Hacker News new | past | comments | ask | show | jobs | submit login
A killer app for Bitcoin? – Spam and Phishing Emails Killer (spamandphishingemailkiller.wordpress.com)
45 points by spek on Nov 21, 2013 | hide | past | favorite | 46 comments



The problem with this idea is that with current Bitcoin protocol implementation it will not work. As of Bitcoin 0.8.2 the network doesn't relay transactions with "dust" outputs - transactions that are sending 0.00005430 BTC or less [1]. This check was implemented, ironically, to prevent spam, because one could flood the network with one-satoshi transactions which could take much space and bandwidth.

[1] https://github.com/bitcoin/bitcoin/pull/2577


As far as I understand, sending amounts smaller than the dust cutoff does not violate the network protocol. The difficulty is in finding miners who are using software that deviates from version 0.8.2 by not performing this check, and will include the micro transaction in blocks they mine. I'm only 85% confident about this though.


Yes. That's correct. You would've had to find a miner to include the transactions in the blockchain. However, in upcoming versions, setting output as OP_RETURN (which basically allows you to embed additional data) will be a standard transaction type, but immediately prunable. Hence, for the 'currency' part of Bitcoin these transactions won't form part of the bloat, but nodes that want to keep track of it, can (https://bitcoinfoundation.org/blog/?p=290).

Miners will still choose whether to include them or not (being a good citizen and adding a tx fee is still recommended), but now other nodes will relay them so they can find their way to a miner. Other Bitcoin developers can correct me if I'm wrong.


Wouldn't it be possible to implement an option to send a really small amount per wallet per week? This would prevent people from flooding the blockchain, but it would still allow to send small amounts.


There's similar ideas being thrown around in the Bitcoin community by using the network to create identities that is cheap for good users, but expensive for spammers, by sacrificing a portion of BTC to the miners of the network (and then showing verifiably proof). See: https://bitcointalk.org/index.php?topic=140711.0


Also see: https://en.bitcoin.it/wiki/Fidelity_bonds

Edit: Also, a comment I wrote a while ago with some more thoughts and information on fidelity bonds: https://news.ycombinator.com/item?id=6048650 (starting on the 3rd paragraph)


The idea seams really nice, really.

Being easily cynical, I will say that those who currently own Bitcoins are those who would make a buck if it was adopted, so it's not surprising to see that kind of idea there. But that's just me being cynical.


This is part of the reason bitcoin is succeeding. Early adopters become like investors. There never was an explicit quid quo pro, but overall people involved early in bitcoin ended up owning a bunch of bitcoin.

Anything that increases adoption increases demand for bitcoin and benefits current owners-participants. Cynical is a cynical way of describing this. Interest doesn't necessarily mean sociopathic. It's good for people to have an interest in making bitcoin successful.

Our current sensitivity to financial bubbles is confusing here. Currencies are basically always bubbles. They have value only because and to the extent that people believe they will have value in the future. To make a new currency, you need to make a bubble.

IMO, what bitcoin needs is exactly something like this: uses as a medium of exchange. A lot of the current interest/value is centered around bitcoin as a value store, if that's what we want to call the current speculative investing. To really make an impact as a currency, people need to be exchanging bitcoin for goods & services.


I call this realism.


Have a look at http://craphound.com/spamsolutions.txt and make sure you can rebut all the points :)


Specifically, "Requires immediate total cooperation from everybody at once"


This have been solved by Bitcoin. People driven by greed are cooperating on building one blockchain. Then, they'd be happy to pay pennies for safe, secure and clean email system.


> Then, they'd be happy to pay pennies for safe, secure and clean email system.

That's a pretty big leap of faith right there.

The truth is we already have a cryptosystem that is not only free, but always works. We have had it for years: PGP. Only the extremely savvy use it. We have a non-free one that does cost a mere "few pennies:" digital certificates. Only the insanely savvy use it.

The existence of a system does not imply its use. That should be common knowledge for people who hang around on a website that is mostly geared toward startups.


Cooperating on a single blockchain is beneficial from a purely functional greed-motivation; cooperating on a safe, secure and clean e-mail system is not beneficial from a purely functional greed-motivation.

See the difference?

"they'd be happy to pay pennies for [safer email system]" is an assumption that has been explored for years and is clearly completely false.


You don't need to "cooperate" on a email system in a way blockchain requires cooperation (with strict following on one protocol). People who need to transfer and store messages will pay for it. Bitcoin makes it easy and cheap. That's it.


People who need to transmit messages will not pay for it, because they can send email for free - in order for a spam-solution to be useful, you need cooperation from a critical mass of recipients that they will not accept or send "old free email" anymore.


( ) Public reluctance to accept weird new forms of money


Public, driven by greed, will also invest in Bitcoin sooner or later as a better "store of wealth" and eventually a more liquid currency. We see it happening in real time. Don't lie to yourself: everyone sees 10x or more increases each year in number of hands willing to hold bitcoins long-term and not throwing them out on every temporary "bubble".



Killer app for Bitcoin is Bitcoin itself. Digital gold that can't be easily confiscated, can be cheaply stored and transferred and resist a lot of censorship.

It can also encode various cool contracts that are impossible to do otherwise: http://blog.oleganza.com/post/58240549599/contracts-without-... (well, this requires a UI, but it's already supported by the network).


In what way is Bitcoin harder to confiscate than fiat currency (in, say, anonymous Swiss account) or gold (hidden in a secret location) ?

In all cases, you apply physical force or legal imprisonment until confiscation succeeds, it's that simple.


1. It's easier to hide even the fact that you have some Bitcoins. When you buy some Bitcoins for cash from random people on the street and keep in different wallets (or even "brain wallets"), no one would know how much do you have. In case of Swiss or any banking account, there's always a paper trace somewhere. And recently US was cooperating with Swiss to uncover all accounts belonging to US citizens.

2. Even if someone knows that you have some bitcoins, they don't know where they are stored and how keys are encrypted. "Civil forfeiture" or "freezing account" are very cheap and are used a lot. Irrelevant against Bitcoin savings.

3. Legal force is more costly than "civil forfeiture", but then you can always multiple ways to preserve the money: by sending it to someone else you trust, by locking it up in multisig transaction with someone you trust (who is outside jurisdiction), by saying "my hard drive has crashed" etc.

In other words, the cost for anyone to confiscate your coins is much higher than with paper cash and even higher than with bank accounts.


"It's easier to hide even the fact that you have some Bitcoins. When you buy some Bitcoins for cash from random people on the street and keep in different wallets (or even "brain wallets"), no one would know how much do you have"

...kind of like a drug dealer who has stashes of paper money hidden in various places around his turf.

"Even if someone knows that you have some bitcoins, they don't know where they are stored"

When the police make an arrest, SOP is to take all computers, all storage devices, and then look in places where tiny storage devices might be hidden. Sure, you could have buried your Bitcoin wallet in the woods, but this is not much different from burying some paper in the woods (yes, people do this).

"locking it up in multisig transaction with someone you trust (who is outside jurisdiction), by saying "my hard drive has crashed" etc."

...so basically the old trick of trying to send your paper money out of the country? Criminals try this all the time, and the police are not deterred by it.

The reality is that if you want to hide money from the government, Bitcoin has no particular advantage over paper money -- and the fact that your transactions are broadcast to the entire network actually makes it a bit harder. In any case, you also have ignored the greatest problem with off-the-books money: taxes. You can send your Bitcoin money through all kinds of convoluted paths, and then when you get out of prison and try to spend it the IRS will just swoop in and you will be re-arrested for failing to declare your income.


Paper stashes can't be put in DarkJPEG and uploaded on Facebook or Dropbox.

I'm not arguing that it's always possible to find stuff out. I'm saying the bar is being raised much higher. Today cops don't go to thousands of people to make a "haircut" on their bank deposits. It's done with a single button click and negotiation with a couple of bankers. With bitcoin you'd have to send around policemen to every house to confiscate stuff. When there's no good justification, people will strongly oppose that.

The reality is that if you want to hide from govt and govt wants to find you, then Bitcoin does not help you much (although it's certainly more mobile than cash). But if thousands of people are hiding from govt routinely, it's much more expensive for govt to go after all of them at once. See what happens with Bittorrent. Some people get caught or threatened, while thousands of others enjoy cheap movies.

Regarding last example: you underestimate the power of good mixing. You can swap coins anonymously so that all the "tainted" coins get distributed to thousands of different hands and never go back to you anymore, while you receive thousands of little coins back, from various sources, not tainted by whatever business you was doing before.


I'm not saying hiding BTC is easy now or will ever be cheap and simple, but it's certainly possible to automate and optimise security measures a lot and there will be a lot of apps doing exactly that. Right now the cost of taxation, inflation and bailing-in is very-very low and hurts most innocent law-abiding citizens (not everyone loves how the taxes are spent, but they are easy to extract anyway). With Bitcoin, huge chunk of money can safely sit in private wallets instead of bank accounts, thus making fractional reserve banking less relevant, bail-in becomes irrelevant too. Those who wish to not declare some income have much safer way to receive/send money, than with modern banking system. Most outstanding tax-evaders will be chased and caught, but 99% may safely enjoy their own little share of global black market without worries. They already enjoy local black market when they pay in cash for lots of things, but it'll expand with Bitcoin to the whole world.


"Addresses being flagged and becoming useless" is not a solution to spam, as it happens already currently.

In particular, even if everyone suddenly agreed to reject every mail without this deposit (which won't happen), the proposal allows for setting up a new address, making the (small) deposit, blasting a million emails before it gets blacklisted, and using a new address the next time.


> If you happen to think this is a good idea, please register your approval by making a tiny payment (eg one satoshi) to 1SPEKXiV6NF9Xg6Ridw2qUtV83T8TRJZZ. Larger donations accepted, if you like ;-)

This is a strange way to cast a vote. Nonetheless I doubt spammers would have any problem with creating bitcoin accounts with a small amount of money in them - one email/bitcoin wallet could send millions of emails before anyone takes action via the "sentinel" route.

And as mark242/craphound points out: "Requires immediate total cooperation from everybody at once"


Interestingly, some of bitcoin's roots originate from spam fighting. The idea of the 'proof of work' function that is used by bitcoin came out of the idea of Hashcash http://www.hashcash.org/ (as referenced in Satoshi's white-paper).

The hashcash idea wasn't so successful, although it's still around and somewhat useful. One problem of hashcash was that the difficulty didn't adjust, making it somewhat unfair, not taking in account GPUs, etc.

On the other hand, bitcoin solves the difficulty problem that was with hashcash. Perhaps in the future, bitcoin could be used for ensuring the smooth delivery of priority emails.


What if someone flags your emails as spam deliberately, just in order to cause harm?


Yup that's what comes to mind. Easy way to get a target's email marked as spam all around the world for a buck or two.


Interestingly, someone has proposed basically this idea in 1997 http://en.wikipedia.org/wiki/Hashcash

The idea is that sender has to do a non-trivial amount of computation to find a hash below a certain threshold for receivers to accept it; but the amount of computation would be low enough to not adversely effect day-to-day email transacting; something like 1/20th of a second worth of computation of a cpu would probably be about right.

As a side note, spam hasn't been an issue for me for a long time.


I like the idea but wouldn't that also make a lot of legitimate use cases of email too costly? Think of a service like MailChimp.com: They allow you to easily manage and send out newsletters. Computing the hashes for 5000 newsletter receivers would take approximately 5 minutes of raw CPU power.

Also: The CPU power needed has to increase every year because CPUs get better and better (and thus cheaper).

Then again: Let's assume Hashcash would eliminate spam or reduce it: This would probably mean a production boost (no humans have to detect spam anymore) and spam filters cost CPU as well...

In the end it may be a 0-sum game but reverse the burden. The more I think about it the smaller the potential problems seem to be...


Hashcash is actually used by some spamfilters, but not as a sole determining factor. For legit services like MaichChimp, they could be whitelisted (so always clear spam filters) or they can get a bonus to their rating (negative? whichever way would make it less likely to get marked as spam) and then determined to be spam on its contents rather than the absence of a Hashcash value.


I completely agree! I implemented Hashcash a couple years ago into my email wrapper and the algorithm is (in it's essence) suprisingly similar to Bitcoin: given this data, find a hash that has a consecutive number of x zeros in it.

I don't know about the 1/20th of a second though. Back in the day, the slowest Hashcash stamp with 5 ending zeros I could produce (with optimized code of course) was marginally above 1 second. But the good thing about Hashcash is that, much like Bitcoin - the difficulty scales. Just ask for more zeroes and it'll take longer. Email clients could also use the number of zeroes to help email classification: 0 zeros = SPAM, 5-8 zeroes = HAM.


HashCash is not surprisingly similar to Bitcoin; HashCash was the inspiration for Bitcoin. HashCash is one of the few examples of previous work that the Bitcoin paper bothers to cite.


I was going to post the same thing, have always like the idea of hashcash and surprised it hasnt recieved greater usage specifically for things like captchas although in the case it would be a reverse captcha :)

edit

Could potentially be used a a form of ddos mitigation if the browser was required to design a tiny bit of hash cash for each request, although I haven't really thought through all the details.


Junk mail was (is) common even in the physical world where there are printing costs and relatively high postage.


The problem as I see it is that, contrary to popular belief, sending spam is not free, so this merely increases the cost a little.

The difference between free and $n is greater than that between $n and ($n * 2).


This sounds very similar to Microsoft Research's concept 'Penny Black' - which, worth noting, was itself inspired by real postage history.

http://research.microsoft.com/en-us/projects/pennyblack/


Arguably a hacker would value a phishing E-Mail landing in a "trusted" folder more than I would want to pay when sending legitimate mails...

Of course, this proposal doesn't have a snowball's chance in hell, so arguing about it is equally pointless and I'm just too bored not to comment...


This is a solid idea, that has been suggested and talked about before ( on shows like letstalkbitcoin ).

However, the world should and most likely is looking to move away from insecure emailing. This is especially true for corporate and political figures.

Free software like bitmessage are revolutionary in nature and are surprisingly easy to use. Bitmessage also uses the bitcoin protocol to relay messages across the network in an untraceable & encrypted manner. A spammer would need your public key in order to send you anything. You can also block messages that do not originate from predetermined public key addresses.

The bitcoin protocol and blockchain have so many innovating alternative uses outside of finance and currecny. Also, bitcoin overlay protocols such as mastercoin are already gaining traction.


And for the lazy ones, here are the "votes/donations" on his Bitcoin address:

https://blockchain.info/address/1SPEKXiV6NF9Xg6Ridw2qUtV83T8...


Anyone care to explain to me why just having X wallets and tranfering ammounts from these wallets to X new ones isn't going to work? Just let some addresses get marked as spam then move to new addresses?


This is a cool idea, but one thing kills it dead.

Requiring the user to pay to report spam will not work. It is just one more point of friction. How many report spam right now in Gmail when it is free?


This sounds like an overcomplicated version of HashCash...which was the inspiration for Bitcoin in the first place.


I don't think spending a satoshi is a problem for any serious spammer.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: