Adi Shamir's paper on bitcoin

[auzaar]

Has this already been posted here?

http://eprint.iacr.org/2012/584.pdf

Abstract. The Bitcoin scheme is a rare example of a large scale global
payment system in which all the transactions are publicly accessible (but
in an anonymous way). We downloaded the full history of this scheme,
and analyzed many statistical properties of its associated transaction
graph. In this paper we answer for the rst time a variety of interesting
questions about the typical behavior of account owners, how they acquire
and how they spend their Bitcoins, the balance of Bitcoins they keep
in their accounts, and how they move Bitcoins between their various
accounts in order to better protect their privacy. In addition, we isolated
all the large transactions in the system, and discovered that almost all
of them are closely related to a single large transaction that took place
in November 2010, even though the associated users apparently tried
to hide this fact with many strange looking long chains and fork-merge
structures in the transaction graph.

[1714812033]

1714812033

[1714812033]

1714812033

[1714812033]

1714812033

[1714812033]

1714812033

[davout]

Not on my watch

[Atlas]

Sponsored by the Citi Foundation

[auzaar]

Sponsored by the Citi Foundation

I was about to paste that, "This research was supported by the Citi Foundation."
so big guys are having a look at it

[stevegee58]

Nonetheless it was an interesting read.

[davidj411]

you guys know who this is right? one of the founding fathers of modern day cryptography. case in point: RSA, the S stands for Shamir.
Who cares about the Citi Foundation link.  On an unrelated note, I think its ironic that Citibank funded the Nazis and Shamir is Israeli.

[Raoul Duke]

Was I the only person to find odd that they would download/parse blockchain.info webpages instead of downloading the blockchain itself and working with it? lol

[auzaar]

Was I the only person to find odd that they would download/parse blockchain.info webpages instead of downloading the blockchain itself and working with it? lol

may be they found official client too cumbersome to use, scraping webpages is much easier

[Atlas]

Was I the only person to find odd that they would download/parse blockchain.info webpages instead of downloading the blockchain itself and working with it? lol

may be they found official client too cumbersome to use, scraping webpages is much easier

The web is always easier to work with.

[LightRider]

Bitcoins are digital coins

Wrong in the first sentence. Does not bode well.

[davout]

The figures about Instawallet do not seem correct, on May 13th there were 103,513 wallets having at least one incoming transaction, the paper mentions 23,649 different addresses (one wallet = one address on Instawallet).

Source :

Code:

SELECT COUNT(*) 
FROM wallets 
WHERE 
  EXISTS (SELECT * FROM transfers WHERE `key` = wallet_key) AND
  created_at <= '2012-05-13'

[Roy Badami]

Having not read (and barely skimmed) the paper:

It does bother me that the paper does not contain the word 'change'.  Any statistical analysis of bitcoin transactions that doesn't discuss the notion of change is likely to be flawed, since the ambiguity as to which output is the 'real' transfer and which is the change is one of the major ambiguities in the public blockchain.

I'm prepared to consider the possibility that they understand the concepts but just not the terminology (and hence my naive search for the word 'change' failed to find their discussion of this).  But I'd like to see some discussion as to whether the patterns they see in these apparently large transactions could adequately be explained simply by the normal mechanisms of change when spending large coins, and a quick skim doesn't find such a discussion.  I worry that their finding of these chains that end up transfering coins back to the sender are just the normal process of sending change back to new addresses in the same wallet, only to be susequently combined again in a future transaction.

Still, intersting to see Shamir as co-author on a paper.  For him to get second billing, I imagine that probably means that the work was primarily done by a student of his who he supervised - but still.  To know that bitcoin is on Adi Shamir's radar is certainly interesting.

roy

[davout]

It does bother me that the paper does not contain the word 'change'.  Any statistical analysis of bitcoin transactions that doesn't discuss the notion of change is likely to be flawed, since the ambiguity as to which output is the 'real' transfer and which is the change is one of the major ambiguities in the public blockchain.

They accounted for it, not specifically as change but seeing those as loops to the same owner in the transactions graph.

[elux]

Interestingly:

Today I saw Adi Shamir (you know, the S in RSA) and managed to talk with him a bit about Bitcoin.

[kaerf]

Here is our first surprising discovery. The total number of BTC’s in the system is linear in the number of blocks. Each block is associated with the generation of 50 new BTC’s and thus there are 9,000,050 BTC’s in our graph of owners (generated from the 180,001 blocks between block number zero and block number 180,000). However, if we sum up the amounts accumulated at the 609,270 addresses which only receive and never send BTC’s, we see that their owners have actually put aside in some kind of “saving accounts” 7,019,100 BTC’s, which are almost 78% of all existing BTC’s. 59.7% of all the coins are “old coins” which were received more than three month before the cut off date (May 13th 2012),
Quantitative Analysis of the Full Bitcoin Transaction Graph 7
and still had not triggered any outgoing transactions. This means that there are much fewer BTC’s in circulation than previously presumed.

7M unused coins...where are the hoarders at?!

[kwukduck]

I wouldn't be surprised if a big number of those were just gone because the user trashed his wallet or lost it in some other way, remember that btc weren't always worth 12 bucks....
my friends trashed a few thousand bitcoins too in the early days...

[bg002h]

What happened November 2010? Anyone remember?

When did Gavin present at the CIA? I had forgotten about Bitcoin between summer 2010 and the bubble...

Addendum 4/27/2011 is when Gavin posted he would preset at the CIA.

[johnyj]

I suppose, in the early days, lot's of coins are lost due to their almost non-existent value, application uninstalled, hard drive erased, computer sold out, etc...

So the total amount of BTC in existing are much lower than advertised 21 million

[jgarzik]

At least one of this paper's fundamental assumptions is flawed.

Posted this gist going into detail:

Peer review of "Quantitative Analysis of the Full Bitcoin Transaction Graph"
https://gist.github.com/3901921

Comments welcome.  If there is further criticism that may be added to the gist, speak up.

[Jan]

Interesting read. Shamir is a BIG name in crypto world. Academia is really digging in now. Had to lol at the html scraping stuff though. I'll contact them to offer them my blockchain observer.