Hello everybody,
This is very exciting work. I would love to be able to work on this effort on a regular basis. But my work and my one year old (he is waking up now...) are at this point almost all I can handle. I'll be a fly in the wall checking what progress is being made. Giving my biased opinion. And reading some documentation and code.
One thought that comes to mind is the HATEOAS constraint on REST. It is an interesting idea, and I have tried in the past to find a way to support it in my applications. There always leaks through the abstraction. Anyone please correct me if I am wrong. The motivation for this constraint is to have a conceptual framework on which new media types may be created. Browsers or other software can implement these new media types and the server can still contain all of the application logic without the need of modifying the client. Pure HTML is the one example of this.
For example a shopping cart application may benefit some by providing a REST api that would enable other media clients. I personally have found that the benefits of functional programming like better composition, less bugs .. ect really have an impact on the user experience. In my mind the HATEOAS constraint does not really have that much of an impact (Pure HTML aside).
I would be most interested on an approach similar to react, Which seems easier to program. Also a programming environment that would allow me to move code execution from client to server whenever it makes sense for the project without having to rewrite the code into a different language (already possible but should be better with a new stack). I really like the idea of using something like HOPAC. I have not had the pleasure of doing so, but it is a very interesting concept. There may be a way to have HOPAC operate over HTTP. That could be very interesting.
When developing websites. I personally struggle on who should maintain HTML. If HTML is not maintained by programmers, Offering a solution to type check HTML templates would be a plus. If HTML is maintained by programmers then the websharper/typed DSL approach seems better. If the budget/time allows I normally deal with this problems by allowing both approaches to coexist.
Anyway, I don't want to create controversy over HATEOAS. I have been proven wrong before in the past. And of all places I think this group may be where I may learn the value of HATEOAS. Just wanted to give my two cents. And I will be reading code, blogs, articles and if time allows I will put in some code of my own.
The one thing I could perhaps help with is with writing some blogs with some examples of how to use all this new stuff.
Arturo Hernandez