idnits 2.17.1 draft-kunze-ark-13.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 16. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 2363. ** The document seems to lack an RFC 3979 Section 5, para. 1 IPR Disclosure Acknowledgement. ** The document seems to lack an RFC 3979 Section 5, para. 2 IPR Disclosure Acknowledgement. ** The document seems to lack an RFC 3979 Section 5, para. 3 IPR Disclosure Invitation. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? ** The document is more than 15 pages and seems to lack a Table of Contents. == No 'Intended status' indicated for this document; assuming Proposed Standard == The page length should not exceed 58 lines per page, but there was 50 longer pages, the longest (page 2) being 63 lines == It seems as if not all pages are separated by form feeds - found 0 form feeds but 51 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. ** There are 8 instances of too long lines in the document, the longest one being 21 characters in excess of 72. ** The abstract seems to contain references ([Qualifier]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. == There are 8 instances of lines with non-RFC2606-compliant FQDNs in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year == Line 1136 has weird spacing: '... regexp repla...' -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (23 February 2007) is 6272 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'Qualifier' is mentioned on line 436, but not defined == Unused Reference: 'MD5' is defined on line 2140, but no explicit reference was found in the text -- Possible downref: Non-RFC (?) normative reference: ref. 'ANVL' -- Possible downref: Non-RFC (?) normative reference: ref. 'ARK' -- Possible downref: Non-RFC (?) normative reference: ref. 'DCORE' -- Possible downref: Non-RFC (?) normative reference: ref. 'DERC' -- Possible downref: Non-RFC (?) normative reference: ref. 'DOI' -- Possible downref: Non-RFC (?) normative reference: ref. 'ERC' -- Possible downref: Non-RFC (?) normative reference: ref. 'Handle' ** Obsolete normative reference: RFC 2616 (ref. 'HTTP') (Obsoleted by RFC 7230, RFC 7231, RFC 7232, RFC 7233, RFC 7234, RFC 7235) ** Downref: Normative reference to an Informational RFC: RFC 1321 (ref. 'MD5') -- Possible downref: Non-RFC (?) normative reference: ref. 'N2T' ** Obsolete normative reference: RFC 2915 (ref. 'NAPTR') (Obsoleted by RFC 3401, RFC 3402, RFC 3403, RFC 3404) -- Possible downref: Non-RFC (?) normative reference: ref. 'NLMPerm' -- Possible downref: Non-RFC (?) normative reference: ref. 'NOID' -- Possible downref: Non-RFC (?) normative reference: ref. 'PURL' ** Obsolete normative reference: RFC 822 (Obsoleted by RFC 2822) -- Possible downref: Non-RFC (?) normative reference: ref. 'TEMPER' -- Possible downref: Non-RFC (?) normative reference: ref. 'THUMP' ** Obsolete normative reference: RFC 2396 (ref. 'URI') (Obsoleted by RFC 3986) ** Downref: Normative reference to an Informational RFC: RFC 2288 (ref. 'URNBIB') ** Obsolete normative reference: RFC 2141 (ref. 'URNSYN') (Obsoleted by RFC 8141) ** Obsolete normative reference: RFC 2611 (ref. 'URNNID') (Obsoleted by RFC 3406) Summary: 18 errors (**), 0 flaws (~~), 8 warnings (==), 18 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Internet-Draft: draft-kunze-ark-13.txt J. Kunze 2 ARK Identifier Scheme University of California (UCOP) 3 Expires 23 August 2007 R. P. C. Rodgers 4 US National Library of Medicine 5 23 February 2007 7 The ARK Persistent Identifier Scheme 9 (http://www.ietf.org/internet-drafts/draft-kunze-ark-13.txt) 11 Status of this Document 13 By submitting this Internet-Draft, each author represents that any 14 applicable patent or other IPR claims of which he or she is aware 15 have been or will be disclosed, and any of which he or she becomes 16 aware will be disclosed, in accordance with Section 6 of BCP 79. 18 Internet-Drafts are working documents of the Internet Engineering 19 Task Force (IETF), its areas, and its working groups. Note that 20 other groups may also distribute working documents as Internet- 21 Drafts. 23 Internet-Drafts are draft documents valid for a maximum of six months 24 and may be updated, replaced, or obsoleted by other documents at any 25 time. It is inappropriate to use Internet-Drafts as reference 26 material or to cite them other than as ``work in progress.'' 28 The list of current Internet-Drafts can be accessed at 29 http://www.ietf.org/ietf/1id-abstracts.txt 31 The list of Internet-Draft Shadow Directories can be accessed at 32 http://www.ietf.org/shadow.html. 34 Distribution of this document is unlimited. Please send comments to 35 jak@ucop.edu. 37 Copyright (C) The IETF Trust (2007). All Rights Reserved. 39 Abstract 41 The ARK (Archival Resource Key) naming scheme is designed to 42 facilitate the high-quality and persistent identification of 43 information objects. A founding principle of the ARK is that 44 persistence is purely a matter of service and is neither inherent in 45 an object nor conferred on it by a particular naming syntax. The best 46 that an identifier can do is to lead users to the services that 47 support persistence. The term ARK itself refers both to the scheme 48 and to any single identifier that conforms to it. An ARK has five 49 components: 51 [http://NMAH/]ark:/NAAN/Name[Qualifier] 53 an optional and mutable Name Mapping Authority Hostport, the "ark:" 54 label, the Name Assigning Authority Number (NAAN), the assigned Name, 55 and an optional and possibly mutable Qualifier supported by the NMA. 56 The NAAN and Name together form the immutable persistent identifier 57 for the object. An ARK is a special kind of URL that connects users 58 to three things: the named object, its metadata, and the provider's 59 promise about its persistence. When entered into the location field 60 of a Web browser, the ARK leads the user to the named object. That 61 same ARK, followed by a single question mark ('?'), returns a brief 62 metadata record that is both human- and machine-readable. When the 63 ARK is followed by dual question marks ('??'), the returned metadata 64 contains a commitment statement from the current provider. Tools 65 exist for minting, binding, and resolving ARKs. 67 1. Introduction 69 This document describes a scheme for the high-quality naming of 70 information resources. The scheme, called the Archival Resource Key 71 (ARK), is well suited to long-term access and identification of any 72 information resources that accommodate reasonably regular electronic 73 description. This includes digital documents, databases, software, 74 and websites, as well as physical objects (books, bones, statues, 75 etc.) and intangible objects (chemicals, diseases, vocabulary terms, 76 performances). Hereafter the term "object" refers to an information 77 resource. The term ARK itself refers both to the scheme and to any 78 single identifier that conforms to it. A reasonably concise and 79 accessible overview and rationale for the scheme is available at 80 [ARK]. 82 Schemes for persistent identification of network-accessible objects 83 are not new. In the early 1990's, the design of the Uniform Resource 84 Name [URNSYN] responded to the observed failure rate of URLs by 85 articulating an indirect, non-hostname-based naming scheme and the 86 need for responsible name management. Meanwhile, promoters of the 87 Digital Object Identifier [DOI] succeeded in building a community of 88 providers around a mature software system [Handle] that supports name 89 management. The Persistent Uniform Resource Locator [PURL] was 90 another scheme that has the unique advantage of working with 91 unmodified web browsers. ARKs represent an approach that attempts to 92 build on the strengths and to avoid the weaknesses of the other 93 schemes. 95 A founding principle of the ARK is that persistence is purely a 96 matter of service. Persistence is neither inherent in an object nor 97 conferred on it by a particular naming syntax. Nor is the technique 98 of name indirection - upon which URNs, Handles, DOIs, and PURLs are 99 founded - of central importance. Name indirection is an ancient and 100 well-understood practice; new mechanisms for it keep appearing and 101 distracting practitioner attention, with the Domain Name System [DNS] 102 being a particularly dazzling and elegant example. What is often 103 forgotten is that maintenance of an indirection table is the 104 overwhelming and unavoidable cost to the organization providing 105 persistence, and the cost is equivalent across naming schemes. That 106 indirection has always been a native part of the web while being so 107 lightly utilized for the persistence of web-based objects is an 108 indication of how unsuited most organizations are to the task of 109 table maintenance and to the overall challenge of digital permanence. 111 Persistence is achieved through a provider's successful stewardship 112 of objects and their identifiers. The highest level of persistence 113 will be reinforced by a provider's robust contingency, redundancy, 114 and succession strategies. It is further safeguarded to the extent 115 that a provider's mission is shielded from marketplace and political 116 instabilities. These are by far the major challenges confronting 117 persistence providers, and no identifier scheme has any direct impact 118 on them. In fact, some schemes may be actual liabilities for 119 persistence because they create short- and long-term dependencies for 120 every object access on complex, special-purpose local and global 121 infrastructures, parts of which are proprietary and all of which 122 increase the carry-forward burden for the preservation community. It 123 is for this reason that the ARK scheme relies only on educated name 124 assignment and light use of general-purpose infrastructures that the 125 entire internet community needs (the DNS, web servers, and web 126 browsers) and that one can reasonably expect many others to help 127 carry forward into the technologically evolving future. 129 1.1. Reasons to Use ARKs 131 If no persistent identifier scheme contributes directly to 132 persistence, why not just use URLs? A particular URL may be as 133 durable an identifier as it is possible to have, but nothing 134 distinguishes it from an ordinary URL to the recipient who is 135 wondering if it is suitable for long-term reference. An ARK is just 136 a URL, distinguished by its form, that provides some of the necessary 137 conditions for credible persistence. An ARK invites access to not 138 one, but to three things: to the object, to its metadata, and to a 139 nuanced statement of commitment from the provider regarding the 140 object. Existence of the two extra services can be probed 141 automatically by appending either `?' or `??' to the ARK. 143 The form of the ARK also supports the natural separation of naming 144 authorities into the original name assigning authority and the 145 diverse multiple name mapping (or servicing) authorities that in 146 succession and in parallel will take over custodial responsibilities 147 from the original assigner for the large majority of a long-term 148 object's archival lifetime. The mapping authority, indicated by the 149 hostname part of the URL that contains the ARK, serves to launch the 150 ARK into cyberspace. Should it ever fail (and there is no reason why 151 a well-chosen hostname of a 100-year-old cultural memory institution 152 shouldn't last as long as the DNS), that host name is considered 153 disposeable and replaceable. Again, the form of the ARK helps 154 because it defines exactly how to recover the core immutable object 155 identity, and several simple algorithms (based on the URN model) are 156 defined for locating another mapping authority. 158 There are tools to assist in generating ARKs and other identifiers, 159 such as [NOID] and "uuidgen", both of which rely for uniqueness on 160 human-maintained registries. This document also contains some 161 guidelines and considerations for managing namespaces and choosing 162 hostnames wisely. 164 1.2. Three Requirements of ARKs 166 The first requirement of an ARK is to give users a link from an 167 object to a promise of stewardship for it. That promise is a multi- 168 faceted covenant that binds the word of an identified service 169 provider to a specific set of responsibilities. No one can tell if 170 successful stewardship will take place because no one can predict the 171 future. Reasonable conjecture, however, may be based on past 172 performance. There must be a way to tie a promise of persistence to 173 a provider's demonstrated or perceived ability - its reputation - in 174 that arena. Provider reputations would then rise and fall as 175 promises are observed variously to be kept and broken. This is 176 perhaps the best way we have for gauging the strength of any 177 persistence promise. Note that over time, current providers have 178 nothing to do with the intentions of the original assigners of names. 180 The second requirement of an ARK is to give users a link from an 181 object to a description of it. The problem with a naked identifier 182 is that without a description real identification is incomplete. 183 Identifiers common today are relatively opaque, though some contain 184 ad hoc clues that reflect brief life cycle periods such as the 185 address of a short stay in a filesystem hierarchy. Possession of 186 both an identifier and an object is some improvement, but positive 187 identification may still be uncertain since the object itself might 188 not include a matching identifier or might not carry evidence obvious 189 enough to reveal its identity without significant research. In 190 either case, what is called for is a record bearing witness to the 191 identifier's association with the object, as supported by a recorded 192 set of object characteristics. This descriptive record is partly an 193 identification "receipt" with which users and archivists can verify 194 an object's identity after brief inspection and a plausible match 195 with recorded characteristics such as title and size. 197 The final requirement of an ARK is to give users a link to the object 198 itself (or to a copy) if at all possible. Persistent access is the 199 central duty of an ARK. Persistent identification plays a vital 200 supporting role but, strictly speaking, it can be construed as no 201 more than a record attesting to the original assignment of a never- 202 reassigned identifier. Object access may not be feasible for various 203 reasons, such as catastrophic loss of the object, a licensing 204 agreement that keeps an archive "dark" for a period of years, or when 205 an object's own lack of tangible existence confuses normal concepts 206 of access (e.g., a vocabulary term might be accessed through its 207 definition). In such cases the ARK's identification role assumes a 208 much higher profile. But attempts to simplify the persistence 209 problem by decoupling access from identification and concentrating 210 exclusively on the latter are of questionable utility. A perfect 211 system for assigning forever unique identifiers might be created, but 212 if it did so without reducing access failure rates, no one would be 213 interested. The central issue - which may be summed up as the "HTTP 214 404 Not Found" problem - would not have been addressed. 216 1.3. Organizing Support for ARKs: Our Stuff vs. Their Stuff 218 An organization and the user community it serves can often be seen to 219 struggle with two different areas of persistent identification: the 220 Our Stuff problem and the Their Stuff problem. In the Our Stuff 221 problem, we in the organization want our own objects to acquire 222 persistent names. Since we possess or control these objects, our 223 organization tackles the Our Stuff problem directly. Whether or not 224 the objects are named by ARKs, our organization is the responsible 225 party, so it can plan for, maintain, and make commitments about the 226 objects. 228 In the Their Stuff problem, we in the organization want others' 229 objects to acquire persistent names. These are objects that we do 230 not own or control, but some of which are critically important to us. 231 But because they are beyond our influence as far as support is 232 concerned, creating and maintaining persistent identifiers for Their 233 Stuff is not especially purposeful or feasible for us to do. There 234 is little that we can do about someone else's stuff except encourage 235 them to find or become providers of persistence services. 237 Co-location of persistent access and identification services is 238 natural. Any organization that undertakes ongoing support of true 239 persistent identification (which includes description) is well-served 240 if it controls, owns, or otherwise has clear internal access to the 241 identified objects, and this gives it an advantage if it wishes also 242 to support persistent access to outsiders. Conversely, persistent 243 access to outsiders requires orderly internal collection management 244 procedures that include monitoring, acquisition, verification, and 245 change control over objects, which in turn requires object 246 identifiers persistent enough to support auditable record keeping 247 practices. 249 Although, organizing ARK services under one roof thus tends to make 250 sense, object hosting can successfully be separated from name 251 mapping. An example is when a name mapping authority centrally 252 provides uniform resolution services via a protocol gateway on behalf 253 of organizations that host objects behind a variety of access 254 protocols. It is also reasonable to build value-added description 255 services that rely on the underlying services of a set of mapping 256 authorities. 258 Supporting ARKs is not for every organization. By requiring 259 specific, revealed commitments to preservation, to object access, and 260 to description, the bar for providing ARK services is higher than for 261 some other identifier schemes. On the other hand, it would be hard 262 to grant credence to a persistence promise from an organization that 263 could not muster the minimum ARK services. Not that there isn't a 264 business model for an ARK-like, description-only service built on top 265 of another organization's full complement of ARK services. For 266 example, there might be competition at the description level for 267 abstracting and indexing a body of scientific literature archived in 268 a combination of open and fee-based repositories. The description- 269 only service would have no direct commitment to the objects, but 270 would act as an intermediary, forwarding commitment statements from 271 object hosting services to requestors. 273 1.4. Definition of Identifier 275 An identifier is not a string of character data - an identifier is an 276 association between a string of data and an object. This abstraction 277 is necessary because without it a string is just data. It's nonsense 278 to talk about a string's breaking, or about its being strong, 279 maintained, and authentic. But as a representative of an 280 association, a string can do, metaphorically, the things that we 281 expect of it. 283 Without regard to whether an object is physical, digital, or 284 conceptual, to identify it is to claim an association between it and 285 a representative string, such as "Jane" or "ISBN 0596000278". What 286 gives a claim credibility is a set of verifiable assertions, or 287 metadata, about the object, such as age, height, title, or number of 288 pages. In other words, the association is made manifest by a record 289 (e.g., a cataloging or other metadata record) that vouches for it. 291 In the complete absence of any testimony (metadata) regarding an 292 association, a would-be identifier string is a meaningless sequence 293 of characters. To keep an externally visible but otherwise internal 294 string from being perceived as an identifier by outsiders, for 295 example, it suffices for an organization not to disclose the nature 296 of its association. For our immediate purpose, actual existence of 297 an association record is more important than its authenticity or 298 verifiability, which are outside the scope of this specification. 300 It is a gift to the identification process if an object carries its 301 own name as an inseparable part of itself, such as an identifier 302 imprinted on the first page of a document or embedded in a data 303 structure element of a digital document header. In cases where the 304 object is large, unwieldy, or unavailable (such as when licensing 305 restrictions are in effect), a metadata record that includes the 306 identifier string will usually suffice. That record becomes a 307 conveniently manipulable object surrogate, acting as both an 308 association "receipt" and "declaration". 310 Note that our definition of identifier extends the one in use for 311 Uniform Resource Identifiers [URI]. The present document still 312 sometimes (ab)uses the terms "ARK" and "identifier" as shorthand for 313 the string part of an identifier, but the context should make the 314 meaning clear. 316 2. ARK Anatomy 318 An ARK is represented by a sequence of characters (a string) that 319 contains the label, "ark:", optionally preceded by the beginning part 320 of a URL. Here is a diagrammed example. 322 http://foobar.zaf.org/ark:/12025/654xz321/s3/f8.05v.tiff 323 \___________________/ \__/ \___/ \______/ \____________/ 324 (replaceable) | | | Qualifier 325 | ARK Label | | (NMA-supported) 326 | | | 327 Name Mapping Authority | Name (NAA-assigned) 328 Hostport (NMAH) | 329 Name Assigning Authority Number (NAAN) 331 The ARK syntax can be summarized, 333 [http://NMAH/]ark:/NAAN/Name[Qualifier] 335 where the NMAH and Qualifier parts are in brackets to indicate that 336 they are optional. 338 2.1. The Name Mapping Authority Hostport (NMAH) 340 Before the "ark:" label may appear an optional Name Mapping Authority 341 Hostport (NMAH) that is a temporary address where ARK service 342 requests may be sent. It consists of "http://" (or any service 343 specification valid for a URL) followed by an Internet hostname or 344 hostport combination having the same format and semantics as the 345 hostport part of a URL. The most important thing about the NMAH is 346 that it is "identity inert" from the point of view of object 347 identification. In other words, ARKs that differ only in the 348 optional NMAH part identify the same object. Thus, for example, the 349 following three ARKs are synonyms for just one information object: 351 http://loc.gov/ark:/12025/654xz321 352 http://rutgers.edu/ark:/12025/654xz321 353 ark:/12025/654xz321 355 Strictly speaking, in the realm of digital objects, these ARKs may 356 lead over time to somewhat different or diverging instances of the 357 originally named object. In an ideal world, divergence of persistent 358 objects is not desirable, but it is widely believed that digital 359 preservation efforts will inevitably lead to alterations in some 360 original objects (e.g, a format migration in order to preserve the 361 ability to display a document). If any of those objects are held 362 redundantly in more than one organization (a common preservation 363 strategy), chances are small that all holding organizations will 364 perform the same precise transformations and all maintain the same 365 object metadata. More significant divergence would be expected when 366 the holding organizations serve different audiences or compete with 367 each other. 369 The NMAH part makes an ARK into an actionable URL. As with many 370 internet parameters, it is helpful to approach the NMAH being liberal 371 in what you accept and conservative in what you propose. From the 372 recipient's point of view, the NMAH part should be treated as 373 temporary, disposable, and replaceable. From the NMA's point of 374 view, it should be chosen with the greatest concern for longevity. A 375 carefully chosen NMAH should be at least as permanent as the 376 providing organization's own hostname. In the case of a national or 377 university library, for example, there is no reason why the NMAH 378 should not be considerably more permanent than soft-funded proxy 379 hostnames such as hdl.handle.net, dx.doi.org, and purl.org. In 380 general and over time, however, it is not unexpected for an NMAH 381 eventually to stop working and require replacement with the NMAH of a 382 currently active service provider. 384 This replacement relies on a mapping authority "resolver" discovery 385 process, of which two alternate methods are outlined in a later 386 section. The ARK, URN, Handle, and DOI schemes all use a resolver 387 discovery model that sooner or later requires matching the original 388 assigning authority with a current provider servicing that 389 authority's named objects; once found, the resolver at that provider 390 performs what amounts to a redirect to a place where the object is 391 currently held. All the schemes rely on the ongoing functionality of 392 currently mainstream technologies such as the Domain Name System 393 [DNS] and web browsers. The Handle and DOI schemes in addition 394 require that the Handle protocol layer and global server grid be 395 available at all times. 397 The practice of prepending "http://" and an NMAH to an ARK is a way 398 of creating an actionable identifier by a method that is itself 399 temporary. Assuming that infrastructure supporting [HTTP] 400 information retrieval will no longer be available one day, ARKs will 401 then have to be converted into new kinds of actionable identifiers. 402 By that time, if ARKs see widespread use, web browsers would 403 presumably evolve to perform this (currently simple) transformation 404 automatically. 406 2.2. The ARK Label Part - ark: 408 The label part distinguishes an ARK from an ordinary identifier. In 409 a URL found in the wild, the string, "ark:/", indicates that the URL 410 stands a reasonable chance of being an ARK. If the context warrants, 411 verification that it actually is an ARK can be done by testing it for 412 existence of the three ARK services. 414 Since nothing about an identifier syntax directly affects 415 persistence, the "ark:" label (like "urn:", "doi:", and "hdl:") 416 cannot tell you whether the identifier is persistent or whether the 417 object is available. It does tell you that the original Name 418 Assigning Authority (NAA) had some sort of hopes for it, but it 419 doesn't tell you whether that NAA is still in existence, or whether a 420 decade ago it ceased to have any responsibility for providing 421 persistence, or whether it ever had any responsibility beyond naming. 423 Only a current provider can say for certain what sort of commitment 424 it intends, and the ARK label suggests that you can query the NMAH 425 directly to find out exactly what kind of persistence is promised. 426 Even if what is promised is impersistence (i.e., a short-term 427 identifier), saying so is valuable information to the recipient. 428 Thus an ARK is a high-functioning identifier in the sense that it 429 provides access to the object, the metadata, and a commitment 430 statement, even if the commitment is explicitly very weak. 432 2.3. The Name Assigning Authority Number (NAAN) 434 Recalling that the general form of the ARK is, 436 [http://NMAH/]ark:/NAAN/Name[Qualifier] 438 the part of the ARK directly following the "ark:" is the Name 439 Assigning Authority Number (NAAN) enclosed in `/' (slash) characters. 440 This part is always required, as it identifies the organization that 441 originally assigned the Name of the object. It is used to discover a 442 currently valid NMAH and to provide top-level partitioning of the 443 space of all ARKs. NAANs are registered in a manner similar to URN 444 Namespaces, but they are pure numbers consisting of 5 digits or 9 445 digits. Thus, the first 100,000 registered NAAs fit compactly into 446 the 5 digits, and if growth warrants, the next billion fit into the 9 447 digit form. In either case the fixed odd numbers of digits helps 448 reduce the chances of finding a NAAN out of context and confusing it 449 with nearby quantities such as 4-digit dates. 451 The NAAN designates a top-level ARK namespace. Once registered for a 452 namespace, a NAAN is never re-registered. It is possible, however, 453 for there to be a succession of organizations that manage of an ARK 454 namespace. 456 2.4. The Name Part 458 The part of the ARK just after the NAAN is the Name assigned by the 459 NAA, and it is also required. Semantic opaqueness in the Name part 460 is strongly encouraged in order to reduce an ARK's vulnerability to 461 era- and language-specific change. Identifier strings containing 462 linguistic fragments can create support difficulties down the road. 463 No matter how appropriate or even meaningless they are today, such 464 fragments may one day create confusion, give offense, or infringe on 465 a trademark as the semantic environment around us and our communities 466 evolves. 468 Names that look more or less like numbers avoid common problems that 469 defeat persistence and international acceptance. The use of digits 470 is highly recommended. Mixing in non-vowel alphabetic characters a 471 couple at a time is a relatively safe and easy way to achieve a 472 denser namespace (more possible names for a given length of the name 473 string). Such names have a chance of aging and traveling well. 474 Tools exists that mint, bind, and resolve opaque identifiers, with or 475 without check characters [NOID]. More on naming considerations is 476 given in a subsequent section. 478 2.5. The Qualifier Part 480 The part of the ARK following the NAA-assigned Name is an optional 481 Qualifier. It is a string that extends the base ARK in order to 482 create a kind of service entry point into the object named by the 483 NAA. At the discretion of the providing NMA, such a service entry 484 point permits an ARK to support access to individual hierarchical 485 components and subcomponents of an object, and to variants (versions, 486 languages, formats) of components. A Qualifier may be invented by 487 the NAA or by any NMA servicing the object. 489 In form, the Qualifier is a ComponentPath, or a VariantPath, or a 490 ComponentPath followed by a VariantPath. A VariantPath is introduced 491 and subdivided by the reserved character `.', and a ComponentPath is 492 introduced and subdivided by the reserved character `/'. In this 493 example, 495 http://foobar.zaf.org/ark:/12025/654xz321/s3/f8.05v.tiff 497 the string "/s3/f8" is a ComponentPath and the string ".05v.tiff" is 498 a VariantPath. The ARK Qualifier is a formalization of some 499 currently mainstream URL syntax conventions. This formalization 500 specifically reserves meanings that permit recipients to make strong 501 inferences about logical sub-object containment and equivalence based 502 only on the form of the received identifiers; there is great 503 efficiency in not having to inspect metadata records to discover such 504 relationships. NMAs are free not to disclose any of these 505 relationships merely by avoiding the reserved characters above. 506 Hierarchical components and variants are discussed further in the 507 next two sections. 509 The Qualifier, if present, differs from the Name in several important 510 respects. First, a Qualifier may have been assigned either by the 511 NAA or later by the NMA. The assignment of a Qualifier by an NMA 512 effectively amounts to an act of publishing a service entry point 513 within the conceptual object originally named by the NAA. For our 514 purposes, an ARK extended with a Qualifier assigned by an NMA will be 515 called an NMA-qualified ARK. 517 Second, a Qualifier assignment on the part of an NMA is made in 518 fulfillment of its service obligations and may reflect changing 519 service expectations and technology requirements. NMA-qualified ARKs 520 could therefore be transient, even if the base, unqualified ARK is 521 persistent. For example, it would be reasonable for an NMA to 522 support access to an image object through an actionable ARK that is 523 considered persistent even if the experience of that access changes 524 as linking, labeling, and presentation conventions evolve and as 525 format and security standards are updated. For an image "thumbnail", 526 that NMA could also support an NMA-qualified ARK that is considered 527 impersistent because the thumbnail will be replaced with higher 528 resolution images as network bandwidth and CPU speeds increase. At 529 the same time, for an originally scanned, high-resolution master, the 530 NMA could publish an NMA-qualfied ARK that is itself considered 531 persistent. Of course, the NMA must be able to return its separate 532 commitments to unqualified, NAA-assigned ARKs, to NMA-qualified ARKs, 533 and to any NAA-qualified ARKs that it supports. 535 A third difference between a Qualifier and a Name concerns the 536 semantic opaqueness constraint. When an NMA-qualified ARK is to be 537 used as a transient service entry point into a persistent object, the 538 priority given to semantic opaqueness observed by the NAA in the Name 539 part may be relaxed by the NMA in the Qualifier part. If service 540 priorities in the Qualifier take precedence over persistence, short- 541 term usability considerations may recommend somewhat semantically 542 laden Qualifier strings. 544 Finally, not only is the set of Qualifiers supported by an NMA 545 mutable, but different NMAs may support different Qualifier sets for 546 the same NAA-identified object. In this regard the NMAs act 547 independently of each other and of the NAA. 549 The next two sections describe how ARK syntax may be used to declare, 550 or to avoid declaring, certain kinds of relatedness among qualified 551 ARKs. 553 2.5.1. ARKs that Reveal Object Hierarchy 555 An NAA or NMA may choose to reveal the presence of a hierarchical 556 relationship between objects using the `/' (slash) character after 557 the Name part of an ARK. Some authorities will choose not to 558 disclose this information, while others will go ahead and disclose so 559 that manipulators of large sets of ARKs can infer object 560 relationships by simple identifier inspection; for example, this 561 makes it possible for a system to present a collapsed view of a large 562 search result set. 564 If the ARK contains an internal slash after the NAAN, the piece to 565 its left indicates a containing object. For example, publishing an 566 ARK of the form, 568 ark:/12025/654/xz/321 570 is equivalent to publishing three ARKs, 572 ark:/12025/654/xz/321 573 ark:/12025/654/xz 574 ark:/12025/654 576 together with a declaration that the first object is contained in the 577 second object, and that the second object is contained in the third. 579 Revealing the presence of hierarchy is completely up to the assigner 580 (NMA or NAA). It is hard enough to commit to one object's name, let 581 alone to three objects' names and to a specific, ongoing relatedness 582 among them. Thus, regardless of whether hierarchy was present 583 initially, the assigner, by not using slashes, reveals no shared 584 inferences about hierarchical or other inter-relatedness in the 585 following ARKs: 587 ark:/12025/654_xz_321 588 ark:/12025/654_xz 589 ark:/12025/654xz321 590 ark:/12025/654xz 591 ark:/12025/654 593 Note that slashes around the ARK's NAAN (/12025/ in these examples) 594 are not part of the ARK's Name and therefore do not indicate the 595 existence of some sort of NAAN super object containing all objects in 596 its namespace. A slash must have at least one non-structural 597 character (one that is neither a slash nor a period) on both sides in 598 order for it to separate recognizable structural components. So 599 initial or final slashes may be removed, and double slashes may be 600 converted into single slashes. 602 2.5.2. ARKs that Reveal Object Variants 604 An NAA or NMA may choose to reveal the possible presence of variant 605 objects or object components using the `.' (period) character after 606 the Name part of an ARK. Some authorities will choose not to 607 disclose this information, while others will go ahead and disclose so 608 that manipulators of large sets of ARKs can infer object 609 relationships by simple identifier inspection; for example, this 610 makes it possible for a system to present a collapsed view of a large 611 search result set. 613 If the ARK contains an internal period after Name, the piece to its 614 left is a base name and the piece to its right, and up to the end of 615 the ARK or to the next period is a suffix. A Name may have more than 616 one suffix, for example, 617 ark:/12025/654.24 618 ark:/12025/xz4/654.24 619 ark:/12025/654.20v.78g.f55 621 There are two main rules. First, if two ARKs share the same base 622 name but have different suffixes, the corresponding objects were 623 considered variants of each other (different formats, languages, 624 versions, etc.) by the assigner (NMA or NAA). Thus, the following 625 ARKs are variants of each other: 627 ark:/12025/654.20v.78g.f55 628 ark:/12025/654.321xz 629 ark:/12025/654.44 631 Second, publishing an ARK with a suffix implies the existence of at 632 least one variant identified by the ARK without its suffix. The ARK 633 otherwise permits no further assumptions about what variants might 634 exist. So publishing the ARK, 636 ark:/12025/654.20v.78g.f55 638 is equivalent to publishing the four ARKs, 640 ark:/12025/654.20v.78g.f55 641 ark:/12025/654.20v.78g 642 ark:/12025/654.20v 643 ark:/12025/654 645 Revealing the possibility of variants is completely up to the 646 assigner. It is hard enough to commit to one object's name, let 647 alone to multiple variants' names and to a specific, ongoing 648 relatedness among them. The assigner is the sole arbiter of what 649 constitutes a variant within its namespace, and whether to reveal 650 that kind of relatedness by using periods within its names. 652 A period must have at least one non-structural character (one that is 653 neither a slash nor a period) on both sides in order for it to 654 separate recognizable structural components. So initial or final 655 periods may be removed, and adjacent periods may be converted into a 656 single period. Multiple suffixes should be arranged in sorted order 657 (pure ASCII collating sequence) at the end of an ARK. 659 2.6. Character Repertoires 661 The Name and Qualifier parts are strings of visible ASCII characters 662 and should be less than 128 bytes in length. The length restriction 663 keeps the ARK short enough to append ordinary ARK request strings 664 without running into transport restrictions (e.g., within HTTP GET 665 requests). Characters may be letters, digits, or any of these six 666 characters: 668 = # * + @ _ $ 670 The following characters may also be used, but their meanings are 671 reserved: 673 % - . / 675 The characters `/' and `.' are ignored if either appears as the last 676 character of an ARK. If used internally, they allow a name assigner 677 to reveal object hierarchy and object variants as previously 678 described. 680 Hyphens are considered to be insignificant and are always ignored in 681 ARKs. A `-' (hyphen) may appear in an ARK for readability, or it may 682 have crept in during the formatting and wrapping of text, but it must 683 be ignored in lexical comparisons. As in a telephone number, hyphens 684 have no meaning in an ARK. It is always safe for an NMA that 685 receives an ARK to remove any hyphens found in it. As a result, like 686 the NMAH, hyphens are "identity inert" in comparing ARKs for 687 equivalence. For example, the following ARKs are equivalent for 688 purposes of comparison and ARK service access: 690 ark:/12025/65-4-xz-321 691 http://sneezy.dopey.com/ark:/12025/654--xz32-1 692 ark:/12025/654xz321 694 The `%' character is reserved for %-encoding all other octets that 695 would appear in the ARK string, in the same manner as for URIs [URI]. 696 A %-encoded octet consists of a `%' followed by two hex digits; for 697 example, "%7d" stands in for `}'. Lower case hex digits are 698 preferred to reduce the chances of false acronym recognition; thus it 699 is better to use "%acT" instead of "%ACT". The character `%' itself 700 must be represented using "%25". As with URNs, %-encoding permits 701 ARKs to support legacy namespaces (e.g., ISBN, ISSN, SICI) that have 702 less restricted character repertoires [URNBIB]. 704 2.7. Normalization and Lexical Equivalence 706 To determine if two or more ARKs identify the same object, the ARKs 707 are compared for lexical equivalence after first being normalized. 708 Since ARK strings may appear in various forms (e.g., having different 709 NMAHs), normalizing them minimizes the chances that comparing two ARK 710 strings for equality will fail unless they actually identify 711 different objects. In a specified-host ARK (one having an NMAH), the 712 NMAH never participates in such comparisons. 714 Normalization of an ARK for the purpose of octet-by-octet equality 715 comparison with another ARK consists of four steps. First, any upper 716 case letters in the "ark:" label and the two characters following a 717 `%' are converted to lower case. The case of all other letters in 718 the ARK string must be preserved. Second, any NMAH part is removed 719 (everything from an initial "http://" up to the next slash) and all 720 hyphens are removed. 722 Third, structural characters (slash and period) are normalized. 723 Initial and final occurrences are removed, and two structural 724 characters in a row (e.g., // or ./) are replaced by the first 725 character, iterating until each occurrence has at least one non- 726 structural character on either side. Finally, if there are any 727 components with a period on the left and a slash on the right, either 728 the component and the preceding period must be moved to the end of 729 the Name part or the ARK must be thrown out as malformed. 731 The fourth and final step is to arrange the suffixes in ASCII 732 collating sequence (that is, to sort them) and to remove duplicate 733 suffixes, if any. It is also permissible to throw out ARKs for which 734 the suffixes are not sorted. 736 The resulting ARK string is now normalized. Comparisons between 737 normalized ARKs are case-sensitive, meaning that upper case letters 738 are considered different from their lower case counterparts. 740 To keep ARK string variation to a minimum, no reserved ARK characters 741 should be %-encoded unless it is deliberately to conceal their 742 reserved meanings. No non-reserved ARK characters should ever be 743 %-encoded. Finally, no %-encoded character should ever appear in an 744 ARK in its decoded form. 746 3. Naming Considerations 748 The most important threats faced by persistence providers include 749 such things as funding loss, natural disaster, political and social 750 upheaval, processing faults, and errors in human oversight. There is 751 nothing that an identifer scheme can do about such things. Still, a 752 few observed identifier failures and inconveniences can be traced 753 back to naming practices that we now know to be less than optimal for 754 persistence. 756 3.1. ARKS Embedded in Language 758 The ARK has different goals from the URI, so it has different 759 character set requirements. Because linguistic constructs imperil 760 persistence, for ARKs non-ASCII character support is unimportant. 761 ARKs and URIs share goals of transcribability and transportability 762 within web documents, so characters are required to be visible, non- 763 conflicting with HTML/XML syntax, and not subject to tampering during 764 transmission across common transport gateways. Add the goal of 765 making an undelimited ARK recognizable in running prose, as in 766 ark:/12025/=@_22*$, and certain punctuation characters (e.g., comma, 767 period) end up being excluded from the ARK lest the end of a phrase 768 or sentence be mistaken for part of the ARK. 770 This consideration has more direct effect on ARK usability in a 771 natural language context than it has on ARK persistence. The same is 772 true of the rule preventing hyphens from having lexical significance. 773 It is fine to publish ARKs with hyphens in them (e.g., such as the 774 output of UUID/GUID generators), but the uniform treatment of hyphens 775 as insignificant reduces the possibility of users transcribing 776 identifiers that will have been broken through unpredictable 777 hyphenation by word processors. Any measure that reduces user 778 irritation with an identifier will increase its chances of survival. 780 3.2. Objects Should Wear Their Identifiers 782 A valuable technique for provision of persistent objects is to try to 783 arrange for the complete identifier to appear on, with, or near its 784 retrieved object. An object encountered at a moment in time when its 785 discovery context has long since disappeared could then easily be 786 traced back to its metadata, to alternate versions, to updates, etc. 787 This has seen reasonable success, for example, in book publishing and 788 software distribution. An identifier string only has meaning when 789 its association is known, and this a very sure, simple, and low-tech 790 method of reminding everyone exactly what that association is. 792 3.3. Names are Political, not Technological 794 If persistence is the goal, a deliberate local strategy for 795 systematic name assignment is crucial. Names must be chosen with 796 great care. Poorly chosen and managed names will devastate any 797 persistence strategy, and they do not discriminate by identifier 798 scheme. Whether a mistakenly re-assigned name is a URN, DOI, PURL, 799 URL, or ARK, the damage - failed access and confusion - is not 800 mitigated more in one scheme than in another. Conversely, in-house 801 efforts to manage names responsibly will go much further towards 802 safeguarding persistence than any choice of naming scheme or name 803 resolution technology. 805 Branding (e.g., at the corporate or departmental level) is important 806 for funding and visibility, but substrings representing brands and 807 organizational names should be given a wide berth except when 808 absolutely necessary in the hostname (the identity-inert) part of the 809 ARK. These substrings are not only unstable because organizations 810 change frequently, but they are also dangerous because successor 811 organizations often have political or legal reasons to actively 812 suppress predecessor names and brands. Any measure that reduces the 813 chances of future political or legal pressure on an identifier will 814 decrease the chances that our descendants will be obliged to 815 deliberately break it. 817 3.4. Choosing a Hostname or NMA 819 Hostnames appearing in any identifier meant to be persistent must be 820 chosen with extra care. The tendency in hostname selection has 821 traditionally been to choose a token with recognizable attributes, 822 such as a corporate brand, but that tendency wreaks havoc with 823 persistence that is supposed to outlive brands, corporations, subject 824 classifications, and natural language semantics (e.g., what did the 825 three letters "gay" mean in 1958, 1978, and 1998?). Today's 826 recognized and correct attributes are tomorrow's stale or incorrect 827 attributes. In making hostnames (any names, actually) long-term 828 persistent, it helps to eliminate recognizable attributes to the 829 extent possible. This affects selection of any name based on URLs, 830 including PURLs and the explicitly disposable NMAHs. 832 There is no excuse for a provider that manages its internal names 833 impeccably not to exercise the same care in choosing what could be an 834 exceptionally durable hostname, especially if it would form the 835 prefix for all the provider's URL-based external names. Registering 836 an opaque hostname in the ".org" or ".net" domain would not be a bad 837 start. Another way is to publish your ARKs with an organizational 838 domain name that will be mapped by DNS to an appropriate NMA host. 839 This makes for shorter names with less branding vulnerability. 841 It is a mistake to think that hostnames are inherently unstable. If 842 you require brand visibility, that may be a fact of life. But things 843 are easier if yours is the brand of long-lived cultural memory 844 institution such as a national or university library or archive. 845 Well-chosen hostnames from organizations that are sheltered from the 846 direct effects of a volatile marketplace can easily provide longer- 847 lived global resolvers than the domain names explicitly or implicitly 848 used as starting points for global resolution by indirection-based 849 persistent identifier schemes. For example, it is hard to imagine 850 circumstances under which the Library of Congress' domain name would 851 disappear sooner than, say, "handle.net". 853 For smaller libraries, archives, and preservation organizations, 854 there is a natural concern about whether they will be able to keep 855 their web servers and domain names in the face of uncertain funding. 856 One option is to form or join a consortium [N2T] of like-minded 857 organizations with the purpose of providing mutual preservation 858 support. The first goal of such a consortium would be to perpetually 859 rent a hostname on which to establish a web server that simply 860 redirects incoming member organization requests to the appropriate 861 member server; using ARKs, for example, a 150-member consortium could 862 run a very small server (24x7) that contained nothing more than 150 863 rewrite rules in its configuration file. Even more helpful would be 864 additional consortial support for a member organization that was 865 unable to continue providing services and needed to find a successor 866 archival organization. This would be a low-cost, low-tech way to 867 publish ARKs (or URLs) under highly persistent hostnames. 869 There are no obvious reasons why the organizations registering DNS 870 names, URN Namespaces, and DOI publisher IDs should have among them 871 one that is intrinsically more fallible than the next. Moreover, it 872 is a misconception that the demise of DNS and of HTTP need adversely 873 affect the persistence of URLs. At such a time, certainly URLs from 874 the present day might not then be actionable by our present-day 875 mechanisms, but resolution systems for future non-actionable URLs are 876 no harder to imagine than resolution systems for present-day non- 877 actionable URNs and DOIs. There is no more stable a namespace than 878 one that is dead and frozen, and that would then characterize the 879 space of names bearing the "http://" prefix. It is useful to 880 remember that just because hostnames have been carelessly chosen in 881 their brief history does not mean that they are unsuitable in NMAHs 882 (and URLs) intended for use in situations demanding the highest level 883 of persistence available in the Internet environment. A well-planned 884 name assignment strategy is everything. 886 3.5. Assigners of ARKs 888 A Name Assigning Authority (NAA) is an organization that creates (or 889 delegates creation of) long-term associations between identifiers and 890 information objects. Examples of NAAs include national libraries, 891 national archives, and publishers. An NAA may arrange with an 892 external organization for identifier assignment. The US Library of 893 Congress, for example, allows OCLC (the Online Computer Library 894 Center, a major world cataloger of books) to create associations 895 between Library of Congress call numbers (LCCNs) and the books that 896 OCLC processes. A cataloging record is generated that testifies to 897 each association, and the identifier is included by the publisher, 898 for example, in the front matter of a book. 900 An NAA does not so much create an identifier as create an 901 association. The NAA first draws an unused identifier string from 902 its namespace, which is the set of all identifiers under its control. 903 It then records the assignment of the identifier to an information 904 object having sundry witnessed characteristics, such as a particular 905 author and modification date. A namespace is usually reserved for an 906 NAA by agreement with recognized community organizations (such as 907 IANA and ISO) that all names containing a particular string be under 908 its control. In the ARK an NAA is represented by the Name Assigning 909 Authority Number (NAAN). 911 The ARK namespace reserved for an NAA is the set of names bearing its 912 particular NAAN. For example, all strings beginning with 913 "ark:/12025/" are under control of the NAA registered under 12025, 914 which might be the National Library of Finland. Because each NAA has 915 a different NAAN, names from one namespace cannot conflict with those 916 from another. Each NAA is free to assign names from its namespace 917 (or delegate assignment) according to its own policies. These 918 policies must be documented in a manner similar to the declarations 919 required for URN Namespace registration [URNNID]. 921 To register for a NAAN, please read about the mapping authority 922 discovery file in the next section and send email to ark@cdlib.org. 924 3.6. NAAN Namespace Management 926 Every NAA must have a namespace management strategy. A time-honored 927 technique is to hierarchically partition a namespace into 928 subnamespaces using prefixes that guarantee non-collision of names in 929 different partition. This practice is strongly encouraged for all 930 NAAs, especially when subnamespace management will be delegated to 931 other departments, units, or projects within an organization. For 932 example, with a NAAN that is assigned to a university and managed by 933 its main library, care should be taken to reserve semantically opaque 934 prefixes that will set aside large parts of the unused namespace for 935 future assignments. Prefix-based partition management is an 936 important responsibility of the NAA. 938 This sort of delegation by prefix is well-used in the formation of 939 DNS names and ISBN identifiers. An important difference is that in 940 the former, the hierarchy is deliberately exposed and in the latter 941 it is hidden. Rather than using lexical boundary markers such as the 942 period (`.') found in domain names, the ISBN uses a publisher prefix 943 but doesn't disclose where the prefix ends and the publisher's 944 assigned name begins. This practice of non-disclosure, borrowed from 945 the ISBN and ISSN schemes, is encouraged in assigning ARKs, because 946 it reduces the visibility of an assertion that is probably not 947 important now and may become a vulnerability later. 949 Reasonable prefixes for assigned names usually consist of consonants 950 and digits and are 1-5 characters in length. For example, the 951 constant prefix "x9t" might be delegated to a book digitization 952 project that creates identifiers such as 954 http://444.berkeley.edu/ark:/28722/x9t38rk45c 956 If longevity is the goal, it is important to keep the prefixes free 957 of recognizable semantics; for example, using an acronym representing 958 a project or a department is discouraged. At the same time, you may 959 wish to set aside a subnamespace for testing purposes under a prefix 960 such as "fk..." that can serve as a visual clue and reminder to 961 maintenance staff that this "fake" identifier was never published. 963 There are other measures one can take to avoid user confusion, 964 transcription errors, and the appearance of accidental semantics when 965 creating identifiers. If you are generating identifiers 966 automatically, pure numeric identifiers are likeley to be 967 semantically opaque enough, but it's probably useful to avoid leading 968 zeroes because some users mistakenly treat them as optional, thinking 969 (arithmetically) that they don't contribute to the "value" of the 970 identifier. 972 If you need lots of identifiers and you don't want them to get too 973 long, you can mix digits with consonants (but avoid vowels since they 974 might accidentally spell words) to get more identifiers without 975 increasing the string length. In this case you may not want more 976 than a two letters in a row because it reduces the chance of 977 generating acronyms. Generator tools such as [NOID] provide support 978 for these sorts of identifiers, and can also add a computed check 979 character as a guarantee against the most common transcription 980 errors. 982 3.7. Sub-Object Naming 984 As mentioned previously, semantically opaque identifiers are very 985 useful for long-term naming of abstract objects, however, it may be 986 appropriate to extend these names with less opaque extensions that 987 reference contemporary service entry points (sub-objects) in support 988 of the object. Sub-object extensions beginning with a digit or 989 underscore (`_') are reserved for the possibilty of developing a 990 future registry of canonical service points (e.g., numeric references 991 to versions, formats, languages, etc). 993 4. Finding a Name Mapping Authority 995 In order to derive an actionable identifier (these days, a URL) from 996 an ARK, a hostport (hostname or hostname plus port combination) for a 997 working Name Mapping Authority (NMA) must be found. An NMA is a 998 service that is able to respond to the three basic ARK service 999 requests. Relying on registration and client-side discovery, NMAs 1000 make known which NAAs' identifiers they are willing to service. 1002 Upon encountering an ARK, a user (or client software) looks inside it 1003 for the optional NMAH part (the hostport of the NMA's ARK service). 1004 If it contains an NMAH that is working, this NMAH discovery step may 1005 be skipped; the NMAH effectively uses the beginning of an ARK to 1006 cache the results of a prior mapping authority discovery process. If 1007 a new NMAH needs to found, the client looks inside the ARK again for 1008 the NAAN (Name Assigning Authority Number). Querying a global 1009 database, it then uses the NAAN to look up all current NMAHs that 1010 service ARKs issued by the identified NAA. The global database is 1011 key, and two specific methods for querying it are given in this 1012 section. 1014 A third very promising method, called the Name-to-Thing [N2T] 1015 Resolver, is being explored. It is a low-cost, highly stable, 1016 consortially maintained NMAH that simply exists to support actionable 1017 HTTP-based URLs for as long as HTTP is used. One of its big 1018 advantages over the other two methods and the URN, Handle, DOI, and 1019 PURL methods, is that N2T addresses the namespace splitting problem. 1020 When objects maintained by one NMA are inherited by more than one 1021 successor NMA, until now one of those successors would be required to 1022 maintain forwarding tables on behalf of the other successors. 1024 In the interests of long-term persistence, however, ARK mechanisms 1025 are first defined in high-level, protocol-independent terms so that 1026 mechanisms may evolve and be replaced over time without compromising 1027 fundamental service objectives. Either or both specific methods 1028 given here may eventually be supplanted by better methods since, by 1029 design, the ARK scheme does not depend on a particular method, but 1030 only on having some method to locate an active NMAH. 1032 At the time of issuance, at least one NMAH for an ARK should be 1033 prepared to service it. That NMA may or may not be administered by 1034 the Name Assigning Authority (NAA) that created it. Consider the 1035 following hypothetical example of providing long-term access to a 1036 cancer research journal. The publisher wishes to turn a profit and 1037 the National Library of Medicine wishes to preserve the scholarly 1038 record. An agreement might be struck whereby the publisher would act 1039 as the NAA and the national library would archive the journal issue 1040 when it appears, but without providing direct access for the first 1041 six months. During the first six months of peak commercial 1042 viability, the publisher would retain exclusive delivery rights and 1043 would charge access fees. Again, by agreement, both the library and 1044 the publisher would act as NMAs, but during that initial period the 1045 library would redirect requests for issues less than six months old 1046 to the publisher. At the end of the waiting period, the library 1047 would then begin servicing requests for issues older than six months 1048 by tapping directly into its own archives. Meanwhile, the publisher 1049 might routinely redirect incoming requests for older issues to the 1050 library. Long-term access is thereby preserved, and so is the 1051 commercial incentive to publish content. 1053 Although it will be common for an NAA also to run an NMA service, it 1054 is never a requirement. Over time NAAs and NMAs will come and go. 1055 One NMA will succeed another, and there might be many NMAs serving 1056 the same ARKs simultaneously (e.g., as mirrors or as competitors). 1057 There might also be asymmetric but coordinated NMAs as in the 1058 library-publisher example above. 1060 4.1. Looking Up NMAHs in a Globally Accessible File 1062 This subsection describes a way to look up NMAHs using a simple name 1063 authority table represented as a plain text file. For efficient 1064 access the file may be stored in a local filesystem, but it needs to 1065 be reloaded periodically to incorporate updates. It is not expected 1066 that the size of the file or frequency of update should impose an 1067 undue maintenance or searching burden any time soon, for even 1068 primitive linear search of a file with ten-thousand NAAs is a 1069 subsecond operation on modern server machines. The proposed file 1070 strategy is similar to the /etc/hosts file strategy that supported 1071 Internet host address lookup for a period of years before the advent 1072 of DNS. 1074 The name authority table file is updated on an ongoing basis and is 1075 available for copying over the internet from the California Digital 1076 Library at http://www.cdlib.org/inside/diglib/ark/natab and from a 1077 number of mirror sites. The file contains comment lines (lines that 1078 begin with `#') explaining the format and giving the file's 1079 modification time, reloading address, and NAA registration 1080 instructions. There is even a Perl script that processes the file 1081 embedded in the file's comments. As of February 2006, currently 1082 registered Name Assigning Authorities are: 1084 12025 National Library of Medicine 1085 12026 Library of Congress 1086 12027 National Agriculture Library 1087 13030 California Digital Library 1088 13038 World Intellectual Property Organization 1089 20775 University of California San Diego 1090 29114 University of California San Francisco 1091 28722 University of California Berkeley 1092 21198 University of California Los Angeles 1093 15230 Rutgers University 1094 13960 Internet Archive 1095 64269 Digital Curation Centre 1096 62624 New York University 1097 67531 University of North Texas 1098 27927 Ithaka Electronic-Archiving Initiative 1099 12148 Bibliotheque nationale de France / National Library of France 1100 78319 Google 1101 88435 Princeton University 1102 78428 University of Washington 1103 89901 Archives of Region of Vastra Gotaland and City of Gothenburg, Sweden 1104 80444 Northwest Digital Archives 1105 25593 Emory University 1106 25031 University of Kansas 1107 17101 Centre for Ecology & Hydrology, UK 1109 A snapshot of the name authority table file appears in an appendix. 1111 4.2. Looking up NMAHs Distributed via DNS 1113 This subsection introduces a method for looking up NMAHs that is 1114 based on the method for discovering URN resolvers described in 1115 [NAPTR]. It relies on querying the DNS system already installed in 1116 the background infrastructure of most networked computers. A query 1117 is submitted to DNS asking for a list of resolvers that match a given 1118 NAAN. DNS distributes the query to the particular DNS servers that 1119 can best provide the answer, unless the answer can be found more 1120 quickly in a local DNS cache as a side-effect of a recent query. 1121 Responses come back inside Name Authority Pointer (NAPTR) records. 1122 The normal result is one or more candidate NMAHs. 1124 In its full generality the [NAPTR] algorithm ambitiously accommodates 1125 a complex set of preferences, orderings, protocols, mapping services, 1126 regular expression rewriting rules, and DNS record types. This 1127 subsection proposes a drastic simplification of it for the special 1128 case of ARK mapping authority discovery. The simplified algorithm is 1129 called Maptr. It uses only one DNS record type (NAPTR) and restricts 1130 most of its field values to constants. The following hypothetical 1131 excerpt from a DNS data file for the NAAN known as 12026 shows three 1132 example NAPTR records ready to use with the Maptr algorithm. 1134 12026.ark.arpa. 1135 ;; US Library of Congress 1136 ;; order pref flags service regexp replacement 1137 IN NAPTR 0 0 "h" "ark" "USLC" lhc.nlm.nih.gov:8080 1138 IN NAPTR 0 0 "h" "ark" "USLC" foobar.zaf.org 1139 IN NAPTR 0 0 "h" "ark" "USLC" sneezy.dopey.com 1141 All the fields are held constant for Maptr except for the "flags", 1142 "regexp", and "replacement" fields. The "service" field contains the 1143 constant value "ark" so that NAPTR records participating in the Maptr 1144 algorithm will not be confused with other NAPTR records. The "order" 1145 and "pref" fields are held to 0 (zero) and otherwise ignored for now; 1146 the algorithm may evolve to use these fields for ranking decisions 1147 when usage patterns and local administrative needs are better 1148 understood. 1150 When a Maptr query returns a record with a flags field of "h" (for 1151 hostport, a Maptr extension to the NAPTR flags), the replacement 1152 field contains the NMAH (hostport) of an ARK service provider. When 1153 a query returns a record with a flags field of "" (the empty string), 1154 the client needs to submit a new query containing the domain name 1155 found in the replacement field. This second sort of record exploits 1156 the distributed nature of DNS by redirecting the query to another 1157 domain name. It looks like this. 1159 12345.ark.arpa. 1160 ;; Digital Library Consortium 1161 ;; order pref flags service regexp replacement 1162 IN NAPTR 0 0 "" "ark" "" dlc.spct.org. 1164 Here is the Maptr algorithm for ARK mapping authority discovery. In 1165 it replace with the NAAN from the ARK for which an NMAH is 1166 sought. 1168 (1) Initialize the DNS query: type=NAPTR, 1169 query=.ark.arpa. 1171 (2) Submit the query to DNS and retrieve (NAPTR) records, 1172 discarding any record that does not have "ark" for the service 1173 field. 1175 (3) All remaining records with a flags fields of "h" contain 1176 candidate NMAHs in their replacement fields. Set them aside, if 1177 any. 1179 (4) Any record with an empty flags field ("") has a replacement 1180 field containing a new domain name to which a subsequent query 1181 should be redirected. For each such record, set 1182 query= then go to step (2). When all such records 1183 have been recursively exhausted, go to step (5). 1185 (5) All redirected queries have been resolved and a set of 1186 candidate NMAHs has been accumulated from steps (3). If there 1187 are zero NMAHs, exit - no mapping authority was found. If there 1188 is one or more NMAH, choose one using any criteria you wish, 1189 then exit. 1191 A Perl script that implements this algorithm is included here. 1193 #!/depot/bin/perl 1195 use Net::DNS; # include simple DNS package 1196 my $qtype = "NAPTR"; # initialize query type 1197 my $naa = shift; # get NAAN script argument 1198 my $mad = new Net::DNS::Resolver; # mapping authority discovery 1200 &maptr("$naa.ark.arpa"); # call maptr - that's it 1202 sub maptr { # recursive maptr algorithm 1203 my $dname = shift; # domain name as argument 1204 my ($rr, $order, $pref, $flags, $service, $regexp, 1205 $replacement); 1206 my $query = $mad->query($dname, $qtype); 1207 return # non-productive query 1208 if (! $query || ! $query->answer); 1209 foreach $rr ($query->answer) { 1210 next # skip records of wrong type 1211 if ($rr->type ne $qtype); 1212 ($order, $pref, $flags, $service, $regexp, 1213 $replacement) = split(/\s/, $rr->rdatastr); 1214 if ($flags eq "") { 1215 &maptr($replacement); # recurse 1216 } elsif ($flags eq "h") { 1217 print "$replacement\n"; # candidate NMAH 1218 } 1219 } 1220 } 1222 The global database thus distributed via DNS and the Maptr algorithm 1223 can easily be seen to mirror the contents of the Name Authority Table 1224 file described in the previous section. 1226 5. Generic ARK Service Definition 1228 An ARK request's output is delivered information; examples include 1229 the object itself, a policy declaration (e.g., a promise of support), 1230 a descriptive metadata record, or an error message. The experience 1231 of object delivery is expected to be an evolving mix of information 1232 that reflects changing service expectations and technology 1233 requirements; contemporary examples include such things as an object 1234 summary and component links formatted for human consumption. ARK 1235 services must be couched in high-level, protocol-independent terms if 1236 persistence is to outlive today's networking infrastructural 1237 assumptions. The high-level ARK service definitions listed below are 1238 followed in the next section by a concrete method (one of many 1239 possible methods) for delivering these services with today's 1240 technology. 1242 5.1. Generic ARK Access Service (access, location) 1244 Returns (a copy of) the object or a redirect to the same, although a 1245 sensible object proxy may be substituted. Examples of sensible 1246 substitutes include, 1248 - a table of contents instead of a large complex document, 1249 - a home page instead of an entire web site hierarchy, 1250 - a rights clearance challenge before accessing protected data, 1251 - directions for access to an offline object (e.g., a book), 1252 - a description of an intangible object (a disease, an event), or 1253 - an applet acting as "player" for a large multimedia object. 1255 May also return a discriminated list of alternate object locators. 1256 If access is denied, returns an explanation of the object's current 1257 (perhaps permanent) inaccessibility. 1259 5.2. Generic Policy Service (permanence, naming, etc.) 1261 Returns declarations of policy and support commitments for given 1262 ARKs. Declarations are returned in either a structured metadata 1263 format or a human readable text format; sometimes one format may 1264 serve both purposes. Policy subareas may be addressed in separate 1265 requests, but the following areas should should be covered: object 1266 permanence, object naming, object fragment addressing, and 1267 operational service support. 1269 The permanence declaration for an object is a rating defined with 1270 respect to an identified permanence provider (guarantor), which will 1271 be the NMA. It may include the following aspects. 1273 (a) "object availability" - whether and how access to the object 1274 is supported (e.g., online 24x7, or offline only), 1276 (b) "identifier validity" - under what conditions the identifier 1277 will be or has been re-assigned, 1279 (c) "content invariance" - under what conditions the content of 1280 the object is subject to change, and 1282 (d) "change history" - access to corrections, migrations, and 1283 revisions, whether through links to the changed objects 1284 themselves or through a document summarizing the change history 1286 One approach to a permanence rating framework, conceived 1287 independently from ARKs, is given in [NLMPerm]. Under ongoing 1288 development and limited deployment at the US National Library of 1289 Medicine, it identifies the following "permanence levels": 1291 Not Guaranteed: No commitment has been made to retain this 1292 resource. It could become unavailable at any time. Its 1293 identifier could be changed. 1295 Permanent: Dynamic Content: A commitment has been made to keep 1296 this resource permanently available. Its identifier will always 1297 provide access to the resource. Its content could be revised or 1298 replaced. 1300 Permanent: Stable Content: A commitment has been made to keep 1301 this resource permanently available. Its identifier will always 1302 provide access to the resource. Its content is subject only to 1303 minor corrections or additions. 1305 Permanent: Unchanging Content: A commitment has been made to 1306 keep this resource permanently available. Its identifier will 1307 always provide access to the resource. Its content will not 1308 change. 1310 Naming policy for an object includes an historical description of the 1311 NAA's (and its successor NAA's) policies regarding differentiation of 1312 objects. Since it the NMA who responds to requests for policy 1313 statements, it is useful for the NMA to be able to produce or 1314 summarize these historical NAA documents. Naming policy may include 1315 the following aspects. 1317 (i) "similarity" - (or "unity") the limit, defined by the NAA, 1318 to the level of dissimilarity beyond which two similar objects 1319 warrant separate identifiers but before which they share one 1320 single identifier, and 1322 (ii) "granularity" - the limit, defined by the NAA, to the level 1323 of object subdivision beyond which sub-objects do not warrant 1324 separately assigned identifiers but before which sub-objects are 1325 assigned separate identifiers. 1327 Subnaming policy for an object describes the qualifiers that the NMA, 1328 in fulfilling its ongoing and evolving service obligations, allows as 1329 extensions to an NAA-assigned ARK. To the conceptual object that the 1330 NAA named with an ARK, the NMA may add component access points and 1331 derivatives (e.g., format migrations in aid of preservation) in order 1332 to provide both basic and value-added services. 1334 Addressing policy for an object includes a description of how, during 1335 access, object components (e.g., paragraphs, sections) or views 1336 (e.g., image conversions) may or may not be "addressed", in other 1337 words, how the NMA permits arguments or parameters to modify the 1338 object delivered as the result of an ARK request. If supported, 1339 these sorts of operations would provide things like byte-ranged 1340 fragment delivery and open-ended format conversions, or any set of 1341 possible transformations that would be too numerous to list or to 1342 identify with separately assigned ARKs. 1344 Operational service support policy includes a description of general 1345 operational aspects of the NMA service, such as after-hours staffing 1346 and trouble reporting procedures. 1348 5.3. Generic Description Service 1350 Returns a description of the object. Descriptions are returned in 1351 either a structured metadata format or a human readable text format; 1352 sometimes one format may serve both purposes. A description must at 1353 a minimum answer the who, what, when, and where questions concerning 1354 an expression of the object. Standalone descriptions should be 1355 accompanied by the modification date and source of the description 1356 itself. May also return discriminated lists of ARKs that are related 1357 to the given ARK. 1359 6. Overview of The HTTP URL Mapping Protocol (THUMP) 1361 The HTTP URL Mapping Protocol (THUMP) is a way of taking a key (a 1362 kind of identifier) and asking such questions as, what information 1363 does this identify and how permanent is it? [THUMP] is in fact one 1364 specific method under development for delivering ARK services. The 1365 protocol runs over HTTP to exploit the web browser's current pre- 1366 eminence as user interface to the Internet. THUMP is designed so 1367 that a person can enter ARK requests directly into the location field 1368 of current browser interfaces. Because it runs over HTTP, THUMP can 1369 be simulated and tested within keyboard-based [TELNET] sessions. 1371 The asker (a person or client program) starts with an identifier, 1372 such as an ARK or a URL. The identifier reveals to the asker (or 1373 allows the asker to infer) the Internet host name and port number of 1374 a server system that responds to questions. Here, this is just the 1375 NMAH that is obtained by inspection and possibly lookup based on the 1376 ARK's NAAN. The asker then sets up an HTTP session with the server 1377 system, sends a question via a THUMP request (contained within an 1378 HTTP request), receives an answer via a THUMP response (contained 1379 within an HTTP response), and closes the session. That concludes the 1380 connected portion of the protocol. 1382 A THUMP request is a string of characters beginning with a `?' 1383 (question mark) that is appended to the identifier string. The 1384 resulting string is sent as an argument to HTTP's GET command. 1385 Request strings too long for GET may be sent using HTTP's POST 1386 command. The three most common requests correspond to three 1387 degenerate special cases that keep the user's learning and typing 1388 burden low. First, a simple key with no request at all is the same 1389 as an ordinary access request. Thus a plain ARK entered into a 1390 browser's location field behaves much like a plain URL, and returns 1391 access to the primary identified object, for instance, an HTML 1392 document. 1394 The second special case is a minimal ARK description request string 1395 consisting of just "?". For example, entering the string, 1397 ark.nlm.nih.gov/12025/psbbantu? 1399 into the browser's location field directly precipitates a request for 1400 a metadata record describing the object identified by 1401 ark:/12025/psbbantu. The browser, unaware of THUMP, prepares and 1402 sends an HTTP GET request in the same manner as for a URL. THUMP is 1403 designed so that the response (indicated by the returned HTTP content 1404 type) is normally displayed, whether the output is structured for 1405 machine processing (text/plain) or formatted for human consumption 1406 (text/html). 1408 In the following example THUMP session, each line has been annotated 1409 to include a line number and whether it was the client or server that 1410 sent it. Without going into much depth, the session has four pieces 1411 separated from each other by blank lines: the client's piece (lines 1412 1-3), the server's HTTP/THUMP response headers (4-7), and the body of 1413 the server's response (8-17). The first and last lines (1 and 17) 1414 correspond to the client's steps to start the TCP session and the 1415 server's steps to end it, respectively. 1417 1 C: [opens session] 1418 C: GET http://ark.nlm.nih.gov/ark:/12025/psbbantu? HTTP/1.1 1419 C: 1420 S: HTTP/1.1 200 OK 1421 5 S: Content-Type: text/plain 1422 S: THUMP-Status: 0.1 200 OK 1423 S: 1424 S: |set: NLM | 12025/psbbantu? | 20030731 1425 S: | http://ark.nlm.nih.gov/ark:/12025/psbbantu? 1426 10 S: here: 1 | 1 | 1 1427 S: 1428 S: erc: 1429 S: who: Lederberg, Joshua 1430 S: what: Studies of Human Families for Genetic Linkage 1431 15 S: when: 1974 1432 S: where: http://profiles.nlm.nih.gov/BB/A/N/T/U/_/bbantu.pdf 1433 S: [closes session] 1435 The first two server response lines (4-5) above are typical of HTTP. 1436 The next line (6) is peculiar to THUMP, and indicates the THUMP 1437 version and a normal return status. The balance of the response 1438 consists of a record set header (lines 8-10) and a single metadata 1439 record (12-16) that comprises the ARK description service response. 1440 The record set header identifies (8-9) who created the set, what its 1441 title is, when it was created, and where an automated process can 1442 access the set; it ends in a line (10) whose respective sub-elements 1443 indicate that here in this communication the recipient can expect to 1444 find 1 record, starting at the record numbered 1, from a set 1445 consisting of a total of 1 record (i.e., here is the entire set, 1446 consisting of exactly one record). 1448 The returned record (12-16) is in the format of an Electronic 1449 Resource Citation [ERC], which is discussed in more detail in the 1450 next section. For now, note that it contains four elements that 1451 answer the top priority questions regarding an expression of the 1452 object: who played a major role in expressing it, what the 1453 expression was called, when is was created, and where the expression 1454 may be found. This quartet of elements comes up again and again in 1455 ERCs. 1457 The third degenerate special case of an ARK request (and no other 1458 cases will be described in this document) is the string "??", 1459 corresponding to a minimal permanence policy request. It can be seen 1460 in use appended to an ARK (on line 2) in the example session that 1461 follows. 1463 1 C: [opens session] 1464 C: GET http://ark.nlm.nih.gov/ark:/12025/psbbantu?? HTTP/1.1 1465 C: 1466 S: HTTP/1.1 200 OK 1467 5 S: Content-Type: text/plain 1468 S: THUMP-Status: 0.1 200 OK 1469 S: 1470 S: |set: NLM | 12025/psbbantu?? | 20030731 1471 S: | http://ark.nlm.nih.gov/ark:/12025/psbbantu?? 1472 10 S: here: 1 | 1 | 1 1473 S: 1474 S: erc: 1475 S: who: Lederberg, Joshua 1476 S: what: Studies of Human Families for Genetic Linkage 1477 15 S: when: 1974 1478 S: where: http://profiles.nlm.nih.gov/BB/A/N/T/U/_/bbantu.pdf 1479 S: erc-support: 1480 S: who: USNLM 1481 S: what: Permanent, Unchanging Content 1482 20 S: when: 20010421 1483 S: where: http://ark.nlm.nih.gov/yy22948 1484 S: [closes session] 1486 Again, a single metadata record (lines 12-21) is returned, but it 1487 consists of two segments. The first segment (12-16) gives the same 1488 basic citation information as in the previous example. It is 1489 returned in order to establish context for the persistence 1490 declaration in the second segment (17-21). 1492 Each segment in an ERC tells a different story relating to the 1493 object, so although the same four questions (elements) appear in 1494 each, the answers depend on the segment's story type. While the 1495 first segment tells the story of an expression of the object, the 1496 second segment tells the story of the support commitment made to it: 1497 who made the commitment, what the nature of the commitment was, when 1498 it was made, and where a fuller explanation of the commitment may be 1499 found. 1501 7. Overview of Electronic Resource Citations (ERCs) 1503 An Electronic Resource Citation (or ERC, pronounced e-r-c) [ERC] is a 1504 simple, compact, and printable record designed to hold data 1505 associated with an information resource. By design, the ERC is a 1506 metadata format that balances the needs for expressive power, very 1507 simple machine processing, and direct human manipulation. 1509 A founding principle of the ERC is that direct human contact with 1510 metadata will be a necessary and sufficient condition for the near 1511 term rapid development of metadata standards, systems, and services. 1512 Thus the machine-processable ERC format must only minimally strain 1513 people's ability to read, understand, change, and transmit ERCs 1514 without their relying on intermediation with specialized software 1515 tools. The basic ERC needs to be succinct, transparent, and 1516 trivially parseable by software. 1518 In the current Internet, it is natural seriously to consider using 1519 XML as an exchange format because of predictions that it will obviate 1520 many ad hoc formats and programs, and unify much of the world's 1521 information under one reliable data structuring discipline that is 1522 easy to generate, verify, parse, and render. It appears, however, 1523 that XML is still only catching on after years of standards work and 1524 implementation experience. The reasons for it are unclear, but for 1525 now very simple XML interpretation is still out of reach. Another 1526 important caution is that XML structures are hard on the eyeballs, 1527 taking up an amount of display (and page) space that significantly 1528 exceeds that of traditional formats. Until these conflicts with ERC 1529 principle are resolved, XML is not a first choice for representing 1530 ERCs. Borrowing instead from the data structuring format that 1531 underlies the successful spread of email and web services, the first 1532 ERC format uses [ANVL], which is based on email and HTTP headers 1533 [RFC822]. There is a naturalness to ANVL's label-colon-value format 1534 (seen in the previous section) that barely needs explanation to a 1535 person beginning to enter ERC metadata. 1537 Besides simplicity of ERC system implementation and data entry 1538 mechanics, ERC semantics (what the record and its constituent parts 1539 mean) must also be easy to explain. ERC semantics are based on a 1540 reformulation and extension of the Dublin Core [DCORE] hypothesis, 1541 which suggests that the fifteen Dublin Core metadata elements have a 1542 key role to play in cross-domain resource description. The ERC 1543 design recognizes that the Dublin Core's primary contribution is the 1544 international, interdisciplinary consensus that identified fifteen 1545 semantic buckets (element categories), regardless of how they are 1546 labeled. The ERC then adds a definition for a record and some 1547 minimal compliance rules. In pursuing the limits of simplicity, the 1548 ERC design combines and relabels some Dublin Core buckets to isolate 1549 a tiny kernel (subset) of four elements for basic cross-domain 1550 resource description. 1552 For the cross-domain kernel, the ERC uses the four basic elements - 1553 who, what, when, and where - to pretend that every object in the 1554 universe can have a uniform minimal description. Each has a name or 1555 other identifier, a location, some responsible person or party, and a 1556 date. It doesn't matter what type of object it is, or whether one 1557 plans to read it, interact with it, smoke it, wear it, or navigate 1558 it. Of course, this approach is flawed because uniformity of 1559 description for some object types requires more semantic contortion 1560 and sacrifice than for others. That is why at the beginning of this 1561 document, the ARK was said to be suited to objects that accommodate 1562 reasonably regular electronic description. 1564 While insisting on uniformity at the most basic level provides 1565 powerful cross-domain leverage, the semantic sacrifice is great for 1566 many applications. So the ERC also permits a semantically rich and 1567 nuanced description to co-exist in a record along with a basic 1568 description. In that way both sophisticated and naive recipients of 1569 the record can extract the level of meaning from it that best suits 1570 their needs and abilities. Key to unlocking the richer description 1571 is a controlled vocabulary of ERC record types (not explained in this 1572 document) that permit knowledgeable recipients to apply defined sets 1573 of additional assumptions to the record. 1575 7.1. ERC Syntax 1577 An ERC record is a sequence of metadata elements ending in a blank 1578 line. An element consists of a label, a colon, and an optional 1579 value. Here is an example of a record with five elements. 1581 erc: 1582 who: Gibbon, Edward 1583 what: The Decline and Fall of the Roman Empire 1584 when: 1781 1585 where: http://www.ccel.org/g/gibbon/decline/ 1587 A long value may be folded (continued) onto the next line by 1588 inserting a newline and indenting the next line. A value can be thus 1589 folded across multiple lines. Here are two example elements, each 1590 folded across four lines. 1592 who/created: University of California, San Francisco, AIDS 1593 Program at San Francisco General Hospital | University 1594 of California, San Francisco, Center for AIDS Prevention 1595 Studies 1596 what/Topic: 1597 Heart Attack | Heart Failure 1598 | Heart 1599 Diseases 1601 An element value folded across several lines is treated as if the 1602 lines were joined together on one long line. For example, the second 1603 element from the previous example is considered equivalent to 1605 what/Topic: Heart Attack | Heart Failure | Heart Diseases 1607 An element value may contain multiple values, each one separated from 1608 the next by a `|' (pipe) character. The element from the previous 1609 example contains three values. 1611 For annotation purposes, any line beginning with a `#' (hash) 1612 character is treated as if it were not present; this is a "comment" 1613 line (a feature not available in email or HTTP headers). For 1614 example, the following element is spread across four lines and 1615 contains two values: 1617 what/Topic: 1618 Heart Attack 1619 # | Heart Failure -- hold off until next review cycle 1620 | Heart Diseases 1622 7.2. ERC Stories 1624 An ERC record is organized into one or more distinct segments, where 1625 where each segment tells a story about a different aspect of the 1626 information resource. A segment boundary occurs whenever a segment 1627 label (an element beginning with "erc") is encountered. The basic 1628 label "erc:" introduces the story of an object's expression (e.g., 1629 its publication, installation, or performance). The label "erc- 1630 about:" introduces the story of an object's content (what it is 1631 about) and "erc-support:" introduces the story of a support 1632 commitment made to it. A story segment that concerns the ERC itself 1633 is introduced by the label "erc-from:". It is an important segment 1634 that tells the story of the ERC's provenance. Elements beginning 1635 with "erc" are reserved for segment labels and their associated story 1636 types. From an earlier example, here is an ERC with two segments. 1638 erc: 1639 who: Lederberg, Joshua 1640 what: Studies of Human Families for Genetic Linkage 1641 when: 1974 1642 where: http://profiles.nlm.nih.gov/BB/A/N/T/U/_/bbantu.pdf 1643 erc-support: 1644 who: NIH/NLM/LHNCBC 1645 what: Permanent, Unchanging Content 1646 # Note to ops staff: date needs verification. 1647 when: 2001 04 21 1648 where: http://ark.nlm.nih.gov/yy22948 1650 Segment stories are told according to journalistic tradition. While 1651 any number of pertinent elements may appear in a segment, priority is 1652 placed on answering the questions who, what, when, and where at the 1653 beginning of each segment so that readers can make the most important 1654 selection or rejection decisions as soon as possible. To make things 1655 simple, the listed ordering of the questions is maintained in each 1656 segment (as it happens most people who have been exposed to this 1657 story telling technique are already familiar with the above 1658 ordering). 1660 The four questions are answered by using corresponding element 1661 labels. The four element labels can be re-used in each story 1662 segment, but their meaning changes depending on the segment (the 1663 story type) in which they appear. In the example above, "who" is 1664 first used to name a document's author and subsequently used to name 1665 the permanence guarantor (provider). Similarly, "when" first lists 1666 the date of object creation and in the next segment lists the date of 1667 a commitment decision. Four labels appearing across three segments 1668 effectively map to twelve semantically distinct elements. Distinct 1669 element meanings are mapped to Dublin Core elements in a later 1670 section. 1672 7.3. The ERC Anchoring Story 1674 Each ERC contains an anchoring story. It is usually the first 1675 segment labeled "erc:" and it concerns an "anchoring" expression of 1676 the object. An "anchoring" expression is the one that a provider 1677 deemed the most suitable basic referent given the audience and 1678 application for which it produced the ERC. If it sounds like the 1679 provider has great latitude in choosing its anchoring expression, it 1680 is because it does. A typical anchoring story in an ERC for a born- 1681 digital document would be the story of the document's release on a 1682 web site; such a document would then be the anchoring expression. 1684 An anchoring story need not be the central descriptive goal of an ERC 1685 record. For example, a museum provider may create an ERC for a 1686 digitized photograph of a painting but choose to anchor it in the 1687 story of the original painting instead of the story of the electronic 1688 likeness; although the ERC may through other segments prove to be 1689 centrally concerned with describing the electronic likeness, the 1690 provider may have chosen this particular anchoring story in order to 1691 make the ERC visible in a way that is most natural to patrons (who 1692 would find the Mona Lisa under da Vinci sooner than they would find 1693 it under the name of the person who snapped the photograph or scanned 1694 the image). In another example, a provider that creates an ERC for a 1695 dramatic play as an abstract work has the task of describing a piece 1696 of intangible intellectual property. To anchor this abstract object 1697 in the concrete world, if only through a derivative expression, it 1698 makes sense for the provider to choose a suitable printed edition of 1699 the play as the anchoring object expression (to describe in the 1700 anchoring story) of the ERC. 1702 The anchoring story has special rules designed to keep ERC processing 1703 simple and predictable. Each of the four basic elements (who, what, 1704 when, and where) must be present, unless a best effort to supply it 1705 fails. In the event of failure, the element still appears but a 1706 special value (described later) is used to explain the missing value. 1707 While the requirement that each of the four elements be present only 1708 applies to the anchoring story segment, as usual these elements 1709 appear at the beginning of the segment and may only be used in the 1710 prescribed order. A minimal ERC would normally consist of just an 1711 anchoring story and the element quartet, as illustrated in the next 1712 example. 1714 erc: 1715 who: National Research Council 1716 what: The Digital Dilemma 1717 when: 2000 1718 where: http://books.nap.edu/html/digital%5Fdilemma 1720 A minimal ERC can be abbreviated so that it resembles a traditional 1721 compact bibliographic citation that is nonetheless completely machine 1722 processable. The required elements and ordering makes it possible to 1723 eliminate the element labels, as shown here. 1725 erc: National Research Council | The Digital Dilemma | 2000 1726 | http://books.nap.edu/html/digital%5Fdilemma 1728 7.4. ERC Elements 1730 As mentioned, the four basic ERC elements (who, what, when, and 1731 where) take on different specific meanings depending on the story 1732 segment in which they are used. By appearing in each segment, albeit 1733 in different guises, the four elements serve as a valuable mnemonic 1734 device - a kind of checklist - for constructing minimal story 1735 segments from scratch. Again, it is only in the anchoring segment 1736 that all four elements are mandatory. 1738 Here are some mappings between ERC elements and Dublin Core [DCORE] 1739 elements. 1741 Segment ERC Element Equivalent Dublin Core Element 1742 --------- ----------- ------------------------------ 1743 erc who Creator/Contributor/Publisher 1744 erc what Title 1745 erc when Date 1746 erc where Identifier 1747 erc-about who 1748 erc-about what Subject 1749 erc-about when Coverage (temporal) 1750 erc-about where Coverage (spatial) 1752 The basic element labels may also be qualified to add nuances to the 1753 semantic categories that they identify. Elements are qualified by 1754 appending a `/' (slash) and a qualifier term. Often qualifier terms 1755 appear as the past tense form of a verb because it makes re-using 1756 qualifiers among elements easier. 1758 who/published: ... 1759 when/published: ... 1760 where/published: ... 1762 Using past tense verbs for qualifiers also reminds providers and 1763 recipients that element values contain transient assertions that may 1764 have been true once, but that tend to become less true over time. 1765 Recipients that don't understand the meaning of a qualifier can fall 1766 back onto the semantic category (bucket) designated by the 1767 unqualified element label. Inevitably recipients (people and 1768 software) will have diverse abilities in understanding elements and 1769 qualifiers. 1771 Any number of other elements and qualifiers may be used in 1772 conjunction with the quartet of basic segment questions. The only 1773 semantic requirement is that they pertain to the segment's story. 1774 Also, it is only the four basic elements that change meaning 1775 depending on their segment context. All other elements have meaning 1776 independent of the segment in which they appear. If an element label 1777 stripped of its qualifier is still not recognized by the recipient, a 1778 second fall back position is to ignore it and rely on the four basic 1779 elements. 1781 Elements may be either Canonical, Provisional, or Local. Canonical 1782 elements are officially recognized via a registry as part of the 1783 metadata vernacular. All elements, qualifiers, and segment labels 1784 used in this document up until now belong to that vernacular. 1785 Provisional elements are also officially recognized via the registry, 1786 but have only been proposed for inclusion in the vernacular. To be 1787 promoted to the vernacular, a provisional element passes through a 1788 vetting process during which its documentation must be in order and 1789 its community acceptance demonstrated. Local elements are any 1790 elements not officially recognized in the registry. The registry 1791 [DERC] is a work in progress. 1793 Local elements can be immediately distinguishable from Canonical or 1794 Provisional elements because all terms that begin with an upper case 1795 letter are reserved for spontaneous local use. No term beginning 1796 with an upper case letter will ever be assigned Canonical or 1797 Provisional status, so it should be safe to use such terms for local 1798 purposes. Any recipient of external ERCs containing such terms will 1799 understand them to be part of the originating provider's local 1800 metadata dialect. Here's an example ERC with three segments, one 1801 local element, and two local qualifiers. The segment boundaries have 1802 been emphasized by comment lines (which, as before, are ignored by 1803 processors). 1805 erc: 1806 who: Bullock, TH | Achimowicz, JZ | Duckrow, RB 1807 | Spencer, SS | Iragui-Madoz, VJ 1808 what: Bicoherence of intracranial EEG in sleep, 1809 wakefulness and seizures 1810 when: 1997 12 00 1811 where: http://cogprints.soton.ac.uk/%{ 1812 documents/disk0/00/00/01/22/index.html %} 1813 in: EEG Clin Neurophysiol | 1997 12 00 | v103, i6, p661-678 1814 IDcode: cog00000122 1815 # ---- new segment ---- 1816 erc-about: 1817 what/Subcategory: Bispectrum | Nonlinearity | Epilepsy 1818 | Cooperativity | Subdural | Hippocampus | Higher moment 1819 # ---- new segment ---- 1820 erc-from: 1821 who: NIH/NLM/NCBI 1822 what: pm9546494 1823 when/Reviewed: 1998 04 18 021600 1824 where: http://ark.nlm.nih.gov/12025/pm9546494? 1826 The local element "IDcode" immediately precedes the "erc-about" 1827 segment, which itself contains an element with the local qualifier 1828 "Subcategory". The second to last element also carries the local 1829 qualifier "Reviewed". Finally, what might be a provisional element 1830 "in" appears near the end of the first segment. It might have been 1831 proposed as a way to complete a citation for an object originally 1832 appearing inside another object (such as an article appearing in a 1833 journal or an encyclopedia). 1835 7.5. ERC Element Values 1837 ERC element values tend to be straightforward strings. If the 1838 provider intends something special for an element, it will so 1839 indicate with markers at the beginning of its value string. The 1840 markers are designed to be uncommon enough that they would not likely 1841 occur in normal data except by deliberate intent. Markers can only 1842 occur near the beginning of a string, and once any octet of non- 1843 marker data has been encountered, no further marker processing is 1844 done for the element value. In the absence of markers the string is 1845 considered pure data; this has been the case with all the examples 1846 seen thus far. The fullest form of an element value with all three 1847 optional markers in place looks like this. 1849 VALUE = [markup_flags] (:ccode) , DATA 1851 In processing, the first non-whitespace character of an ERC element 1852 value is examined. An initial `[' is reserved to introduce a 1853 bracketed set of markup flags (not described in this document) that 1854 ends with `]'. If ERC data is machine-generated, each value string 1855 may be preceded by "[]" to prevent any of its data from being 1856 mistaken for markup flags. Once past the optional markup, the 1857 remaining value may optionally begin with a controlled code. A 1858 controlled code always has the form "(:ccode)", for example, 1860 who: (:unkn) Anonymous 1861 what: (:791) Bee Stings 1863 Any string after such a code is taken to be an uncontrolled (e.g., 1864 natural language) equivalent. The code "unkn" indicates a 1865 conventional explanation for a missing value (stating that the value 1866 is unknown). The remainder of the string makes an equivalent 1867 statement in a form that the provider deemed most suitable to its 1868 (probably human) audience. The code "791" could be a fixed numeric 1869 topic identifier within an unspecified topic vocabulary. Any code 1870 may be ignored by those that do not understand it. 1872 There are several codes to explain different ways in which a required 1873 element's value may go missing. 1875 (:unac) temporarily inaccessible 1876 (:unal) unallowed, suppressed intentionally 1877 (:unap) not applicable, makes no sense 1878 (:unas) value unassigned (e.g., Untitled) 1879 (:unav) value unavailable indefinitely 1880 (:unkn) unknown (e.g., Anonymous, Inconnue) 1881 (:etal) too numerous to list (I). 1882 (:none) never had a value, never will 1883 (:null) explicitly empty 1884 (:tba) to be assigned or announced later 1886 Once past an optional controlled code, the remaining string value is 1887 subjected to one final test. If the first next non-whitespace 1888 character is a `,' (comma), it indicates that the string value is 1889 "sort-friendly". This means that the value is (a) laid out with an 1890 inverted word order useful for sorting items having comparably laid 1891 out element values (items might be the containing ERC records) and 1892 (b) that the value may contain other commas that indicate inversion 1893 points should it become necessary to recover the value in natural 1894 word order. Typically, this feature is used to express Western-style 1895 personal names in family-name-given-name order. It can also be used 1896 wherever natural word order might make sorting tricky, such as when 1897 data contains titles or corporate names. Here are some example 1898 elements. 1900 who: , van Gogh, Vincent 1901 who:,Howell, III, PhD, 1922-1987, Thurston 1902 who:, Acme Rocket Factory, Inc., The 1903 who:, Mao Tse Tung 1904 who:, McCartney, Paul, Sir, 1905 what:, Health and Human Services, United States Government 1906 Department of, The, 1908 There are rules to use in recovering a copy of the value in natural 1909 word order, if desired. The above example strings have the following 1910 natural word order values, respectively. 1912 Vincent van Gogh 1913 Thurston Howell, III, PhD, 1922-1987 1914 The Acme Rocket Factory, Inc. 1915 Mao Tse Tung 1916 Sir Paul McCartney 1917 The United States Government Department of Health and Human Services 1919 7.6. ERC Element Encoding and Dates 1921 Some characters that need to appear in ERC element values might 1922 conflict with special characters used for structuring ERCs, so there 1923 needs to be a way to include them as literal characters that are 1924 protected from special interpretation. This is accomplished through 1925 an encoding mechanism that resembles the %-encoding familiar to [URI] 1926 handlers. 1928 The ERC encoding mechanism also uses `%', but instead of taking two 1929 following hexadecimal digits, it takes one non-alphanumeric character 1930 or two alphabetic characters that cannot be mistaken for hex digits. 1931 It is designed not to be confused with normal web-style %-encoding. 1932 In particular it can be decoded without risking unintended decoding 1933 of normal %-encoded data (which would introduce errors). Here are 1934 the one-character (non-alphanumeric) ERC encoding extensions. 1936 ERC Purpose 1937 --- ------------------------------------------------ 1938 %! decodes to the element separator `|' 1939 %% decodes to a percent sign `%' 1940 %. decodes to a comma `,' 1941 %_ a non-character used as syntax shim 1942 %{ a non-character that begins an expansion block 1943 %} a non-character that ends an expansion block 1945 One particularly useful construct in ERC element values is the pair 1946 of special encoding markers ("%{" and "%}") that indicates a 1947 "expansion" block. Whatever string of characters they enclose will 1948 be treated as if none of the contained whitespace (SPACEs, TABs, 1949 Newlines) were present. This comes in handy for writing long, multi- 1950 part URLs in a readable way. For example, the value in 1952 where: http://foo.bar.org/node%{ 1953 ? db = foo 1954 & start = 1 1955 & end = 5 1956 & buf = 2 1957 & query = foo + bar + zaf 1958 %} 1960 is decoded into an equivalent element, but with a correct and intact 1961 URL: 1963 where: 1964 http://foo.bar.org/node?db=foo&start=1&end=5&buf=2&query=foo+bar+zaf 1966 In a parting word about ERC element values, a commonly recurring 1967 value type is a date, possibly followed by a time. ERC dates use the 1968 [TEMPER] format, taking on one of the following forms: 1970 1999 (four digit year) 1971 2000 12 29 (year, month, day) 1972 2000 12 29 235955 (year, month, day, hour, minute, second) 1974 In dates, all internal whitespace is squeezed out to achieve a 1975 normalized form suitable for lexical comparison and sorting. This 1976 means that the following dates 1978 2000 12 29 235955 (recommended for readability) 1979 2000 12 29 23 59 55 1980 20001229 23 59 55 1981 20001229235955 (normalized date and time) 1983 are all equivalent. The first form is recommended for readability. 1984 The last form (shortest and easiest to compute with) is the 1985 normalized form. Hyphens and commas are reserved to create date 1986 ranges and lists, for example, 1988 1996-2000 (a range of four years) 1989 1952, 1957, 1969 (a list of three years) 1990 1952, 1958-1967, 1985 (a mixed list of dates and ranges) 1991 20001229-20001231 (a range of three days) 1993 7.7. ERC Stub Records and Internal Support 1995 The ERC design introduces the concept of a "stub" record, which is an 1996 incomplete ERC record intended to be supplemented with additional 1997 elements before being released as a standalone ERC record. A stub 1998 ERC record has no minimum required elements. It is just a group of 1999 elements that does not begin with "erc:" but otherwise conforms to 2000 the ERC record syntax. 2002 ERC stubs may be useful in supporting internal procedures using the 2003 ERC syntax. Often they rely on the convenience and accuracy of 2004 automatically supplied elements, even the basic ones. To be ready 2005 for external use, however, an ERC stub must be transformed into a 2006 complete ERC record having the usual required elements. An ERC stub 2007 record can be convenient for metadata embedded in a document, where 2008 elements such as location, modification date, and size - which one 2009 would not omit from an externalized record - are omitted simply 2010 because they are much better supplied by a computation. A separate 2011 local administrative procedure, not defined for ERC's in general, 2012 would effect the promotion of stubs into complete records. 2014 While the ERC is a general-purpose container for exchange of resource 2015 descriptions, it does not dictate how records must be internally 2016 stored, laid out, or assembled by data providers or recipients. 2017 Arbitrary internal descriptive frameworks can support ERCs simply by 2018 mapping (e.g., on demand) local records to the ERC container format 2019 and making them available for export. Therefore, to support ERCs 2020 there is no need for a data provider to convert internal data to be 2021 stored in an ERC format. On the other hand, any provider (such as 2022 one just getting started in the business of resource description) may 2023 choose to store and manipulate local data natively in the ERC format. 2025 8. Advice to Web Clients 2027 This section offers some advice to web client software developers. 2028 It is hard to write about because it tries to anticipate a series of 2029 events that might lead to native web browser support for ARKs. 2031 ARKs are envisaged to appear wherever durable object references are 2032 planned. Library cataloging records, literature citations, and 2033 bibliographies are important examples. In many of these places URLs 2034 (Uniform Resource Locators) currently stand in, and URNs, DOIs, and 2035 PURLs have been proposed as alternatives. 2037 The strings representing ARKs are also envisaged to appear in some of 2038 the places where URLs currently appear: in hypertext links (where 2039 they are not normally shown to users) and in rendered text (displayed 2040 or printed). Internet search engines, for example, tend to include 2041 both actionable and manifest links when listing each item found. A 2042 normal HTML link for which the URL is not displayed looks like this. 2044 Click Here 2046 The same link with an ARK instead of a URL: 2048 Click Here 2050 Web browsers would in general require a small modification to 2051 recognize and convert this ARK, via mapping authority discovery, to 2052 the URL form. 2054 Click Here 2056 A browser that knows how to make that conversion could also 2057 automatically detect and replace a non-working NMAH. 2059 An NAA will typically make known the associations it creates by 2060 publishing them in catalogs, actively advertizing them, or simply 2061 leaving them on web sites for visitors (e.g., users, indexing 2062 spiders) to stumble across in browsing. 2064 9. Security Considerations 2066 The ARK naming scheme poses no direct risk to computers and networks. 2067 Implementors of ARK services need to be aware of security issues when 2068 querying networks and filesystems for Name Mapping Authority 2069 services, and the concomitant risks from spoofing and obtaining 2070 incorrect information. These risks are no greater for ARK mapping 2071 authority discovery than for other kinds of service discovery. For 2072 example, recipients of ARKs with a specified hostport (NMAH) should 2073 treat it like a URL and be aware that the identified ARK service may 2074 no longer be operational. 2076 Apart from mapping authority discovery, ARK clients and servers 2077 subject themselves to all the risks that accompany normal operation 2078 of the protocols underlying mapping services (e.g., HTTP, Z39.50). 2079 As specializations of such protocols, an ARK service may limit 2080 exposure to the usual risks. Indeed, ARK services may enhance a kind 2081 of security by helping users identify long-term reliable references 2082 to information objects. 2084 10. Authors' Addresses 2086 John A. Kunze 2087 California Digital Library 2088 University of California, Office of the President 2089 415 20th St, 4th Floor 2090 Oakland, CA 94612-3550, USA 2092 Fax: +1 510-893-5212 2093 EMail: jak@ucop.edu 2095 R. P. C. Rodgers 2096 US National Library of Medicine 2097 8600 Rockville Pike, Bldg. 38A 2098 Bethesda, MD 20894, USA 2100 Fax: +1 301-496-0673 2101 EMail: rodgers@nlm.nih.gov 2103 11. References 2105 [ANVL] J. Kunze, B. Kahle, et al, "A Name-Value Language", work 2106 in progress, 2107 http://www.cdlib.org/inside/diglib/ark/anvlspec.pdf 2109 [ARK] J. Kunze, "Towards Electronic Persistence Using ARK 2110 Identifiers", Proceedings of the 3rd ECDL Workshop on Web 2111 Archives, August 2003, (PDF) 2112 http://bibnum.bnf.fr/ecdl/2003/proceedings.php?f=kunze 2114 [DCORE] Dublin Core Metadata Initiative, "Dublin Core Metadata 2115 Element Set, Version 1.1: Reference Description", July 2116 1999, http://dublincore.org/documents/dces/. 2118 [DERC] J. Kunze, "Dictionary of the ERC", work in progress within 2119 the Dublin Core Metadata Initiative's Kernel Working 2120 Group, http://dublincore.org/groups/kernel/ 2122 [DNS] P.V. Mockapetris, "Domain Names - Concepts and 2123 Facilities", RFC 1034, November 1987. 2125 [DOI] International DOI Foundation, "The Digital Object 2126 Identifier (DOI) System", February 2001, 2127 http://dx.doi.org/10.1000/203. 2129 [ERC] J. Kunze, "A Metadata Kernel for Electronic Permanence", 2130 Journal of Digital Information, Vol 2, Issue 2, January 2131 2002, ISSN 1368-7506, (PDF) 2132 http://jodi.ecs.soton.ac.uk/Articles/v02/i02/Kunze/ 2134 [Handle] L. Lannom, "Handle System Overview", ICSTI Forum, No. 30, 2135 April 1999, http://www.icsti.org/forum/30/#lannom 2137 [HTTP] R. Fielding, et al, "Hypertext Transfer Protocol -- 2138 HTTP/1.1", RFC 2616, June 1999. 2140 [MD5] R. Rivest, "The MD5 Message-Digest Algorithm", RFC 1321, 2141 April 1992. 2143 [N2T] CDL, "Name-to-Thing Resolover", work in progress, August 2144 2006, http://n2t.info 2146 [NAPTR] M. Mealling, Daniel, R., "The Naming Authority Pointer 2147 (NAPTR) DNS Resource Record", RFC 2915, September 2000. 2149 [NLMPerm] M. Byrnes, "Defining NLM's Commitment to the Permanence of 2150 Electronic Information", ARL 212:8-9, October 2000, 2151 http://www.arl.org/newsltr/212/nlm.html 2153 [NOID] J. Kunze, "Nice Opaque Identifiers", February 2005, 2154 http://www.cdlib.org/inside/diglib/ark/noid.pdf 2156 [PURL] K. Shafer, et al, "Introduction to Persistent Uniform 2157 Resource Locators", 1996, 2158 http://purl.oclc.org/OCLC/PURL/INET96 2160 [RFC822] D. Crocker, "Standard for the format of ARPA Internet text 2161 messages", RFC 822, August 1982. 2163 [TELNET] J. Postel, J.K. Reynolds, "Telnet Protocol Specification", 2164 RFC 854, May 1983. 2166 [TEMPER] J. Kunze, "Temporal Enumerated Ranges", work in progress, 2167 http://www.cdlib.org/inside/diglib/ark/temperspec.pdf 2169 [THUMP] K. Gamiel, J. Kunze, "The HTTP URL Mapping Protocol", work 2170 in progress, http://www.ietf.org/internet-drafts/draft- 2171 kunze-thump-00.txt 2173 [URI] T. Berners-Lee, et al, "Uniform Resource Identifiers 2174 (URI): Generic Syntax", RFC 2396, August 1998. 2176 [URNBIB] C. Lynch, et al, "Using Existing Bibliographic Identifiers 2177 as Uniform Resource Names", RFC 2288, February 1998. 2179 [URNSYN] R. Moats, "URN Syntax", RFC 2141, May 1997. 2181 [URNNID] L. Daigle, et al, "URN Namespace Definition Mechanisms", 2182 RFC 2611, June 1999. 2184 12. Appendix: ARK Implementations 2186 Currently, the primary implementation activity is at the California 2187 Digital Library (CDL), 2189 http://ark.cdlib.org/ 2191 housed at the University of California Office of the President, where 2192 over 200,000 ARKs have been assigned to objects that the CDL owns or 2193 controls. Some experimentation in ARKs is taking place at JSTOR, the 2194 Digital Curation Centre, WIPO and at the University of California's 2195 San Diego, San Francisco, and Berkeley campuses. 2197 The US National Library of Medicine (NLM) also has an experimental, 2198 prototype ARK service under development. It is being made available 2199 for purposes of demonstrating various aspects of the ARK system, but 2200 is subject to temporary or permanent withdrawal (without notice) 2201 depending upon the circumstances of the small research group 2202 responsible for making it available. It is described at: 2204 http://ark.nlm.nih.gov/ 2206 Comments and feedback may be addressed to rodgers@nlm.nih.gov. 2208 13. Appendix: Current ARK Name Authority Table 2210 This appendix contains a copy of the Name Authority Table (a file) at 2211 the time of writing. It may be loaded into a local filesystem (e.g., 2212 /etc/natab) for use in mapping NAAs (Name Assigning Authorities) to 2213 NMAHs (Name Mapping Authority Hostports). It contains Perl code that 2214 can be copied into a standalone script that processes the table (as a 2215 file). Because this is still a proposed file, none of the values in 2216 it are real. 2218 # 2219 # Name Assigning Authority / Name Mapping Authority Lookup Table 2220 # Last change: 2006.08.22 2221 # Reload from: http://ark.nlm.nih.gov/etc/natab 2222 # Mirrored at: http://www.cdlib.org/inside/diglib/ark/natab 2223 # To register: mailto:ark@cdlib.org?Subject=naareg 2224 # Process with: Perl script at end of this file (optional) 2225 # 2226 # Each NAA appears at the beginning of a line with the NAA Number 2227 # first, a colon, and an ARK or URL to a statement of naming policy 2228 # (see http://ark.cdlib.org for an example). 2229 # All the NMA hostports that service an NAA are listed, one per 2230 # line, indented, after the corresponding NAA line. 2231 # 2232 # National Library of Medicine 2233 12025: http://www.nlm.nih.gov/xxx/naapolicy.html 2234 ark.nlm.nih.gov USNLM 2235 foobar.zaf.org UCSF 2236 # 2237 # Library of Congress 2238 12026: http://www.loc.gov/xxx/naapolicy.html 2239 foobar.zaf.org USLC 2240 # 2241 # National Agriculture Library 2242 12027: http://www.nal.gov/xxx/naapolicy.html 2243 foobar.zaf.gov:80 USNAL 2244 # 2245 # California Digital Library 2246 13030: http://www.cdlib.org/inside/diglib/ark/ 2247 ark.cdlib.org CDL 2248 # 2249 # World Intellectual Property Organization 2250 13038: http://www.wipo.int/xxx/naapolicy.html 2251 www.wipo.int WIPO 2252 # 2253 # University of California San Diego 2254 20775: http://library.ucsd.edu/xxx/naapolicy.html 2255 ucsd.edu UCSD 2256 # 2257 # University of California San Francisco 2258 29114: http://library.ucsf.edu/xxx/naapolicy.html 2259 ucsf.edu UCSF 2260 # 2261 # University of California Berkeley 2262 28722: http://library.berkeley.edu/xxx/naapolicy.html 2263 berkeley.edu UCB 2264 # 2265 # University of California Los Angeles 2266 21198: http://library.ucla.edu/xxx/naapolicy.html 2267 ucla.edu UCLA 2268 # 2269 # Rutgers University 2270 15230: http://rci.rutgers.edu/xxx/naapolicy.html 2271 rutgers.edu RU 2272 # 2273 # Internet Archive 2274 13960: http://www.archive.org/xxx/naapolicy.html 2275 archive.org IA 2276 # 2277 # Digital Curation Centre 2278 64269: http://www.dcc.ac.uk/xxx/naapolicy.html 2279 dcc.ac.uk DCC 2280 # 2281 # New York University 2282 62624: http://library.nyu.edu/xxx/naapolicy.html 2283 nyu.edu NYU 2284 # 2285 # University of North Texas 2286 67531: http://www.library.unt.edu/xxx/naapolicy.html 2287 unt.edu UNT 2288 # 2289 # Ithaka Electronic-Archiving Initiative 2290 27927: http://www.ithaka.org/xxx/naapolicy.html 2291 ithaka.org ITHAKA 2292 # 2293 # Bibliothque nationale de France / National Library of France 2294 12148: http://www.bnf.fr/xxx/naapolicy.html 2295 bnf.fr BNF 2296 # 2297 # Princeton University 2298 88435: http://diglib.princeton.edu/xxx/naapolicy.html 2299 princeton.edu PU 2300 # 2301 # University of Washington 2302 78428: http://u.washington.edu/xxx/naapolicy.html 2303 u.washington.edu UW 2304 # 2305 # Archives of Region of Vstra Gtaland and City of Gothenburg, Sweden 2306 89901: http://www.arkivnamnden.org/xxx/naapolicy.html 2307 arkivnamnden.org AVGG 2308 # 2309 # Northwest Digital Archives 2310 80444: http://nwda.wsulibs.wsu.edu/xxx/naapolicy.html 2311 nwda.wsulibs.wsu.edu NWDA 2312 # 2313 # Emory University 2314 25593: http://id.library.emory.edu/xxx/naapolicy.html 2315 id.library.emory.edu EMORY 2316 # 2317 # University of Kansas 2318 25031: http://www.lib.ku.edu/xxx/naapolicy.html 2319 www.lib.ku.edu UKANSAS 2321 # 2322 # Google 2323 78319: http://www.google.com/xxx/naapolicy.html 2324 www.google.com GOOGLE 2325 # 2326 # UK Centre for Ecology and Hydrology 2327 17101: http://www.ceh.ac.uk/xxx/naapolicy.html 2328 www.ceh.ac.uk CEH 2329 # 2330 #12345: reserved for examples 2331 # 2332 #--- end of data --- 2333 # The following Perl script takes an NAA as argument and outputs 2334 # the NMAs in this file listed under any matching NAA. 2335 # 2336 # my $naa = shift; 2337 # while (<>) { 2338 # next if (! /^$naa:/); 2339 # while (<>) { 2340 # last if (! /^[#\s]./); 2341 # print "$1\n" if (/^\s+(\S+)/); 2342 # } 2343 # } 2344 # 2345 # Create a g/t/nroff-safe version of this table with the UNIX command, 2346 # 2347 # expand natab | sed 's/\\/\\\e/g' > natab.roff 2348 # 2349 # end of file 2351 14. Copyright Notice 2353 Copyright (C) The IETF Trust (2007). This document is subject to the 2354 rights, licenses and restrictions contained in BCP 78, and except as 2355 set forth therein, the authors retain all their rights. 2357 This document and the information contained herein are provided on an 2358 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 2359 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 2360 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 2361 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 2362 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 2363 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 2365 Expires 23 August 2007 2366 Table of Contents 2368 Status of this Document . . . . . . . . . . . . . . . . . . . . . . 1 2369 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2370 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2371 1.1. Reasons to Use ARKs . . . . . . . . . . . . . . . . . . . . . 4 2372 1.2. Three Requirements of ARKs . . . . . . . . . . . . . . . . . . 4 2373 1.3. Organizing Support for ARKs: Our Stuff vs. Their Stuff . . . 5 2374 1.4. Definition of Identifier . . . . . . . . . . . . . . . . . . . 7 2375 2. ARK Anatomy . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2376 2.1. The Name Mapping Authority Hostport (NMAH) . . . . . . . . . . 8 2377 2.2. The ARK Label Part - ark: . . . . . . . . . . . . . . . . . . 9 2378 2.3. The Name Assigning Authority Number (NAAN) . . . . . . . . . . 10 2379 2.4. The Name Part . . . . . . . . . . . . . . . . . . . . . . . . 10 2380 2.5. The Qualifier Part . . . . . . . . . . . . . . . . . . . . . . 11 2381 2.5.1. ARKs that Reveal Object Hierarchy . . . . . . . . . . . . . 12 2382 2.5.2. ARKs that Reveal Object Variants . . . . . . . . . . . . . . 13 2383 2.6. Character Repertoires . . . . . . . . . . . . . . . . . . . . 14 2384 2.7. Normalization and Lexical Equivalence . . . . . . . . . . . . 15 2385 3. Naming Considerations . . . . . . . . . . . . . . . . . . . . . 16 2386 3.1. ARKS Embedded in Language . . . . . . . . . . . . . . . . . . 16 2387 3.2. Objects Should Wear Their Identifiers . . . . . . . . . . . . 17 2388 3.3. Names are Political, not Technological . . . . . . . . . . . . 17 2389 3.4. Choosing a Hostname or NMA . . . . . . . . . . . . . . . . . . 17 2390 3.5. Assigners of ARKs . . . . . . . . . . . . . . . . . . . . . . 19 2391 3.6. NAAN Namespace Management . . . . . . . . . . . . . . . . . . 20 2392 3.7. Sub-Object Naming . . . . . . . . . . . . . . . . . . . . . . 21 2393 4. Finding a Name Mapping Authority . . . . . . . . . . . . . . . . 21 2394 4.1. Looking Up NMAHs in a Globally Accessible File . . . . . . . . 22 2395 4.2. Looking up NMAHs Distributed via DNS . . . . . . . . . . . . . 23 2396 5. Generic ARK Service Definition . . . . . . . . . . . . . . . . . 26 2397 5.1. Generic ARK Access Service (access, location) . . . . . . . . 26 2398 5.2. Generic Policy Service (permanence, naming, etc.) . . . . . . 26 2399 5.3. Generic Description Service . . . . . . . . . . . . . . . . . 28 2400 6. Overview of The HTTP URL Mapping Protocol (THUMP) . . . . . . . 28 2401 7. Overview of Electronic Resource Citations (ERCs) . . . . . . . . 31 2402 7.1. ERC Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . 33 2403 7.2. ERC Stories . . . . . . . . . . . . . . . . . . . . . . . . . 34 2404 7.3. The ERC Anchoring Story . . . . . . . . . . . . . . . . . . . 35 2405 7.4. ERC Elements . . . . . . . . . . . . . . . . . . . . . . . . . 36 2406 7.5. ERC Element Values . . . . . . . . . . . . . . . . . . . . . . 38 2407 7.6. ERC Element Encoding and Dates . . . . . . . . . . . . . . . . 40 2408 7.7. ERC Stub Records and Internal Support . . . . . . . . . . . . 41 2409 8. Advice to Web Clients . . . . . . . . . . . . . . . . . . . . . 42 2410 9. Security Considerations . . . . . . . . . . . . . . . . . . . . 43 2411 10. Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 43 2412 11. References . . . . . . . . . . . . . . . . . . . . . . . . . . 44 2413 12. Appendix: ARK Implementations . . . . . . . . . . . . . . . . 45 2414 13. Appendix: Current ARK Name Authority Table . . . . . . . . . . 46 2415 14. Copyright Notice . . . . . . . . . . . . . . . . . . . . . . . 49