11. Landing page encoding

Landing page web resources can be written in any format (e.g. HTML, XML). Although not obligatory, ideally resources should be encoded in formats that not only improve syntactic interpretation of information but semantic understanding of the information. In other words, machines can not only read but understand the meaning of the information presented in web resources, enhancing interoperability and integration between systems. Below are some examples of landing page encodings.

11.1. Examples

11.1.1. JSON-LD

There is a strong relation between PIDs with values of types that are defined in a data type registry (DTR) as for instance in the NERC example in Table 5.1 and linked data. First of all a PID with a type value is a triple where the PID plays the role of the subject, the type definition is the predicate and the value is the object. Secondly the type definition itself can refer to sub types also defined in a DTR. If this construction of types out of other types is done in a consistent and machine actionable way, as it is done for instance in the ePC DTR, these subtypes may be referred by human readable names. The names are disambiguated by the type definition, because each subtype used in a type is identified by a PID. Such PIDs with types defined upon sub types span a graph of metadata around the PID. PIDs with types are in other words a specific representation of linked data.

It is therefore obvious to ask for other, more a common linked data representation like RDF or JSON-LD of such PIDs with types. Such a conversion can be done by a simple backtracking algorithm that crawls from the PID through all its type and subtypes definitions to identify the used names by the type PIDs and to collect this information for the LD representation. This way the whole graph is explored and this graph can be mapped into a linked data representation. In the following a respective representation in JSON-LD of the schema example shown in Table 5.1 is shown in Snippet 11.1.

Snippet 11.1 representation in JSON-LD of the NERC example of Table 5.1.
    {
    "@context" : {
     "ARK-Identifier" : "dti:21.T11148/7af6f46512fb4c190d01",
     "AlternateIdentifier" : "dti:21.T11148/d87a75c52c68b06e9a18",
     "AlternateIdentifierValue" : "dti:21.T11148/38330bcc6a40ca85e5b4",
     "AlternateIdentifiers" : "dti:21.T11148/eb3c713572f681e6c4c3",
     "Bibcode-Identifier" : "dti:21.T11148/6c2fc7682e48ac7160b5",
     "DOI-Identifier-General" : "dti:21.T11148/d93427e3c56173e9dc08",
     "Date" : "dti:21.T11148/eb9a4bc1c0c153e4e4b0",
     "Dates" : "dti:21.T11148/22c62082a4d2d9ae2602",
     "Description" : "dti:21.T11148/55f8ebc805e65b5b71dd",
     "Handle-Identifier-ASCII" : "dti:21.T11148/3626040cadcac1571685",
     "ISAN-Identifier" : "dti:21.T11148/48cfce4482166a103c50",
     "ISBN-Identifier" : "dti:21.T11148/2ff8ad6cdd4e46622944",
     "ISNI-Identifier" : "dti:21.T11148/cff32964e132c14fc56f",
     "ISRC-Identifier" : "dti:21.T11148/2719170925ff2bfb5157",
     "ISSN-Identifier" : "dti:21.T11148/7e689432354610f388c0",
     "ISTC-Identifier" : "dti:21.T11148/1f0df9ef66774b2e2aa1",
     "ISWC-Identifier" : "dti:21.T11148/698fba7e1c659fcfdcdd",
     "InstrumentType" : "dti:21.T11148/f76ad9d0324302fc47dd",
     "LandingPage" : "dti:21.T11148/9a15a4735d4bda329d80",
     "Manufacturer" : "dti:21.T11148/7adfcd13b3b01de0d875",
     "Manufacturers" : "dti:21.T11148/1f3e82ddf0697a497432",
     "MeasuredVariable" : "dti:21.T11148/1fcb0dad9aced457d67e",
     "MeasuredVariables" : "dti:21.T11148/72928b84e060d491ee41",
     "Name" : "dti:21.T11148/709a23220f2c3d64d1e1",
     "Owner" : "dti:21.T11148/89ff31225c5f042fff61",
     "Owners" : "dti:21.T11148/4eaec4bc0f1df68ab2a7",
     "PMCID-Identifier" : "dti:21.T11148/e94bec7d7f1c63dd00cd",
     "PMID-Identifier" : "dti:21.T11148/234c084bac48480bfe5d",
     "RelatedIdentifier" : "dti:21.T11148/ec9f00af0761a065dbd0",
     "RelatedIdentifierType" : "dti:21.T11148/015dc79a77940fb65aa4",
     "RelatedIdentifierValue" : "dti:21.T11148/38330bcc6a40ca85e5b4",
     "RelatedIdentifiers" : "dti:21.T11148/178fb558abc755ca7046",
     "URN-Identifier" : "dti:21.T11148/d22b6854df3503df7831",
     "VariableMeasured" : "dti:21.T11148/f1627ce85386d8d75078",
     "alternateIdentifierType" : "dti:21.T11148/015dc79a77940fb65aa4",
     "arXiv-Identifier" : "dti:21.T11148/d66f8368c3d305941a2e",
     "date" : "dti:21.T11148/be707495360a234ef049",
     "dateType" : "dti:21.T11148/2f0e608b621a5a97e0d9",
     "dti" : "http://hdl.handle.net/",
     "identifier-general-with-type" : "dti:21.T11148/8eb858ee0b12e8e463a5",
     "identifierType" : "dti:21.T11148/015dc79a77940fb65aa4",
     "identifierValue" : "dti:21.T11148/38330bcc6a40ca85e5b4",
     "manufacturerIdentifier" : "dti:21.T11148/5b240e16ea32ea25cf65",
     "manufacturerIdentifierType" : "dti:21.T11148/015dc79a77940fb65aa4",
     "manufacturerIdentifierValue" : "dti:21.T11148/38330bcc6a40ca85e5b4",
     "manufacturerName" : "dti:21.T11148/798588c5a1ec532f737b",
     "modelName" : "dti:21.T11148/798588c5a1ec532f737b",
     "other" : "dti:21.T11148/f40cb15558a7c1546c91",
     "ownerContact" : "dti:21.T11148/a88b7dcd1a9e3e17770b",
     "ownerIdentifier" : "dti:21.T11148/1e3c17ac2a3e7ebf466a",
     "ownerIdentifierType" : "dti:21.T11148/015dc79a77940fb65aa4",
     "ownerIdentifierValue" : "dti:21.T11148/38330bcc6a40ca85e5b4",
     "ownerName" : "dti:21.T11148/798588c5a1ec532f737b",
     "relationType" : "dti:21.T11148/292a53bd9ee27a242082"
    },
    "@id" : "dti:21.T11998/0000-001A-3905-F",
    "AlternateIdentifiers" : [
     {
     "AlternateIdentifier" : {
        "AlternateIdentifierValue" : "2490",
        "alternateIdentifierType" : "serialNumber"
     }
     }
    ],
    "Dates" : [
     {
     "date" : {
        "date" : "1999-11-01",
        "dateType" : "Commissioned"
     }
     }
    ],
    "Description" : "A high accuracy conductivity and temperature recorder with an optional pressure sensor designed for deployment on moorings. The IM model has an inductive modem for real-time data transmission plus internal flash memory data storage.",
    "InstrumentType" : "http://vocab.nerc.ac.uk/collection/L22/current/TOOL0022/",
    "LandingPage" : "https://linkedsystems.uk/system/instance/TOOL0022_2490/current/",
    "Manufacturers" : [
     {
     "Manufacturer" : {
        "manufacturerIdentifier" : {
           "manufacturerIdentifierType" : "URL",
           "manufacturerIdentifierValue" : "http://vocab.nerc.ac.uk/collection/L35/current/MAN0013/"
        },
        "manufacturerName" : "Sea-Bird Scientific",
        "modelName" : "SBE 37-IM"
     }
     }
    ],
    "MeasuredVariables" : [
     {
     "MeasuredVariable" : {
        "VariableMeasured" : "http://vocab.nerc.ac.uk/collection/P01/current/CNDCPR01/"
     }
     },
     {
     "MeasuredVariable" : {
        "VariableMeasured" : "http://vocab.nerc.ac.uk/collection/P01/current/PSALPR01/"
     }
     },
     {
     "MeasuredVariable" : {
        "VariableMeasured" : "http://vocab.nerc.ac.uk/collection/P01/current/TEMPPR01/"
     }
     },
     {
     "MeasuredVariable" : {
        "VariableMeasured" : "http://vocab.nerc.ac.uk/collection/P01/current/PREXMCAT/"
     }
     }
    ],
    "Name" : "Sea-Bird SBE 37-IM MicroCAT C-T Sensor",
    "Owners" : [
     {
     "Owner" : {
        "ownerContact" : "louise.darroch@bodc.ac.uk",
        "ownerIdentifier" : {
           "ownerIdentifierType" : "URL",
           "ownerIdentifierValue" : "http://vocab.nerc.ac.uk/collection/B75/current/ORG00009/"
        },
        "ownerName" : "National Oceanography Centre"
     }
     }
    ],
    "RelatedIdentifiers" : [
     {
     "RelatedIdentifier" : {
        "RelatedIdentifierType" : "URL",
        "RelatedIdentifierValue" : "https://www.bodc.ac.uk/data/documents/nodb/pdf/37imbrochurejul08.pdf",
        "relationType" : "IsDescribedBy "
     }
     }
    ],
    "identifier-general-with-type" : {
     "identiferType" : "MeasuringInstrument",
     "identifierValue" : "http://hdl.handle.net/21.T11998/0000-001A-3905-F"
    }
  }

As one can see in this result the context is over complete in the sense that all possible sub types are resolved and referred in @context, but not all of them are actually used by the types occuring in the PID. This could be pruned by an additional step of the algorithm to a version reduced to the necessary and sufficient sub types. Such a pruning is also automatically done by LD converters[1] as one can see in the following snippet with a conversion into Turtle Terse RDF that results into the following serialization (Snippet 11.2), where only the values remain and the names used in the type definitions are replaced by their type PID suffixes:

Snippet 11.2 representation in Turtle Terse RDF of the NERC example of Table 5.1 that was generated by a JSON-LD to RDF converter from the JSON-LD in Snippet 11.1.
  @prefix ns0: <http://hdl.handle.net/21.T11148/> .
  @prefix xsd: <http://www.w3.org/2001/XMLSchema#> .

  <http://hdl.handle.net/21.T11998/0000-001A-3905-F>
    ns0:178fb558abc755ca7046 [ ns0:ec9f00af0761a065dbd0 [
     ns0:015dc79a77940fb65aa4 "URL"^^xsd:string ;
     ns0:292a53bd9ee27a242082 "IsDescribedBy "^^xsd:string ;
     ns0:38330bcc6a40ca85e5b4 "https://www.bodc.ac.uk/data/documents/nodb/pdf/37imbrochurejul08.pdf"^^xsd:string
     ] ] ;
    ns0:1f3e82ddf0697a497432 [ ns0:7adfcd13b3b01de0d875 [
     ns0:5b240e16ea32ea25cf65 [
        ns0:015dc79a77940fb65aa4 "URL"^^xsd:string ;
        ns0:38330bcc6a40ca85e5b4 "http://vocab.nerc.ac.uk/collection/L35/current/MAN0013/"^^xsd:string
     ] ;
     ns0:798588c5a1ec532f737b "Sea-Bird Scientific"^^xsd:string, "SBE 37-IM"^^xsd:string
     ] ] ;
    ns0:22c62082a4d2d9ae2602 [ ns0:be707495360a234ef049 [
     ns0:2f0e608b621a5a97e0d9 "Commissioned"^^xsd:string ;
     ns0:be707495360a234ef049 "1999-11-01"^^xsd:string
     ] ] ;
    ns0:4eaec4bc0f1df68ab2a7 [ ns0:89ff31225c5f042fff61 [
     ns0:1e3c17ac2a3e7ebf466a [
        ns0:015dc79a77940fb65aa4 "URL"^^xsd:string ;
        ns0:38330bcc6a40ca85e5b4 "http://vocab.nerc.ac.uk/collection/B75/current/ORG00009/"^^xsd:string
     ] ;
     ns0:798588c5a1ec532f737b "National Oceanography Centre"^^xsd:string ;
     ns0:a88b7dcd1a9e3e17770b "louise.darroch@bodc.ac.uk"^^xsd:string
     ] ] ;
    ns0:55f8ebc805e65b5b71dd "A high accuracy conductivity and temperature recorder with an optional pressure sensor designed for deployment on moorings. The IM model has an inductive modem for real-time data transmission plus internal flash memory data storage."^^xsd:string ;
    ns0:709a23220f2c3d64d1e1 "Sea-Bird SBE 37-IM MicroCAT C-T Sensor"^^xsd:string ;
    ns0:72928b84e060d491ee41 [ ns0:1fcb0dad9aced457d67e [ ns0:f1627ce85386d8d75078 "http://vocab.nerc.ac.uk/collection/P01/current/CNDCPR01/"^^xsd:string ] ], [ ns0:1fcb0dad9aced457d67e [ ns0:f1627ce85386d8d75078 "http://vocab.nerc.ac.uk/collection/P01/current/PSALPR01/"^^xsd:string ] ], [ ns0:1fcb0dad9aced457d67e [ ns0:f1627ce85386d8d75078 "http://vocab.nerc.ac.uk/collection/P01/current/TEMPPR01/"^^xsd:string ] ], [ ns0:1fcb0dad9aced457d67e [ ns0:f1627ce85386d8d75078 "http://vocab.nerc.ac.uk/collection/P01/current/PREXMCAT/"^^xsd:string ] ] ;
    ns0:8eb858ee0b12e8e463a5 [ ns0:38330bcc6a40ca85e5b4 "http://hdl.handle.net/21.T11998/0000-001A-3905-F"^^xsd:string ] ;

11.1.2. Sensor web enablement (SWE)

Global standards have been developed which enable the web-based discovery, exchange and processing of sensors and their observations. Many developers using standards, such as the Open Geospatial Consortium’s (OGC) Sensor Web Enablement (SWE), publish formal, machine-readable descriptions of sensors and their technical information as web resources using URLs, identifying the instrument locally. Web-based sensor descriptions published using SensorML, part of the SWE specifications, and may be used as a URL to the landing page of the instrument registered at a PID provider. A SensorML landing page example has been published at the British Oceanographic Data Centre (BODC) via the ePIC PID provider service (http://hdl.handle.net/21.T11998/0000-001A-3905-F). To view the Handle record directly see http://hdl.handle.net/21.T11998/0000-001A-3905-F?noredirect or Table 5.1 in this document.

In SensorML (version 2.0), sensors are identified using a unique ID within the gml:identifier element and institutions may choose to use an instrument PID to assure uniqueness. Alternatively, an instrument PID may be declared as metadata within a SensorML description using the sml:identifier property (Snippet 11.3). While the latter is simpler to implement, it may limit the global discoverability of sensors and their observations within the Sensor Observation Service (SOS) web Application Programming Interface (API), part of the SWE standard. Web-based enquiries, requests or transactions made for sensors using this service are typically based on gml:identifier element in SensorML (expressed as a procedure), thus identifying sensors using local identifiers rather than global instrument PIDs directly. The link between local identifiers and instrument PIDs can be found indirectly using a combination of GetCapabilities and DescribeSensor operational requests to a SOS server.

Snippet 11.3 An example of expressing an instrument PID (http://hdl.handle.net/21.T11998/0000-001A-3905-F) as identifying metadata within a SensorML technical description using the sml:identifier property for a SeaBird Scientific SBE 37 Conductivity, temperature and depth sensor.
  <sml:identifier>
    <sml:Term definition="http://www.example.com/definitions/pidinst/">
       <sml:label>Instrument persistent identifier</sml:label>
       <sml:value>http://hdl.handle.net/21.T11998/0000-001A-3905-F</sml:value>
    </sml:Term>
  </sml:identifier>

The list of properties that can be expressed in SensorML to describe sensors is not particularly restrictive and it is recommended that institutional instrument providers follow the PIDINST guidance on landing page content (see Landing page content). Recently, the Marine SWE Profiles initiative specified a comprehensive metadata profile to improve the semantic interoperability of SensorML in the Earth Science marine domain by developing sets of sensor specific terminologies.

11.2. Content negotiation

We recommend using content negotiation where instrument landing pages are not easily consumed for human reading (such as XML schemas). PIDINST does not specify the form of negotiation that produces human-readable content from machine-readable representations. Other groups, such as the W3C Dataset Exchange Working Group (DXWG) are currently drafting recommendations on content negotiation from different information models.[2]

[1]as for instance: http://www.easyrdf.org/converter
[2]https://www.w3.org/TR/dx-prof-conneg/#dfn-data-profile