For an end user "spam" is "email I don't want".

"But" I hear you screaming, "That's their credit card bill, they even confirmed their email address and specifically opted into receiving it by email - that's not spam". It's unwanted, they don't have $850 to pay off the balance and they don't want to think about the interest charges, so it's spam as far as the user is concerned, some non-trivial fraction of users will click "Spam".

You've probably seen people who have piles of unopened real world postal mail, they feel the same way about that. Unsolicited catalogue. Bank statement showing I am overdrawn. Nagging letter from my mother. All unwanted: Spam.

Worse, some of these same users will anyway be angry that the important message they marked as spam is treated as spam. Why are facts true? They shouldn't be true if I don't want them to be. Is that irrational? Sure, that's humans for you.

This is a losing proposition, and the end-user facing providers are aware of that, but "spam blocking" is something users demand from such providers even while being very angry that it doesn't do what they want. "Humans are irrational" is not actionable.

last time I checked (been a few years since I used Google) BigTech is perfectly able to distinguish a bad actor from an opt-in type of service. They even presented me with a little notification whenever I moved stuff into SPAM that didn't belong there asking "do you want Gmail to unsubscribe you from X" ... I can't exactly recall the text - but all their lofty claims about their "AI" should account for something?

I agree with you that the _user_ is the biggest wildcard. But it's not that we should allow Google/Microsoft get way with this as an excuse. They are currently right in the middle of killing open protocols like imap in favor of their own flavored shit (probably something AMP like but for email). They do not deserve the benefit of a doubt. They do not even deserve to have their side of the argument heard at this stage.

So if you mark a legitimate subscription as spam and don't unsubscribe when gmail prompts you to, it's likely that it'll get sent to your spam in the future, and perhaps contribute to the sender's spam score which affects other people's inboxes too.

[1]: de-facto standard? I can't find any RFC for it.

This is true, from both sides of the equation. The biggest single problem with email today, IME at least, is the number of major mail providers who are acting as gatekeepers for their users and doing a bad job of it. In particular, they block way too much legitimate mail, and often do it silently, so the sender is not even aware of the problem or able to contact them about getting it fixed.

Then the sender of the missing mail gets the customer support requests about the missing password reset emails, or the complaints that someone didn't know they were still subscribed despite the receipt emails being sent for each payment, or...

At this point, there really ought to be a blacklist for unreliable mail services on the receiving side analogous to the spam blacklists, so businesses can warn their users if given an address on a bad service and invite them to choose another.

If you only send a few emails per month, you are fine. When you host an email server for others, even if it's only your family and friends, it can quickly become a nightmare. And if I ran a business, I'd never do it myself.

Landing on a blacklist of those large providers effectively shuts your email server down and it's way too easy to get blacklisted. There are so many techs around email now (DKIM, TLS, SPF, HELO testing), that coordinating them becomes a huge responsibility. Misconfiguring a single one can lead to rejection at the large providers that deal with billions of incoming emails a day, with hyper aggressive spam filters. The worst thing that can happen is that your emails were to be classified as spam but you didn't know it. A flat out rejection is relatively rare so you are often not notified that your configuration is incorrect.

Getting off those blacklists is extremely hard. Especially if you don't land on GMail's blacklist but on those community driven ones that only allow your IP to be removed from it once. If you land on them a second time, you need a new IP and hope that this IP isn't already blacklisted.

Many large providers are blacklisting entire ISP IP ranges due to past so waves. E.g. GMail is blocking the entire AWS EC2 range, I believe.

We need to start over and build a truly open internet.

I want to agree, but email is actually an example of an open and federated protocol achieving success. It's pretty much everything that could have been hoped for as far as openness and interoperability go at the protocol level. Unfortunately, balkanization still occurs at the social (ie network) level.

Open alone is insufficient.

As such, the spammer would only have to store each unique message and the list of recipients associated with each unique message.

1. Sender canot simply mass-cram messages to recipients, as many spam engines today do. (Qmail for a time was notorious for doing this).

2. There's good chance that any bulk spammer will accrue a negative reputation by the time most recipients bother to request messages. So they won't.

3. Spammer must sustain a stable online presence --- not be knocked offline or blocked by ISP or hosting provider.

4. Spammer has to track successful deliveries vs. messages that are complete.

Upshot: the spammer's workfactor increases.

If you want to send me an email, you have to send me an SMTPCoin. If my user doesn't mark it as spam, I give you the coin back. If my user marks it as spam, I keep the coin.

https://en.wikipedia.org/wiki/Hashcash