Topic: 20,000 transactions - New Bitcoin Network Record (Read 3193 times)

Remember way back, way way back  ( ... about 4 weeks ago  ) when this was a longshot bet?

"Bet: There will be more than 50,000 Bitcoin transactions in one day in 2012"
 - http://betsofbitco.in/item?id=351

Let's not forget to keep an eye on this one as well:



http://www.blockchain.info/charts/blocks-size

And the hits keep coming!  

40,000 Transactions in 24 hours (per Blockchain.info)



 - http://bit.ly/K15Kmb

400,000 transactions cumulative on the blockchain (and 12.3% of all data ever put into the blockchain) due to SatoshiDICE.

Statistics from:
 - https://bitcointalksearch.org/topic/satoshi-dice-statistical-analysis-80312

BitcoinWatch shows 27,273 transactions in the past 24 hours.



 - http://blockchain.info/charts/n-transactions
 - http://blockchain.info/charts/n-transactions?timespan=60days

In my last post, I agreed with evoorhees's 'we fucking one' philosophy in general, but I do take some exceptions with certain of your suggestions as quoted.

Firstly, I think it likely that a shift to Bitcoin would coincide with a loss of confidence in other currencies (and conversely, that Bitcoin will continue as a fringe spec of dust on the shop floor of the global economy absent such a thing.)  Such a loss of confidence could happen fairly rapidly and fairly soon (though hopefully it won't.)  What it is unlikely to do is to wait for technology to develop to the point of supporting a much more resource intensive Bitcoin network.

Secondly it is clear that technology marches forward, but the latest advances are typically available to select entities for a long period of time (and often indefinitely.  Think about satellite.  Sure, we enjoy PPV porn on demand, but we do so at the pleasure of large corporate and government agencies.  Ping me when a terrorist network (which the Bitcoin community is one legal decision away from) launches it's own communications satellite and uses effectively for more than a few hours.

I'll go ahead and agree with that as well.  But I don't see any reason not to try to stretch the life of Bitcoin-I out as long as practicable, however, especially if it just takes a certain amount of planning and is considered in determining development priorities.

When Bitcoin-I's race is run, I imagine that what will form is a deliberately feather-weight foundation upon which as many 'bitcoin-1 like' shards as are needed can exist.  Each 'shard' would contribute some small portion of it's resources toward securing this foundation solution.  In doing so all would help support one another.

Trading value between the 'shards' when needed should be fairly straightforward and reasonably easy on the users.

In this way I would hope that a scalable peer-to-peer based economy can exist and require nothing more than the most primitive of network resources thus making it that much more difficult to attack.

I outlined some ideas along these lines a while ago before losing interest.  If anyone is interested, they are here:  http://bakcoin.org

Agreed!

If we break Bitcoin by eventually convincing half the world to use it as their primary currency and payment system, then we fucking won.

If half the world were using bitcoin everyday we would be living in a completely different world so those numbers are meaningless. That world would likely have vastly superior technology. If bitcoin breaks because its too popular/useful thats a good thing. Bitcoin2 will fix the issues and we will be better off in the long run.

I for one think the extra traffic is REALLY needed for the bitcoin ecosystem. If it ever starts really catching as a medium of dominant exchange, even locally, 20,000 transactions per 24 hour period is only a small portion of what is actually needed.

I think we will need to move to an increasingly thin-client based model, but perhaps allow blockchain verification to occur occasionally, e.g. once a month, locally running through all the transactions since my last verification locally, and verify that it all makes sense (e.g. hashes line up, transaction input is all available, etc.). Then the thin-client could have a 'last trusted block hash' and servers can provide merkle trees to provide verified balances, etc.  This would make the system more like an open-book, transactionally peer-to-peer, and crypto-hard-to-forge version of a banking system, but it makes sense to me to move in that direction.

-s

From another recent thread I did a quicky estimate that if half of the worlds population did 2 tx/day, that's about 70,000 tx/sec.  I'd say to account for peaks (but ignoring DOS potential) one should have 250,000 tx/sec capacity.

I calculated our recent tx record as around .25 tx/sec.  About one millionth of the above scenario.  So, absent significant optimization development, a very simplistic back-of-the-envelope (and in many cases, wrong) calc about the resources needed to meet the aforementioned target is to multiply the current need by a million. 

A suggested solution was that most transactions would happen in bank-like entities offloading the block chain work.  That's fine, but it is NOT Bitcoin, and not unlike a lot of the solutions that I use and distrust today.

A pretty much universal suggestion is to rely on resources which are accessible currently mainly to large entities (network bandwidth, 'supernode' clustering, etc.)  Moore's law is also commonly suggested.  I, seemingly alone, have limited confidence that such resources would remain widely available to support the Bitcoin network if there were a strong incentive for TPTB to wish them not to be.

Lastly, various people (including myself) would hope that the infrastructure required to operate a (large) Bitcoin network would be at least tolerated and protected by enough governments around the world to keep the system going.  My concern here is that 'hope' is not a very reliable strategy and I could see many scenarios where this would not likely happen.

If there is a rapid loss of confidence in various world currencies, I could see the demands on Bitcoin growing quite rapidly.  The bip16/bip17 struggles indicate to me that the rather significant development work needed to optimize along the various theorized paths would not be especially rapid and/or reliable.

I suggest to the development team putting some thought into how to discourage users rather than how encourage them.  I sense that to this point the focus has been on the latter.  Similarly, it might be worthwhile to focus some efforts on how to make it more likely that various kinds of failures are likely transient and not be misconstrued as terminal failures of the whole concept of Bitcoin itself (unless, of course, they are.)

well said, man. I'd rather see multi-terabyte blk0001.dats than bitcoin use declining. It's been clear for a while that running a full node will not be for everyone and his grandma in the future.

Erhm.. what?
You mean the developers currently working on the client that Satoshi started?
The so called "main developers" is an ever changing list of people that have decided to spend time on the implementation that Satoshi started.
I trust and hope that we will see more than one fully independent client going forward. And, I see absolutely no reason why dudes working on one implementation would have to be the same dudes working on another implementation, be it thin client or not.

I can't speak for everyone, but moving some values in a database somewhere isn't going to do much in the field of sexual arousal. It'd probably work quite well for a service like SatoshiDice, on the other hand.

I am concerned about the size of the block chain as well.  In the end it doesn't make sense.  There's this great thing called bitcoin, but don't use it.  Bitcoin will need to adapt to the growing block chain or the whole thing is doomed to fail.  I'm just hoping that the main developers plan on releasing an official thin client sometime soon. 

I will try out blockchain.info's client next month, hopefully I like it. 

So building a revolutionary new monetary system that may help free the world from coercion, deceit, and fraud is "abuse of precious resources" but the porn you've downloaded and stored over the years is legitimate usage? 

Welcome to growth?

...so let's abuse the fuck out of it.

The world is full of resources which do not grow proportionately with human demand.

Fortunately, hard drive space, bandwidth, and processing power has not proven to be one of them.

Consider more serious people jump in.

The real cost of blockchain bloat is that it reduces the proportion of Bitcoin users who can run their own client.  The larger the blockchain is, the fewer people will have the bandwidth and hard disk to download and store it.

You only need the blockchain if you are generating your own work to mine.  You don't need the blockchain to mine on deepbit because deepbit sends you all the information you need to hash.