Bitcoin as the anti-spam measure that hashcash aspired to be

[eMansipater]

So, here's a fairly undeveloped idea:  could bitcoin be used to address spam in ways the hashcash model couldn't?

The greatest problem I see with hashcash is that it, like most anti-spam mechanisms, failed to differentiate between spam and legitimate bulk email--the difference being that the latter is presumably desired by the recipient.  So why not actualise that consent as a transaction?

Here's the outline.  Small bitcoin payments to the email provider contained in a header automatically mark email as "not spam" (a simple registry allows email providers to post their public keys).  If the email is received as normal by the recipient, then after a set amount of time the money is returned to sender.  If, however, the recipient marks the email as spam/undesired, the payment is kept.  When a legitimate bulk emailer has a payment of theirs kept, they treat it as an unsubscribe request--this also tells them which email was marked as undesired if they use different addresses for each mailing.  The level of payment required is set by the email provider in order to create the desired equilibrium.  So, for example, you might offer a nospam.com email address that requires .1 bitcoins per email (extreme example) and nobody will send you bulk email unless they have a darn good reason to believe you want it.  On the other hand your hotmail.com address may only require .01 bitcoins per email and some bulk emailers might consider it worth the risk.

What do you think the problems and advantages would be for such a system?  One unique feature is that the more spam a provider has to deal with, the higher their reimbursement for it.  You can also cancel out the one-payment loss of an unsubscription neatly by requiring a subscriber to send you an email at sign-up (something true spam won't be able to do).  Spam statistics within the system also become public per-provider in the blockchain, allowing interesting metrics.  Potential downsides include the increased risk of someone using your email service (a webmail, for example) to send spam, but there are ways to mitigate this as well (for example, emails sent from a webmail get no payment sent with them and thus have to navigate the normal anti-spam mechanisms).

[genjix]

I like the idea, but it means that a) people need bitcoins b) I would need 2 emails; 1 for people using this system and 1 for the plebs that don't (I'd recommend them to switch to this).

But why wasn't hash cash ever implemented? Newsletters and mailing lists do count as spam to me, and I maintain separate emails for all that junk. Something like hashmail would be great for personal emails and spam free! (so I would never be worried about posting my email all over the net)

[Timo Y]

I have a more elegant solution, which would not only prevent spam but also DDoS attacks (both are caused by the fact that the internet is a commons).

Every IP packet is signed with the private key of the sender's bitcoin address.

Routers will only route packets as long as the packet's sender has a sent bitcoins to that individual router's bitcoin address, and as long as the balance remains positive.

Of course the whole process is automated and instantaneous.  So for example, if your packet get stuck somewhere in a router in UK on its way to Germany, that router in UK would request a few microBitcoins from your client, so that it can sent it to the next router.  Say that the same packet is also stuck in a router in Sweden, but that this router offers a lower bid for sending on the packet.  The client could then decide to pay the router in Sweden, and drop the router in the UK (for minimising price) or to pay both routers (for maximising latency and reliability).

Perhaps bitcoin transactions can be included in the packet directly for routers the client already trusts.

Another advantage of this scheme is that your bandwidth and ping would not be determined by your ISP contract. You can instantly purchase bandwidth on-demand, directly from world market of routers, and the limit would only be your physical connection.

The high liquidity of the world router market, as well as the increased competition and absence of contracts and middlemen, would also reduce everyone’s cost per mbit/s.

Only problem is that this would involve building an Internet v2 from scratch.  The bitcoin network would also need to be expanded to somehow allow instant transactions.

[error]

Another advantage of this scheme is that your bandwidth and ping would not be determined by your ISP contract. You can instantly purchase bandwidth on-demand, directly from world market of routers, and the limit would only be your physical connection.

Thus ends "net neutrality".

[error]

...
The greatest problem I see with hashcash is that it, like most anti-spam mechanisms, failed to differentiate between spam and legitimate bulk email--the difference being that the latter is presumably desired by the recipient.  So why not actualise that consent as a transaction?
...

My response would be: tough titty 'legitimate bulk emailers' find another medium for your legitimate communications or pay up. I bet most email recipients would agree they do not want those 'legitimate bulk emails' that much after all.

I think this might be workable if the recipient is the one who gets paid to receive the bulk email.

[eMansipater]

I like the idea, but it means that a) people need bitcoins b) I would need 2 emails; 1 for people using this system and 1 for the plebs that don't (I'd recommend them to switch to this).

But why wasn't hash cash ever implemented? Newsletters and mailing lists do count as spam to me, and I maintain separate emails for all that junk. Something like hashmail would be great for personal emails and spam free! (so I would never be worried about posting my email all over the net)

My response would be: tough titty 'legitimate bulk emailers' find another medium for your legitimate communications or pay up. I bet most email recipients would agree they do not want those 'legitimate bulk emails' that much after all.

I think this might be workable if the recipient is the one who gets paid to receive the bulk email.

Actually, only the service providers need to purchase a small store of bitcoins, equivalent to the amount of email they send during the designated waiting period.  I envisioned the system being overlayed on existing methodology rather than requiring the payment, at least initially--though nothing would prevent someone from going whole hog right away if they thought it was viable.  Same deal for paying recipients:  service providers are welcome to go for it if they want, or people who run their own email server.

I used to work in IT though, and imho the reason hashcash was never implemented was that the same people as myself making the technical call (running most organisations' email servers) found the cost-benefit analysis lacking.  Legit bulk email, which you should never be receiving unless you both signed up for it and are too lazy to click the unsubscribe button, does have a lot of good reasons for its use and most orgs send as much as they receive--why should they foot the bill for spam?

Also, changing something as huge as email is hard.  You need to be able to introduce it gradually and make sure it works.  An initial agreement between major players such as the top 3 webmailers or even a coalition of enterprises would give it a solid basis, and then it could grow based on their success.  There would be a strong incentive for it to expand to any organisation doing business with one that uses the method.  At any time before full implementation though, most people are going to want to receive email from people not on the system yet, so doing a full on requirement for specific emails is a bad idea across the board.  You also have the technical problem of how you send the payment in the first place--it's easier (and smarter) to have public keys listed per domain than per email addy.

Since the service provider can track which emails payments are sent too, this also means that "bait" emails could be given to people whose full-time job is to ensnare email harvesters.  Then they get paid by the spammer, via the email provider!  So we keep the blacklisting pros, just applying their expertise in a different direction.

As a complete aside, I think a technical application of Bitcoin in a practical scenario is one of the best ways to establish its value initially.  There's no need to build out the whole economy for something like this to succeed, and it's not dependent on the particular value of bitcoin--only that it has some value.  Perhaps someone reading this can come up with other applications too?


@forever-d sometimes the free market is a lousy solution to a technical problem.  Can you imagine the overhead of your proposed system?  You're also forgetting that botnets would merely cost their victims and DDOS all the same.  Not to mention, as error commented, the social implications.  Personally I think the one tier internet was designed just right.

[error]

@forever-d sometimes the free market is a lousy solution to a technical problem.  Can you imagine the overhead of your proposed system?  You're also forgetting that botnets would merely cost their victims and DDOS all the same.  Not to mention, as error commented, the social implications.  Personally I think the one tier internet was designed just right.

Huh? I didn't see anything particular to a free market in that system.

[grondilu]

Let's assume it's a good idea.

How exactly is this supposed to work?

Ok, so I have an email address that I publish while saying: "if you want to send me a message to this address, it will cost you 0.01 BTC.  I won't read any message from people who will not have paid."

How exactly is someone supposed to prove that he has actually paid for the message he has sent??

[eMansipater]

@forever-d sometimes the free market is a lousy solution to a technical problem.  Can you imagine the overhead of your proposed system?  You're also forgetting that botnets would merely cost their victims and DDOS all the same.  Not to mention, as error commented, the social implications.  Personally I think the one tier internet was designed just right.

Huh? I didn't see anything particular to a free market in that system.

I don't mean that the system was designed by the free market.  I mean that it would try to make packet routing into one.  More correctly I should have said "a free market".

How exactly is someone supposed to prove that he has actually paid for the message he has sent??

As a simple solution the sending person's software looks up the email provider's public key, then it transfers the amount, and references the payment in an email header.  The receiving server verifies that this payment has not been referenced before, and presto.  This is of course a degenerate solution:  there are much more optimal ways to do this.  But it gives you the idea.

[grondilu]

As a simple solution the sending person's software looks up the email provider's public key, then it transfers the amount, and references the payment in an email header.  The receiving server verifies that this payment has not been referenced before, and presto.  This is of course a degenerate solution:  there are much more optimal ways to do this.  But it gives you the idea.

Ok, that makes sense.

[Anonymous]

We  implemented micro transactions at witcoin.com

So far no spam 

If you want to send someone a message on the site you "donate" to them any amount you want , even .00000001  .

To reply the other person "donates" back to you.

See my sig for the link.

Spam is only spam when its unwelcome. Attaching a micro payment compensates my time for opening it and I dont mind so much.

[Timo Y]

@forever-d sometimes the free market is a lousy solution to a technical problem.  Can you imagine the overhead of your proposed system?  You're also forgetting that botnets would merely cost their victims and DDOS all the same.  Not to mention, as error commented, the social implications.  Personally I think the one tier internet was designed just right.

You're correct, sometimes a "centrally planned" solution is better.

Not sure about the social implacations. As it stands the internet is terribly unequal, bandwidth wise. The bandwidth you can get for a Dollar can differ by a factor of 1000 depending where in the world you live, sometimes even in the same country.

This inequallity is caused by too much centralisation and lack of competition. If everyone could be "their own ISP", as it were, it would even out a little.

[genjix]

centrally planned = facebook
federated/decentralised = email

[Anonymous]

witcoin is a great idea and service. If only it was decentralised as bitcoin itself.

You need somewhere to post the content and also enable the sending of tiny amounts such as .00000001   

[mizerydearia]

You need somewhere to post the content and also enable the sending of tiny amounts such as .00000001  

yes... you need it!  come to witcoin and get it! ... wait, what?  O_X

[hazek]

This is actually a really good idea but I don't think anyone mentioned the proper execution yet. I'm not a programmer so I don't understand how one could code it but I do think the logic behind such a system should be as follows:

If you want to send me an email I should have an option to charge you with a tiny fee for receiving your email and actually reading it. For example, right now if I don't know you and I have no business with you but you know my email you can send me whatever for free and I may or may not read your mail. This system has two problems: 1st I was waiting for an important email but it got buried under spam, or even got categorized as spam and I missed it wasting extra time while waiting and 2nd I didn't know the receiver beforehand but I really wanted him to read my mail even though he may think it's just spam.

Well how to solve this problem? Simple! Incentivize it with BitCoins.

Say I get the option to charge you for receiving your email, for allowing your email to land in my main email folder where I get all the important mails that I always read, you pay a small fee to me which then becomes the key to my mail folder. You still get the option to send a free mail but it will land straight in the spam folder. So if you pay me a small fee you get preferential treatment. And 2ndley if I have a small fee setup I am certain that all the important mail I'm waiting for will land in my main folder and there's no chance for me to miss such an email if the sender payed the fee. It's a win-win.


Basically the fee should open the door to my main email folder. If you don't pay, you don't get in and both parties have something to gain! And besides if you're just writing back and forth between you and your say business partner none of you will get any richer. The only time it will cost you is if you don't get a reply and the only time you'll gain something is when people email you stuff you didn't really want to receive and you won't reply.


That's my idea, now you figure out how to make it into a code

[eMansipater]

@hazek There are a number of reasons that the return of the payment is important.  For example, in your proposed system, how can the sender verify that paid emails are truly ending up in the main folder and/or being read?  In practice this end of things would be very difficult to differentiate from a get-paid-to-receive-spam concept and I'm pretty sure if that was a workable business model it would already be out there.

Second of all, there are a lot of situations where messages/replies between two people will be asymmetrical, yet it's undesirable for the frequent sender to be paying more.  Consider that maybe one person sends small, frequent emails while their friend or business acquaintance sends larger, more encompassing emails less often.  I'm sure you can think of more examples.

With the return of the payment automatically behind-the-scenes, most people don't have to worry about money until there exists some special circumstance that they themselves typically initiate.  And companies or individuals can decide for themselves how to treat an unreturned payment, so interactions like those you describe are still possible if desired.


It's also worth mentioning, as I realise I forgot to in my first post, that there may be a simple way to keep the myriad transactions involved from overwhelming the bitcoin network.  Instead of actually sending the payment with the email, the sender could simply send a signed, valid transaction that they themselves have not submitted to the bitcoin network.  Rather than needing to return a payment for every legit email, the sender would only have to submit the transactions from emails marked as spam.  I'm unsure, however, of the technical specifics for this.  How long would such transactions remain valid?  If it's a long time or indefinitely, then would the need for the sender to re-spend those coins in order to secure them wipe out the advantage, or could the latter be done in aggregate or somehow more optimally? You would also need to account for middlemen vandals since emails are sent in plain text.  Then there's the problem of someone potentially sending out a million conflicting transactions for the same small quantity of bitcoins, though this isn't as much of an issue as you might think since the first receiver who marks the item as spam by submitting that transaction immediately notifies all the others.  Perhaps with suitable inter-provider agreements this could even be the way the system functions by design since I suspect most non-bulk senders are almost never marked as spam by their recipients.  That could drastically reduce the amount of bitcoins most providers need to actually hold on hand to participate in the scheme.  I don't know the statistics on that though, and they would be very relevant.

[NghtRppr]

Another advantage of this scheme is that your bandwidth and ping would not be determined by your ISP contract. You can instantly purchase bandwidth on-demand, directly from world market of routers, and the limit would only be your physical connection.

Thus ends "net neutrality".

Good riddance. Drop all regulations and let competition thrive. Then anyone that tries to charge outrages prices for delivering packets will be routed around.