Hacker News Reader

Comment by ajaygoel | Hacker News Reader

You're confusing a couple of issues. It's not that we're looking the other way. It's that I wanted to design a system where we don't have to look at all. GMass doesn't actually send any emails through its own servers. Emails are sent either through users' own Gmail accounts or through a third party SMTP service like SendGrid. Those actual email server operators (Gmail or SendGrid) have pretty sophisticated systems to detect spammers. THEY will shut down a user's account before I could ever even notice it. I don't have to look, because the owner of the email server will do the monitoring.

View on HN

> Those actual email server operators (Gmail or SendGrid) have pretty sophisticated systems to detect spammers.

Yes, and your product helps spammers bypass those detection systems. You even explain it in the post I linked:

> GMass will automatically send 500 emails/day or 2,000 emails/day if your email campaign has more than 500 or 2,000 recipients, respectively. You can also control, however, how many go out per day with the new Spread out setting under the GMass Settings arrow. If left blank, GMass will use 500 (regular Gmail) or 2,000 (G Suite) automatically, but you can override this by setting your own value.

And...

> It's not that we're looking the other way... we don't have to look at all.

That is what the phrase means.

The text you're quoting has to do with how we distribute volume, NOT how we enable spammers. High volume campaigns does not equate to spamming. You can send a 10-email campaign that IS spam and a 10,000-email campaign that is NOT spam.

I hate this argument. People always resort to this type of argument when they want to divert responsibility. Have you ever heard of how the environment enables certain behaviors?

Supporting a system where people can mass-send emails = enabling spam.

How absurd will you extend this argument? "Oh, yeah, 99% of my customers are spammers, but I'm not the one making them click send! They could've just paid, twiddled their thumbs and not do anything with my product!"

This is effectively structuring for email. Would it be immoral to sell a system that allows people to bypass money laundering restrictions by automating structuring? I don't see how this is any different.

Structuring email isn’t illegal.

If you found most of the campaigns fell in to the spam category would you shut down the business?

Maybe that's a convenient reason not to check.

> High volume campaigns does not equate to spamming

In what instances is high volume not spam when using this service? Given one legitimate use case where a customer isn't better off using a cheap MailChimp account.

I used to run ops for a mailing list with over 1 million subscribers (all double-opt-in — perhaps you should google that?).

We never had to bother avoiding any of the measures you deliberately subvert, because our mail was all legitimate.

Granted, doing a 'mail run' with that volume took about a week, but whatever.

If your (client's) recipients aren't all double-opt in, then what you're running could very easily be seen as irresponsible (and that's me being polite), no matter how much money it's making.

It is a spam tool plain & simple. Stop defending yourself and keep counting your money instead. Congratulations on your financial success.

I don't see how any of the citations you've presented incriminate GMass in helping spammers bypass their detection systems.

And the phrase "It's not that we're looking the other way... we don't have to look at all" simply means he's delegating spam detection to the Gmail/SendGrid platforms instead.

"(Gmail or SendGrid) have pretty sophisticated systems to detect spammers"

Yes, but, the evil genius of the idea is that the email will be from a gmail domain. Emails from gmail to gmail with your own 5 year old account will have a ridiculous conversion rate. In my tests with email, if the email is from the actual gmail domain other carriers TRUST it more. Spam systems rely heavily on DNS/IPaddresses -- and this bypasses all that.

I feel for the smaller email providers. Gmail wont be able to patch this anytime soon, thus, the smaller ESPs without resources will have to change their rules to make the gmail domain less trusted.

I don't know about other servers, but on the systems I manage gmail itself is the #1 spam source since 3 years at the very least. If you look at my previous posts, I commented on this two days ago.

There's literally _Zero_ value in DNSBL/DKIM/SPF because of this. Email sent from reputable sources is actually _more_ likely to be spam than small mail servers.

Content-based filters are the only thing that work against this sort of spam.

Following your logic, his idea wouldn't work, and he doesn't make 115k a month. But, clearly, it does [if he isn't lying] - so I am not following.

I'm not contradicting his assertions. I actually confirm them: there is now a lot of spam coming from gmail. He's clearly one of the reasons.

You are correct. :)

I feel uneasy with your reaction.

Maybe reducing the max number of recipients to the hundreds would help reduce the chance for abuse while still keeping your core customer base happy? It might also reduce the risk of getting booted off the Chrome Store.

You're assuming their "core customer base" aren't spammers.

Google keeps the number pretty low (2000 daily recipients per GSuite user. Less for gmail?)

True spammers are sending spam to many orders of magnitude more recipients. They are not going to use this tool.

My reaction was simply validating that sending from-gmail to-gmail has high open/conversion rates, as the poster stated. However, you can only do this with a limited number of emails (500/day for Gmail or 2,000/day for G Suite). That's not to say that Google will let ANYTHING be sent between Gmail and G Suite accounts. There are lots of cases where Gmail has blocked its own users from sending to each other.

From your documentation:

>We recommend that an unsubscribe link be added to all mass emails containing promotional or marketing content, but the decision is up to you.

An unsubscribe link is a requirement of the CAN-SPAM Act and the GDPR. Your own docs effectively say "We advise you not to use our product to break federal and EU law, but the decision is up to you". I can't speak for anyone else, but that doesn't really accord with my sense of ethics.

https://www.gmass.co/blog/unsubscribe-link/

Technically CAN-SPAM does not require an unsubscribe link. It requires providing instructions to be removed. That could be using the text, "reply back with the word 'unsubscribe'". We put the burden of legal compliance on our users, and the tradeoff is complete control over your message content.

That's like saying a cigarette company saying "We're not the reason lung cancer exists. We just create them, and if people choose to smoke it, that's their decision!".

Which is fine, but admit you have an immoral business, and are just doing this for lots of money. Don't justify anything.

It'd be more like a gas station saying that, not the cigarette company saying that.

I mean they’re not wrong... it’s like the hospital refusing to help you give birth because having another child will worsen the ecological catastrophe on Earth. That’s ridiculous. Ultimately, the decision is yours, as it should be.

If there are no people left on earth, who are we preventing the catastrophe for?

Aka I’m going to assume no liability

Huh. Does it matter if you shoot someone with a gun labeled "Gmail" vs. one named "SendGrid"?

Depends on sendgrid's perceived customer base.

To answer your analogy with another analogy: If you're gonna get in a shootout under questionable circumstances you'd be well advised not to pick the hundred dollar hi-point as your weapon of choice.

Reply on news.ycombinator.com