4 ms·
Ask HN: Guarantee an email is HUMAN with a CRYPTO-approach
I am tired of spam. Can we create a protocol where in the header there is a X-Human-Blob and X-Human-Hash
If the Sha256 of the email content together with the blob start with Y zeros, then the receiving email client knows the creator of such email INVESTED energy to create the email. Each receiving email client can demand different number of zeros as compute becomes cheaper.
That way, when I send a cold email to someone in which the X-Human-Hash starts with a 1000 zeros is IT CLEAR THAT IT IS A F-ING IMPORTANT EMAIL!! NOT SPAM. This would important for legal teams, engineering teams, etc.
It is the 21st century, I have a SaaS that makes $1.5m per year, I live in a super advanced nation, use best practices, and I CAN'T GUARANTEE an important email INBOXES!? How did we get here.
Thoughts?
If you wanna work on this with me email me at somid3-at-gm@il-D0t-c0m
- sebmellen 5y agoI know the folks over at Vereign https://www.vereign.com/ https://www.vereign.com/ are working on something similar to this. Very smart people!
- somid3 5y agoAnd they are a decade late, this should have come out back in 2010 when crypto became a thing
- schoen 5y agoThis idea was proposed in almost exactly this form by Adam Back and called "hashcash", predating blockchain-based cryptocurrencies: https://en.wikipedia.org/wiki/Hashcash https://en.wikipedia.org/wiki/Hashcash
- somid3 5y agodo you wanna build an extension to solve this problem wiht me?
- somid3 5y agobasically when you send the email it adds the blob and hash under the email message. When you receive an email the extension checks to see if it contains that info and it gives the reader a % likehilhood this is a human
- schoen 5y agoI'd encourage you to read about the history of hashcash, because there were various debates and implementations that might inform your ideas. I think it's a great idea at one level, but a huge practical problem is that given people's different levels of access to hardware resources, it's difficult to tell how much a given sender actually "spent" to produce a particular hash -- which is made even more complicated by botnets (illicitly spending other people's computing power) and dedicated hash hardware (investing capital to gain a huge advantage in hashrate compared to an ordinary PC).
- schoen 5y agoI would like to add belatedly that I'm glad you're interested in this and don't want to discourage you from researching this area; I just think you should read the hashcash/rpow history including problems that people ran into, because -- like most things having to do with spam -- I think it turned out to be a much thornier issue than anyone expected. So I think we don't have a clear path to a solution based on this research at present, because of things like spammers buying what we would now call mining hardware in order to outcompete regular people using a desktop PC, or spammers buying time on botnets for the same reason. But, it's very worth thinking about whether there are approaches that don't have this problem. In a way, that question is parallel to the ongoing debate in cryptocurrency mining about whether there's a way to do a mining-based blockchain that will continue to be decentralized and involve regular people's devices somehow, and that won't even up giving an advantage to centralized miners who invest a lot of capital into server farms and custom hardware. I know one recent entrant in this area is Chia which uses a proof-of-storage mechanism; I don't know whether that actually succeeds in reversing economic pressures toward centralization, but it could be interesting to look into whether its structure and incentives could somehow be bootstrapped into a "apparently, someone probably dedicated some moderately expensive resources that are probably on an ordinary PC, for a moderate period of time, in order to send you this message". And again, without ultimately getting hijacked by botnets that can make third parties' PCs participate illicitly. :-(