If we all built our APIs with a registered content type, it would really go a long way for standardizing APIs. Throw in some HATEOAS and you've got standards and links.
If we all built our APIs with a registered content type, it would really go a long way for standardizing APIs. Throw in some HATEOAS and you've got standards and links.
There's no reason why I shouldn't be able to select an existing data model (from some popular source, or microformats), and generate a REST endpoint in 1 click. There's no reason for anyone to manually create local classes that mimic these models, code the logic that convert the models in and out, and do a bunch of other manipulations using installed libraries. I just want to select the data, plug it through third-party libraries as a service (online APIs/functions, some of which being open source could be cached locally within the code), and be done with it.
We need to be able to reuse everything in a seamless way. Currently, reusing something often means more work and frustration, and most people end up coding their own lesser clones/forks that "does the job". Sigh.
REST is an architecture that can be implemented in many different ways, so I doubt there will ever be a one size fits all for REST. I think the real win will be when web frameworks allow us to work primarily with resources and representations instead of an MVC architecture.