Hi all,
Im wanting to restructure my existing polyglot app which is a mix of
MongoDB and Postgres, to using OrientDB. The reason is that I want to
rewrite the site to allow for a more fluid social model with the follow /
following and be able to infer connections. On top of that, given it's film
related, which most of the GraphDB models demo, it makes great sense!
One the features is to be able to recommend a film to a friend. I thought
at first this would be a straight forward relationship but as soon as I
started to give it some thought, I've found myself asking too many
questions that I can answer by myself.
The key aspect is that a recommendation can be approved or rejected,
perhaps even have a back and forth comments exchange attached between the
friends involved.
My current plan is to have Vertex called Recommendation, which has 3 main
edges, FriendA, Film (the entity being recommended, though could be a
person, or another user), and FriendB
Friend -> recommends -> Recommendation
-> is_recommended -> AnotherFriend
-> recommended -> Film
The Approve / Reject action would be a property of the recommendation.
Comments would be joined by an Edge to the Recommendation too
<https://lh3.googleusercontent.com/-4xgHoQnWLKw/VIDTeqCAPxI/AAAAAAAALKU/3eHhZuOf86c/s1600/Screen%2BShot%2B2014-12-04%2Bat%2B21.34.13.png>
So the model looks like the above image
Has anyone experience of modelling a recommendation? Is the above a
feasible solution? I actually found modelling recommendations hard in all
three, RDMB, GraphDB and Document when it comes to performing queries on
them too.
Whilst if two different users recommend the same film at different time,
this would create two cases, but there are times when a group of friends
make a single recommendation at once (planned feature)
Looking forward to getting involved heavily with OrientDB and helping where
I can.
Many thanks
Sky
--
---
You received this message because you are subscribed to the Google Groups
"OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
