Stefan Tilkov wrote: > I'm reposting something I sent earlier (http://www.mail-archive.com/ > <http://www.mail-archive.com/> > service-orientated-architecture@yahoogroups.com > <mailto:service-orientated-architecture%40yahoogroups.com>/msg05227.html) > > I didn't get any reaction from anyone, not even Gregg, who is usually > certain to disagree ... maybe it didn't make it?
I don't recall seeing it. > IMO, there's a big difference between an interface with just one > generic operation and the uniform interface of REST - and it's not > just "one" vs. "four" (or whatever number), but a difference in quality. Okay, one of the issues that I want to clarify, is that INVOKE is not the useful operation, just like GET and PUT and DELETE and POST are not the useful operations. The specification of these operations is important. What is useful, is the actions that these operations stimulate. GET and PUT and DELETE and POST are meaningless without resources to act on. INVOKE is meaningless without resources to act on. > Compare an OO interface like this This is where I trying to lead you away from in your thinking. I am not talking about the programmatic interfaces. I am talking about the remote application interface that allows remoting to occur. My web browser can open an HTML document that I have on my desktop, equally well to downloading one from the network. The network transfer is enabled by HTTP. With the RMI programming model, I can write Java code that works in a single JVM with exactly the same model as I write Java code that uses the INVOKE operation to let me talk to a remote JVM. Gregg Wonderly
