arubi 2017-01-19 02:25:18
cannon-c, bare multisig doesn't have that issue btw
cannon-c 2017-01-19 02:25:37
hello arubi
arubi 2017-01-19 02:25:41
o/
cannon-c 2017-01-19 02:25:45
What do you mean by bare multisig?
arubi 2017-01-19 02:26:03
non p2sh(multisig) , just the multisig itself as scriptpubkey
arubi 2017-01-19 02:26:14
"ye olde multisig"
GAit 2017-01-19 02:30:32
bare multisig howerver has some limitations too: most wallets out there can't send to it, sender pays for the security of the receiver
arubi 2017-01-19 02:31:34
well specifically here it's for personal use with premeditated intention, that's why I'm suggesting
cannon-c 2017-01-19 02:31:53
Where can I learn about "bare multisig"?
cannon-c 2017-01-19 02:32:05
WHy IJ am looking into fragmented backup of xpriv,
cannon-c 2017-01-19 02:32:08
more simplicity
cannon-c 2017-01-19 02:32:27
xpriv --> Shamirs secret sharing scheme
arubi 2017-01-19 02:32:29
it's not very simple though, in either case
GAit 2017-01-19 02:32:34
arubi: makes sense. Although the tooling is probably not the most battle tested/userfriendly
arubi 2017-01-19 02:33:25
GAit, yea, 'bitcoin-tx' could do it nicely but it's all cli. cannon-c, why fragment the privkey? it's not really a backup if each part isn't enough for the whole key
arubi 2017-01-19 02:34:30
cannon-c, https://bitcoin.org/en/glossary/multisig
piqure 2017-01-19 02:34:31
^^^ WARNING: any URL may lead directly or indirectly to COIN-STEALING MALWARE! ^^^
cannon-c 2017-01-19 02:34:37
dont need all parts
cannon-c 2017-01-19 02:34:50
SSSS only need say 3 of 5 fragments to re-assemble
cannon-c 2017-01-19 02:34:52
for example
cannon-c 2017-01-19 02:35:30
basically Im thinking of using SSSS in form of words
cannon-c 2017-01-19 02:36:07
like the mnemonic backup for seed used by wallets
cannon-c 2017-01-19 02:36:29
but in fragments of pre-defined thresholds for re-assemble
arubi 2017-01-19 02:37:21
hmm, so wait, you want this in case something happens to you right?
cannon-c 2017-01-19 02:37:26
yeah
cannon-c 2017-01-19 02:37:33
so for example
cannon-c 2017-01-19 02:37:44
5 trusted people have a list of words
arubi 2017-01-19 02:37:49
and you plan on 3/5? what if people collude to steal?
arubi 2017-01-19 02:38:03
trusted, okay
cannon-c 2017-01-19 02:38:04
3 of them can combine their list of words to recover xpriv
cannon-c 2017-01-19 02:38:10
why they should be trusted
arubi 2017-01-19 02:38:51
right, I think a cleaner solution is a time locked transaction that you maintain unspendable
cannon-c 2017-01-19 02:38:55
or secure backup
cannon-c 2017-01-19 02:39:01
since if one backup discovered not fully compromised
cannon-c 2017-01-19 02:39:22
because adversary would need to recover 3 backups each from different hiding spots
arubi 2017-01-19 02:41:09
cannon-c, if you tell your trustees to give you their addresses, you can keep an up to date, unspendable time locked tx that you spend to a new time lock say every few months
arubi 2017-01-19 02:41:39
that would be p2sh, but a couple or so months probably still isn't enough to crack
cannon-c 2017-01-19 02:41:59
that would work
cannon-c 2017-01-19 02:42:07
although my usecase in mind also is backups
arubi 2017-01-19 02:42:30
give them all an encrypted backup of the key that they can't access
arubi 2017-01-19 02:42:37
in case you want it, grab it from them
arubi 2017-01-19 02:42:53
in case you're gone, they can spend the time locked tx after 2 months
arubi 2017-01-19 02:44:10
cannon-c, there's no reason for a group to be able to recover one single private key because that probably doesn't go well when money is involved
arubi 2017-01-19 02:44:34
especially in time like when a tragedy happens and not everyone are thinking straight
