ietf-openproxy
[Top] [All Lists]

RE: P work in new charter

2004-07-02 05:42:01

All,

interesting so far.
I really suggest that you read about WSDL, CDL, BPEl etc and draw lines on
what can be done in OPES vs Not. Mnay of your concerns regarding service
discovery, security, composition, binding etc has been addressed there.

Abbie

-----Original Message-----
From: Geetha Manjunath [mailto:geetham(_at_)india(_dot_)hp(_dot_)com] 
Sent: Friday, July 02, 2004 7:38 AM
To: Markus Hofmann
Cc: ietf-openproxy(_at_)imc(_dot_)org
Subject: Re: P work in new charter



(d) defining mechanisms by which a user can communicate 
rulesets 
to the OPES processor.

Such mechanism is needed (one might default to existing 
ones), but 
out of scope of the WG.

Can you please give me specific pointers to existing mechanisms 
here? I would like
to evaluate their sufficiency  for the intended usage.

For example, use a CLI to load a rules file locally on the machine, 
use FTP to transfer a rules file, you might use HTTP....

-Markus

Oh! I guess these are fine! I was wondering whether there 
were any protocols along the lines of WPAD (Web Proxy 
Auto-Discovery Protocol) for this.

The reason I raised the point about "mechanisms to 
communicate the rules" was based on my trials of using IRML 
in a practical OPES env. Though IRML did talk about rule sets 
that included 3 classes of rules - (i)Rules set by OPES 
administrator, (ii)Rules set by Content Provider and (iii) 
Rules set by the user/client  - there were several practical 
problems  in effecting the latter two cases. I guess the 
question then boils down to  "who are the targetted authors 
that would use P language to write rulesets?". Clearly it is 
(i) but do we include (ii) and (iii) ?

Thanks and regards
geetha



<Prev in Thread] Current Thread [Next in Thread>