Hacker News Reader

Comment by threepipeproblm | Hacker News Reader

I only skimmed this so forgive me if I missed something. But could this simply be what happens to you when a sufficient number of email recipients classify you as spam, and you then reach some critical threshold?

>> I changed my sender email address to see if would help

But if the pattern of content got flagged as spam by Google users, the sender address might not be the only determinant.

>> I was relieved to discover that others with newsletters were having similar troubles very recently.

But if this results from crossing some sort of user-classified spam threshold, one would probably always be able to find some recent "victims".

This is not intended as a criticism -- I'm sensitive to Google's seemingly arbitrary power -- just an idea to consider or rule out.

View on HN

Similarly, it could have been 3-4 consecutive newsletters with particularly low engagement that put him across some threshold. Not people marking as spam, just not opening or opening and quickly deleting.

Personally, I can't remember the last time gmail marked anything important spam. Maybe a newsletter whose absence I hadn't even noticed, but nothing actually important to me.

I just wish I could get gmail spam filtering on my cell phone. Robocalls are out of control -- 4-5 / day. (I've looked at some of the apps / subscriptions for this, but I'm not ready to give a relatively unknown 3rd party access to my contacts or to adding contacts, which they generally require.)

Try making google voice your main number? They had solid robo filtering if I remember right. I'm on Fi now and I can't remember the last time I received a robo call.

Adding myself to the no-call list and marking the numbers as spam in the phone made them disappear for me.

> a sufficient number of email recipients classify you as spam

From Google's end, I can understand this decision. But from the consumer end, I only want that logic used for de novo email sources.

Even if many people think a newsletter is spam, I'd want my personal patterns to override that. If I've opened letters from the same sender repeatedly and never marked one as spam, it shouldn't much matter what other users have to say. If I've ever explicitly taken a sender out of spam and not put them in spam, that should basically settle the matter.

I regularly lose emails I've requested to Gmail spam, and even when I've un-spammed the same sender multiple times. I appreciate that Google achieves pre-emptive blocking via a large userbase, but I don't appreciate when the wisdom of the crowd overrules my own behavior.

That seems an unlikely thing to have happen to multiple high-profile newsletters at once in a short span of time. Particularly, if these people aren't seeing big changes in their users, why would the spam threshold be hit now? Did a large chunk of email readers suddenly decide enough is enough?

What if it happens to some high-profile newsletters over every/most any short span of time you might pick?

Then you'd expect this article to appear on the FP every week, surely? It seems more likely to me that whenever people write about something out of the ordinary, it would not happen ordinarily - otherwise there wouldn't be anything to write about.

No you'd expect people to be complaining about it somewhere at some frequencly...and it hasn't been established that isn't the case. However, if you search for "Google newsletter suddenly spam folder", you can easily find lots posts that are superficially similar-looking, from various time periods.

Maybe Google changed the threshold? Lets say 20% of recipients have historically classified the list as spam, and Google just adjusted the spam threshold from 25% to 15%.

That would be exactly what the article is describing. GGP was suggesting an event which Google didn't directly control.

Another possibility is that the spammers have adapted and rolled out the next phase of their plans. After creating a million or three Gmail accounts (aged, so they're all months old, and staggered, so they're not all created at the same time, and with some amount of fake "organic" traffic such that they seem they're owned by real people), the obvious first phase, as a spammer, is to sign up for your own newsletters, and mark those as "not spam" from your three million fake accounts.

As a spammer, you've got to assume Google's going to catch onto your first three million fake accounts and mark both those accounts, the email servers' IPs you use to send, and the email list itself as spam. Consider your first three million gmail account burned by that point.

The less obvious second phase then, using your next three million fake, aged, but-with-fake-email-traffic-to-make-them-seem-real Gmail accounts, is to sign up for not-spam newsletters, but mark those as spam. The wishful-thinking hope is that with enough of those marked as spam, the wider public will see more articles like the linked article, and ever so barely push back on Google, who then have to figure out another way to distinguish signal from noise in a vastly complicated and ML-enchanted algorithm.

I suspect it's a consequence of Google not caring whether or not you're actively subscribed to the newsletter, and treating it as spam based on all other signs.

Checkboxing "[ ] I want your newsletter" on some website doesn't inform Gmail to whitelist that site.

Reply on news.ycombinator.com