... The HTML Web works because the "smart client" is a human driving the browser. (Or a dumb crawler that doesn't care what it gets.) I don't see how it's at all related to APIs driven by software.
And the HTML Web is NOT RESTful. Plenty of sites return different results for a URL, break links, have non-idempotent GET, don't return the created item's URL on POST, etc.
So when people talk like this about API-driven stuff, it sounds like UDDI or some magic dream. Can you draw a direct line to what kind of code is gonna be written? Or is this like a "well if we invent AI-like stuff then it'd be nice because AI won't be able to figure out RPC?" Maybe I'm being dumb and unimaginative.
And the HTML Web is NOT RESTful. Plenty of sites return different results for a URL, break links, have non-idempotent GET, don't return the created item's URL on POST, etc.
So when people talk like this about API-driven stuff, it sounds like UDDI or some magic dream. Can you draw a direct line to what kind of code is gonna be written? Or is this like a "well if we invent AI-like stuff then it'd be nice because AI won't be able to figure out RPC?" Maybe I'm being dumb and unimaginative.