So, the use case we were interested in was when legally-mandated filtering was happening; apologies if I haven't used the WG's terminology correctly.
Other forms of filtering are less interesting (at least to me), and I'd be concerned that supporting them could be seen as encouraging opinionated (rather than legally-driven) filtering by intermediaries. Cheers, > On 23 Feb 2026, at 6:19 pm, Michael Richardson <[email protected]> wrote: > > > Mark Nottingham <[email protected]> wrote: >> So, vendor extension namespaces make sense when they need a >> 'playground' to operate within for instance-to-instance >> interoperability. How would that work here -- what's the case for a >> "vendor-specific" database that *isn't* referred to by others? > > I thought that this was dealing with the part: > > mcr> 2) we want to allow the set of errors to be easily extended by many > different > mcr> resolvers > > and databases > (as you clarified) > > mcr> who have an increasing and many obtuse set of filtering > mcr> conditions. > > So when we get a new filtering that says that domains with adjacent vowels > forbidden, rule 4Q8NQ from a coalition of western (Albertan, Saskatchewan) > separatists attacking the other 7 provinces (the other 7 have adajacent > vowels, but ya gotta think bilingually. I should be asleep). > > When non-Canadian users wonder, they get a nice explanation in their > language. > > Unless my work is wrong, I think it's a really good^Wbad, absolutely insane > example. > > -- > ] Never tell me the odds! | ipv6 mesh networks [ > ] Michael Richardson, Sandelman Software Works | IoT architect [ > ] [email protected] http://www.sandelman.ca/ | ruby on rails > [ > -- Mark Nottingham https://www.mnot.net/ _______________________________________________ DNSOP mailing list -- [email protected] To unsubscribe send an email to [email protected]
