Email Stamps to Solve Spam | Hacker News Reader

Email Stamps to Solve Spam

adam.blog.heroku.com

5 pointssant0sk15 comments

Wow, it's been a while since a pay-to-send email mechanism has been floated about. They all seem to fail long before implementation, and for many of the same reasons.

One of the biggest reasons is that you can already get 99% of this effect for free today. There are a number of services and options where any email that comes from an unknown sender is quarantined by a system, which sends an email to the unknown sender. This email contains some instructions, perhaps to click a link, or solve a simple math problem, or some other such gating factor that makes it near impossible for the common spammer to invest the effort to circumvent the system in any reliable way.

And systems like qsecretary tend to provoke reactions like this:

http://www.ops.ietf.org/lists/namedroppers/namedroppers.2003...

Yes. Any mishandled whitelisting system is going to produce responses similar to the one in your link.

I imagine the author would be even more outraged if he had to pay 5cents multiple times to get past a poorly implemented spam block.

> perhaps to click a link, or solve a simple math problem

This approach has rather high false positive rate, i.e. it ends up filtering out something that's not spam. Lots of people loathe callback requests and never bother following up on them. This has little to do with how hard/simple the challenge is, and a lot with the fact that the recipient considers it OK to offload his mail handling headaches onto everyone else.

That's not to mention that this approach basically doubles the volume of spam traffic because From headers are routinely forged.

Honestly, even if this sorta works and you somehow magically get everyone into it, we had and still have spam in the form of physical mail. At 5 cents it sounds like email spam will still be cheaper than that. ...I am now very tempted to scream 'fail'.

Reply on news.ycombinator.com