Model Data Without Making Tables — A Pervasive Linked Data Stack

Accepted Session
Short Form
Scheduled: Tuesday, June 26, 2012 from 2:30 – 3:15pm in B301


Want to be agile? Why bother modeling your data with a static table, declaring classes, and setting up mapping from tables to objects and finally to HTML and back again? The linked data standard presents a more powerful data model, and lets you use your website itself as a database.


MVC, ORM, OO, RDBMS — too many letters, yuck. Why try modeling your world with square tables that you pretend are an object graph inserted into an HTML tree? Why write so much boilerplate for the common cases that you have generators for everything? And why not have the option of starting with instance data OR a schema?

Starting with describing things piecemeal in a graph store, I’ll show how to maintain the flexibility such a model has to offer throughout the stack, keeping things ultra-DRY, normalized, and addressable as syncable web resources.

Also to be glossed over: Querying the web as if it were a database, and using logic rules on the model, rather than encoding business logic in low-level app code.

Speaking experience

Talking about E4X at PDXJS:
Talking about some of the same stuff last year with Brian:
And more of that at PDXScala: