Clojurians Log v2

> joined #rdf along with 13 others. Wow, it’s like RDF became popular all of a sudden!

it is popular (among 65 people).

There were a lot less people yesterday tho! 🙂

Out of curiosity, How many people here besides @rickmoynihan are using RDF in production, or something that supports something in production?

Thanks @ivan.kanak_clojurians, that was a good talk.

We use a couple of RDF datasets to benchmark #crux against other DBs: https://dsg.uwaterloo.ca/watdiv/ http://swat.cse.lehigh.edu/projects/lubm/ (the bench runs nightly)

Yeah, you kind of need both of these at the same time, so putting one or the other 'first' seems a little like saying 'right foot first' vs. 'left-foot-first'. 🙂

I really love Clojure's map abstraction, and a lot of what motivated the IGraph project was a desire to extend a similar abstraction to the expression of relationships.

Maps are great but have their limits; analagous to structured databases vs relational. At some point you realise you’re building a database with your datastructures, e.g. when you start computing lookup tables, and having lists of denormalised/sorted keys etc; which you then need bespoke functions to handle. Rich Hickey: Every class is an island. Rick Moynihan: Your datastructure’s become islands too 🙂 At that point you want a query language; and graphs are super flexible In all fairness Rich has known this since pre clojure 1.0, and he spoke a lot about it back in the early days of clojure… A lot of that thinking I think manifested itself into datomic.