Should be enough to (1) talk to the Java API and (2) provide
a JSON API for the world.

This is really just experimental. MMT should be integrated directly
w/o JSON if possible to avoid nonsensical serialization.

Honestly Play is way too much for what I need. I just need
a simple JSON endpoint.

- I'll need some REST endpoint for the data

- GrahDB uses RDF4J -> JVM language prefered

- Let's try out Play. It works w/ Scala and Java.

- It was the seperate Importer compontent written in Java.

- It is not necessary anymore as we do it directly from REST
  in luoweb.

- I might merge it all in Java if it proves to be too annoying
  to do the RPC

- Better put all Go code into one repo.

- I've now wasted an hour on stupid file shuffling.

- It's the command line tool that will use the ulocollect lib.

- It's a Git submodule.

- code goes into /src

- prose in /doc

- /experimental is a playground

- Name interfaces in the canonical Golang way.

- Skip the complicated logging madness for now. I do want
  dedicated logging for each job, but not right now...

- This now works just fine.

- Next up is (1) refactoring, (2) cleanup of the received ULO/RDF
  files and (3) better state information (for the web interface)

We now run jobs w/o limits. This is probably bad in a hosted setup but
we can worry about that later should it ever become a problem. As it
stands the implementation is easier with a simple goroutine rather than
a complicated pool of threads.

- Not perfect yet. I think Job should be a struct so
  Collector can modify its members easily.

Now we are getting started for real :)

- This uses the RDF4J library which is what GraphDB is based on.
  Though it also supports other backends.

- The API is kinda nice tbh.

- By the end of the week I should know about the API.

- We could use SPARQL, but the RDF4J interface is essentialy SPARQL on
  steroids.