[AFS3-std] Standardization of GetCapabilties RPCs for AFS3 client and services

[Derrick J Brashear]

On Sat, 25 Feb 2006, Jeffrey Hutzelman wrote:

>
>
> On Friday, February 24, 2006 07:41:30 PM -0500 
> Derrick J Brashear <shadow@dementia.org> wrote:
>
>> On Fri, 24 Feb 2006, Chaskiel M Grundman wrote:
>>> Why is it a good idea to put client policy 
>>> configuration in the server?
>>> 
>> 3 words: imap magic plus
>
> I fail to see how this is in any way relevant to 
> Chaskiel's question.
> Magic plus is not putting client policy 
> configuration in the server.
> I know of no IMAP client anywhere that changes its 
> behavior based on whether a server advertises 
> support for magic plus, and it certainly wasn't 
> the intent that any clients ever would do so.  In 
> fact, the nature of that particular feature is 
> such that it is most useful with clients that 
> don't even know it exists.

it exists because some purist decided it was bad for clients to identify 
themselves.

it's wretched. and the purist probably hates it, but we got here because 
of them, not in spite of them. you reap what you sow. i'd rather sow what 
jeff altman wants to plant. i may have even suggested it. tom's suggestion 
also works. the 'distribute policy out of band to clients' does not. if 
you can't understand why not, then we have an unbridgeable gulf.

now, as to 'per volume encryption requirements', if we do vldb extensions 
that becomes very easy.

but now, i'm going to eat, so I get out of Nashville in a timely manner.

Derrick