Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I have a genius idea how to make a p2p pseudo-twitter which I have shared with only one person yet, but why not share it now on HN (maybe someone can build it instead of me, since I don't have the time at the moment because of school stuff).

Basically, what you need to solve in something like p2p twitter is: how to authentificate the user, how to send the message around, how to archive the message that it is not lost.

Now the genius idea: all of this is exactly what bitcoin solves.

Bitcoin already has mechanisms for authenticating. Bitcoin already has mechanisms for sharing messages (altough it is called "transactions"). Bitcoin has a mechanism for archiving those messages (it's called "blockchain").

And the kicker - bitcoin already has an ability to enter arbitrary data into the transaction. (sure, not much, but about the 160 letters tweets have.) You just add it into the script and add OP_DROP; since the transaction including script is signed, noone can just remove it.

My idea was to hijack the already existing bitcoin infrastructure and hack Twitter-like P2P messaging thing on top of it; let's call it "bitmessage".

How would sending a "bitmessage" look like? Well, you can, for example, send money from your address to the same address (send money to yourself) and add the wanted string into the script.

You can then look for a message of a given user (if you take bitcoin address as an ID of the user).

Now of course, this wouldn't be free for long (if it catched up). Right now, bitcoin miners take just anything the P2P network throws at them, but if the network was flooded with people sending money to themselves with a message in script, they would just throw these away so it doesn't enlarge the already giant blockchain. However, you can add fees to any transaction. If you add fees high enough, some block miner will eventually take your "bitmessage" and add it to the blockchain.

Would the resulting thing be equivalent of twitter? No, it would be slow (it takes about 10 minutes to get transaction to blockchain) and it would most likely not be for free (if it catches on, miners will throw away the messages with low fees). On the other hand, every message would be recorded forever, without any chance of anyone deleting it, censoring it or getting your IP address. Also, the Bitcoin seems to be going strong and doesn't seem like it will go away any time soon, and the infrastructure is super-strong.

I want to hack a prototype... maybe... one day.



Just a tip. When you start a sentence with "I have a genius idea" people will not think it's a genius idea; or at least judge your comments to an unreasonably high standard of your own creation.


well, I didn't mean it that seriously, but thanks


That is a great idea--it's funny because I was thinking along similar lines where bitcoin has the blockchain and maybe a similar scheme could be used to have a central, p2p verified stream/history of tweets.

I was thinking the downside would be that the blockchain would end up being HUGE if it got even 5% of the traffic that twitter sees every day. That said, maybe you could get around it by breaking the chain into randomly distributed sub-segments or something like that. Maybe I'm overestimating the storage requirements and it wouldn't be so different than what bitcoin has to deal with already. And ah yes, the miners... Not as much motivation for miners to verify transactions (or posts or whatever) with a service like that, and kind of a pita for the users if they have to pay something for each tweet... then again, would probably cut down on twitter bots/spam...

It seems to me that a project like that would be really difficult as Twitter has had enough problems scaling an infrastructure they control 100%. But yeah, I think there's a lot that could be learned from bitcoin to do something like that.


About the size of the blockchain: yeah, it would probably explode. On the other hand, if the dreams of BitCoin proponents are to come true and BitCoin will become the new online currency it wants to be, it will explode anyway.

About the scaling problems.... the bitcoin ecosystem is altogether the biggest supercomputer ever. I think it will handle it somehow.

Of course, the problems like searching in the blockchain for a given address would still be there. But again, thin clients (and, therefore, a slight centralization) are the future for bitcoin anyway.


There are extensions to "Namecoin", a bitcoin alternate chain, that does things like this. For example: http://dot-bit.org/forum/viewtopic.php?p=2273




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

Search: