Rails application architecture to work with RDF storage

Question

I need to make an application, that saves the data in the RDF storage for subsequent processing. I have several solutions and don't know what to choose:

1. Use classical application architecture with AR/Mongoid + RDF rendering + crawler, that agregates RDF data and saves it to the storage. It's comfortable, but I need to create a crawler and feed him data on every save.
2. Use spira ORM and save directly to the RDF storage. Here I don't need to write a crawler, but I need to make backend for auth plugin and to implement everything, that don't support spira.
3. Use classical architecture + additional spira models, so that I would have 2 model classes for each entity, ex: User in AR + UserSpira in Spira, where on save event of User Spira object is saved too.

What application architecture to choose? May be some other solutions?

Answer 1

I've found ActiveRDF to be a joy to work with. It does a great job of making working with RDF data Ruby-like. It supports a variety of RDF stores, and the abilities it has are based largely on the capability of the back-end that you use.

Answer 2

I've made my application with Mongoid(for accounts+devise) and spira(with sesame storage) for other models. It works fine, but spira is not very good, because doesn't support models exporting from owl onthologies and even doesn't support activemodel, but it seems it's the best we have in rails.