- From: Timothy Holborn <timothy.holborn@gmail.com>
- Date: Thu, 23 Mar 2017 23:13:05 +0000
- To: "public-schema-gen@w3.org" <public-schema-gen@w3.org>, Hans Polak <info@polak.es>
- Cc: Adam Lake <creatinglake@gmail.com>, Manu Sporny <msporny@digitalbazaar.com>, Dan Brickley <danbri@google.com>
- Message-ID: <CAM1Sok16FH3UR8LeFzr051XZhugjQFwb1_FNQDYnjQ37eC2Dtg@mail.gmail.com>
Basically, Schema is much like a machine readable word-list rather than creating sentences, that can the be checked for reputation (ie: do the claims make sense once processed by search magic). CONCEPTS A site exists that allows someone to use a UI/UX to create mark-up. They're able to use a series of RDF Statements to create a RDF 'sentence', that may otherwise not be easily covered by the existing 'word-list' provided by mark-up. 1. The ability to create concepts (and save them): ie: MedicalSpa = Place (daySpa + medicalClinic), staffedBy (medicalPractitioner, beautyTherapist) 2. The ability to use existing Concepts to add to records (ie: medicalSpa + Location, Opening Times, etc.) 3. The ability to generate a URL/URI that acts as a schemaURI (triple) to describe the whole concept. Other examples; NewsArticle HistoricalDocument (place, about(wikidatalink pertaining to wikipedia and/or other page on historical subjects) In this way, it's a bit like an external CSS file. User can either download the generated file, or simply use the URI/URL in a document as part of a triple to describe what the document is attempting to describe. USE OF 'VERIFIABLE CLAIMS' (seeAlso: https://www.w3.org/community/credentials ) The creator of concepts could 'sign' them, and they could be stacked. Ie: a publisher might create a statement that says it's a publisher, et.al. A journalist may have a statement that pertains to the publisher; and the article may be authorized by the journalist. In-effect you get 'concepts' as 'verifiable claims', that can be used as elements to describe other (broader) concepts (graphs) that may in-turn pertain to a multitude of actors (ie: some medicalSpa association says that medicalSpa is a thing and that the nominated 'subject/actor' is a member). Other examples of similar concepts; Google Services: - Google My Maps: https://www.google.com/maps/d/ - allows a user to upload a CSV file with stuff in it, then generate a URI to represent the custom map. - http://cse.google.com/ - Allows a user to generate a custom search (inclusively to the use of schemaorg terms) then generate a public. NOTE: TBL's argument the HTTP URIs (without "#") should be understood as referring to documents, not cars.[1] This would be a machine-readable document, however the underlying notion it's providing is a set of machine-readable declarations about the document it is intended to be asserted to. CONSIDERATIONS *REFERENCE DESIGN* A reference design be produced that is available via GitHub as a work-product of this group. I can see enormous benefits (including the insertion of advertising and/or tracking codes) for commercial applications of the concept; however i also think it's kinda important it's not produced simply as a web 2.0 styled silo, and i'm as yet unsure of how interop requirements may grow if the idea was successfully developed. EXPORT OF CONCEPTS Existing 'word-lists' in schemaorg (for instance) are based on discussions and are slowly developed. This tool may find that a particular concept is readily used and therefore better added to the core 'dictionary' rather than being a constructed concept[2]. This is similar to the manner in which language develops in society, in that a concept ends-up getting added to a dictionary[3]. TOOLS Document For anyone who hasn't checked it out, or isn't aware - the Tools[4] Sheet has been developing and contributions are welcomed. Tim.H. [1] https://www.w3.org/2001/tag/issues#httpRange-14 [2] https://github.com/schemaorg/schemaorg/issues/1493 [3] https://en.oxforddictionaries.com/word-of-the-year/word-of-the-year-2016 [4] https://docs.google.com/spreadsheets/d/19IEgvdvwl_EOGhmIFinVQu4OerRojeje8PaZWGvoO4Q/edit#gid=0
Received on Thursday, 23 March 2017 23:13:52 UTC