Re: I-D ACTION:draft-ietf-ldup-lcup-04.txt

Kurt D. Zeilenga wrote:

>At 03:14 PM 3/10/2003, John Merrells wrote:
>  
>
>>Kurt D. Zeilenga wrote:
>>    
>>
>>>Well, I thin that if we believe that certain implementation
>>>approaches would lead to implementations which would do harm,
>>>that we should state these beliefs and, where appropriate,
>>>detail implementation-approach-neutral imperatives of what
>>>behaviors are considered harmful.  
>>>      
>>>
>>In other words you're suggesting that the specification document that implementors should implement the specification correctly.
>>    
>>
>
>I'm suggesting the specification discuss areas where the
>implementor should take care not to do harm.
>
>  
>
>>>Lack of "eventual convergence data consistency" 
>>>      
>>>
>>What does LCUP currently state?
>>    
>>
>
>Presently, the I-D is a bit lacking in its data consistency
>statements.  It needs to be more explicit in stating that
>servers are to either generate messages which will lead to
>convergence or return an error.  This, I think, can be resolved
>by rewording portions of 3.3.7.
>
Reword to say what?  Change the SHOULD in the first paragraph of 3.3.7 
to a MUST?

>>You're looking for a guarentee from LCUP that the client state be
>>exactly the same as a server state at some point?
>>    
>>
>
>I am looking for the LCUP to be designed to support content
>synchronization with eventual convergent data consistency.
>
>I'm not yet convinced that LCUP provides this, hence this
>dialog.
>
>  
>
>>>and/or "overly chattiness" would be harmful.
>>>      
>>>
>>That's implementation defined.
>>    
>>
>
>Chattiness should be defined in terms of the protocol
>and the operations they perform.  Not only should
>protocols be designed to avoid overly chattiness,
>where the protocol implementation allows for
>chatty protocol exchanges, the specification should
>discuss chattiness issues and state appropriate
>imperatives.
>
>The LCUP I-D makes some statements in this area, I believe
>more is necessary and will over specific suggestions
>at a later date.
>
>Kurt
>
>
>  
>