Pay a few BitCoins, get access to my main email folder and my prime reading time

[hazek]

I know there have been a few topics on using BitCoins to avoid spam and other meaningless and useless emails already but I just couldn't resists but make a new one just for my idea.

And it goes like this:

I want to be able to have a folder for receiving email that will charge a few BitCoins, maybe as few as 0.02BTC or eventually even less, to gain access to it. If you send me an email you get the option of either attaching those few BitCoins to it and your mail will receive special access and treatment by me or you can choose to not attach anything and your mail will get placed with the rest of the emails I usually receive but don't really pay attention to.

Also if I decide that the content was worth my time I spent reading I can just click a button and I refund you the price of my special attention. If I reply to you this happens automatically as I also want for my returning mail to land in your special folder.



Now I have no clue how to implement this service but I sure would use it. What do you think?


EDIT: Feasible implementation posted by ffe

[rebuilder]

I can't help but think if you're going to go to all this trouble, you may as well just require all incoming e-mail to be PGP-signed and only accept mail from people whose keys you know. Any mail signed with a key you don't know would go into a sorting folder you would periodically check for legit mails, adding the associated keys to your list of known keys.

Both methods have the problem that you have to convince people who send you mail to go to the trouble of using them.

[Jered Kenna (TradeHill)]

This has been talked about a little. I'd link you the post but I can't remember where. Something about using btc to lower spam.
I'd track it down but I have to get off here right now. Search around and you should be able to find it.

I like the idea though.

[ffe]

Here’s a use case: The client creates a special kind of transaction that you could attach to the email. This transaction transfers a certain amount of Bitcoin from your public key to the recipients public key. This transaction is not published and therefore not yet final. Furthermore, it has a time to live, say 48 hours.

At the receiver the software checks the validity of the transaction. If it is valid and the money is there in Bitcoin it presents the email to you. If you like the email you do nothing and the transaction expires. No funds transfer. After the transaction expires the sending wallet deletes its copy of the transaction and credits the sender with the coin.

If you don’t like the email you click the button and the coin is collected. (The transaction is published and bundled into the next Bitcoin block.) The sending wallet sees the transaction and makes the deduction of funds from the wallet permanent.

This is like a check in the mail. It’s safe because only the recipient public key can collect it and it can’t be lost because of the expiration date.

[hazek]

I can't help but think if you're going to go to all this trouble, you may as well just require all incoming e-mail to be PGP-signed and only accept mail from people whose keys you know. Any mail signed with a key you don't know would go into a sorting folder you would periodically check for legit mails, adding the associated keys to your list of known keys.

Both methods have the problem that you have to convince people who send you mail to go to the trouble of using them.

The difference is that I financially benefit if someone decides to spend money on my prime email reading time. Remember it's my choice to refund and I don't have to do it if I think I want my reading time back. That's the main difference between using BitCoins and any other filter method.

[hazek]

Here’s a use case: The client creates a special kind of transaction that you could attach to the email. This transaction transfers a certain amount of Bitcoin from your public key to the recipients public key. This transaction is not published and therefore not yet final. Furthermore, it has a time to live, say 48 hours.

At the receiver the software checks the validity of the transaction. If it is valid and the money is there in Bitcoin it presents the email to you. If you like the email you do nothing and the transaction expires. No funds transfer. After the transaction expires the sending wallet deletes its copy of the transaction and credits the sender with the coin.

If you don’t like the email you click the button and the coin is collected. (The transaction is published and bundled into the next Bitcoin block.) The sending wallet sees the transaction and makes the deduction of funds from the wallet permanent.

This is like a check in the mail. It’s safe because only the recipient public key can collect it and it can’t be lost because of the expiration date.

Nice! Perfect logic for my idea. I like it.

[rebuilder]

Perhaps you'd like to compare your idea to this checklist:
http://craphound.com/spamsolutions.txt
I don't feel snarky enough to fill it out myself.

So, feel free to try to implement this. You have a huge hurdle in getting people to adopt this, I think any other concerns pale in comparison to that.

Now, a question: Your scheme seems to require the user to read every e-mail they get in order to choose whether to refund or not. Is this a good idea? I know I'm a lazy e-mail reader; I think a lot of people would be pissed off with me for getting their bitcents stuck in limbo because I neglect reading their mails.

[casascius]

I would call the idea doomed for one reason only: won't be long before transfers of 0.02 BTC are infeasible because there's no space in blocks for them.  Even if there was, why would I and every other BTC user want to download 1 KB of permanent cruft in our block chain for a very long time just so one person could to avoid reading 1KB of spam.

Bitcoin microtransfers are unsustainable with the current design - it just so happens they're feasible for the moment due to a lack of other activity.  It would be safe to assume they are pretty much infeasible for long-term purposes now.

[ffe]

Now, a question: Your scheme seems to require the user to read every e-mail they get in order to choose whether to refund or not. Is this a good idea? I know I'm a lazy e-mail reader; I think a lot of people would be pissed off with me for getting their bitcents stuck in limbo because I neglect reading their mails.

Nah. If you ignore the email the transactions expire. If you do nothing the transactions expire. Nothing gets stuck in limbo. The coins are either claimed through an active act or they revert if ignored.

[hazek]

I would call the idea doomed for one reason only: won't be long before transfers of 0.02 BTC are infeasible because there's no space in blocks for them.  Even if there was, why would I and every other BTC user want to download 1 KB of permanent cruft in our block chain for a very long time just so one person could to avoid reading 1KB of spam.

Bitcoin microtransfers are unsustainable with the current design - it just so happens they're feasible for the moment due to a lack of other activity.  It would be safe to assume they are pretty much infeasible for long-term purposes now.

If that's the case then BitCoins has such a huge flaw from the get-go that it'll never see any major success. I hope you are wrong though.

[Jered Kenna (TradeHill)]

I would call the idea doomed for one reason only: won't be long before transfers of 0.02 BTC are infeasible because there's no space in blocks for them.  Even if there was, why would I and every other BTC user want to download 1 KB of permanent cruft in our block chain for a very long time just so one person could to avoid reading 1KB of spam.

Bitcoin microtransfers are unsustainable with the current design - it just so happens they're feasible for the moment due to a lack of other activity.  It would be safe to assume they are pretty much infeasible for long-term purposes now.

If that's the case then BitCoins has such a huge flaw from the get-go that it'll never see any major success. I hope you are wrong though.

I suppose it depends if you put a fee on those transfers. even if half of the  .02 btc went to the one that solved it then it's not  a hell of a lot though I guess.
I'm going to search microtransfers and see what comes up.

[casascius]

I suppose it depends if you put a fee on those transfers. even if half of the  .02 btc went to the one that solved it then it's not  a hell of a lot though I guess.
I'm going to search microtransfers and see what comes up.

Even if you put a fee on them, that doesn't mean they'll get accepted.  Fees are to compete for a limited amount of block space.  The block space is still a limited resource, paying a fee doesn't change that.  So long you wish to create more transactions than would fit in blocks (which this arrangement would be sure to create), the transaction fee you have to offer would rise to make such transactions no longer worthwhile.

[Jered Kenna (TradeHill)]

I suppose it depends if you put a fee on those transfers. even if half of the  .02 btc went to the one that solved it then it's not  a hell of a lot though I guess.
I'm going to search microtransfers and see what comes up.

Even if you put a fee on them, that doesn't mean they'll get accepted.  Fees are to compete for a limited amount of block space.  The block space is still a limited resource, paying a fee doesn't change that.  So long you wish to create more transactions than would fit in blocks (which this arrangement would be sure to create), the transaction fee you have to offer would rise to make such transactions no longer worthwhile.

So how do you fix it? Not trying to be a smartass honestly interested. I'm sure it's been talked about though so I'll save some space and search.

[casascius]

So how do you fix it? Not trying to be a smartass honestly interested. I'm sure it's been talked about though so I'll save some space and search.

It's not broken by design.  The way Bitcoin will have to handle micropayments in the future will have to be through Bitcoin banks.  There is a very specific number of transactions per minute that Bitcoin can sustain network-wide.  It is probably in the hundreds.  It is far smaller than the number of important e-mails that go out per minute.

Bitcoin transactions aren't "free" in a resource sense: they cost the disk space of ALL bitcoin users to store the transactions (some forever, and in the current implementation, everyone forever), as well as the bandwidth of ALL bitcoin users to relay them over the network.  If the average size of a transaction is 1 kilobyte, I simply don't have 1 kilobyte of internet bandwidth to download and relay the transaction each time someone spends 0.02 to send an e-mail, neither do you.  You'd need dedicated fiber to keep up.

You might remember the peer to peer network Gnutella.  WHen it first came out, it was great, soon it became so popular that nobody could use it, downloads of popular files were often impossible, it was crushed by the weight of its own popularity.  Bitcoin uses the transaction fee to avoid that.  You are competing for the available transaction slots which are very finite.  When there's more transactions then slots, some transactions won't make it into blocks.  Only the transactions important enough to compete for those limited slots (read: the payer is willing to pay a big enough transaction fee, and it won't be a penny) will make it into blocks.

Millions of e-mails go out every minute, millions of BTC transactions vastly exceeds the current peer to peer network's capabilities, this transaction volume with the current client with the internet as we know it today is impossible.  Plus we'd have an unsustainable growing block chain nobody will be able to download because it will grow faster than anyone can possibly keep up with.  Who would want to use Bitcoin if the starter block chain won't even fit on their 2 Terabyte hard drive?  That would take all day just to copy from one drive to another, let alone download.

That's why there's a limit.

The bitcoin network as we know it, when transaction volume picks up, will only be suitable for large transactions or interchanges between Bitcoin "banks" (not banks in the usual sense, but rather, sites like MyBitcoin.com, who privately handles small transactions so they don't consume resources in the block chain).



Hope this helps.

[hazek]

The bitcoin network as we know it, when transaction volume picks up, will only be suitable for large transactions or interchanges between Bitcoin "banks" (not banks in the usual sense, but rather, sites like MyBitcoin.com, who privately handles small transactions so they don't consume resources in the block chain).

What? Are you serious? So it's going to be a decentralized currency but in the hands of "banks" and normal ordinary users wont even have their own wallets?! If that's the future of BitCoins I'm sorry but I don't like it and I wont use it.

I mean the way I envisioned BitCoins being used way down the road is that maybe at an exchange rate of $50k/BTC millions of users would constantly create very small transactions of no more then say 0.00005BTC. If that's not going to be possible by design then I can't really see it every getting that big.

[eMansipater]

This has been talked about a little. I'd link you the post but I can't remember where. Something about using btc to lower spam.
I'd track it down but I have to get off here right now. Search around and you should be able to find it.

I like the idea though.

Here you go.

Here’s a use case: The client creates a special kind of transaction that you could attach to the email. This transaction transfers a certain amount of Bitcoin from your public key to the recipients public key. This transaction is not published and therefore not yet final. Furthermore, it has a time to live, say 48 hours.

At the receiver the software checks the validity of the transaction. If it is valid and the money is there in Bitcoin it presents the email to you. If you like the email you do nothing and the transaction expires. No funds transfer. After the transaction expires the sending wallet deletes its copy of the transaction and credits the sender with the coin.

If you don’t like the email you click the button and the coin is collected. (The transaction is published and bundled into the next Bitcoin block.) The sending wallet sees the transaction and makes the deduction of funds from the wallet permanent.

This is like a check in the mail. It’s safe because only the recipient public key can collect it and it can’t be lost because of the expiration date.

Yup, this is the method offered in that thread, though I don't think we ever got an answer to the question of whether unsubmitted transactions expire on their own within the bitcoin protocol--can someone more familiar with the guts answer that?

@casascius and hazek I'd say the scenario is a little bit better than that, or can be with a little work.  But as an upper limit, if a transaction is only processed when someone clicks the button the volume here wouldn't be so bad.  Consider that even if you read email 24 hours a day, how often do you normally have to click "report spam"?  Multiply that by 7 billion, and that's a pretty decent upper limit.  Remember--this could be a complement to existing filtering, at least at first.

One of the things to note about spam is that the presence of an effective, non-probabilistic solution for it would reduce the amount of spam sent in the first place, thereby reducing the overhead of fighting spam itself.  The equilibrium for the scenario would rest pretty heavily on the side of minimal undesired email.  When you start to compare the remaining cost with the amount of resources many organisations already spend fighting spam, the proposition starts to look very attractive even if you do pay a small slice off the top to miners.

[casascius]

What? Are you serious? So it's going to be a decentralized currency but in the hands of "banks" and normal ordinary users wont even have their own wallets?! If that's the future of BitCoins I'm sorry but I don't like it and I wont use it.

I mean the way I envisioned BitCoins being used way down the road is that maybe at an exchange rate of $50k/BTC millions of users would constantly create very small transactions of no more then say 0.00005BTC. If that's not going to be possible by design then I can't really see it every getting that big.

It has nothing to do with the transaction size in Bitcoins... it has to do with the transaction size in kilobytes and how many total transactions the system allows per block (more related to the transaction count, or the number of kilobytes, not the number of BTC)... Bitcoin will have no problem being be scaled.  No need to panic.

In the hands of "banks"... the difference being that anyone can start their own bank, just like anyone can start their own mining operation, both would simply require an investment in computing resources.  What's gone is the monopoly.  

Sure, people can still create small transactions.  They're just not going to be able to do it with Bitcoin.exe version 0.3.20 or have it added to the block chain.  If I want to send you 0.02, my choices might be: free if I send it to your mybitcoin.com address through mybitcoin.com, or I might have to pay a BTC 0.20 fee to make it go through the block chain if I want the transaction to clear anytime soon.  On the other hand, mybitcoin settles the micro transaction locally for free (and probably faster too) without consuming anyone else's bandwidth or disk space.

Think of blockchain transfers like the way bank wires are today: wires are fast, convenient, but don't make sense for buying a sodapop (fees $20+).  They make sense if you're buying a house, the Fedwire network is the current banking system's payment network backbone: only meant for transactions important enough to merit a $20 fee.  It wouldn't work if every sodapop were paid for with a bank wire.

You'll still be just as able to have your own wallet.  Imagine the fee is 0.20 BTC.  The point of the fee (besides encouraging mining) is to separate the transactions that deserve the limited slots from those that don't.  That fee is sensible if you want to put 100 BTC in your personal wallet (way better than an ATM fee), but not if you want to dispense 2 cents each time you send an e-mail.  This is the fee doing its job prioritizing access to a finite resource.

[casascius]

Yup, this is the method offered in that thread, though I don't think we ever got an answer to the question of whether unsubmitted transactions expire on their own within the bitcoin protocol--can someone more familiar with the guts answer that?

This idea (I have brought up before) has Satoshi's condemnation.  Per Satoshi it's not acceptable to have a transaction expiration time because it interferes with the network's fundamental need to be able to reorder transactions in the block chain if needed. If part of the chain must be chopped and the valid transactions brought into the surviving chain, it's the same as them being "submitted" again.  This could cause a transaction to retroactively "expire" which would invalidate and undo all the downstream recipients of the coins.  It is the consensus of most everyone involved in maintaining the project that this is unfair and unacceptable, so it's pretty much a doomed idea.

[eMansipater]

Yup, this is the method offered in that thread, though I don't think we ever got an answer to the question of whether unsubmitted transactions expire on their own within the bitcoin protocol--can someone more familiar with the guts answer that?

This idea (I have brought up before) has Satoshi's condemnation.  Per Satoshi it's not acceptable to have a transaction expiration time because it interferes with the network's fundamental need to be able to reorder transactions in the block chain if needed. If part of the chain must be chopped and the valid transactions brought into the surviving chain, it's the same as them being "submitted" again.  This could cause a transaction to retroactively "expire" which would invalidate and undo all the downstream recipients of the coins.  It is the consensus of most everyone involved in maintaining the project that this is unfair and unacceptable, so it's pretty much a doomed idea.

That makes sense.

[Anonymous]

1.Sign up to http://witcoin.com

2.Visit someones donate page http://witcoin.com/witizens/donate/2


3.Set a rate to send a message.







4. Profit!