[FoRK] Resource-oriented computing?

Joseph S. Barrera III joe-ml at barrera.org
Tue Jun 3 01:04:30 PDT 2008


Dr. Ernie Prabhakar wrote:

 > Still, inquiring minds want to know: what do FoRKers think of NetKernel?

OK, so I finally clicked on the link (as opposed to just commenting on 
the descriptions).

What do I think? Existing ideas covered with a thick slime of marketing.

Time prevents a detailed analysis of each claim.
But I'll give a stab at a couple of sentences.

"Simply stated, resource-oriented computing is a logical computing model 
in which information is modeled as abstract resources and each resource 
is identified by a logical URI addresses. The microkernel sits between 
the logical and physical computing layers and mediates all resource 
requests, binding the request to the physical level at the moment of the 
request and then unbinding it when the result is returned."

How is this any different than Unix file names vs. physical blocks? Or 
more generally, the Unix principle that everything is a file (or is 
named by a filename)?

"NetKernel implements a full resource-oriented computing model. In fact, 
all resources within an application are addressed by URI. Even code itself!"

s/URI/filename/g -> yawn

"The research traces its heritage back even further, to the World Wide 
Web, and even indirectly to the foundations of the Unix operating system."

Yeah -- no shit

"Other results of resource oriented computing approach include:

"* Functions and services decoupled from languages - processes can be 
scripted rapidly with any supported dynamic language (Ruby, Python, 
Groovy, BeanShell, Javascript or dynamically compiled Java) or with a 
domain specific language (XSL, E4X, XQuery, RDFQ, STM).
"

VAX/VMS was doing that about 18 fucking billion years ago. And IBM was 
probably doing it earlier.

Yawn fucking god damn yawn. Next?




More information about the FoRK mailing list