Topic: Anybody want to get a collection of... (Read 1675 times)

Create offline wallet address. Copypasta wallet into file. Done.

Damn. We really need to do this as a free service for the community.

What about the private keys leaking as has been discovered with the encryption bug ?

You can embed anything in any binary file. You can even do it with a simple DOS command.

Bittorrent may be a bad protocol for this, because the torrent can't be changed for more people to join in or add additional keys.

Man, I really need to create a protocol where you can pay others with BTC for software tasks like storing or relaying data, or offer your computer for others at BTC fees. Every day, I see a new usage scenario for it. I already have a madly awesome design, but some of the hard parts are missing; especially, I could really use an established .NET/Mono BTC Client library.

Maybe I can set the project in motion once Bitcoin software support has matured a little more. There's a lot of unused potential in the "offer your computer's resources for BTC" business model, not just mining.

What if nodes arrange themselves in pools in sizes such as 100. To back up your wallet, you must run the software and it backs up 100 other wallets onto your disk. That way there is a motivation to run the software- you get to back up your wallet to all the other nodes as well. Also, this enables us to create a mechanism to delete old information- if a node doesn't sign on for an entire year for some length of time, then everyone will delete his wallet from their disk.

Oh! What if individual nodes connect to other nodes on an individual basis and then trade data, and mutually store each others data. In order to recover your data, you will have to provide the other node with theirs at the same time to prove that you kept it. Or how about this even better way to prove that you are keeping data: one node sends the other two values, such as 5 and 872, and the other has to send the SHA256 hash of the data between bytes 5 and 872. This method can also be used to verify that the nodes you think are holding your data really are. This would not rely on pooling like the above method does, would be able to store an arbitrary amount of data, is decentralized, and provides for a variable amount of redundancy- if 5 nodes are enough for you then you only store 5 times as much data. If you want 50 copies out in the cloud, then you can store 50 units of other people's data. That will be up to you. I don't know how a node could relocate peers... perhaps with each peer maintaining a list of unique identities along with IP addresses and exchanging them similar to the way bitcoin exchanges addr messages?

Yes, I think this would work. Upon starting your client, it would bootstrap and download addr messages (which contain IPs and Unique IDs) and connect to them until one said it was willing to trade the proper amount of data. You then have one partner. Once you have the proper amount of partners, you stop and only give the addr messages to the new people who show up. Whenever a host disconnects and then reconnects the next day with a different IP, they broadcast their new addr message and all the other nodes relay it (which also contains their Unique ID which is unchanged). Whichever partners are looking for this peer will hear the addr message and thus learn of its new IP.

Or we could just scrap the whole addr message thing and as long as only one of the nodes has their IP address change, then they'll be able to reconnect as long as they both have their firewalls forwarded properly.

Although I guess this wasn't what the OP was proposing which was being able to show up in a different country with nothing but the clothes on your back and a password in your head and be able to recover your Bitcoins. Oh well.

The wallets are pretty negligible in size also...checking mine currently, only ~2.3 MB.  Traffic would/should be manageable.  Could easily develop a p2p network to house a database of the wallets, I would think.

Im sure there is technology out there that can make all this very easy yes.

Not quite sure how to go about this but i can throw up my server for this if anyone takes on the project.

EDIT: Maybe we can use namecoin for the domain name, just incase bitcoin gets made illegal and our domain name is seized

And we could write open source software which would run on the servers of volunteers. The various servers replicate data to each other for when volunteers drop out. Optionally, the people running the servers could collect the 1 BTC fee when the person wants to do a recovery. Recoveries during the first 30 days should be free so that people can experiment and build confidence. Each person running a server would be given a subdomain name of the primary.

For example:
a.encryptedwallets.com
b.encryptedwallets.com
c.encryptedwallets.com
etc...

So that when you want to recover your wallet, you just trial-and-error domain names until one works. I suppose someone would need to be in charge who owns the primary domain name and lends out subdomain names.

Or it could be a non-web-based stand-alone program. Wouldn't that just be a distributed database? Don't these things exist?

Cant we just drum up a site that holds wallets? if they are all encrypted should be fine

Only catch is it cost 1BTC to get it back from the cloud 

Im sure will be worth if you just lost that one that has 2.5k BTC in it haha

Putting one's wallet(s) in the cloud is a good, but rather obvious idea.  They are then available from anywhere that one might find themselves even if one has been stripped of one's belongings.

Grouping the wallets with other peoples is also a good and rather less obvious idea.  If, as a for-instance, I am on the lamb and my wallet gets pulled out of an obscure corner of the cloud, that is an instant lead.  Of course I could (and would) be clever enough to make that a false lead, but perhaps not everyone would.  Grouping the wallets into a batch with a lot of others would create a lot of dead-end leads.

Probably the best way to achieve the goal of having one's wallet.dat file(s) available would be to hide them in images which are chronically found via Google searches, or simply put the images on a popular website if one is available.  Then random people from all over the world would be downloading the wallet.dat files regularly and one could do it without being suspicious.

It goes without saying (or should go without saying) that the wallet.dat files would be encrypted.  Most stego programs probably do a serviceable job at that anyway, but I tend to prefer openssl since the cryptography is more widely studied.

Is it possible to embed an encrypted file in the header of a movie file?

I actually like this idea.  It seems intuitively awful at first, but the more I think about it, the more I like it.

Don't backup your spendy wallet like this. In the extreme it only needs one key, more is better so the 99% doesn't lynch you when they see your total, but whatever.

A savings wallet might never be used 10 times in a lifetime.

At first I thought this was 100% bad.

But maybe not. If it's a bunch of wallets together lots of people have incentive to keep the file available, sort of anyway...

But I wouldn't count on it and it seems better to put it in more reliable storage.

Now if you could make the wallets not-easily-separable from some desirable content you'd probably be golden but it hardly seems like the most efficient way.

Well seriously.. rather than just describing the method of execution of whatever monstrosity this is, perhaps he could describe the intention?

If it's for a group of people to back up wallets - it sure seems batshit crazy to me.
Wallets as I understand it produce roughly 100 extra private keys at a time by default - so what is going to happen here when some users do a lot of transactions and want to back up a new wallet?
There will be multiple versions that need to be seeded and various people clamouring to add theirs to the set..  etc etc, and some people who were reliably seeding earlier versions decide not to because they don't use those old ones any more but suddenly someone is hunting for it..

I'm no p2p expert (maybe that's obvious) but it sure sounds like a 'hey.. let's herd cats for fun and profit'  post to me.

I lolled.

It's not a bad idea. Many people store their wallets in the cloud. It might cause a handful of non bitcoiners to ask questions, but I doubt they would find anything good about bitcoin.

^^ lol wtf

Are the words 'idea' and 'brainfart' synonymous for you?

Anybody want to get a collection encrypted wallets together and put them on a P2P torrent on PirateBay? Just an idea.