Is this the death of REST?

(Confession: the title was sqlite in js [673,971 bytes gzipped; via emscripten] but nobody cared, so I made it more linkbaity.)

So we've got sqlite in JS now. It's a little heavy, but that's amazing for some applications. Meanwhile, there's the W3C's IndexedDB under proposal:

http://www.w3.org/TR/IndexedDB/

https://developer.mozilla.org/en-US/docs/IndexedDB

... optionally with https://github.com/axemclion/IndexedDBShim for greater browser reach.

(I really do wonder how the W3C justified IndexedDB given SQLite's excellent codebase and small size? I guess they mainly wanted event hooks because I can't see any other impressive features.)

So, how long until we see an implementation of `rsync` for some of these things? Why bother with developing complex manual-integration APIs if you can just perform a gzipped batch-sync of your client-exposed DB portion right down to the client app, and offload all that server processing and round-trip overhead, SQL injection risk, etc.?

Potentially fun project for someone else: learn IndexedDB + LLVM + asm.js + emscripten + rsync batch mode source + generalized javascript serialization + sync the crap out of some DB's over slow links + benchmark performance/reliability/integration implementation speed changes vs. current era REST APIs. (Alternatively, https://github.com/ttezel/anchor appears to be an rsync implementation in node.js)

Perhaps for mobile applications and other rich client scenarios (browser games, data-heavy financial or monitoring frontends, etc.) REST is now on the way out.