AU2010101079A4

AU2010101079A4 - Domain Name Identifier and Directory

Info

Publication number: AU2010101079A4
Application number: AU2010101079A
Authority: AU
Inventors: Peter Bernard West
Original assignee: Johnson Philip Ian; West Peter Bernard Mr
Current assignee: Johnson Philip Ian; West Peter Bernard Mr
Priority date: 2009-10-02
Filing date: 2010-10-02
Publication date: 2010-11-11
Anticipated expiration: 2018-10-02

Description

DOMAIN NAME IDENTIFIER & DIRECTORY DESCRIPTION FIELD OF THE INVENTION This invention relates to electronic information systems utilising domain-name 5 technology for identification. In particular, the technological field of this invention is that of groups of electronic information systems serving a common national or global community such that each member of the community must be uniquely denoted by one or more commonly recognised identifiers to facilitate interaction with diverse national or global systems, using the Internet. 10 Such a larger community is itself comprised of a hierarchy of smaller communities which may reflect geographical, professional, functional or other constraints, such that the creation of a commonly recognised identifier may be governed by different policies and practices in different sub-communities. Each identifier must reflect a relationship of a member to a particular sub-community to 15 which the member belongs and reflect the relationship of the sub-community to the larger community. Identification across the diverse systems is based on the ubiquitous naming system of the internet, the Domain Name System (DNS). BACKGROUND OF THE INVENTION 20 Nameable entities often belong to a variety of communities. Australian citizens, for example, may study at a tertiary institution, work in a small or large company, belong to many clubs and groups, subscribe to many publications, be an office holder in community organisations, etc. Using a single identifier, like an electoral identifier, in all of these places would be absurd. 25 There are however some groupings which encompass a broad range of people, skills, qualifications, places and things. An example of such a grouping is the health sector in Australia. This sector encompasses the entire population of the country, because every person will interact with the health sector at some time. It encompasses all of the professional health workers, and all of the business and organisations that are connected 30 in any way with health care. The data that is associated with an identified entity will vary. Persons and organisations will be associated with different operational systems; similar variation will occur between different types of organisation, between health workers and consumers, and between different categories of health worker. The sensitivity of information associated 35 with each of these different categories of entity will also vary. In some circumstances, data related to an identifier will necessarily be publicly accessible, for example a professional's accreditation; in other cases it will necessarily be private, for example a patient's health records. Because it is far safer to add accessibility to an inherently private system, than to add privacy to an inherently public system, such cross-sector identifiers should by default preserve privacy to the greatest possible extent. In spite of these great differences in requirements and systems, there is a strong case for using a common identifier across the entire health sector, not least because it helps to guard against mistaken identity, and therefore mistaken therapy or mistaken 5 accreditation. A common identifier with cross-sector semantics enables the flow of electronic data between consumers, clinicians and institutions with a high level of confidence. In sensitive cross-sector applications like health, identities require accreditation at many levels. When a professional identifier is created and used, it is critical that the validity of 10 the identifier be guaranteed. It would be dangerous to create professional identifiers for professionals who do not exist. Similarly, it is vital that, when a patient identifier is passed from a general practitioner to a specialist or a hospital, the receiver can be confident that the identifier does denote a real individual who has been referred on from the general practitioner. In this and other senses, all identifiers must carry a level of 15 accreditation. Some prior art takes advantage of the Domain Name System (DNS) as a database to store various items of information. The Internet is a global communication platform that universally employs the DNS as its addressing technology. DNS is a database containing information keyed on a DNS 20 domain name. A domain name is a hierarchical set of simple names, known as labels, which are informative for computers and for human users, with the added advantage that they are mnemonic for human users. This characteristic greatly assists the accurate person-to-person communication of domain names. While DNS databases can contain various types of records, there are two record types which are dominantly employed. 25 The most important of these is the A (Address) record. It provides the multi-part numeric IP address associated with a given domain name. They are used only for devices which are attached to the Internet. The second most important record is the MX (Mail Exchange) record. It associates a mail server with the domain name. A hierarchical DNS name reads left to right from the most particular to the most general 30 sub-domain name. In the domain name A.B. C, for example, the most specific component is A, and the least specific is C. The DNS name A.B. C is a sub-domain of the domain B.C. Domain B.C is the parent domain of A.B.C. It is generally the case that, for each label from left to right, that label has been assigned by the authority controlling the next label to the right. No sub-domain name can be assigned in the absence of its parent 35 domain. For example, the name A.B. C cannot be created until the name B. C has been created. US Patent Applications 20080133471, 20080288774 and 20080320015 have some features relating to identifiers, and make extensive use of the DNS database. Patent application 20080133471 concerns a communication system which simplifies the 40 association of an entity with one or more data sets associated with that entity. The system comprises three databases, one of which is modelled on DNS. One database contains identifying information about the entity, which is used to find a domain name, on the basis of which associated information is obtained from the DNS database. This associated information is directed towards a third database, from which the final results 45 of a query against the first database (which is not a DNS database) are returned. The patent describes the DNS elements in terms of the top-level domain tel and notes at para 0032 that the top-level domain could, in fact, be broken into sub-domains, for example au.tel and us.tel, to provide for the repetition of names in each sub-domain; for example johnsmith.us.tel and johnsmith.au.tel. Also at para 0032, patent 20080133471 5 notes, "It will be readily appreciated that although the foregoing discussion is limited to the use of content or contact datasets being associated with a user identifier, other types or categories of data can be used without changing the nature of the invention." Patent 20080133471 describes a generalised searchable database set. It assumes that the set of domain names in the DNS database are accessible through a search of the first 10 database. It further assumes that the useful information in the DNS database is provided through NAPTR mapping records. Patent application 20080288774 describes an application of DNS to communities of "friends". This patent builds on the ideas of 20080133471, though the "first database" of that patent is not included in 20080288774. That is, knowledge of a user's domain name 15 is assumed, as it is in the current invention. However, a "sponsoring organisation", which is not itself a part of the DNS structure described in the patent, and which has knowledge of all registrations within the DNS database, is essential to establishing the relationships between "friends". In 20080288774, some of the information associated with an entity through the DNS 20 domain name assigned to that entity is encrypted. For each other system user who is authorised to access the encrypted information, the authorising user is assigned a new sub-domain. Information about the authorising user is encrypted in a way the authorised user can decipher, and is associated with the authorised user's subdomain. For example, for user A, with domain A.tel, when user B, with domain B.tel, is authorised to access 25 A's data, a new subdomain is created under A.tel, for instance, Adata-forB.A.tel. When B requests A's encrypted data, the data associated with Adata-forB.A.tel is sent to B. When user C is authorised to access A's data, and new subdomain Adata-forC.A.tel is created, and so on. Note that in order to define the relationship "B is a friend of A", the domains A.tel, B.tel 30 and Adata-forB.A.tel must be created. Note also that the relationship is not transitive; when B becomes a friend of A, A does not thereby become a friend of B. However, as "friendship" is generally transitive, the relationship is mapped is such a way that the other side of the transitive relationship can also be mapped. For example, the sub domain Bdata-forA.B.tel can be created. The relationships mapped are not hierarchical, 35 as for example a parent-child, or collection-member relationship would be. This design cannot provide a generalised map of hierarchical relationships. The encryption used in this relationship is external to the native operation of DNS. DNS merely carries a package of encrypted data which it treats like any other data. It does not provide a guarantee of the validity of the source or the content of messages from the 40 DNS server. Patent application 20080320015 builds on 20080288774, which it includes by reference (as European Patent Application 07252529.0) in para 0001. 20080320015 describes a search engine, search engine indexer, and structured data contained in DNS TXT resource records. While the patent particularly addresses the contact information storage 45 system described in the earlier patent, it notes that the system may be extended as a more general search indexing mechanism. There is a caveat in the patent at para 0122 concerning this broader application. "When using the invention in such a conventional Internet system, it is necessary to know the domain names before search terms for those domain names may be retrieved. In one embodiment, a DNS registry is queried for domain names which are then used to construct and/or maintain the search index. 5 However, in the general case this is very inefficient as the entity querying the registry has to guess which domains are registered, and so might have to make many queries with random names before detecting a domain name which had been registered." OBJECT OF THE INVENTION An object of the present invention is to seek to provide a novel and innovative solution 10 to combine the local responsibility of naming authorities for the names under their immediate control, with global reach that guarantees the creation of identities which cannot conflict with any other identities in the overall system, and seeks to allow simple and immediate access to important information relating to the identities, thereby seeking to overcome the disadvantages or limitations of the prior art, or to provide the public 15 with a useful choice. STATEMENT OF THE INVENTION Accordingly, in one aspect the invention resides in a domain-name identifier and directory (DIAD) system which provides electronic communication identifiers for a dispersed hierarchical community of communities, where the members of each 20 community are named entities, and where named entities are either simple entities or naming authority entities responsible for naming the members of a particular sub community, and where the population of named entities is potentially global in scale, such that each identifier names a single entity, and only that entity and where naming authority entities operate DNS registries; 25 wherein in operation the assignment of an identifier is a verification by the naming authority of the existence and identity of the named entity and of the relationship between the naming authority and the named entity, the precise nature of which will vary from naming authority to naming authority; and wherein in operation the existence of a particular identifier can be readily 30 determined by any party. Preferably, the identifiers are domain-names. Preferably, the responsibility for creation and maintenance of the said identifier resides in local naming authorities, without reference to any central naming authority. Preferably, the local naming authorities are responsible for verifying the identity of 35 entities to whom they provide an identifier. One possible implementation of this invention may see identifiers created under a model where appropriate fees might be applied to different type or levels of participant. This may take the form of a licence or subscription or other fee model. Preferably, information concerning the entity denoted by a particular identifier is 40 directly associated with the identifier, wherein in operation, such information is readily accessible to any party to whom the identifier is known.

Preferably, information about the location of services relating to the entity denoted by a particular identifier is directly associated with the identifier, wherein in operation, such information is readily accessible to any party to whom the identifier is known. Preferably, the set of identifiers issued by any one naming authority is known only to 5 authorised operators of the naming authority's registry, wherein in operation the complete population of identifiers issued by any one naming authority is known only to the authority's authorised staff, and it is infeasible to conduct an exhaustive search to independently determine that population. In another aspect the invention resides in the DIAD system delivered over a hierarchical 10 group of servers which are connected to the Internet and utilise common Internet devices, for example PCs and mobile phones. In another aspect the invention resides in technology modalities including a hierarchical group of DNS servers and a group of interface systems which maintain databases of information created and delivered by the DIAD system in operation. 15 In an alternative use of the DNS system to provide identifiers for a particular community, the root of the tree of identifiers for said particular community in the DNS system can be specified as the root of an alternative DNS system by modifying the location of the root servers to correspond to the servers for the root of the tree of identifiers of the community, such that all software systems wishing to access the 20 community identifiers must use the alternative DNS system, with the alternative root servers, and such that domain names in the alternative DNS tree will not be accessible from software using the standard DNS root servers. DETAILED DESCRIPTION OF THE INVENTION In a preferred embodiment, the invention is a method of providing electronic identifiers 25 for large populations of nameable entities, be they persons, organisations, places or objects. These electronic identifiers will hereinafter be referred to simply as identifiers. The population to be named has the following characteristics. The population can readily be arranged into a hierarchy of entities. In one aspect, each entity falls into one of two categories; it is either a simple entity or a naming authority 30 entity. Naming authority entities will hereinafter be referred to as naming authorities. Any nameable entity can be a simple entity, but in general only organisations will be naming authorities. Each entity except one, the root naming authority, is assigned its identifier by a responsible naming authority. Each naming authority also has an identifier. There is no 35 difference in kind between the identifiers assigned to naming authorities and to simple entities. Each naming authority may issue identifiers to simple entities, and it may issue identifiers to other naming authorities. There may, consequently, be a multi-level hierarchy of naming authorities and simple entities in the form of a tree, with branches 40 emanating from each naming authority, and with the end points of all branches being simple entities. These end points are known as the leaf nodes of the tree. The identifier of the naming authority that issues any particular identifier is known as the parent of the particular identifier. Correspondingly, the identifiers which are issued by a naming authority are known as the children of the authority's identifier. Note that, strictly speaking, the terms parent and children refer in this discussion to identifiers, not entities as such. However, the entities may loosely be referred to as parents and children, while keeping the strict meaning in mind. 5 Only one naming authority is responsible for any identifier. The root naming authority issues an identifier to itself. Note that while a naming authority is responsible for all of the identifiers it issues, it is not responsible for identifiers issued by any naming authorities it has identified. Such children are in turn responsible for identifiers they issue. 10 As a consequence of this structure, each entity can be uniquely named by combining the identifier of the issuing naming authority with a name component which is unique amongst the children of the issuing authority's identifier. That is, the problem of unique naming is, by this naming structure, reduced to the problem of uniquely naming only the direct children of each naming authority. Naming, and associated responsibility, is 15 thereby localised to each naming authority. While the naming responsibility is localised, in order to be functional as a population wide identifier, the identifier must be able to be checked by anyone with an interest in the population. This invention addresses these concerns by a novel application of an existing 20 technology - the Domain Name System (DNS). The DNS is a distributed database that associates small sets of data with a key. The key is a hierarchically-organised name which is readily comprehensible and communicable by both human beings and computers. DNS names are ubiquitous on the Internet and on the World Wide Web. They form part of all modern email addresses, and part of every textual Universal 25 Resource Locator (URL) used on the Web. Without DNS, the Web could not exist, and email could not function. DNS is consulted across the world many millions of times every day. The key to the success of DNS is its hierarchical and distributed nature. There is no central repository of DNS names. Nor could there be. No centralised system could keep 30 track of the constantly evolving space of DNS names. Technical and commercial experience of DNS systems is correspondingly widely distributed. DNS was designed for a specific purpose: to associate human-comprehensible and human-communicable names with the numerical Internet addresses that underlie all Internet communication. These associations are described by A (address) and AAAA 35 (IPv6 address) records within the DNS database. Many other types of records have been defined, and continue to be defined for DNS. To discover the Internet address associated with a given domain name, the DNS database is queried using the name as a key, and, if the name is present in the database, the address is returned. The novelty of this invention is that the database for an entire sub-tree, or zone, of the 40 DNS name space associates no Internet addresses with its domain names. A query using the domain name as a key will return a possibly empty set of directory data, not including A or AAAA records. The most important information returned from the query of a name within the zone is the fact of its existence or non-existence. The domain names are just names; identifiers associated with human and corporate agents or other 45 nameable entities. In a name of the form A.B.C, the structure of the identifier implicitly asserts that the entity named A.B. C has been named by the entity named B. C. This implication provides a foundation for the hierarchical assertion of identity, because only the controller of the domain B.C can issue subdomains such as A.B.C. The directory associated with each name is optional. If used, it is defined primarily in 5 TXT (text) and NAPTR (naming authority pointer) resource records in the DNS database. Other resource records, for example SR V (server) and LOC (location) records may also be used. All information in the directory is public. However, such information may be represented directly and publicly, as in the form of TXT records, or may be represented indirectly, as in the form of NAPTR records, which re-direct enquiries to 10 systems entirely external to DNS, where appropriate levels of access security may be enforced. That is, whilst the re-direction details are public, the information available from the re-directed query may be public or private. The hierarchy of names would be based on functional, legal and practical categories determined by the nature of the application domain. Individual entities (be they persons, 15 organisations, places, objects or any other nameable entity) would obtain or be assigned identifiers at their point or points of contact with the application domain. The characteristics of individual names would, again, be determined by the requirements of individual components of the application domain. For example, in the e-health system, individual patients could have "opaque" identifiers congruent with their need for 20 privacy. On the other hand, registered professionals may be required to use an identifier which could be readily confirmed as belonging to a particular person. DNS offers security at a number of levels. All of the names and associated information for a zone is regularly transferred between the master and slave nameservers of each zone. It is common practice to validate requests for this transfer by signing such 25 requests, and to validate the transferred zone data by signing the return message. The signing is done with a key shared by the master and slaves. This process ensures that the complete zone data cannot be readily obtained by parties other than zone administrators. As a result, whilst the zone can be queried for individual names, the complete zone contents could only be discovered by exhaustive queries for all possible names within 30 the zone. In standard DNS, guaranteeing the validity of all of the data returned in response to queries is more difficult. It can be achieved by applying DNSSEC.bis (DNSSEC hereafter) security to the zones. Whether DNSSEC is used would depend on the security requirements of each application domain. However, in the case of the e-health 35 application domain, it would be essential. When used, DNSSEC would have to be applied to all zones within the application domain sub-tree. Examples N.B. All of the following examples are notional. The particulars of domain names, and the syntax and semantics of directory information are beyond the scope of the invention. 40 They would be determined as part of the specification of the application domain. Example: GP Clinic and Patients Identifiers are not restricted to persons; organisations may also possess them. For example, a medical general practice within the Sunshine Coast Division of General Practice, may have the identifier: dh-clinic.scdgp.gpq.healthid.au A patient called James M. Brown, attending a general practice clinic might have the identifier: thegecko.dh-clinic.scdgp.gpq.healthid.au 5 The patient chooses a handle, i.e. pseudonym, as the individual component of the identifier. Consequently, the identity of the patient is not externally evident,. Such pseudonyms are now familiar from social networking sites on the Web. At the clinic, the usual medical records will be maintained, with the addition of this identifier. The patients will be known to the staff as usual, and they will have access, as required, to the 10 patient's identifier. However, no casual observer would be able to make that association. The identifying entity for the identifier is dh-clinic.scdgp.gpq.healthid.au. Example: Professional Study and Registration Professional health care workers will generally have different levels of authorisation at different stages of their careers. For example, student, intern, registrar, specialist. If an 15 identifier were to be associated with each stage, the identifying entity would be different in each case. Assume health-care professionals are registered by a national body, known as National Registration and Accreditation Scheme (NRAS), and that accrediting educational institutions and training hospitals are affiliated with the NRAS. 20 If a student named John Stephen Smith were to enrol in medicine at Queensland University, for example, he might be assigned an identifier of the form: smith-john-s- 19731106-01 .uq.nras.healthid.au On graduation, John Smith obtains an internship at Princess Alexandra Hospital, a recognised tertiary teaching hospital. He might be assigned a new identifier of the form: 25 smith-john-s- 19731106-01 .pah.nras.healthid.au Once his internship has been successfully completed, John Smith is eligible for registration with NRAS, when he might receive the identifier: smith-js- 19731106-0 l.mbbs.nras.healthid.au The person is identified as smith-js-19731106-01, whose identifier was provided by the 30 entity mbbs.nras within the Australian health-care application domain. In the case of professional identifiers, it would probably be considered advantageous to have an identifier which is readily associated with the person identified, as in this example. Example: Directory Linking Multiple Identifiers The professional registration example shows that multiple identifiers may both be 35 required, and required to be linked. It would be necessary to be able to trace the path to accreditation of professionals within the system. Linking could be achieved in a number of ways through DNS records. One possible method is to use NAPTR records to establish a double-linked list. The Application Unique String for this application would the identifier for which links were required. 40 The First Well Known Rule would use the identifier unchanged to request NAPTR records. The expected service in the NAPTR records would be alias with either next or prev as secondary service protocols. The possible values of the service field would then be alias+ next or alias+prev An application program can then derive the complete set of links starting from any point in the chain. If there were no aliases, no appropriate NAPTR records would be returned. If there were 5 one alias only, both the next and prev entries would transform to the same alias. In other cases, the chain could be followed in either direction. The same process allows the optional linking of multiple identifiers that individual patients may choose to have. Example: Location capability for service or care facilities location finding 10 Where a specific location is required, for example the exact location of a hospital, DNS provides a Location (LOC) Record function (known as LOC RR) that might be assigned or attributed to the e-identifier of the service. The LOC RR permits the addition of latitude, longitude and altitude to e-identifier - should this be required. Example: Directory Publicly Available Health Information for Patient 15 If a patient chose to have some demographic or medical information associated with his or her identifier, the TXT records associated with the identifier might contain the following strings: birthyear= 1960 eyes=blue 20 height=170cm blood=Oneg allergies=penicillin,bee-sting medication=Warfarin:2mg:updated-2010-11-05:exp-2011-02-05 Such information is public; it would be available to anyone with Internet access. 25 However, only someone knowing the association between the individual and the identifier would be able to associate the person with the data. Example: Directory Private Health Information for Patient If a patient chose to participate in a service supplying certain information from his or her medical records as held by the clinic, in an Internet-accessible service whose 30 interface were well-known, a combination the patient-specific label (the first label of the domain-name) and the clinic identifier (the remainder of the domain-name) could be used in conjunction with NAPTR and SRV records associated with the clinic domain name to access the service with a query about the patient. Note that the NAPTR and SRV records are publicly readable, but the service to which they point can be public or 35 private. If private, this access would be subject to the authentication and access policies governing such sensitive data. Example: Directory Information for Professionals For a professional identifier, some publicly accessible information may be associated with the identifier by means of TXT records. It might, for example, be considered that a 40 professional registration number should be published with the identifier. In addition, because the professional may also use his or her professional identifier as a patient identifier, the same basic personal medical facts may also be recorded. Professional registration bodies may define interfaces to web services with confidential information concerning the professional's registration and history. Such services might be accessed through NAPTR and SRV records on the registration body's identifier, using that identifier and the professional's identifier to locate and query the service. As with patient data, this access would be subject to the authentication and access policies 5 governing such sensitive data.

Claims

1. A domain-name identifier and directory (DIAD) system which provides electronic communication identifiers for a dispersed hierarchical community of 5 communities, where the members of each community are named entities, and where named entities are either simple entities or naming authority entities responsible for naming the members of a particular sub-community, and where the population of named entities is potentially global in scale, such that each identifier names a single entity, and only that entity and where naming authority 10 entities operate DNS registries; wherein in operation the assignment of an identifier is a verification by the naming authority of the existence and identity of the named entity and of the relationship between the naming authority and the named entity, the precise nature of which will vary from naming authority to naming authority; and wherein in operation the existence of a particular 15 identifier can be readily determined by any party.

2. The system of claim 1 wherein a small set of data is associated with the identifier by means of DNS TXT records, such that the data can be accessed in any location by any person having a system for internet access and a knowledge of the identifier. 20

3. The system of claims I wherein an individual identifier can be linked to a number of other identifiers by means of DNS NAPTR records associated with the identifier, indicating that the linked records all denote the same entity.

4. The system of claim 1 wherein further information systems can be accessed through an identifier by means of DNS NAPTR records associated with the 25 identifier.