@prefix this: . @prefix sub: . @prefix fairmi: . @prefix dct: . @prefix dce: . @prefix np: . @prefix pav: . @prefix rdfs: . @prefix xsd: . @prefix fair: . @prefix foaf: . @prefix dcat: . @prefix orcid: . @prefix void: . sub:Head { this: np:hasAssertion sub:assertion; np:hasProvenance sub:provenance; np:hasPublicationInfo sub:pubinfo; a np:Nanopublication . } sub:assertion { fairmi:Gen2_MI_F2B a fairmi:FAIR-Maturity-Indicator; rdfs:label "FAIR Maturity Indicator Gen2-MI-F2B"; foaf:primaryTopic fair:F2; fairmi:comments """BEWARE: Apache Tika is capable of extracting metadata, in the form of Linked Data, from a wide range of opaue file-types such as PDFs and images. This process will therefore return Linked Data that can only be found using a special tool. Therefore, passing this Maturity Indicator does not mean that the publisher has *actively* made grounded metadata available."""; fairmi:examples ""; fairmi:measuring """Whether the metadata of the record contains \"structured\" elements that are \"grounded\" in shared vocabularies. For example, in one of the various forms of linked data (JSON-LD, RDFa, Turtle, etc.)"""; fairmi:procedure """Metadata is harvested by: 1) resolving the GUID (following all redirects) with a Content-Type header specifically searching for some form of structured data. e.g. 'Accept: text/turtle, application/n3, application/rdf+n3, application/turtle, application/x-turtle,text/n3,text/turtle, text/rdf+n3, text/rdf+turtle,application/json+ld, text/xhtml+xml,application/rdf+xml,application/n-triples' 2) resolving any Link 'meta' HTTP Headers (processed independently according to this same process, but not iteratively) 3) parsing the response body either as a hash (for non-linked data) or as a Graph for linked data, or both. 4) All other data is passed to the 'extruct' tool (https://github.com/scrapinghub/extruct) or to the Apache Tika tool (https://tika.apache.org/) for deep exploration 5) Any linked or hash-type data found by those tools are merged with the existing Hash or Graph data 6) The Graph is interrogated v.v. if it contains any data"""; fairmi:rationale """Structured, grounded data is inherently easier for machines to accurately process and interpret, in particular by generic agents, who are able to precisely determine the meaning of an element based on it being a GUID (and thus, more FAIR)"""; fairmi:relevance "All"; fairmi:requirements "The Metadata GUID."; fairmi:validation "Graph contains data." . } sub:provenance { sub:_1 dce:format "text/markdown"; a void:Dataset, dcat:Distribution; dcat:downloadURL fairmi:Gen2_MI_F2B.md . sub:assertion pav:authoredBy orcid:0000-0001-5306-5690, orcid:0000-0001-6960-357X, orcid:0000-0001-8888-635X, orcid:0000-0002-1164-1351, orcid:0000-0003-4727-9435; dcat:distribution sub:_1 . } sub:pubinfo { orcid:0000-0001-6960-357X foaf:name "Mark Wilkinson" . this: dct:created "2019-02-26"^^xsd:dateTime; dct:rights ; dct:rightsHolder ; pav:authoredBy orcid:0000-0001-6960-357X; pav:createdBy orcid:0000-0002-1267-0234 . }