- From: Timothy Holborn <timothy.holborn@gmail.com>
- Date: Sat, 18 Feb 2017 08:31:02 +0000
- To: W3C Credentials Community Group <public-credentials@w3.org>
- Cc: "public-schema-gen@w3.org" <public-schema-gen@w3.org>
- Message-ID: <CAM1Sok2_UdytQn_4d8qrn7muoGxYsL_xHDS7RMqMC_aFGyspVw@mail.gmail.com>
I've been doing some work on figuring out an RDF discovery protocol in relation to schema-gen[1]. Plan / hope is to figure out a method that will mean providers of 'stuff' can advertise, that can be indexed somehow and that the indexing method is delivered in a way that can be queried, decentralised, stored, etc. Concept thereafter is that this is kinda like replacing robots.txt or sitemap.xml. Websites generally have primary purpose, as is increasingly defined by the range of extensions available. Yet, if site-servers were to have some sort of mechanism to advertise themselves as part of some sort of decentralised-ledger (discovery method / protocol, that includes 'append' or similar); then that is the underlying context of a ".rdf" upgrade. At the moment, json-ld contains a @context method to provide the means to illustrate a documents contents refers to a particular ontological definition. What is json-ld contained a @ledger (or whatever) to direct that the site should be added to whatever decentralised ledger is used for that type of offer / website. - software - data tool (ie: ckan / datahub / wikidata) - sparql endpoint - blog - news - sns - personal Website - banking website - Search Website - etc. this idea would likely make more sense with the use of an agent in relation to the server of the offer to add the information based on the @ledger context. noting also, that this may well contain information about reputation via either an incorporated or independent graph, that could in-turn sort those who are bad actors (ie: spam) vs. others (which in-turn could be supported via browser plugin for instance). Thoughts? Am i making this too hard? The underlying consideration is that whilst the RDF document states that it should be read in relation to x ontological framework, it does not declare at present; that it is directed to be indexed as part of the 'web of data' offerings in which the context then isolates what it is suggesting to be an additional source of information for the web. therefore perhaps rather than blockchains - we could do Quins (ie: Triples, Quads, Quins). the additional information being a means to identify which to add to a decentralised index and which not to add, as to segment down the total filesize for any-such indexing as a facet that may also includes concepts such as language, etc. Tim.H. [1] https://www.w3.org/community/schema-gen/
Received on Saturday, 18 February 2017 08:31:47 UTC