Re: [dhcwg] Re: I-D ACTION:draft-ietf-geopriv-dhcp-civil-07.txt

I'm fine with that approach. Certainly simpler to describe...

Before I draft text, anybody objecting to this?

Bernie Volz (volz) wrote:
> Ted:
> 
> Excellent solution!
> 
> - Bernie 
> 
> 
>>-----Original Message-----
>>From: Ted Hardie [mailto:hardie@qualcomm.com] 
>>Sent: Monday, November 21, 2005 2:49 PM
>>To: Bernie Volz (volz); Henning Schulzrinne
>>Cc: Ralph Droms (rdroms); DHCP discussion list; geopriv@ietf.org
>>Subject: RE: [dhcwg] Re: I-D 
>>ACTION:draft-ietf-geopriv-dhcp-civil-07.txt
>>
>>At 9:12 AM -0500 11/21/05, Bernie Volz \(volz\) wrote:
>>
>>>If you want to keep this information private, DON'T SEND IT via DHCP.
>>>Period. End of story! That's why clients shouldn't be sending this
>>>unless configured to.
>>>
>>
>>This suggests that we might get past this by reversing 
>>Henning's original
>>proposal.  Do not make this a protocol error, but state in 
>>the document
>>that sending it from client to server is equivalent to 
>>sending a pidf-lo
>>with unlimited distribution and retention permitted.   Make 
>>it a SHOULD NOT,
>>obviously, but note that if you do send it that is the effect.
>>
>>My reasoning here is that it is nigh impossible to determine what
>>administrative domain means in many deployments, and the agreement
>>to share among servers in a single administrative domains translates
>>into "you don't know where this is going".  If we state in 
>>the document
>>that it is SHOULD NOT unless the client is willing to set the 
>>ravish-me bits,
>>that is at least clear on what the consequences are.
>>
>>How would that work for folks?
>>				Ted
>>

_______________________________________________
dhcwg mailing list
dhcwg@ietf.org
https://www1.ietf.org/mailman/listinfo/dhcwg