Comment by glenngillen | Hacker News Reader

Comment by glenngillen | Hacker News Reader

Parent Full thread

I don’t think this is a legitimate concern any more. There’s basically zero value in “confirming” an email address is legitimate. Between all of the data breaches and various other ways to get actual email addresses this isn’t a problem. It’s also so cheap to send email there isn’t an operational cost where you need to optimise for sending only to know addresses.

There is definitely a punitive cost for sending emails that are repeatedly marked as spam though. You also can’t just cycle IPs because a brand new IP with zero sender reputation is treated with almost as much suspicion by the big player as one that is known to be a spammer.

It’s much better to give people an option to opt out, and to honour it. Most of the email sending providers (e.g., SendGrid, mailchimp, etc) force you to include the link and automatically block future sending to that address. Some will even provide you the option to provide a reason, where you can specify “I did not sign up for this” which in sufficient number will flag the sender account. I suspect the vast majority of cases where people unsubscribe but continue to get email is actually some incompetence from not having multiple disparate email systems sync back to a shared do not contact list (rather each system is maintaining its own).

Click the unsubscribe button.

there's a line in the movie the incredibles where robert (the hero) is meeting with the designer (edna) of super suits to design him a new suit even though such activities are technically illegal.

The dialogue goes something like this:

robert: you know I'm retired from hero work.

Edna:As am I, Robert, yet here we are.

so now it's 2023. you're telling me it's now safe to click on unsubscribe to the spam emails.

yet here we are.

no, the strait forward response is to ignore and mark as spam any unsolicited emails you did not explicitly sign up for. don't try to interact through the desired or expected channels of any entity that spams you.

When you mark an email as spam, the mailing service emails the sender along with all the original headers. See https://en.m.wikipedia.org/wiki/Feedback_loop_(email)

Your reasons are not actually rational.

surely this depends on the actual mail provider you use.

i appreciate its impossible to prove a negative (everyone could be doing something they have no evidence or documentation of doing), but given my mail provider both says that you have to mark a selection before they'll share such information with partners and that marking emails as spam still trains your own user specific spam filter, i don't think (and an really hoping) this is not a universal thing.

It’s universal as far as I know.

I worked on an email system that sent billions of emails a month. We used these messages from providers to ensure we never sent them an email again to prevent hurting our reputation. (Marking an email as spam, is by itself, a very low signal on reputation, unless some massive % of recipients mark it as spam. Sending an email to someone who has already indicated you are sending them spam is a high signal that you’re sending spam, however).

It doesn’t even matter when you do it. We had people (outliers) who would go back and hit every single email we sent them for the last 6 years as spam, after a bad customer service interaction, not getting a refund, or whatever pissed them off. We actually investigated all outliers. Most people didn’t report spam on anything older than 6 months.

Reply on news.ycombinator.com