On 01/08/2019 11:33 AM, Michael Peddemors wrote:
We have long been planning on a IPv6 MTA registry, where those wanting to run MTA's on IPv6 could register as a certified operator, eg. a legitimate party with proper abuse contacts etc..
Okay. I question how large the value is in that.I say this because I can see a died in the wool spammer actually registering, paying the (nominal) fee(s), fulfilling the requirements, etc. Much like some spammers were early adopters of SPF / DKIM / DMARC.
The key thing is that the list as described doesn't differentiate between good operators and bad operators that both play by all the standard operating rules.
Hence, what value does the registry provide?I guess it does provide a first line of identifying someone that's IPv6 address hopping and sending spam to avoid black lists.
DNSBL on IPv4 works well enough, but for the IPv6 size, it is more effective to use a registry model, eg where an operator can say we only accept MTA traffic from IPv6 addresses that are contained in the registry.
The black hat in me worries that an unscrupulous mailbox operator could operate their own registry and charge people to be white listed. - But nothing prevents that behavior now.
But justifying allocating resources for this project, it is hard to come up with a model that would work, we only have so much free cash flow for altruistic projects that benefit the community..And while it would not really be nice to have any form of payment preventing parties from conducting email services on IPv6, the idea of a 'paid' registry might make such a project more viable.
In my (not so) humble opinion, it all comes down the price point. I'd like to equate it to a co-operative utility company. You pay enough to help support the necessary infrastructure. As long as you're not /for/ /profit/ (via exploitation), I think that something like this can be reasonable.
As for fees, I as a stingy private individual would be willing to pay $1 ~ $5 per sending IP per year. I might also be willing to pay $1 ~ $5 to be allowed to slave an RBL zone. (I don't know how such would be licensed.) Assuming that the registry was used by enough people to be worth while.
Finally, I support the justification of some nominal amount of money exchanged via credit cards (or maybe other services) as a test to make sure that senders are (more likely to be) who they claim to be.
-- Grant. . . . unix || die
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ mailop mailing list mailop@mailop.org https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
