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

At 06:56 AM 3/11/2003, Mark C Smith wrote:
>John Merrells wrote:
>>In this case the server is using some historical information to optimize the
>>amount of information that needs to be sent back to the client. The more
>>historical information stored the more you can optimize the amount of data
>>that needs to be sent back.
>
>Exactly. There is a big tradeoff here between complexity of implementation and the efficiency of the LCUP protocol exchange. It is important to allow implementors to have some flexibility while still assuring that the protocol works reliably. In many cases, efficiency tradeoffs can and should be left in the hands of the implementors.

I would agree that in many (if not most) cases, such tradeoffs can
be and should be left in the hands of the implementors.

However, where behaviors of one implementor is likely to cause
harm to others (either its peer or to other components of the
Internet), we need to place restrictions upon all implementor
to prevent that harm from occurring.

Kurt