|
Title: z Post by: znort987 on March 08, 2012, 11:51:06 AM z
Title: Re: Estimating the number of bitcoin users from the block chain ? Post by: kjlimo on March 08, 2012, 12:03:13 PM I thought it was easier to just look for IP addresses of people running the bitcoin client. There are also people monitoring the # of transactions.
Users is a good measure of how bitcoin is growing; however, transaction & economic activity are more important. I'm assuming you're trying to prove that bitcoin is growing? Anything else you are trying to support with this information? Title: Re: Estimating the number of bitcoin users from the block chain ? Post by: vuce on March 08, 2012, 12:26:55 PM . Find connected subgraphs of said graph (transitive closures) this would also connect users together, via transactions between them, wouldn't it? Since there is no easy way to distinguish which output goes back to the user and which don't I don't think this would be a pretty good estimate without some really clever logic behind it... Title: Re: Estimating the number of bitcoin users from the block chain ? Post by: kjlimo on March 08, 2012, 01:21:51 PM I thought it was easier to just look for IP addresses of people running the bitcoin client. There are also people monitoring the # of transactions. IPs tell you very little: - Folks connect thru tor - Folks connect thru firewalls (esp. mobile users) - Folks have dynamic IPs - They tell you nothing of actual usage And, at best, they give you a current snapshot, not the whole history. I'm assuming you're trying to prove that bitcoin is growing? Anything else you are trying to support with this information? Yes, partly, but also get an idea of the actual number of users rather than the volume or the number of TX. This is a rather different thing. If we get a snapshot, that may be better than a history. If people get bitcoins and then never use them, this does nothing for the bitcoin economy other than reduce the number of bitcoins in circulation. For similar reasons, the number of people transacting in bitcoins is likely more important than that number of people who own bitcoins. For that I would likely just use Mtgox's information provided of approxmiatly 30,000 people with accounts that have traded bitcoins for currency within the past 30 days. Assuming this is 25-75% of the number of people who "own" bitcoins, that puts us at around 40,000-100,000ish "users." What are you going to do with this information? I personally would like to know the number of bitcoin users in each local market. This would be a great statistic to use to explain how much of a market there is for brick & mortar stores to start accepting bitcoins. |