[ocrow]

The rate of spam to a form is roughly constant over time, whereas the rate of spam to a published email address goes up over time as the site is repeatedly scanned by spam robots and the address added to more and more spammer lists. While spam detection is good, it isn't perfect. As your total volume of spam goes up, so does the amount that sneaks through the filters. Additionally, at a certain point it becomes impossible to look in your spam filter for misclassified real email. Eventually you're overwhelmed and have to change emails. If you're going to publish an email address you have to consider it a burnable resource that you will replace once the volume of spam is too high.

If the author hasn't experienced this, I think it must be because they haven't done the exercise of leaving a live email address on a public website for years.

[squirrel]

My email address has been on my public website for at least 15 years, and my spam level is constant and manageable. That may simply indicate that I’m not popular enough to encounter the problem, of course.

[seabass-labrax]

Same for me - I do get spam, and it's frustrating, but the level seems to be broadly constant over time. I wonder if it's deliberate co-operation between scammers and other spam senders, perhaps in order to keep the total amount of spam just under the threshold that would cause people to actually crack down on the issue. Certainly, receiving only single-digit numbers of spam emails each day keeps it just slightly away from being my personal number one priority to get some proper filters installed.

[byyll]

> That may simply indicate that I’m not popular enough to encounter the problem, of course.

I get no spam and no non-spam email so that probably indicates I am even less popular.

[jltsiren]

In my experience, peak email spam was 15-20 years ago. At some point, I got ~500 messages/day delivered to my spam folder. Today the average for the same address is maybe 2 messages/day. Spam filters flagging legitimate emails as spam or not delivering them at all has long been a much bigger issue than any spam that gets through.

[domdomegg]

Thanks for reading the article! I agree it doesn't consider this point, and I actually hadn't thought of that.

Semi-empirically, I've run some websites with emails and contact forms sitting on them for 5+ years and I haven't noticed this effect. Although I must admit I haven't studied it quantitively well enough to determine this for certain - I'd love to look over the data to see if this is true. Unfortunately on all these inboxes spam is deleted automatically after some time so I no longer have records. If you do have data here, it'd be great to see someone publish this and would happily add a link to this analysis!

And theoretically, would a contact form link not also be a thing that gets added to more and more lists over time and have the same problem? (Although I also didn't notice this pattern on contact forms, so I'm not claiming this does happen - just a thought experiment on this logic!)