Re: Attributes

New Message Reply About this list Date view Thread view Subject view Author view Attachment view

From: Mike Eisler (mre@jurassic.eng.sun.com)
Date: 03/24/97-12:14:16 PM Z


Date: Mon, 24 Mar 1997 10:14:16 -0800 (PST)
From: Mike Eisler <mre@jurassic.eng.sun.com>
Subject: Re: Attributes
Message-ID: <Roam 0.9.4.859227256.9226.mre@jurassic.eng.sun.com>


> > So we might have:
> > 
> > 	- Protocol defined attributes.
> > 
> > 	  An enumerated set of attributes with clearly defined semantics.
> > 	  The set be extensible via standards-track RFCs and registered
> > 	  with IANA.

Adding attributes via standards-track RFCs and IANA seems redundant. 
I prefer just the standards-track RFC approach and associated each
successive RFC with a new NFS V4 minor version #.

> > 	- Application defined attributes.
> > 
> > 	  Name/value pairs.  Supported by the protocol but names
> > 	  and values are opaque.
> > 
> 
> I agree with most of this but I don't think Names should be opaque. I
> prefer the method used for numbering RPC programs. Range x-y is for
> registered attributes. Range 1000000+x - 1000000+y is for client
> unofficial attributes.

The value in using string names for the uncontrolled attribute name space
is that client applications that don't understand the named attributes can at
least display them to a human user who may understand the attribute
because its got an obvious name.

	-mre


New Message Reply About this list Date view Thread view Subject view Author view Attachment view

This archive was generated by hypermail 2.1.2 : 03/04/05-01:45:38 AM Z CST