idnits 2.17.1 draft-ietf-httpbis-client-hints-09.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The abstract seems to contain references ([2], [3], [1]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 18, 2020) is 1529 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 469 -- Looks like a reference, but probably isn't: '2' on line 471 -- Looks like a reference, but probably isn't: '3' on line 473 == Unused Reference: 'RFC6265' is defined on line 458, but no explicit reference was found in the text == Outdated reference: A later version (-19) exists of draft-ietf-httpbis-header-structure-15 ** Obsolete normative reference: RFC 7230 (Obsoleted by RFC 9110, RFC 9112) ** Obsolete normative reference: RFC 7231 (Obsoleted by RFC 9110) ** Obsolete normative reference: RFC 7234 (Obsoleted by RFC 9111) Summary: 4 errors (**), 0 flaws (~~), 3 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 HTTP Working Group I. Grigorik 3 Internet-Draft Y. Weiss 4 Intended status: Experimental Google 5 Expires: August 21, 2020 February 18, 2020 7 HTTP Client Hints 8 draft-ietf-httpbis-client-hints-09 10 Abstract 12 HTTP defines proactive content negotiation to allow servers to select 13 the appropriate response for a given request, based upon the user 14 agent's characteristics, as expressed in request headers. In 15 practice, clients are often unwilling to send those request headers, 16 because it is not clear whether they will be used, and sending them 17 impacts both performance and privacy. 19 This document defines an Accept-CH response header that servers can 20 use to advertise their use of request headers for proactive content 21 negotiation, along with a set of guidelines for the creation of such 22 headers, colloquially known as "Client Hints." 24 Note to Readers 26 Discussion of this draft takes place on the HTTP working group 27 mailing list (ietf-http-wg@w3.org), which is archived at 28 https://lists.w3.org/Archives/Public/ietf-http-wg/ [1]. 30 Working Group information can be found at http://httpwg.github.io/ 31 [2]; source code and issues list for this draft can be found at 32 https://github.com/httpwg/http-extensions/labels/client-hints [3]. 34 Status of This Memo 36 This Internet-Draft is submitted in full conformance with the 37 provisions of BCP 78 and BCP 79. 39 Internet-Drafts are working documents of the Internet Engineering 40 Task Force (IETF). Note that other groups may also distribute 41 working documents as Internet-Drafts. The list of current Internet- 42 Drafts is at https://datatracker.ietf.org/drafts/current/. 44 Internet-Drafts are draft documents valid for a maximum of six months 45 and may be updated, replaced, or obsoleted by other documents at any 46 time. It is inappropriate to use Internet-Drafts as reference 47 material or to cite them other than as "work in progress." 48 This Internet-Draft will expire on August 21, 2020. 50 Copyright Notice 52 Copyright (c) 2020 IETF Trust and the persons identified as the 53 document authors. All rights reserved. 55 This document is subject to BCP 78 and the IETF Trust's Legal 56 Provisions Relating to IETF Documents 57 (https://trustee.ietf.org/license-info) in effect on the date of 58 publication of this document. Please review these documents 59 carefully, as they describe your rights and restrictions with respect 60 to this document. Code Components extracted from this document must 61 include Simplified BSD License text as described in Section 4.e of 62 the Trust Legal Provisions and are provided without warranty as 63 described in the Simplified BSD License. 65 Table of Contents 67 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 68 1.1. Notational Conventions . . . . . . . . . . . . . . . . . 4 69 2. Client Hint Request Header Fields . . . . . . . . . . . . . . 4 70 2.1. Sending Client Hints . . . . . . . . . . . . . . . . . . 4 71 2.2. Server Processing of Client Hints . . . . . . . . . . . . 5 72 3. Advertising Server Support . . . . . . . . . . . . . . . . . 5 73 3.1. The Accept-CH Response Header Field . . . . . . . . . . . 5 74 3.1.1. Interaction with Caches . . . . . . . . . . . . . . . 6 75 4. Security Considerations . . . . . . . . . . . . . . . . . . . 6 76 4.1. Information Exposure . . . . . . . . . . . . . . . . . . 6 77 5. Cost of Sending Hints . . . . . . . . . . . . . . . . . . . . 8 78 5.1. Deployment and Security Risks . . . . . . . . . . . . . . 8 79 5.2. Abuse Detection . . . . . . . . . . . . . . . . . . . . . 9 80 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 81 6.1. Accept-CH . . . . . . . . . . . . . . . . . . . . . . . . 9 82 7. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 83 7.1. Normative References . . . . . . . . . . . . . . . . . . 9 84 7.2. Informative References . . . . . . . . . . . . . . . . . 10 85 7.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 10 86 Appendix A. Interaction with Variants Response Header Field . . 11 87 Appendix B. Changes . . . . . . . . . . . . . . . . . . . . . . 11 88 B.1. Since -00 . . . . . . . . . . . . . . . . . . . . . . . . 11 89 B.2. Since -01 . . . . . . . . . . . . . . . . . . . . . . . . 11 90 B.3. Since -02 . . . . . . . . . . . . . . . . . . . . . . . . 11 91 B.4. Since -03 . . . . . . . . . . . . . . . . . . . . . . . . 11 92 B.5. Since -04 . . . . . . . . . . . . . . . . . . . . . . . . 11 93 B.6. Since -05 . . . . . . . . . . . . . . . . . . . . . . . . 11 94 B.7. Since -06 . . . . . . . . . . . . . . . . . . . . . . . . 12 95 B.8. Since -07 . . . . . . . . . . . . . . . . . . . . . . . . 12 96 B.9. Since -08 . . . . . . . . . . . . . . . . . . . . . . . . 12 97 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . 12 98 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 12 100 1. Introduction 102 There are thousands of different devices accessing the web, each with 103 different device capabilities and preference information. These 104 device capabilities include hardware and software characteristics, as 105 well as dynamic user and client preferences. Applications that want 106 to allow the server to optimize content delivery and user experience 107 based on such capabilities have, historically, had to rely on passive 108 identification (e.g., by matching User-Agent (Section 5.5.3 of 109 [RFC7231]) header field against an established database of client 110 signatures), used HTTP cookies and URL parameters, or use some 111 combination of these and similar mechanisms to enable ad hoc content 112 negotiation. 114 Such techniques are expensive to setup and maintain, are not portable 115 across both applications and servers, and make it hard to reason for 116 both client and server about which data is required and is in use 117 during the negotiation: 119 o User agent detection cannot reliably identify all static 120 variables, cannot infer dynamic client preferences, requires 121 external device database, is not cache friendly, and is reliant on 122 a passive fingerprinting surface. 123 o Cookie based approaches are not portable across applications and 124 servers, impose additional client-side latency by requiring 125 JavaScript execution, and are not cache friendly. 126 o URL parameters, similar to cookie based approaches, suffer from 127 lack of portability, and are hard to deploy due to a requirement 128 to encode content negotiation data inside of the URL of each 129 resource. 131 Proactive content negotiation (Section 3.4.1 of [RFC7231]) offers an 132 alternative approach; user agents use specified, well-defined request 133 headers to advertise their capabilities and characteristics, so that 134 servers can select (or formulate) an appropriate response. 136 However, proactive content negotiation requires clients to send these 137 request headers prolifically. This causes performance concerns 138 (because it creates "bloat" in requests), as well as privacy issues; 139 passively providing such information allows servers to silently 140 fingerprint the user agent. 142 This document defines a new response header, Accept-CH, that allows 143 an origin server to explicitly ask that clients send these headers in 144 requests. It also defines guidelines for content negotiation 145 mechanisms that use it, colloquially referred to as Client Hints. 147 Client Hints mitigate the performance concerns by assuring that 148 clients will only send the request headers when they're actually 149 going to be used, and the privacy concerns of passive fingerprinting 150 by requiring explicit opt-in and disclosure of required headers by 151 the server through the use of the Accept-CH response header. 153 This document defines the Client Hints infrastructure, a framework 154 that enables servers to opt-in to specific proactive content 155 negotiation features, which will enable them to adapt their content 156 accordingly. However, it does not define any specific features that 157 will use that infrastructure. Those features will be defined in 158 their respective specifications. 160 1.1. Notational Conventions 162 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 163 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 164 "OPTIONAL" in this document are to be interpreted as described in BCP 165 14 [RFC2119] [RFC8174] when, and only when, they appear in all 166 capitals, as shown here. 168 This document uses the Augmented Backus-Naur Form (ABNF) notation of 169 [RFC5234] with the list rule extension defined in [RFC7230], 170 Appendix B. It includes by reference the DIGIT rule from [RFC5234] 171 and the OWS and field-name rules from [RFC7230]. 173 2. Client Hint Request Header Fields 175 A Client Hint request header field is a HTTP header field that is 176 used by HTTP clients to indicate configuration data that can be used 177 by the server to select an appropriate response. Each one conveys 178 client preferences that the server can use to adapt and optimize the 179 response. 181 2.1. Sending Client Hints 183 Clients control which Client Hints are sent in requests, based on 184 their default settings, user configuration, and server preferences. 185 The client and server can use an opt-in mechanism outlined below to 186 negotiate which fields should be sent to allow for efficient content 187 adaption, and optionally use additional mechanisms to negotiate 188 delegation policies that control access of third parties to same 189 fields. 191 Implementers should be aware of the passive fingerprinting 192 implications when implementing support for Client Hints, and follow 193 the considerations outlined in "Security Considerations" section of 194 this document. 196 2.2. Server Processing of Client Hints 198 When presented with a request that contains one or more client hint 199 header fields, servers can optimize the response based upon the 200 information in them. When doing so, and if the resource is 201 cacheable, the server MUST also generate a Vary response header field 202 (Section 7.1.4 of [RFC7231]) to indicate which hints can affect the 203 selected response and whether the selected response is appropriate 204 for a later request. 206 Further, depending on the hint used, the server can generate 207 additional response header fields to convey related values to aid 208 client processing. 210 3. Advertising Server Support 212 Servers can advertise support for Client Hints using the mechnisms 213 described below. 215 3.1. The Accept-CH Response Header Field 217 The Accept-CH response header field or the equivalent HTML meta 218 element with http-equiv attribute ([HTML]) indicate server support 219 for particular hints indicated in its value. 221 Accept-CH is a Structured Header [I-D.ietf-httpbis-header-structure]. 222 Its value MUST be an sh-list (Section 3.1 of 223 [I-D.ietf-httpbis-header-structure]) whose members are tokens 224 (Section 3.7 of [I-D.ietf-httpbis-header-structure]). Its ABNF is: 226 Accept-CH = sh-list 228 For example: 230 Accept-CH: Sec-CH-Example, Sec-CH-Example-2 232 When a client receives an HTTP response advertising support for 233 provided list of Clients Hints, it SHOULD process it as origin 234 ([RFC6454]) opt-in to receive Client Hint header fields advertised in 235 the field-value, for subsequent same-origin requests. 237 o The opt-in MUST be delivered over a secure transport. 239 o The opt-in SHOULD be persisted and bound to the origin to enable 240 delivery of Client Hints on subsequent requests to the server's 241 origin, and MUST NOT be persisted for an origin that isn't HTTPS. 243 Accept-CH: Sec-CH-Example, Sec-CH-Example-2 244 Accept-CH: Sec-CH-Example-3 246 For example, based on the Accept-CH example above, which is received 247 in response to a user agent navigating to "https://example.com", and 248 delivered over a secure transport: a user agent SHOULD persist an 249 Accept-CH preference bound to "https://example.com" and use it for 250 user agent navigations to "https://example.com" and any same-origin 251 resource requests initiated by the page constructed from the 252 navigation's response. This preference SHOULD NOT extend to resource 253 requests initiated to "https://example.com" from other origins. 255 3.1.1. Interaction with Caches 257 When selecting an optimized response based on one or more Client 258 Hints, and if the resource is cacheable, the server needs to generate 259 a Vary response header field ([RFC7234]) to indicate which hints can 260 affect the selected response and whether the selected response is 261 appropriate for a later request. 263 Vary: Sec-CH-Example 265 Above example indicates that the cache key needs to include the Sec- 266 CH-Example header field. 268 Vary: Sec-CH-Example, Sec-CH-Example-2 270 Above example indicates that the cache key needs to include the Sec- 271 CH-Example and Sec-CH-Example-2 header fields. 273 4. Security Considerations 275 4.1. Information Exposure 277 Request header fields used in features relying on this document 278 expose information about the user's environment to enable proactive 279 content negotiation. Such information may reveal new information 280 about the user and implementers ought to consider the following 281 considerations, recommendations, and best practices. 283 The underlying assumption is that exposing information about the user 284 as a request header is equivalent to the capability of that request's 285 origin to access that information by other means and transmit it to 286 itself. 288 Therefore, features relying on this document to define Client Hint 289 headers MUST NOT provide new information that is otherwise not 290 available to the application via other means, such as existing 291 request headers, HTML, CSS, or JavaScript. 293 Such features SHOULD take into account the following aspects of the 294 information exposed: 296 o Entropy 298 * Exposing highly granular data may help identify users across 299 multiple requests to different origins. Reducing the set of 300 field values that can be expressed, or restricting them to an 301 enumerated range where the advertised value is close but is not 302 an exact representation of the current value, can improve 303 privacy and reduce risk of linkability by ensuring that the 304 same value is sent by multiple users. 305 o Sensitivity 307 * The feature SHOULD NOT expose user sensitive information. To 308 that end, information available to the application, but gated 309 behind specific user actions (e.g. a permission prompt or user 310 activation) SHOULD NOT be exposed as a Client Hint. 311 o Change over time 313 * The feature SHOULD NOT expose user information that changes 314 over time, unless the state change itself is also exposed (e.g. 315 through JavaScript callbacks). 317 Different features will be positioned in different points in the 318 space between low-entropy, non-sensitive and static information (e.g. 319 user agent information), and high-entropy, sensitive and dynamic 320 information (e.g. geolocation). User agents SHOULD consider the 321 value provided by a particular feature vs these considerations, and 322 MAY have different policies regarding that tradeoff on a per-feature 323 basis. 325 Implementers ought to consider both user and server controlled 326 mechanisms and policies to control which Client Hints header fields 327 are advertised: 329 o Implementers SHOULD restrict delivery of some or all Client Hints 330 header fields to the opt-in origin only, unless the opt-in origin 331 has explicitly delegated permission to another origin to request 332 Client Hints header fields. 333 o Implementers MAY provide user choice mechanisms so that users may 334 balance privacy concerns with bandwidth limitations. However, 335 implementers should also be aware that explaining the privacy 336 implications of passive fingerprinting to users may be 337 challenging. 338 o Implementations specific to certain use cases or threat models MAY 339 avoid transmitting some or all of Client Hints header fields. For 340 example, avoid transmission of header fields that can carry higher 341 risks of linkability. 343 Implementers SHOULD support Client Hints opt-in mechanisms and MUST 344 clear persisted opt-in preferences when any one of site data, 345 browsing history, browsing cache, cookies, or similar, are cleared. 347 5. Cost of Sending Hints 349 While HTTP header compression schemes reduce the cost of adding HTTP 350 header fields, sending Client Hints to the server incurs an increase 351 in request byte size. Servers SHOULD take that into account when 352 opting in to receive Client Hints, and SHOULD NOT opt-in to receive 353 hints unless they are to be used for content adaptation purposes. 355 Due to request byte size increase, features relying on this document 356 to define Client Hints MAY consider restricting sending those hints 357 to certain request destinations [FETCH], where they are more likely 358 to be useful. 360 5.1. Deployment and Security Risks 362 Deployment of new request headers requires several considerations: 364 o Potential conflicts due to existing use of field name 365 o Properties of the data communicated in field value 367 Authors of new Client Hints are advised to carefully consider whether 368 they should be able to be added by client-side content (e.g., 369 scripts), or whether they should be exclusively set by the user 370 agent. In the latter case, the Sec- prefix on the header field name 371 has the effect of preventing scripts and other application content 372 from setting them in user agents. Using the "Sec-" prefix signals to 373 servers that the user agent - and not application content - generated 374 the values. See [FETCH] for more information. 376 By convention, request headers that are client hints are encouraged 377 to use a CH- prefix, to make them easier to identify as using this 378 framework; for example, CH-Foo or, with a "Sec-" prefix, Sec-CH-Foo. 379 Doing so makes them easier to identify programmatically (e.g., for 380 stripping unrecognised hints from requests by privacy filters). 382 5.2. Abuse Detection 384 A user agent that tracks access to active fingerprinting information 385 SHOULD consider emission of Client Hints headers similarly to the way 386 it would consider access to the equivalent API. 388 Research into abuse of Client Hints might look at how HTTP responses 389 that contain Client Hints differ from those with different values, 390 and from those without. This might be used to reveal which Client 391 Hints are in use, allowing researchers to further analyze that use. 393 6. IANA Considerations 395 This document defines the "Accept-CH" HTTP response field, and 396 registers it in the Permanent Message Header Fields registry. 398 6.1. Accept-CH 400 o Header field name: Accept-CH 401 o Applicable protocol: HTTP 402 o Status: standard 403 o Author/Change controller: IETF 404 o Specification document(s): Section 3.1 of this document 405 o Related information: for Client Hints 407 7. References 409 7.1. Normative References 411 [FETCH] van Kesteren, A., "Fetch", n.d., 412 . 414 [HTML] Hickson, I., Pieters, S., van Kesteren, A., Jaegenstedt, 415 P., and D. Denicola, "HTML", n.d., 416 . 418 [I-D.ietf-httpbis-header-structure] 419 Nottingham, M. and P. Kamp, "Structured Headers for HTTP", 420 draft-ietf-httpbis-header-structure-15 (work in progress), 421 January 2020. 423 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 424 Requirement Levels", BCP 14, RFC 2119, 425 DOI 10.17487/RFC2119, March 1997, 426 . 428 [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax 429 Specifications: ABNF", STD 68, RFC 5234, 430 DOI 10.17487/RFC5234, January 2008, 431 . 433 [RFC6454] Barth, A., "The Web Origin Concept", RFC 6454, 434 DOI 10.17487/RFC6454, December 2011, 435 . 437 [RFC7230] Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer 438 Protocol (HTTP/1.1): Message Syntax and Routing", 439 RFC 7230, DOI 10.17487/RFC7230, June 2014, 440 . 442 [RFC7231] Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer 443 Protocol (HTTP/1.1): Semantics and Content", RFC 7231, 444 DOI 10.17487/RFC7231, June 2014, 445 . 447 [RFC7234] Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, 448 Ed., "Hypertext Transfer Protocol (HTTP/1.1): Caching", 449 RFC 7234, DOI 10.17487/RFC7234, June 2014, 450 . 452 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 453 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 454 May 2017, . 456 7.2. Informative References 458 [RFC6265] Barth, A., "HTTP State Management Mechanism", RFC 6265, 459 DOI 10.17487/RFC6265, April 2011, 460 . 462 [VARIANTS] 463 Nottingham, M., "HTTP Representation Variants", draft- 464 ietf-httpbis-variants-06 (work in progress), November 465 2019. 467 7.3. URIs 469 [1] https://lists.w3.org/Archives/Public/ietf-http-wg/ 471 [2] http://httpwg.github.io/ 473 [3] https://github.com/httpwg/http-extensions/labels/client-hints 475 Appendix A. Interaction with Variants Response Header Field 477 Client Hints may be combined with Variants response header field 478 [VARIANTS] to enable fine-grained control of the cache key for 479 improved cache efficiency. Features that define Client Hints will 480 need to specify the related variants algorithms as described in 481 Section 6 of [VARIANTS]. 483 Appendix B. Changes 485 B.1. Since -00 487 o Issue 168 (make Save-Data extensible) updated ABNF. 488 o Issue 163 (CH review feedback) editorial feedback from httpwg 489 list. 490 o Issue 153 (NetInfo API citation) added normative reference. 492 B.2. Since -01 494 o Issue 200: Moved Key reference to informative. 495 o Issue 215: Extended passive fingerprinting and mitigation 496 considerations. 497 o Changed document status to experimental. 499 B.3. Since -02 501 o Issue 239: Updated reference to CR-css-values-3 502 o Issue 240: Updated reference for Network Information API 503 o Issue 241: Consistency in IANA considerations 504 o Issue 250: Clarified Accept-CH 506 B.4. Since -03 508 o Issue 284: Extended guidance for Accept-CH 509 o Issue 308: Editorial cleanup 510 o Issue 306: Define Accept-CH-Lifetime 512 B.5. Since -04 514 o Issue 361: Removed Downlink 515 o Issue 361: Moved Key to appendix, plus other editorial feedback 517 B.6. Since -05 519 o Issue 372: Scoped CH opt-in and delivery to secure transports 520 o Issue 373: Bind CH opt-in to origin 522 B.7. Since -06 524 o Issue 524: Save-Data is now defined by NetInfo spec, dropping 525 o PR 775: Removed specific features to be defined in other 526 specifications 528 B.8. Since -07 530 o Issue 761: Clarified that the defined headers are response 531 headers. 532 o Issue 730: Replaced Key reference with Variants. 533 o Issue 700: Replaced ABNF with structured headers. 534 o PR 878: Removed Accept-CH-Lifetime based on feedback at IETF 105 536 B.9. Since -08 538 o PR 985: Describe the bytesize cost of hints. 539 o PR 776: Add Sec- and CH- prefix considerations. 540 o PR 1001: Clear CH persistence when cookies are cleared. 542 Acknowledgements 544 Thanks to Mark Nottingham, Julian Reschke, Chris Bentzel, Ben 545 Greenstein, Tarun Bansal, Roy Fielding, Vasiliy Faronov, Ted Hardie, 546 Jonas Sicking, Martin Thomson, and numerous other members of the IETF 547 HTTP Working Group for invaluable help and feedback. 549 Authors' Addresses 551 Ilya Grigorik 552 Google 554 Email: ilya@igvita.com 555 URI: https://www.igvita.com/ 557 Yoav Weiss 558 Google 560 Email: yoav@yoav.ws 561 URI: https://blog.yoav.ws/