--- Michael Thomas <mike(_at_)mtcc(_dot_)com> wrote:
There's nothing optional about the KRS cache in the IIM
code. I'm not sure what sort of argument somebody could
offer up to change my mind on making it optional.
Again, we're not talking about the same thing. As the subject line suggests,
I'm talking about the protocol whereas you seem to be referring to your
personal implementation of that protocol.
Your basis for caches seems to be speculation that other developers are
motivated in the same way you are. I think it would be less risk if your
protocol ensured it.
I have shown that developers, including myself, have a large
performance incentive to implement the cache. I have also,
through experience, seen that implementing the cache is a
big win. Do you have experience to the contrary? I have no
problem making implementation of the cache a MUST
requirement if that's what you're after. But you seem to be
implying that they would ignore that requirement regardless
which seems nonsensical to me.