Topic: Can BTC flaw be it's own success? (Read 1725 times)

If someone tries to "play" with lots of transactions, they will only bloat the wallet client's memory. All transactions that happen get sent to everyone on the network, and are held in memory until they get included in a block (or something like that). Remember, a typical block full of transactions right now is about 0.5MB. Since people typically have 1 GB of memory of more, an extra 5mb to even 100mb is memory being taken up by transactions probably wouldn't even be noticed by people running the wallet software. If all those transactions are free, they may crowd out other free transactions for a while, and make it harder for others to send bitcoins for free, but it won't stop it entirely. Plus really small transactions will just be ignored as spam. If those transactions include fees, they will just pay the miners, and will only do this until they run out of money.
FYI, each transaction gets assigned a priority. That priority depends on the size of the fee, but also on how long ago the transaction was created. So if you create a free transaction, and someone else floods the network with their transactions, even if they include a fee, eventually, with enough time, your free transaction will still end up with a higher priority, and will get included.

Satoshi and developers have discussed the idea of DDoS and food attacks very extensively years ago, and made sure it's impossible to attack Bitcoin with that.

Didnt know anarchism is a sexual preference  

Don't want to offend anyone in general, tried to be funny.

Rassah, jace, thanks for the great posts!


All these manual actions... I dont know enough but I would say that this should be decided as some kind of a future plan and automate it. If somebody decides to clog the blockchain, this can happen tomorrow... Today, when there is a reward in action, do miners ignore transactions?

Lets say that somone/(get ready for consperacy in 3 2 1)some government decides to play with lots of transactions to test the system's capabilities. There should be a way to submit the block size change (securely) without a need to actually upgrade the wallets.

I still believe that this is some kind of vulnerability of the system.

This is only a software limit, built into the current version of certain mining clients (probably not even all). Whenever this issue appears at the horizon and expected to become an actual problem within the foreseeable future, the software will simply be updated with a higher tx-per-block limit.

Furthermore, some miners (not even necessarily all) may decide to ignore transactions with fees below a certain threshold. In that case, sending 10k transactions per second becomes a pretty expensive hobby.

Great post very informative.  Thank you!

Well, it can, using current hardware, it's just that no one wants that right now, because we're all worried about spamming the system.

Right. I thought "the system can process around the 10k transactions per second" was about the present time.

It's just a setting in the client, not a hardcoded feature of the blockchain. We can change it if we want.

Just Google "wiki bitcoin scalability" whenever you want to look these claims up

https://en.bitcoin.it/wiki/Scalability

10000 transactions per seconds? Could u give me the link, plz?

This is about as offensive as asking gay people not to post in your thread.

The number of miners has absolutely no effect on transactions. You can have 1M times the number of miners we have today, or 1M times the hashing power, and transactions will still be limited to 10 minutes per block, and ~500KB of transactions in every block. I believe the hard limit on block size is 5MB. So, if Bitcoin gets popular and hits that limit, we'll have only two options:

1) Increase the block size. This sin't something the developers do. This is something, like with all proposed changes currently happening with Bitcoin, someone (anyone) can propose on the developer's site, other people can discuss and argue about until they figure something out to settle on, and then developers, or anyone, really, can write code to increase the block size, and submit it to the same site. After the actual code itself is discussed and argued about, and no bugs are found, THEN the developers can include it in their next release of Bitcoin-QT, as can everyone else who makes Bitcoin wallets, and typically they pic some date way in the future to make the switch. Basically, everybody will be running Bitcoin clients that have code embeded in it that will do nothing until X date, and as soon as X hits, will start to run the code, causing a "hard fork" (older clients will not be compatible with the new fork). last few times it was a few months to a year that X was set in advance.

2) Bypass the block chain completely when making transactions. This is essentially what all the exchanges to when people trade money. No bitcoin transfer actually hits the blockchain when it's traded - it's all done on their own local software. The only time something is written to the blockchain is when someone deposits or withdraws bitcoin. Similar services will likely pop up that let you deposit bitcoin, and then use the online wallet or a credit card of sorts to spend it, with all transactions being recorded only on the service provider's accounting books. Then, at the end of every day, they will just make one big transaction, sending the coins everywhere they need to go.

Sweet Blueberry408,

I know lots of people using BTC do it because they have tons of anarchy pheromones (or anarchymones) to spray on all of us but I am a bit alergic to any kind of anarchymones so please spare me the feds analogy. I only used the limitation example to show that sometimes decisions made in the BTC world without asking the public opinion. I gave the example without giving my own opinion (which is that it is OK if it's done as improvement for the public or as a greater-good/less-evil solution).

The biggest difference between the Fed and BTC is that the majority of the population trust the Fed and the dollar (knowing the high percantage of supperdollars among us) where the majority of population doesn't even know that they should trust BTC more.

Now that I covered this thread with some anarchymones anti-spray, lets get this discussion back on track. so realisticaly, how easy is it to make block larger/shorten the interval between blocks?

Can the developers shape and mold the overall cc economy and thereby act as the Federal Reserve Board does for the us.

Stuff the Fed does for the US economy:
Sets interest rate (a bank may borrow money from the Fed at x%)
Sets reserve requirement (a bank must have x% holdings in federal reserve notes)
Plays the market, talks to legislature, analyzes macroeconomy, reacts to changes in economic climate, deals openly in bond sector in conjunction with the treasury (printer of federal reserve notes), according to the orders placed by acts of congress and executive authorities.

Are there isometries that apply to cc econ?

The interest rate is akin to transaction fees, which is why having a wallet program running is akin to being your own bank. You are also you're own Fed in that the setting of the transaction fee is akin to setting the interest rate. The network is like a competitive consortium of banking agents who compete by processing the more profitable transactions more rapidly than the less profitable transactions.

The reserve requirement is confusing and could have a lot of parallels. The reserve requirement is a result of money checking. Money is created when a person goes to the bank and deposits a paycheck. If the reserve requirement is 20%, then the money supply grows by the remainder percentage of the paycheck. When the Fed makes a change to this number, banks transact large sums of notes, and fluctuations known as money supply inflation and deflation can be well tamed.

There is a physical cryptocurrency note: it is a string of entropic information. The reserve requirement is that 100% of the string must be verified against a chain of blocks of such strings which contain the history since the creation of the string's integral value quantua, that is, when it was first mined. The popular term for such a string is a hash, since it is the result of math known as hashing.

100% reserve requirement is hefty compared to the yankee dollar.

If you receive 51% of someone's federal reserve note, you can take it to the bank and exchange it for 100%. Plus, the guy left holding the 49% is out of luck. 49% of a federal reserve note is worth 0%. The physical bitcoin, that string of entropic information with integral value quanta and ownership vectors, is hefty compared to the yankee dollar.

The Fed might be interested in profiting from the ordeal... they could provide a fairly trustworthy proof-of-burn, where you send them federal reserve notes, and they send you bitcoins with the USD serial numbers encrypted in the hash. That's kind of futuristic, but most futurists have some sort of digital cash in the works.  I can almost go to 7-11 and get bitcoins.  Seems like they have a whole wall of digital currency.

So my next question is - realistically, how fast will it take to make the needed change in the block size? As I see it, after the last change in minimum transaction size, Ideals equal nothing when the developers want to make the change.

...which is something commuters do regularly to save money, whereas travelers may not. I think the number of ASICs working on bitcoins by the time it succeeds will have us in the megatransfers per second range easily. Question remains what bandwidth does bitcoin need, and at what point if any does a litecoin fill the gap?

Afaik the block size could be changed, if enough users would do the patch. And if everything is stuck for 72h, I think the users would want to do that.

The problem I see is less in the amount of mining power but the ability to join the next X amount of blocks. I might be wrong but it seams to me like buying a subway ticket knowing i'll be able to get on the train only later this week.

I think that as the transaction amounts increase, so will the number of miners to process these transactions. I don't see a limit on the number of either of these two things. Correct me if i'm wrong, though.

This is a something that is running in my head for a week already and I assume I'm wrong but I cannot figure out why and how.

There is a finite limit to the amount of transactions per block. We know (according to the wiki) that the system can process around the 10k transactions per second, which is way better than any other banking system but still finite. We also know that the difficulty is there to maintain the block production interval on 10 minutes.

So the question is what would happen if bitcoin succeeds? what would happen when the amount of transactions per block is reached? Lets assume that transaction fee rockets to the sky as a result of a traffic jam in conformations, It could cause the users to look for cheaper solutions.

More important question is what would happen if someone (or something) will decide to produce enough transactions (~4000x24x6x3~17.3M transactions in three days) to make other money transfer options look better than bitcoin in the aspect of conformation time? (meaning it would take more than 72 hours to confirm your transaction).

What I am basically trying to say is that in order to make bitcoin fall you don't need to hack the system but only to make it look bad enough by clogging it. am I correct? what am I missing?