SAFE Network Dev Update - January 31, 2019

[krnelson]

@maidsafe

The next step will be to reuse RDF’s triple hashing so that we can more efficiently store triples as key/value pairs in Mutable Data

From Nov 22nd dev update:

@maidsafe

How to store the RDF triples on the SAFE Network efficiently? There are many known ways of doing this (and the front-end team already used one of them in the safe_app_nodejs RDF emulation layer ), but we’ll need to find an approach that works best while meeting the network fundamentals. This involves storing and indexing RDF triples in Mutable Data, and most importantly, storing triples in the encrypted form.

Related Safe dev forum post.

Some layman questions: Will this mean that indices must be maintained to access the six different RDF triple patterns for each entry[1] and time-space trade off will that bring? Query complexity[2] vs say a native Triple store or graphDB[3]? Is a native RDF triple store (or Graph database[4]) overlay feasible for the Safe Network?

I found some other interesting related topics questioning how RDF and storage/query methods could be simplified for developers while doing a quick background research on this question, have posted to a new topic here.

[1] CumulusRDF: Linked Data Management on Nested Key-Value Stores
http://iswc2011.semanticweb.org/fileadmin/iswc/Papers/Workshops/SSWS/Ladwig-et-all-SSWS2011.pdf

[2] NoSQL Databases for RDF: An Empirical Evaluation

nosqlrdf.pdf

[3] A survey of RDF storage approaches (“The problem of indexing RDF graphs”)
https://hal.inria.fr/hal-01299496/document

[4] What-are-the-differences-between-a-Graph-database-and-a-Triple-store?

What are the differences between a Graph database and a Triple store ?