bryant.coleman
Legendary
Offline
Activity: 3780
Merit: 1219
|
|
June 02, 2014, 08:17:53 AM |
|
Right now I am using USB offline storage. I am seriously considering shifting to paper wallet. But the problem is that I don't have a printer. So I will have to depend upon third party, which increases the risk.
|
|
|
|
Light
|
|
June 02, 2014, 10:02:51 AM |
|
Right now I am using USB offline storage. I am seriously considering shifting to paper wallet. But the problem is that I don't have a printer. So I will have to depend upon third party, which increases the risk. I would advise against that unless those paper wallets are BIP38 encrypted that you do not give them to someone else to print out. Unless it's your family/most trusted friend that your willing to trust your money with I would just be content with a USB stored version. If you have a pen and paper you could also just write it down the old fashioned way - sure it doesn't look so fancy but it really comes to the same thing.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
June 02, 2014, 10:33:09 AM Last edit: June 02, 2014, 11:19:37 AM by johnyj |
|
My current solution is electrum offline wallet installed on an Aegis secure key (Ubuntu as OS), so it is both physically and password protected and can do offline transactions when using the usb key to boot any computer with the network unplugged And I'm researching the most fundamentally secure way: Generate a private key using dice. This can ensure the security for most critical step - key generation. If you generate a key using a program, it is very difficult to examine the process to make sure it is back-door free (For example that program only generate 1 million different addresses, or generate same address due to a weakness in some of the library it calls.) The problem with a dice-generated hex private key is that you need to convert it to WIF format to be imported into wallet, and that involves some kind of calculation which is not straightforward and can not be done by a simple bash script without calling other external functions like SHA256 hashing http://www.swansontec.com/bitcoin-dice.html
|
|
|
|
Parazyd
|
|
June 02, 2014, 10:35:52 AM |
|
Paper wallets. The private keys are encrypted with a simple passphrase too
|
|
|
|
bryant.coleman
Legendary
Offline
Activity: 3780
Merit: 1219
|
|
June 02, 2014, 10:43:34 AM |
|
I would advise against that unless those paper wallets are BIP38 encrypted that you do not give them to someone else to print out. Unless it's your family/most trusted friend that your willing to trust your money with I would just be content with a USB stored version. If you have a pen and paper you could also just write it down the old fashioned way - sure it doesn't look so fancy but it really comes to the same thing.
Well... thought about it for a bit... and I have decided to hold on to my USB wallets. Everything is fine as of now, and I don't want to change the equilibrium. So, for now, it is going to be 100% USB wallets for me.
|
|
|
|
ranochigo
Legendary
Offline
Activity: 3052
Merit: 4443
Crypto Swap Exchange
|
|
June 02, 2014, 02:41:20 PM |
|
7 paper wallets, encrypted in a 14 Alphanumeric password stored inside a safe at 7 different locations.
|
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
June 02, 2014, 03:24:19 PM |
|
7 paper wallets, encrypted in a 14 Alphanumeric password stored inside a safe at 7 different locations.
You told me you didn't have any bitcoins.
|
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
June 02, 2014, 03:26:15 PM |
|
The problem with a dice-generated hex private key is that you need to convert it to WIF format to be imported into wallet, and that involves some kind of calculation which is not straightforward and can not be done by a simple bash script without calling other external functions like SHA256.
So you call an external function. Why is that a problem?
|
|
|
|
keithers
Legendary
Offline
Activity: 1456
Merit: 1001
This is the land of wolves now & you're not a wolf
|
|
June 02, 2014, 04:13:39 PM |
|
Blockchain.info and Coinbase.com don't really qualify as cold storage options, no?
|
|
|
|
|
zetaray
|
|
June 02, 2014, 04:54:03 PM |
|
I do not need a cold storage address yet. I do not own enough to worry about that. If I do, I think I will use electrum. I like the predetermined addresses and the ease of recovery.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
June 03, 2014, 10:50:52 AM |
|
The problem with a dice-generated hex private key is that you need to convert it to WIF format to be imported into wallet, and that involves some kind of calculation which is not straightforward and can not be done by a simple bash script without calling other external functions like SHA256.
So you call an external function. Why is that a problem? An external function might be planted with some back door, so that the WIF format key generated only have limited numbers, like millions of keys maximum, so even if you don't generate the same key twice, the hacker can easily scan those millions of keys and see if someone has put bitcoins in them
|
|
|
|
patricktim
|
|
June 03, 2014, 10:56:34 AM |
|
I use a USB key with a fresh Linux install + electrum, used online only to setup.
|
|
|
|
bryant.coleman
Legendary
Offline
Activity: 3780
Merit: 1219
|
|
June 03, 2014, 11:51:32 AM |
|
You told me you didn't have any bitcoins.
Lol.. that is the best thing to do here. Once I posted here that I was having some 4-5 coins with me. My inbox was flooded with emails from beggers very soon. So better keep silent about your stash.
|
|
|
|
jonald_fyookball
Legendary
Offline
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
|
|
June 03, 2014, 12:31:52 PM |
|
The problem with a dice-generated hex private key is that you need to convert it to WIF format to be imported into wallet, and that involves some kind of calculation which is not straightforward and can not be done by a simple bash script without calling other external functions like SHA256.
So you call an external function. Why is that a problem? An external function might be planted with some back door, so that the WIF format key generated only have limited numbers, like millions of keys maximum, so even if you don't generate the same key twice, the hacker can easily scan those millions of keys and see if someone has put bitcoins in them It's not that easy to hide a backdoor in something as a simple as a cryptographic hash function script. There are already public implementations of this and should be peer reviewed. But more importantly, you can easily verify it using a few random inputs and compare against a separate implementation! If the code looks normal and verifies successfully the hash outputs against other SHA-256 implementations, then it would be extremely unlikely a backdoor would be possible. If you are familiar with the avalanche effect where changing one character in the input changes the entire output completely, the same thing would happen if you tried to manipulate the output so your verification tests would fail. That's the beauty of the hash function.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
June 03, 2014, 02:57:22 PM |
|
The problem with a dice-generated hex private key is that you need to convert it to WIF format to be imported into wallet, and that involves some kind of calculation which is not straightforward and can not be done by a simple bash script without calling other external functions like SHA256.
So you call an external function. Why is that a problem? An external function might be planted with some back door, so that the WIF format key generated only have limited numbers, like millions of keys maximum, so even if you don't generate the same key twice, the hacker can easily scan those millions of keys and see if someone has put bitcoins in them It's not that easy to hide a backdoor in something as a simple as a cryptographic hash function script. There are already public implementations of this and should be peer reviewed. But more importantly, you can easily verify it using a few random inputs and compare against a separate implementation! If the code looks normal and verifies successfully the hash outputs against other SHA-256 implementations, then it would be extremely unlikely a backdoor would be possible. If you are familiar with the avalanche effect where changing one character in the input changes the entire output completely, the same thing would happen if you tried to manipulate the output so your verification tests would fail. That's the beauty of the hash function. Thanks, that is a very good point! If I can verify it gives a correct output for a defined input like other conversion tools, I can even rely on an offline webpage to do the conversion Adding this verification step will complete the whole process: Dice generate the key and script convert to WIF format, then import into any wallet to get address to send coin to
|
|
|
|
|
El Cabron
Gnomo
VIP
Hero Member
Offline
Activity: 840
Merit: 1000
|
|
June 03, 2014, 03:01:44 PM |
|
i gave mine to satoshi, if he cant be trusted, who can?
|
|
|
|
Parazyd
|
|
June 03, 2014, 03:03:37 PM |
|
i gave mine to satoshi, if he cant be trusted, who can?
I can be trusted. Or maybe pirate
|
|
|
|
Clegg
|
|
June 03, 2014, 04:22:49 PM |
|
I like to keep mine backed up to a few USBs. I dont really trust paper wallets.
|
|
|
|
|