On Jan 4, 2008 8:13 AM, Gregg Wonderly <[EMAIL PROTECTED]> wrote: > This comment represents one of the things that I find most often in > conversations about Jini. The supposition is that an underlying > transport/application communications technology is all that matters in > examining the difference between Jini and something else.
I'm afraid that's your own supposition; I know nothing about Jini, so I cannot assume anything about it. I was merely wanting to know *what* in terms of REST the problem was, but so far I'm getting vague notions that you're right and I should just shut up. > Please, please, please go create a Javaspace application using 2 or more > javaspaces with 2 or more lookup servers on 2 or more machines with 10s of > producers and consumers. [...] > Then do the same thing with an HTTP RESTful application and report back > here how things went. Let us know how you managed failover, resource > locking/unlocking, expiration of transient operations etc. Oh god, not the "as soon as you've grown up" and "once you join the *real* world" argument. Look, that's a cheap cop-out more than an argument for Jini or anything else. I'm obviously not going to create such a beastly system to prove or disprove that your own design is a good or bad one. Is it really that hard to point out why REST failed as opposed to, say, the developers who put the system together? Alex -- --------------------------------------------------------------------------- Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps ------------------------------------------ http://shelter.nu/blog/ --------
