On Wed, Jan 15, 2020 at 11:02 AM Jeffrey Race <jrace@post.harvard.edu> wrote: [...]
Aside from the reciprocity issue, it's a basic engineering rule that systems target their goal only when a corrective feedback path exists.
That feedback path does not need to be a personally written e-mail. Instead, it is possible to use signals like the absence of a reliable reporting mechanism to make decisions about not accepting some or all traffic from an abusing network. My main concern with proposal 2019-04 is that it would make everyone look the same. It then takes time and effort to distinguish the networks that will actually use abuse reports to fix problems from those that won't or just don't have the ability to do so. While I would accept Gert's proposal for making abuse-c an optional attribute, the reason I offered a counter proposal for publishing "a statement to the effect that the network operator does not act on abuse reports" is to add clarity at a high level. In the first case, it avoids wasting resources lodging reports that will be ignored. Secondly, it provides reliable statistical information about the networks whose operators claim to use abuse reports to clean things up. This would provide a metric that could be used both by other network operators to guide operational policies and governments or regulators to set theirs. Finally, we don't yet know what the RIPE Database Requirements TF will recommend. But I think that building a new business process on the existing model for publishing contact information assumes they won't recommend changes. Let's wait until they report before asking the RIPE NCC to build new workflows on a model that the community might want to change. Kind regards, Leo Vegoda