It seems to me there are 3 ways the world could go, and no 4th option:
1. Something OPES-like will be standardised in the IETF
2. Something OPES-like will be standardised in another organisation, perhaps
one with less clue than the IETF about scaling and security.
3. Non-standardised OPES-like things will be widely deployed.
Pick the one you prefer.
I'd say that we're likely to end up with some of #2 and #3 in any case.
The fact that these exist, or are likely to exist, does not by itself compel
action by IETF, because it's not immediately clear whether IETF's work on #1
would reduce the abuses present with #2 and #3, or whether IETF's work would
facilitate a significant increase in abuse. And the argument that providing
a standardized interface that facilitiates abuse will increase the amount of
abuse, is not easily dismissed.
We need to have a better understanding of whether, and under what conditions,
IETF standardization of something OPES-like would benefit the Internet
community.
The currently-proposed charter completely ignores the issue.
I believe that it is possible to define an OPES-like service in such a way that
it
can only be used with the explicit consent of at least one of the endpoints.
If
those constraints were designed in and properly implemented, then IETF's OPES
work
could indeed be beneficial and would be unlikely to do harm.
However, I doubt that IETF's work on OPES (even if so constrained) would lessen
the
amount of abuse due to #2 and #3. So I don't see the "external threat" as
being a
justification for doing OPES work in IETF.
We should judge this proposed group on its own merits, and not approve this
group
merely because other folks might be working on similar things for ill purposes.
Keith