Discussion Forum

Setup possible with Core? Distributed development of instances, published publically

I am new user, here to evaluate grakn for our academic project.

We want to create a hypergraph domain knowledge model, to be edited by multiple domain experts (students for now), to be made available as a public ressource. Does grakn core support this quite typical academic use case, or is this use case reserved for enterprises paying for multi-user management?

I can imagine to host the server in a cloud with limited access, synchronize it to several local machines and run it there for populating the schema with instances (I would have to instruct every user not to touch the schema and hope for the best…). Then, export it to some Semantic Web format (JSON-LD?) and upload the current state to a public server. However, I’m not sure whether accidentally parallel access to the graph will cause problems and whether export is possible without programming skills (In the documentation, I found only migration code examples for the Client APIs - But I’m an end user, not an code developer: It’s hard enough for me to fill model a domain by by console insert code.

Talking of which, as a side note: I do not quite understand why the nice, handy, GUI (the only one for hypergraphs I could find :+1:) is limited for use by a much more capable developer of a small schema, whereas domain modellers have to create the many instances of that schema by coding insert queries in the console… :thinking: I would have thought that functions to create and (yes, actually) edit entities, attributes, and relations should be more or less the same on schema and instance level :confused:

I could also imagine to export schema and data after each edit, upload it to a code management system (like Github), and checkout/import it for further editing. I’m not sure how to approach that, though (which files contain the schema / data, or do I have to program that via an API - which I could not do).