@flatfly I did some more digging why I couldn't change servers, and I think I figured it out. If I change the server using electrum the wallet updates as 'server': u'electrum.novit.ro:50001:t' , where the (working) default was 'server': 'ecdsa.org:50001:t' . If I make adjustments to the wallet by hand so that "u" is not appended everything works perfectly (I am able to connect to other servers like that just fine). edit: confirmed by another friend, but only for tcp edit 2: found out "u" stands for unicode string. I have no idea why this wouldn't work then.  |
|
|
|
Google indexing links to Instawallet wallets which people voluntarily published online
This is not so Care to elaborate on that? All this time you're denying everything we say without explanation. |
|
|
|
@vuce
The structure of a trie has no dependence on insert order. Given a set of data, there is only one trie that can hold it. The same goes for Patricia tries (which are level-compressed tries). And given that its query, insert and delete times are based strictly on key size (which will be bounded as above), there are no balance issues at all: it always takes exactly 32 "hops" to get from the root to the leaf you want, regardless of whether you are querying, inserting or deleting. So given fixed key-length, all those operations are actually O(1).
I was citing the patricia trie wiki, where it's pretty obvious that new inserts are inserted as leaves of the tree, therefore making them insert order dependent. If you would direct me to a better explanation I would appreciate it.
nevermind, I misunderstood how it works  FWIW, I agree, I think this might be the best choice. What other data structures am I missing that could be considered? I know B-trees would be a good choice if we are going with insert-order-dependent structure: they are easy to keep balanced with fairly simple rules for balancing.
AVL tree is the mother of balanced binary trees. They have the smallest "worst case height", so the fastest query, but a bit slower insert/delete than red-black trees. They are also very easy to implement. 2-3-4 tree might also be worth considering. Don't know if it's insert-order-independent or not but by a quick look it might be. Or maybe plain 2-3 tree, that one has data in leaves only so it looks kind of like a merkle tree, but does have quite a bit of overhead. |
|
|
|
I'm voting for level-compressed trie structures (so, a variant of a patricia trees) which have no balance issues at all, are insert-order-independent, and O(1) query/insert/delete. To insert a string, we search the trie until we can make no further progress. At this point we either add a new outgoing edge labeled with all remaining characters in the input string, or if there is already an outgoing edge sharing a prefix with the remaining input string, we split it into two edges (the first labeled with the common prefix) and proceed. So new insertions go to the leaves of the tree, I think this would make it insert dependent - just like any other tree. I'd suggest avl instead of r-b, since it has lower worst height. |
|
|
|
As rjk said, if you keep your url safe (not post it anywhere online) google can't find it.
I gave an example. The right person will understand this. I can also randomly try 1000 different instawallet url-s and publish them on a website, but that doesn't mean any of them are legitimate. Actually, it is a lot more likely you're going to be killed by lightning the second after you read this than it is for any one of those 1000 accounts having any bitcoins on it. |
|
|
|
As rjk said, if you keep your url safe (not post it anywhere online) google can't find it. |
|
|
|
I don't understand what you're trying to say. When you visit a site over https the url is encrypted, no one can see what it is. |
|
|
|
I am confused by only one point. URL-s easy to sniff.
Only if you're not using https. |
|
|
|
I don't see how this shows me that (n^((p+1)/4)) ^ 2 = n (which is what we want).
I can live without understanding that. But it's kind of annoying. Can anyone enlighten me?
(n^((p+1)/4)) ^ 2 = (n^((p+1)/2)) = (n^((p-1 + 2)/2)) = (n^((p-1)/2))*n =(*) n (*) By the definition of legendre symbol, if n is a quadratic residue (which in this case it is), (n^((p-1)/2)) = 1 (mod p) (proving this is good practice  ) The reason p = 3 (mod 4) has to hold is so that (p+1)/4 is an integer. |
|
|
|
I can't say I'm overly familiar with the pairing-based crypto, but I think it's mainly used in multi-party key agreement protocols.
Pretty much every crypto used today relies either on integer factorisation or discrete logarithm problem. Those still haven't been cracked.
Is it possible that it could affect multisignature transactions in any way? No, ECDSA is built into bitcoin and it doesn't use anything else, as far as signing is concerned. Implementing something like a pairing-based algorithm would cause a fork in the chain. I think we're still pretty safe for a good while |
|
|
|
Does Bitcoin even use any of the cryptography mentioned? I didn't see a specific technology mentioned, just "pairing-based cryptography".
No. Bitcoin uses ECDSA, not pairing based crypto. Very good. I was fairly sure that ECDSA was not a subset of any "pairing-based cryptos", but wasn't certain. However, how does this bode for RSA/SSL/PGP/GPG/etc? I can't say I'm overly familiar with the pairing-based crypto, but I think it's mainly used in multi-party key agreement protocols. (basically reducing the number of exchanges needed between people from diffie-hellman protocol) Pretty much every crypto used today relies either on integer factorisation or discrete logarithm problem. Those still haven't been cracked. |
|
|
|
Does Bitcoin even use any of the cryptography mentioned? I didn't see a specific technology mentioned, just "pairing-based cryptography".
No. Bitcoin uses ECDSA, not pairing based crypto. |
|
|
|
|
Does anyone else have problems with their clients not updating (on ecdsa.org server)?
|
|
|
|
I am interested in placing a message into the blockchain for the purpose of timestamping the message (eg adding the hash of a document to the blockchain so that the document can be verified as existing sometime before the timestamp of the block) I've looked at how btcmsg does it but the comment in this link suggests that btcmsg isn't the right way to do it. http://bitcoin.stackexchange.com/questions/2616/how-to-decode-messages-embedded-with-btcmsgPlease don't encourage this abuse of Bitcoin. If someone would like to implement a message service correctly, there are many developers on IRC willing to help. – Luke-Jr Can someone propose how I may best achieve placing a message in the blockchain? I've had some ideas but would like to hear how the pros think it should be done. I expect flames that 'bitcoin isnt meant for that' - if so please explain the alternative you'd use instead. I have made a beta implementation of CommitCoin protocol. If you're interested PM me. I'm not ready to make it publicly available just yet but it's fully functional. An easier alternative if you don't mind paying 0.0005 btc fee is to use hash to address utility of blockexplorer to convert the hash of your message to a btc address, and then create a transaction that sends 0 btc to that address. (this way no money is wasted) |
|
|
|
What's with all the sub comments? Anyway, after some thinking it's probably just an address collision due to my own stupidity. You mentioned a potential protocol vulnerability, so of course many people are interested. Just to make this clear, this is commitcoin protocol vulnerability, not bitcoin one. |
|
|
|
What's with all the sub comments? Anyway, after some thinking it's probably just an address collision due to my own stupidity. |
|
|
|
Yeah this sounds like a waste of time to look considering i don't know what this has to do with turning in a paper
The paper is on commitcoin implementation, which this double spend is a part of. If this is what I fear it is it might uncover a vulnerability in the protocol that authors shrug off as being practically unexploitable. Anyway it's not like someone could do this by accident so if you don't know you did it don't bother checking. |
|
|
|
If anyone does I would very much like to hear from you (it's kind of important, as an answer I would need could turn a paper that is due tomorrow upside down) |
|
|
|
I cracked it.. Revalins code up above would have done it in an hour.. EDIT: Don't know if OP was impressed though, haven't heard from him, no reply to the PM where I sent him the right pwd. He transferred the bitcoins to a new address, except the 0.01 BTC I stole when testing (hope you're not angry about that 1zbtc, I'll give it back if you want  ) and no 50-100 BTC reward that was promised?  |
|
|
|
@vuce: try: electrum -o balance
that works, what exactly does it do? I assume it doesn't need to connect to the server, since that obviously doesn't work? |
|
|
|
|
Show Posts
