Hacker News Reader

Comment by georgemcbay | Hacker News Reader

I've been doing this for years and don't feel the least bit bad about it, and I think everyone should adopt this practice.

If marking as spam or unsubscribing are approximately equal effort, I will unsubscribe; if unsubscribing is even slightly more difficult than marking as spam, I mark as spam with no regrets. Even if you don't force me to login but make me do some monkey trick like type "CANCEL" into a box => spam! One click is the only acceptable method, and don't hide the link in a bunch of small print legalese, because I'm not going to expend much effort looking for it while the nice, inviting "Mark as Spam" button is just sitting there waiting for me to click it.

View on HN

It's not because you have no regrets that it's ethical. Even if it's a company like LinkedIn that makes it hard to unsubscribe, your input pollutes the anti-spamming algorythms with semi-legit emails. There are clearly millions of people who are fine with getting LinkedIn messages, and you're just being an ass, like putting your half-eaten burger into the recycling bin because the garbage bin was further out. Someone on the other side of your transaction is going to be tasked with taking your now-rotten burger outside of the recycling.

That is a problem for the anti-spam algorithms that need to figure out who will consider what to be spam.

Spam is in the eye of the recipient. If you send me marketing emails that I do not want, it is spam. Even if someone else would want it.

If people didn't mark stuff as spam, then there would be absolutely no reason for marketing people not to send spam to everyone. Even if only one person in 10,000 wants to read it, why not? But the option of marking stuff as spam is a way of making my attention as not free for you.

Seems to me that it's helping the anti-spam algorithms by marking spam as spam

The data helps solve the challenge. But the real challenge for the anti-spam algorithms is that what is considered spam is recipient-dependent.

The nice thing about Bayesian algorithms is that they are probailistic. If only a couple people flag as spam, it will have little effect on everyone's filtering algorithms. If, on the other hand, enough people have trouble unsubscribing to make this a serious "problem" then the email is arguably spam.

You should generally do Report Spam,at least I do that. If the email was sent from something you didn't actually subscribe, hitting unsubscribe tells the spammer that the email id is in fact genuine which will ultimately get you more spam.

Got any recent evidence for this? Because I'm not sure if this is actually still the case.

I don't have the link anymore because this was already a couple of years ago, was someone taking this ages-old advice to the test. Clicking the "unsubscribe" link on all spam delivered to an old spam-infested email address, he found that it reduced the amount of spam received to that address by a significant amount.

Reply on news.ycombinator.com