On Mon, 21 Jan 2013, Melinda Shore wrote:
And too bad we can't have the IESG implement it while
reviewing it.
There we go. That would cut down on the "I'll support your
awful idea if you support my awful idea" horse trading.
No, it just means that I have to implement my awful idea and
you have to implement yours and then we agree to cross license
the source.
The thing that is easy to forget when viewing a specification
outside of the context of when it was written is that many/most
IETF specs are written to imporve the interoperability of
existing implementations. Part of improving the interoperability
is to incorporate features and/or extensibility to avoid
future hacking. AND we tend to need to grandfather existing
features which we then have to work-around for ever.
And our most popular specifications are the hardest to improve
because of the weight of existing usage.