[Home] [By Thread] [By Date] [Recent Entries]


Alaric Snell wrote:
> 
>...
> 
> I would be inclined to agree with Simon here... CORBA's IIOP, RMI, and good
> 'ole ONC RPC can all be described as 'requests directed to resources using a
> generic interface with standard semantics that can be interpreted by
> intermediaries almost as well as by the machines that originate services" :-)

If the interface to HTTP is basically

class Resource{
  Representation GET()
  void PUT(Representation)
  Representation POST(Representation)
  void DELETE()
}

Then what is the equivalent for IIOP, RMI, etc?

> There are CORBA/RMI/ONCRPC proxies and gateways and debugging tools and so
> on...

Nobody doubts their existence: but read the rest of the sentence.

> The one thing that HTTP has and most RPC protocols lack is that there is a
> way of flagging a request as being cacheable (making it a GET), and it's not
> even that well done in HTTP.

Caching is merely a side effect of the "rest of the sentence." IMHO, it
isn't even the most important side effect, especially in the web
services space.

 Paul Prescod

Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member