e.g. TOR, VPNs) when (s)he 1537 encounters the 451 status code. 1539 Status code 451 impact human rights by making censorship more 1540 transparent and measurable. The status code increases transparency 1541 both by signaling the existence of censorship (instead of a much more 1542 broad HTTP error message like HTTP status code 404) as well as 1543 providing details of the legal restriction, which legal authority is 1544 imposing it, and what class of resources it applies to. This 1545 empowers the user to seek redress. 1547 5.2.10. Middleboxes 1549 On the current Internet, transparency on how packets reach a 1550 destination is no longer a given. This is due to the increased 1551 presence of firewalls, spam filters, and network address translators 1552 networks (NATs) - or middleboxes as these hosts are often called - 1553 that make use of higher-layer fields to function [Walfish]. This 1554 development is contentious. The debate also unfolded at the IETF, 1555 specifically at the Session Protocol Underneath Datagrams (SPUD) 1556 Birds of a Feather (BOF) meeting held at the IETF conference in March 1557 2015. The discussion at the BOF focused on questions about adding 1558 meta-data, or other information to traffic flows, to enable the 1559 sharing of information with middleboxes in that flow. During the 1560 sessions two competing arguments were distilled. On the one hand 1561 adding additional data would allow for network optimization, and 1562 hence improve traffic carriage. On the other hand, there are risks 1563 of information leakage and other privacy and security concerns. 1565 Middleboxes, and the protocols guiding them, influence individuals' 1566 ability to communicate online freely and privately. Repeatedly 1567 mentioned in the discussion was the danger of censorship that comes 1568 with middleboxes, and the IETF's role to prevent such censorship from 1569 happening. Middleboxes are becoming a proxy for the debate on the 1570 extent to which commercial interests are a valid reason to undermine 1571 the end-to-end principle. The potential for abuse and censoring, and 1572 thus ultimately the impact of middleboxes on the Internet as a place 1573 of unfiltered, unmonitored freedom of speech, is real. It is 1574 impossible to make any definitive statements about the direction the 1575 debate on middleboxes will take at the IETF. The opinions expressed 1576 in the SPUD BOF and by the various interviewees indicate that a 1577 majority of engineers are trying to mitigate the negative effects of 1578 middleboxes on freedom of speech, but their ability to act is limited 1579 by their larger commercial context that is expanding the use of 1580 middleboxes. 1582 5.2.11. DDOS attacks 1584 Are Distributed Denial of Service (DDoS) attacks a legitimate form of 1585 online protest protected by the right to freedom of speech and 1586 association? Can they be seen as the equivalent to 'million-(wo)men 1587 marches', or sit-ins? Or are they a threat to freedom of expression 1588 and access to information, by limiting access to websites and in 1589 certain cases the freedom of speech of others? These questions are 1590 crucial in our day and age, where political debates, civil 1591 disobedience and other forms of activism are increasingly moving 1592 online. 1594 Many individuals, not excluding IETF engineers, have argued that DDoS 1595 attacks are fundamentally against freedom of speech. Technically 1596 DDoS attacks are when multiple computers overload the bandwidth or 1597 resources of a website (or other system) by flooding it with traffic, 1598 causing it to temporarily stop being available to users. In their 1599 2010 report Zuckerman et al argue that DDoS attacks are a bad thing 1600 because they are increasingly used by governments to attack and 1601 silence critics. Their research demonstrates that in many countries 1602 independent media outlets and human rights organizations are the 1603 victim of DDoS attacks, which are directly or indirectly linked to 1604 their governments. These types of attacks are particularly 1605 complicated because attribution is difficult, creating a situation in 1606 which governments can effectively censor content, while being able to 1607 deny involvement in the attacks [Zuckerman]. DDoS attacks can thus 1608 stifle freedom of expression, complicate the ability of independent 1609 media and human rights organizations to exercise their right to 1610 (online) freedom of association, while facilitating the ability of 1611 governments to censor dissent. When it comes to comparing DDoS 1612 attacks to protests in offline life, it is important to remember that 1613 only a limited number of DDoS attacks involved solely willing 1614 participants. In most cases, the clients are hacked computers of 1615 unrelated parties that have not consented to being part of a DDoS 1616 (for exceptions see Operation Abibil [Abibil] or the Iranian Green 1617 Movement DDoS [GreenMovement]). 1619 In addition, DDoS attacks are increasingly used as an extortion 1620 tactic, with criminals flooding a website - rendering it inaccessible 1621 - until the owner pays them a certain amount of money to stop the 1622 attack. The costs of mitigating such attacks, either by improving 1623 security to prevent them or paying off the attackers, ends up being 1624 paid by the consumer. 1626 All of these issues seem to suggest that the IETF should try to 1627 ensure that their protocols cannot be used for DDoS attacks. 1628 Decreasing the number of vulnerabilities in the network stacks of 1629 routers or computers, reducing flaws in HTTPS implementations, and 1630 depreciating non-secure HTTP protocols could address this issue. The 1631 IETF can clearly play a role in bringing about some of these changes, 1632 and has indicated in [RFC7258] its commitment to mitigating 1633 'pervasive monitoring (...) in the design of IETF protocols, where 1634 possible.' This means the use of encryption should become standard. 1635 Effectively, for the web this means standardized use of HTTPS. The 1636 IETF could redirect its work such that HTPPS becomes part-and-parcel 1637 of its standards. However, next to the various technical trade-offs 1638 that this might lead to it is important to consider that DDoS attacks 1639 are sometimes seen as a method for exercising freedom of speech. 1641 DDoS although disruptive, and silencing at times, can also enable as 1642 protest and speech. Or as Sauter [Sauter] argues: 'though DDoS as a 1643 tactic is still relatively novel, it fits within a centuries- long 1644 tradition of breaking laws and disrupting business as usual to make a 1645 political point. These actions aren't simply disruption for 1646 disruption's sake. Rather they serve to help the activist or 1647 dissenter to direct the attention of the public through the 1648 interpolation of difference into routine.' (30-31). An often heard 1649 argument against DDoS attacks is that you cannot construe it as a 1650 means to exercise your right to freedom of speech, when the means 1651 used effectively impede the right of the party on the receiving end 1652 of the attack to exercise that same right. The problem with this 1653 line of argumentation is that it conveniently ignores the fact that 1654 online DDoS attacks are often one of the few effective ways for 1655 activists to gain the attention of the media, the government or other 1656 parties of interest. Simply putting up a website for a cause won't 1657 garner the same amount of attention as directly confronting the issue 1658 via the website of the individual or organization at the heart of the 1659 issue. The ability of activists to do so should be protected, 1660 especially considering the fact that as Sauter (2014:4) explains: 1661 'Collectively, we have allowed the construction of an entire public 1662 sphere, the Internet, which by accidents of evolution and design, has 1663 none of the inherent free speech guarantees we have come to expect. 1664 Dissenting voices are pushed out of the paths of potential audiences, 1665 effectively removing them from the public discourse. There is 1666 nowhere online for an activist to stand with her friends and her 1667 sign. She might set up a dedicated blog--which may or may not ever 1668 be read--but it is much harder for her to stand collectively with 1669 others against a corporate giant in the online space.' Although the 1670 Internet is often compared to public space, it is not. Rather the 1671 opposite. The Internet is almost entirely owned by private entities. 1672 And the IETF plays a crucial role in developing this privatized 1673 commercialized Internet. 1675 From a legal and political perspective, the IETF does not have the 1676 legitimacy to determine when a DDoS is legitimate (in legal or 1677 political terms). It does not have the capability to make this 1678 judgment as a matter of public policy and subsequently translate it 1679 to code. Nor should the IETF try to do so. From a technical 1680 perspective, the difference between a 'legitimate' and 'illegitimate' 1681 DDoS attack is meaningless because it would be extremely difficult 1682 for the IETF to engineer a way to detect that difference. In 1683 addition, there is a need for the IETF to be consistent in the face 1684 of attacks (an attack is an attack is an attack) to maintain the 1685 viability of the network. Arguing that some DDoS attacks should be 1686 allowed, based on the motivation of the attackers complicates the 1687 work of the IETF. Because it approaches PM regardless of the 1688 motivation of the attackers (see [RFC7258]) for reasoning), taking 1689 the motivation of the attackers into account for DDoS would 1690 indirectly undermine the ability of the IETF to protect the right to 1691 privacy because it introduces an element of inconsistency into how 1692 the IETF deals with attacks. 1694 David Clark recently published a paper warning that the future of the 1695 Internet is in danger. He argues that the private sector control 1696 over the Internet is too strong, limiting the myriad of ways in which 1697 it can be used [Daedalus], including for freedom of speech. But just 1698 because freedom of speech, dissent, and protest are human rights, and 1699 DDoS is a potential expression of those rights, doesn't mean that 1700 DDoS in and of itself is a right. To widen the analogy, just because 1701 the Internet is a medium through which the right to freedom of 1702 expression can be exercised does not make access to the Internet or 1703 specific ICTs or NCTs a human right. Uses of DDoS might or might not 1704 be legitimate for political reasons, but the IETF has no means or 1705 methods to assess this, and in general enabling DDoS would mean a 1706 deterioration of the network and thus freedom of expression. 1708 In summation, the IETF cannot be expected to take a moral stance on 1709 DDoS attacks, or create protocols to enable some attacks and inhibit 1710 others. But what it can do is critically reflect on its role in 1711 creating a commercialized Internet without a defacto public space or 1712 inherent protections for freedom of speech. 1714 5.3. Model for developing human rights protocol considerations 1716 Having established how human rights relate to standards and 1717 protocols, a common vocabulary of technical concepts that impact 1718 human rights and how these technical concept can be combined to 1719 ensure that the Internet remains an enabling environment for human 1720 rights means the contours of a model for developing human rights 1721 protocol considerations has taken shape. This subsection provides 1722 the last step by detailing how the technical concepts identified 1723 above relate to human rights, and what questions engineers should ask 1724 themselves when developing or improving protocols. 1726 5.3.1. Human rights threats 1728 The human rights threats on the Internet come in a myriad of forms. 1729 Protocols and standards can harm or enable the right to freedom of 1730 expression, right to non-discrimination, right to equal protection, 1731 right to be presumed innocence, right to participate in cultural 1732 life, arts and science, right to freedom of assembly and association, 1733 and the right to security. An end-user who is denied access to 1734 certain services, data or websites may be unable to disclose vital 1735 information about the malpractices of a government or other 1736 authority. A person whose communications are monitored may be 1737 prevented from exercising their right to freedom of association. In 1738 a worst-case scenario, protocols that leak information can lead to 1739 physical danger. A realistic example to consider is when opposition 1740 leaders in totalitarian regimes are subjected to torture on the basis 1741 of information gathered by the regime through information leakage in 1742 protocols. 1744 This sections details several 'common' threats to human rights, 1745 indicating how each of these can lead to human rights violations/ 1746 harms and present several examples of how these threats to human 1747 rights materialize on the Internet. This threat modeling is inspired 1748 by [RFC6973] Privacy Considerations for Internet Protocols, which 1749 bases itself on security threat analysis. This method is by no means 1750 a perfect solution for assessing human rights risks in Internet 1751 protocols and systems; it is however the best approach currently 1752 available. Certain human rights threats are indirectly considered in 1753 Internet protocols as part of the standard privacy and security 1754 considerations [RFC3552]. Others suggested are tailored specifically 1755 to human rights, and represents considerations not currently 1756 considered in other RFCs. 1758 Many threats, enablers and risks are linked to different rights. 1759 This is not unsurprising if one takes into account that human rights 1760 are interrelated, interdependent and universal. Here however we're 1761 not discussing all human rights because not all human rights are 1762 relevant to ICTs in general and protocols and standards in 1763 particular. This is by no means an attempt to cherry picks rights, 1764 if other rights seem relevant, please contact the authors and/or the 1765 hrpc mailinglist. 1767 5.3.2. Guidelines for human rights considerations 1769 This section provides guidance for document authors in the form of a 1770 questionnaire about a protocol being designed. The questionnaire may 1771 be useful at any point in the design process, particularly after 1772 document authors have developed a high-level protocol model as 1773 described in [RFC4101]. 1775 Note that the guidance provided in this section does not recommend 1776 specific practices. The range of protocols developed in the IETF is 1777 too broad to make recommendations about particular uses of data or 1778 how human rights might be balanced against other design goals. 1779 However, by carefully considering the answers to each question 1780 mentioned under 7.3, document authors should be able to produce a 1781 comprehensive analysis that can serve as the basis for discussion of 1782 whether the protocol adequately protects against human rights 1783 threats. This guidance is meant to help the thought process of a 1784 human rights analysis; it does not provide specific directions for 1785 how to write a human rights protocol considerations section 1786 (following the example set in [RFC6973]). 1788 5.3.2.1. Technical concepts as they relate to human rights 1790 5.3.2.1.1. Connectivity 1792 Does your protocol honor the end-to-end principle? 1794 Impacts: 1796 - Right to freedom of expression 1797 - Right to freedom of assembly and association 1799 5.3.2.1.2. Privacy 1801 Did you have a look at the Guidelines in the Privacy Considerations 1802 for Internet Protocols [RFC6973] section 7? Does your protocol in 1803 any way impact the confidentiality of protocol metadata? Does your 1804 protocol countering traffic analysis, or data minimisation? 1806 Impacts: 1808 - Right to freedom of expression 1810 - Right to non-discrimination 1812 - Right to be presumed innocent 1814 5.3.2.1.3. Content agnosticism 1816 If your protocol impacts packet handling, does it look at the packet 1817 content? Is it making decisions based on the content of the packet? 1818 Is the protocol transparent about its decision? Does your protocol 1819 prioritize certain content or services over others? 1821 Impacts: 1823 - Right to freedom of expression 1825 - Right to non-discrimination 1827 - Right to equal protection 1829 - Right to be presumed innocent 1831 5.3.2.1.4. Security 1833 Did you have a look at Guidelines for Writing RFC Text on Security 1834 Considerations [RFC3552]? 1836 Impacts: 1838 - Right to freedom of expression 1840 - Right to freedom of assembly and association 1842 - Right to non discrimination & Right to be presumed innocent) 1844 5.3.2.1.5. Internationalization 1846 Does your protocol have text strings that are readable or entered by 1847 humans? Does your protocol allow Unicode encoded in UTF-8 only, 1848 thereby shifting conversion issues away from individual choices? Did 1849 you have a look at [RFC6365]? 1851 Impacts: 1853 - Right to freedom of expression 1855 - Right to political participation 1857 - Right to participate in cultural life, arts and science 1859 - Right to political participation 1861 5.3.2.1.6. Censorship resistance 1863 Does this protocol introduce new identifiers that might be associated 1864 with persons or content? Does your protocol make it apparent or 1865 transparent when filtering happens? 1867 Identifiers of content exposed within a protocol might be used to 1868 facilitate censorship, as in the case of HTTP in this particular 1869 scenario [...]. 1871 Impacts: 1873 - Right to freedom of expression 1875 - Right to political participation 1877 - Right to participate in cultural life, arts and science 1879 - Right to freedom of assembly and association 1881 5.3.2.1.7. Open Standards 1883 Is your protocol fully documented in a way that it could be easily 1884 implemented, improved, build upon and/or further developed. Is there 1885 any proprietary code needed for the implementation, running or 1886 further development of your protocol? 1888 Impacts: 1890 - Right to freedom of expression 1891 - Right to participate in cultural life, arts and science 1893 5.3.2.1.8. Heterogeneity Support 1895 Does your protocol support heterogeneity by design? Does your 1896 protocol allow for multiple types of hardware? Does your protocol 1897 allow for multiple types of application protocols? 1899 Impacts: 1901 - Right to freedom of expression 1903 5.3.2.1.9. Anonymity 1905 Did you have a look at the Privacy Considerations for Internet 1906 Protocols [RFC6973], especially section 6.1.1 ? 1908 Impacts: 1910 - Right to non-discrimination 1912 - Right to political participation 1914 - Right to freedom of assembly and association 1916 - Right to security 1918 - Right to be presumed innocent 1920 5.3.2.1.10. Pseudonymity 1922 Did you have a look at the Privacy Considerations for Internet 1923 Protocols [RFC6973], especially section 6.1.2 ? 1925 Impacts: 1927 - Right to non-discrimination 1929 - Right to freedom of assembly and association 1931 5.3.2.1.11. Accessibility 1933 When websites, web technologies, or web tools are badly designed, 1934 they can create barriers that exclude people from using the Web. Is 1935 your protocol designed to provide an enabling environment for people 1936 with disabilities? It might be relevant to look at the W3C Web 1937 Accessibility Initiative for examples and guidance.Is your protocol 1938 optimized for low bandwidth and high latency connections? Could your 1939 protocol also be developed in a stateless manner? 1941 Impacts: 1943 - Right to non-discrimination 1945 - Right to freedom of assembly and association 1947 - Right to education 1949 - Right to political participation 1951 5.3.2.1.12. Localization 1953 Does your protocol live up to standards of internationalization? 1954 Have you considered localizing your protocol for relevant audiences? 1956 Impacts: 1958 - Right to non-discrimination 1960 - Right to participate in cultural life, arts and science 1962 5.3.2.1.13. Decentralization 1964 Does your protocol contribute to more centralized points of control? 1965 Can your protocol be implemented without one single point of control? 1966 If applicable, can your protocol be deployed in a federated manner? 1967 What is the potential for discrimination against users of your 1968 protocol? How can use of your protocol be used to implicate users? 1970 Impacts: 1972 - Right to freedom of assembly and association 1974 - Right to be presumed innocent 1976 5.3.2.1.14. Reliability 1978 Is your protocol fault tolerant? Does it degrade gracefully? Do you 1979 have a documented way to announce degradation? Do you have measures 1980 in place for recovery or partial healing from failure? Is your 1981 protocol able to maintain dependability and performance in the face 1982 of unanticipated changes or circumstances? 1984 Impacts: 1986 - Right to security 1988 5.3.2.1.15. Confidentiality 1990 (cf [RFC6973] ) Which information related to identifiers or data is 1991 exposed to each other protocol entity (i.e., recipients, 1992 intermediaries, and enablers)? Are there ways for protocol 1993 implementers to choose to limit the information shared with each 1994 entity? Are there operational controls available to limit the 1995 information shared with each entity? 1997 What controls or consent mechanisms does the protocol define or 1998 require before personal data or identifiers are shared or exposed via 1999 the protocol? If no such mechanisms or controls are specified, is it 2000 expected that control and consent will be handled outside of the 2001 protocol? 2003 Does the protocol provide ways for initiators to share different 2004 information with different recipients? If not, are there mechanisms 2005 that exist outside of the protocol to provide initiators with such 2006 control? 2008 Does the protocol provide ways for initiators to limit which 2009 information is shared with intermediaries? If not, are there 2010 mechanisms that exist outside of the protocol to provide users with 2011 such control? Is it expected that users will have relationships that 2012 govern the use of the information (contractual or otherwise) with 2013 those who operate these intermediaries? 2015 Does the protocol provide ways for initiators to express individuals' 2016 preferences to recipients or intermediaries with regard to the 2017 collection, use, or disclosure of their personal data? 2019 Impacts: 2021 - Right to security 2023 5.3.2.1.16. Integrity 2025 Does your protocol maintain and assure the accuracy of data? Does 2026 your protocol maintain and assure the consistency of data? Does your 2027 protocol in any way allow for the data to be (intentionally or 2028 unintentionally) altered? 2030 Impacts: 2032 - Right to security 2034 5.3.2.1.17. Authenticity 2036 Do you have enough measures to confirm the truth of an attribute of a 2037 single piece of data or entity? Can the attributes get garbled along 2038 the way (see security)? If relevant have you implemented IPsec and 2039 other Standard Security Best Practices? 2041 Impacts: 2043 - Right to security 2045 5.3.2.1.18. Acceptability 2047 Do your protocols adhere to the principle of non-discrimination? Do 2048 your protocols adhere to the principle of content agnosticism? 2049 Impacts: 2051 - Right to education 2053 5.3.2.1.19. Availability 2055 Do your protocols use or depend on proprietary code? Also see 'Open 2056 Standards' above. Also see 'Connectivity' above. 2058 Impacts: 2060 - Right to education 2062 5.3.2.1.20. Adaptability 2064 Could your protocol stifle or hinder permissionless innovation in any 2065 way? See 'Connectivity' above 2067 Impacts: 2069 - Right to education 2071 6. Acknowledgements 2073 A special thanks to all members of the hrpc RG who contributed to 2074 this draft. The following deserve a special mention: 2076 - Joana Varon for helping draft the first iteration of the 2077 methodology, previous drafts and the direction of the film Net of 2078 Rights and working on the interviews at IETF92 in Dallas. 2080 - Daniel Kahn Gillmor (dkg) for helping with the first iteration of 2081 the glossary as well as a lot of technical guidance, support and 2082 language suggestions. 2084 - Claudio Guarnieri for writing the first iterations of the case 2085 studies on VPN, HTTP, and Peer to Peer. 2087 - Will Scott for writing the first iterations of the case studies on 2088 DNS, IP, XMPP. 2090 - Avri Doria for proposing writing a glossary in the first place, 2091 help writing the initial proposals and Internet Drafts and 2092 contributing to the glossary. 2094 and Stephane Bortzmeyer, Barry Shein, Joe Hall, and Tim Sammut who 2095 made a lot of excellent suggestions, many of which found their way 2096 directly into the text. We would also like to thank Molly Sauter, 2097 Arturo Filasto, Eleanor Saitta and all others who provided input on 2098 the draft or the conceptualization of the idea. 2100 7. Security Considerations 2102 As this document concerns a research document, there are no security 2103 considerations. 2105 8. IANA Considerations 2107 This document has no actions for IANA. 2109 9. Research Group Information 2111 The discussion list for the IRTF Human Rights Protocol Considerations 2112 proposed working group is located at the e-mail address hrpc@ietf.org 2113 [3]. Information on the group and information on how to subscribe to 2114 the list is at https://www.irtf.org/mailman/listinfo/hrpc 2116 Archives of the list can be found at: https://www.irtf.org/mail- 2117 archive/web/hrpc/current/index.html 2119 10. References 2121 10.1. Normative References 2123 [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 2124 Morris, J., Hansen, M., and R. Smith, "Privacy 2125 Considerations for Internet Protocols", RFC 6973, DOI 2126 10.17487/RFC6973, July 2013, 2127 . 2129 10.2. Informative References 2131 [Abbate] Abbate, J., "Inventing the Internet", MIT Press , 2000, 2132 . 2134 [Abibil] Danchev, D., "Dissecting 'Operation Ababil' - an OSINT 2135 Analysis", 2012, . 2138 [Appelbaum] 2139 Appelbaum, J., Gibson, A., Kabish, V., Kampf, L., and L. 2140 Ryge, "NSA targets the privacy-conscious", 2015, 2141 . 2144 [Babbie] Babbie, E., "The Basics of Social Research", Belmont CA 2145 Cengage , 2010. 2147 [Benkler] Benkler, Y., "The wealth of Networks - How social 2148 production transforms markets and freedom", New Haven and 2149 London - Yale University Press , 2006, 2150 . 2152 [Berners-Lee] 2153 Berners-Lee, T. and M. Fischetti, "Weaving the Web,", 2154 HarperCollins p 208, 1999. 2156 [Bless] Bless, R. and C. Orwat, "Values and Networks", 2015. 2158 [Blumenthal] 2159 Blumenthal, M. and D. Clark, "Rethinking the design of the 2160 Internet: The end-to-end arguments vs. the brave new 2161 world", ACM Transactions on Internet Technology, Vol. 1, 2162 No. 1, August 2001, pp 70-109. , 2001. 2164 [Bray] Bray, T., "A New HTTP Status Code for Legally-restricted 2165 Resources", 2016, . 2168 [Broeders] 2169 Broeders, D., "The public core of the Internet", WRR , 2170 2015, 2171 . 2174 [Brown] Brown, I. and M. Ziewitz, "A Prehistory of Internet 2175 Governance", Research Handbook on Governance of the 2176 Internet. Cheltenham, Edward Elgar. , 2013. 2178 [BrownMarsden] 2179 Brown, I. and C. Marsden, "Regulating code", MIT Press , 2180 2013, . 2182 [Brownetal] 2183 Brown, I., Clark, D., and D. Trossen, "Should specific 2184 values be embedded in the Internet Architecture?", Sigcomm 2185 , 2010, . 2188 [Clark] Clark, D., "The Design Philosophy of the DARPA Internet 2189 Protocols", Proc SIGCOMM 88, ACM CCR Vol 18, Number 4, 2190 August 1988, pp. 106-114. , 1988. 2192 [Clarketal] 2193 Clark, D., Wroclawski, J., Sollins, K., and R. Braden, 2194 "Tussle in cyberspace - defining tomorrow's Internet", ACM 2195 Digital Library , 2005, . 2198 [Collins] Collins, K., "Hacking Team's oppressive regimes customer 2199 list revealed in hack", 2015, 2200 . 2203 [Daedalus] 2204 Clark, D., "The Contingent Internet", Daedalus Winter 2205 2016, Vol. 145, No. 1. p. 9-17 , 2016, 2206 . 2208 [Davidsonetal] 2209 Davidson, A., Morris, J., and R. Courtney, "Strangers in a 2210 strange land", Telecommunications Policy Research 2211 Conference , 2002, 2212 . 2214 [Denardis14] 2215 Denardis, L., "The Global War for Internet Governance", 2216 Yale University Press , 2014, 2217 . 2219 [Denardis15] 2220 Denardis, L., "The Internet Design Tension between 2221 Surveillance and Security", IEEE Annals of the History of 2222 Computing (volume 37-2) , 2015, . 2224 [Denzin] Denzin, N. and Y. Lincoln, "Handbook of Qualitative 2225 Research", Thousand Oaks CA Sage , 2000, 2226 . 2229 [Doty] Doty, N., "Automated text analysis of Requests for Comment 2230 (RFCs)", 2014, . 2232 [Douceur] Douceur, J., "The Sybil Attack", 2002, 2233 . 2236 [Elahi] Elahi, T. and I. Goldberg, "CORDON - A taxonomy of 2237 Internet Censorship Resistance Strategies", 2012, 2238 . 2241 [FIArch] "Future Internet Design Principles", January 2012, 2242 . 2245 [FRAMEWORK] 2246 ISO/IEC, ., "Information technology - Framework for 2247 internationalization, prepared by ISO/IEC JTC 1/SC 22/WG 2248 20 ISO/IEC TR 11017", 1997. 2250 [Geertz] Clifford, G., "Kinship in Bali", Chicago University of 2251 Chicago Press. , 1975, 2252 . 2255 [Googlepatent] 2256 Google, ., "Method and device for network traffic 2257 manipulation", 2012, . 2260 [GreenMovement] 2261 Villeneuve, N., "Iran DDoS", 2009, 2262 . 2264 [HRC2012] United Nations Human Rights Council, "UN General Assembly 2265 Resolution "The right to privacy in the digital age" 2266 (A/C.3/68/L.45)", 2011, 2267 . 2269 [Haagsma] Haagsma, L., "Deep dive into QUANTUM INSERT", 2015, 2270 . 2273 [ICCPR] United Nations General Assembly, "International Covenant 2274 on Civil and Political Rights", 1976, 2275 . 2278 [ICESCR] United Nations General Assembly, "International Covenant 2279 on Economic, Social and Cultural Rights", 1966, 2280 . 2283 [Jabri] Jabri, V., "Discourses on Violence - conflict analysis 2284 reconsidered", Manchester University Press , 1996. 2286 [King] King, C., "Power, Social Violence and Civil Wars", 2287 Washington D.C. United States Institute of Peace Press , 2288 2007. 2290 [Lessig] Lessig, L., "Code - And Other Laws of Cyberspace, Version 2291 2.0.", New York Basic Books , 2006, . 2293 [Marcak] Marcak, B., Weaver, N., Dalek, J., Ensafi, R., Fifield, 2294 D., McKune, S., Rey, A., Scott-Railton, J., Deibert, R., 2295 and V. Paxson, "China's Great Fire Cannon", 2015, 2296 . 2298 [Marquis-Boire] 2299 Marquis-Boire, M., "Schrodinger's Cat Video and the Death 2300 of Clear-Text", 2014, . 2303 [Mueller] Mueller, M., "Networks and States", MIT Press , 2010, 2304 . 2306 [Musiani] Musiani, F., "Giants, Dwarfs and Decentralized 2307 Alternatives to Internet-based Services - An Issue of 2308 Internet Governance", Westminister Papers in Communication 2309 and Culture , 2015, . 2311 [NETmundial] 2312 NETmundial, "NETmundial Multistakeholder Statement", 2014, 2313 . 2316 [PETS2015VPN] 2317 Pera, V., Barbera, M., Tyson, G., Haddadi, H., and A. Mei, 2318 "A Glance through the VPN Looking Glass", 2015, 2319 . 2322 [Peterson] 2323 Peterson, A., Gellman, B., and A. Soltani, "Yahoo to make 2324 SSL encryption the default for Webmail users. Finally.", 2325 2013, . 2328 [RFC0226] Karp, P., "Standardization of host mnemonics", RFC 226, 2329 DOI 10.17487/RFC0226, September 1971, 2330 . 2332 [RFC0760] Postel, J., "DoD standard Internet Protocol", RFC 760, DOI 2333 10.17487/RFC0760, January 1980, 2334 . 2336 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, DOI 2337 10.17487/RFC0791, September 1981, 2338 . 2340 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, RFC 2341 793, DOI 10.17487/RFC0793, September 1981, 2342 . 2344 [RFC0894] Hornig, C., "A Standard for the Transmission of IP 2345 Datagrams over Ethernet Networks", STD 41, RFC 894, DOI 2346 10.17487/RFC0894, April 1984, 2347 . 2349 [RFC1035] Mockapetris, P., "Domain names - implementation and 2350 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 2351 November 1987, . 2353 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 2354 Communication Layers", STD 3, RFC 1122, DOI 10.17487/ 2355 RFC1122, October 1989, 2356 . 2358 [RFC1631] Egevang, K. and P. Francis, "The IP Network Address 2359 Translator (NAT)", RFC 1631, DOI 10.17487/RFC1631, May 2360 1994, . 2362 [RFC1958] Carpenter, B., Ed., "Architectural Principles of the 2363 Internet", RFC 1958, DOI 10.17487/RFC1958, June 1996, 2364 . 2366 [RFC1984] IAB and , "IAB and IESG Statement on Cryptographic 2367 Technology and the Internet", BCP 200, RFC 1984, DOI 2368 10.17487/RFC1984, August 1996, 2369 . 2371 [RFC2277] Alvestrand, H., "IETF Policy on Character Sets and 2372 Languages", BCP 18, RFC 2277, DOI 10.17487/RFC2277, 2373 January 1998, . 2375 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2376 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 2377 December 1998, . 2379 [RFC2606] Eastlake 3rd, D. and A. Panitz, "Reserved Top Level DNS 2380 Names", BCP 32, RFC 2606, DOI 10.17487/RFC2606, June 1999, 2381 . 2383 [RFC2775] Carpenter, B., "Internet Transparency", RFC 2775, DOI 2384 10.17487/RFC2775, February 2000, 2385 . 2387 [RFC3365] Schiller, J., "Strong Security Requirements for Internet 2388 Engineering Task Force Standard Protocols", BCP 61, RFC 2389 3365, DOI 10.17487/RFC3365, August 2002, 2390 . 2392 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 2393 Text on Security Considerations", BCP 72, RFC 3552, DOI 2394 10.17487/RFC3552, July 2003, 2395 . 2397 [RFC3724] Kempf, J., Ed., Austein, R., Ed., and IAB, "The Rise of 2398 the Middle and the Future of End-to-End: Reflections on 2399 the Evolution of the Internet Architecture", RFC 3724, DOI 2400 10.17487/RFC3724, March 2004, 2401 . 2403 [RFC4084] Klensin, J., "Terminology for Describing Internet 2404 Connectivity", BCP 104, RFC 4084, DOI 10.17487/RFC4084, 2405 May 2005, . 2407 [RFC4101] Rescorla, E. and IAB, "Writing Protocol Models", RFC 4101, 2408 DOI 10.17487/RFC4101, June 2005, 2409 . 2411 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", RFC 2412 4303, DOI 10.17487/RFC4303, December 2005, 2413 . 2415 [RFC4906] Martini, L., Ed., Rosen, E., Ed., and N. El-Aawar, Ed., 2416 "Transport of Layer 2 Frames Over MPLS", RFC 4906, DOI 2417 10.17487/RFC4906, June 2007, 2418 . 2420 [RFC4949] Shirey, R., "Internet Security Glossary, Version 2", FYI 2421 36, RFC 4949, DOI 10.17487/RFC4949, August 2007, 2422 . 2424 [RFC5944] Perkins, C., Ed., "IP Mobility Support for IPv4, Revised", 2425 RFC 5944, DOI 10.17487/RFC5944, November 2010, 2426 . 2428 [RFC6120] Saint-Andre, P., "Extensible Messaging and Presence 2429 Protocol (XMPP): Core", RFC 6120, DOI 10.17487/RFC6120, 2430 March 2011, . 2432 [RFC6365] Hoffman, P. and J. Klensin, "Terminology Used in 2433 Internationalization in the IETF", BCP 166, RFC 6365, DOI 2434 10.17487/RFC6365, September 2011, 2435 . 2437 [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an 2438 Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 2439 2014, . 2441 [RFC7574] Bakker, A., Petrocco, R., and V. Grishchenko, "Peer-to- 2442 Peer Streaming Peer Protocol (PPSPP)", RFC 7574, DOI 2443 10.17487/RFC7574, July 2015, 2444 . 2446 [RFC7626] Bortzmeyer, S., "DNS Privacy Considerations", RFC 7626, 2447 DOI 10.17487/RFC7626, August 2015, 2448 . 2450 [RSF] RSF, ., "Syria using 34 Blue Coat Servers to spy on 2451 Internet users", 2013, . 2454 [Rachovitsa] 2455 Rachovitsa, A., "Engineering "Privacy by Design" in the 2456 Internet Protocols - Understanding Online Privacy both as 2457 a Technical and a Human Rights Issue in the Face of 2458 Pervasive Monitoring", International Journal of Law and 2459 Information Technology , 2015, . 2462 [Richie] Richie, J. and J. Lewis, "Qualitative Research Practice - 2463 A Guide for Social Science Students and Researchers", 2464 London Sage , 2003, . 2468 [Rideout] Rideout, A., "Making security easier", 2008, 2469 . 2472 [Saltzer] Saltzer, J., Reed, D., and D. Clark, "End-to-End Arguments 2473 in System Design", ACM TOCS, Vol 2, Number 4, November 2474 1984, pp 277-288. , 1984. 2476 [Sauter] Sauter, M., "The Coming Swarm", Bloomsbury, London , 2014. 2478 [Schillace] 2479 Schillace, S., "Default https access for Gmail", 2010, 2480 . 2483 [Schneier] 2484 Schneier, B., "Attacking Tor - how the NSA targets users' 2485 online anonymity", 2013, 2486 . 2489 [Schroeder] 2490 Schroeder, I. and B. Schmidt, "Introduction - Violent 2491 Imaginaries and Violent Practice", London and New York 2492 Routledge , 2001, . 2496 [UDHR] United Nations General Assembly, "The Universal 2497 Declaration of Human Rights", 1948, 2498 . 2500 [UNGA2013] 2501 United Nations General Assembly, "UN General Assembly 2502 Resolution "The right to privacy in the digital age" 2503 (A/C.3/68/L.45)", 2013, 2504 . 2506 [W3Ci18nDef] 2507 W3C, "Localization vs. Internationalization", 2010, 2508 . 2510 [WP-Debugging] 2511 "Debugging", n.d., . 2514 [WP-Stateless] 2515 "Stateless protocol", n.d., 2516 . 2518 [Walfish] Walfish, M., Stribling, J., Krohn, M., Balakrishnan, H., 2519 Morris, R., and S. Shenker, "Middleboxes No Longer 2520 Considered Harmful", 2004, . 2522 [Zittrain] 2523 Zittrain, J., "The Future of the Internet - And How to 2524 Stop It", Yale University Press , 2008, 2525 . 2528 [Zuckerman] 2529 Zuckerman, E., Roberts, H., McGrady, R., York, J., and J. 2530 Palfrey, "Report on Distributed Denial of Service (DDoS) 2531 Attacks", The Berkman Center for Internet and Society at 2532 Harvard University , 2010, 2533 . 2537 [ars] Anderson, N., "P2P researchers - use a blocklist or you 2538 will be tracked... 100% of the time", 2007, 2539 . 2543 [bbc-wikileaks] 2544 BBC, "Whistle-blower site taken offline", 2008, 2545 . 2547 [bitmessage] 2548 Bitmessage, "Bitmessage Wiki?", 2014, 2549 . 2551 [caida] Dainotti, A., Squarcella, C., Aben, E., Claffy, K., 2552 Chiesa, M., Russo, M., and A. Pescape, "Analysis of 2553 Country-wide Internet Outages Caused by Censorship", 2013, 2554 . 2557 [draft-hall-censorship-tech-01] 2558 Hall, J., Aaron, M., and B. Jones, "A Survey of Worldwide 2559 Censorship Techniques", 2015, 2560 . 2563 [freenet1] 2564 Freenet, "What is Freenet?", n.d., 2565 . 2567 [freenet2] 2568 Ian Clarke, ., "The Philosphy behind Freenet?", n.d., 2569 . 2571 [greatfirewall] 2572 Anonymous, ., "Towards a Comprehensive Picture of the 2573 Great Firewall's DNS Censorship", 2014, 2574 . 2577 [namecoin] 2578 Namecoin, "Namecoin - Decentralized secure names", 2015, 2579 . 2581 [natusage] 2582 Maier, G., Schneider, F., and A. Feldmann, "NAT usage in 2583 Residential Broadband networks", 2011, 2584 . 2587 [pidgin] js, . and Pidgin Developers, "-XMPP- Invisible mode 2588 violating standard", July 2015, 2589 . 2591 [quic] The Chromium Project, "QUIC, a multiplexed stream 2592 transport over UDP", 2014, . 2595 [spdy] The Chromium Project, "SPDY - An experimental protocol for 2596 a faster web", 2009, . 2599 [spiegel] SPIEGEL, "Prying Eyes - Inside the NSA's War on Internet 2600 Security", 2014, 2601 . 2604 [techyum] Violet, ., "Official - vb.ly Link Shortener Seized by 2605 Libyan Government", 2010, . 2609 [torproject] 2610 The Tor Project, ., "Tor Project - Anonymity Online", 2611 2007, . 2613 [torrentfreak1] 2614 Van der Sar, E., "Proposal for research on human rights 2615 protocol considerations", 2015, . 2619 [torrentfreak2] 2620 Andy, ., "LAWYERS SENT 109,000 PIRACY THREATS IN GERMANY 2621 DURING 2013", 2014, . 2625 [turkey] Akguel, M. and M. Kirlidoğ, "Internet censorship in 2626 Turkey", 2015, 2627 . 2630 [ververis] 2631 Vasilis, V., Kargiotakis, G., Filasto, A., Fabian, B., and 2632 A. Alexandros, "Understanding Internet Censorship Policy - 2633 The Case of Greece", 2015, 2634 . 2637 [wikileaks] 2638 Sladek, T. and E. Broese, "Market Survey - Detection & 2639 Filtering Solutions to Identify File Transfer of Copyright 2640 Protected Content for Warner Bros. and movielabs", 2011, 2641 . 2644 [xmppmanifesto] 2645 Saint-Andre, P. and . XMPP Operators, "A Public Statement 2646 Regarding Ubiquitous Encryption on the XMPP Network", 2647 2014, 2648 . 2651 10.3. URIs 2653 [1] mailto:node@domain/home 2655 [2] mailto:node@domain/work 2657 [3] mailto:hrpc@ietf.org 2659 Authors' Addresses 2661 Niels ten Oever 2662 Article19 2664 EMail: niels@article19.org 2666 Corinne Cath 2667 Oxford Internet Institute 2669 EMail: corinne.cath@oii.ox.ac.uk