Are bitcoins indestructible?

[12648430]

If that did happen, miners would likely stop relaying transactions from the old address type before it became economical to harvest them en masse.

Why?  Also 100.000000000000000000000000000000000000000000% of miners?  Even if the tx includes a giant tx fee? I doubt it

Of course not voluntarily. In the scenario where ECDSA has been broken and its use has long since been phased out, there would doubtless eventually be a discussion among the community about whether to make a hard-forking change that after block X such accounts cannot be spent from. This would of course be a massive heated debate that we don't need to have here and now, but it's my opinion that imposing a rule preventing such spends would be for the good of the community, and I'd go out on a limb and speculate that that would end up being the consensus - although if mining hardware at such a time were capable of the cracking operation, miners' personal interests may prevail and a lot of hashing power would end up moving from safeguarding the network to cracking keys.

[1714049301]

1714049301

[DeathAndTaxes]

Well the nice thing about ASICs is the hashpower can't "move" to other tasks.   The hardware is also a sunk cost, it really never makes sense to turn off an ASIC once purchased until it is no longer profitable to run, at which point it would be turned off no matter what.

[beetcoin]

i don't see how you could say they are indestructible.. the blockchain relies on the users who download them. it could go FUBAR with a 51% attack, no?

[teukon]

The protocol limits the miners reward to be less than or equal to the amount of the subsidy plus fees.

Likely due to a poorly designed miner in the blockchain there are a few blocks where the coinbase is less than subsidy + fees and thus coins have already been destroyed.

I can't remember clearly but didn't someone once accept a 10 BTC block subsidy instead of the usual 50 BTC as a test?

Do you have any examples of such blocks?  I can't seem to find any with a quick web search.  If not, don't worry; if I care enough I'll write a script.

[BurtW]

I destroyed 0.00000004 BTC in this transaction:

https://blockchain.info/tx/bf40e4a1c2546747bc800a085e7145d921a9f402aaf4040c155ff5d0df9cc999

Code:

11When1DieBuryMeDeepLayTwoXVEY5jv - (Unspent) 0.00000001 BTC
11SpeakersAtMyFeetAPairofXXTyrHor - (Unspent) 0.00000001 BTC
11HeadphonesonMyHeadAndXXXXYUSvnd - (Unspent) 0.00000001 BTC
11ALwaysPLayTheGratefuLDeadWdq4Xo - (Unspent) 0.00000001 BTC

[pand70]

It's impossible to send them to an invalid address, BUT it's entirely possible to send them to an address for which no one has the key.

Take for example: 1BitcoinEaterAddressDontSendf59kuE

Are you sure that this wallet does not have a private key?

Because it was created as a donation wallet by someone named holgero, and I can see a 0.559 BTC donation coming to it two months ago.

Let me put it that way. If i ever see this address sending bitcoins to anyone i 'll sell all my coins the very next second.

[bitintheghetto]

Yes. Once mined they are indestructible.

Being abstract constructs, Bitcoins are no more destructible than the number 2 is destructible.
They can only be transferred.
You cannot "unmine" them.

A bitcoin however can be taken out of circulation like every other means of exchange if the owner just refuses to spend it. Or the wallet is lost. But there is no equivalent of burning a $1 note in bitcoins.

Can someone correct me if I'm wrong?

[DeathAndTaxes]

Your wrong.  I already pointed out how through mining coins can be destroyed permanently.

[bitintheghetto]

Are you referring to pools?
I'm new to this can you explain the subsidy and fees?
I was under the impression subsidy and fees would be transferred to a node for solving the block or something.
MY question is simple, aren't those bitcoins transferred to some other client on the node?

[DeathAndTaxes]

The rules only require the amount paid to the block solver to be LESS THAN (but not necessarily equal to) the total tx fees + block subsidy.

[BurtW]

The rules only require the amount paid to the block solver to be LESS THAN (but not necessarily equal to) the total tx fees + block subsidy.

I think this may need further clarification for the noobs:

When a miner (or pool of miners) successfully mines a block the protocol lets them put a special transaction into the block that gives them 25 BTC "out of thin air" - this is where Bitcoins come from.

What D&T is saying is that the protocol specifies that in each block the miner or pool who first mines it gets to give themselves up to 25 BTC.  So if they give themselves only 10 BTC then 15 BTC are lost forever because they will never be mined.

It would be silly for a pool or miner to do this to themselves but if they did that block is done and it cannot be changed.  Since it only created 10 out of the 25 possible BTC for that block the total number of BTC that will ever be created is now 15 less.

Make sense?

[DeathAndTaxes]

Thanks BurtW.  However it is more than just the subsidy.  The same concept applies to fees as well.

The protocol simply requires the value of the coinbase (reward to miner) to be LESS THAN the amount of fees + block subsidy.   So imagine in the future when block subsidy is zero and fees and a block has x BTC in fees.  A miner solves the block and has the coinbase set to zero.  12 BTC of existing coins will be permanently destroyed.  Users will pay 12 BTC in fees and the miner won't collect them.  This can happen at any time.

Lets look at a recent block.
https://blockchain.info/block-index/445116/000000000000000564bb70e7d7b56c9e0e399aea3c82ec84154f84aee94cc7cd

The current block subsidy = 25 BTC
The value of tx fees in block = 0.01678413 BTC
The amount paid to miner (see first tx) = 25.01678413 BTC.

So this is obviously a valid block.  
If the coinbase was 25.01678414 BTC or higher it would be an invalid block.
However if the coinbase was 25.01678412 BTC or less it would still be a valid block and one or more satoshis would be permanently "deleted".

There are at least six blocks where the coinbase payout is less than block subsidy + fees, so some coins have already been destroyed in this manner although I assume this was due to a miner error rather than any malicious intent.  I am not particularly worried about it but the answer to the question "are bitcoins indestructible?" the answer is a definitive no.

[manuel]

It's impossible to send them to an invalid address, BUT it's entirely possible to send them to an address for which no one has the key.

Take for example: 1BitcoinEaterAddressDontSendf59kuE

Check it out on blockchain. If you can brute force the private key, the coins are yours. Is it impossible? Theoretically, no, but practically...

Let's say you had a super computer that was guessing 999 trillion keys per second. It would take you 3.5 billion years to exhaust just 10% of the keyspace, which means in 3.5 billion years you would have a 10% chance of having guessed the key. Good luck with those odds!

Watch that address. If you see someone spending its outputs, it is probably time to get out of bitcoin.

How do people know somebody doesn't have the private key to that address all along and they're just sitting on the coins?

What does "coinbase reward mean"?

[BurtW]

How do people know somebody doesn't have the private key to that [1BitcoinEaterAddressDontSendf59kuE] address all along and they're just sitting on the coins?

Because the person that created the address 1BitcoinEaterAddressDontSendf59kuE never had the private key.  They simply started with the string "1BitcoinEaterAddressDontSend" and then added the correct checksum "f59kuE" onto the end of the string (it is a bit more complicated than that but you get the point).

Since they never had the private key no one will ever have the private key so any coins sent to that address are lost forever.

[justusranvier]

Since they never had the private key no one will ever have the private key so any coins sent to that address are lost forever.

inb4 somebody quibbles about the odds not being exactly zero.

[Desensitizer]

As they are digital I am going to assume you do not mean indestructible in the literal sense of the word. They cannot practically be bruteforced, nor can they be sent to an invalid address. So the only way they can be lost is via human error. One has to forget where they left them, or forget the key to access them, or send them to a random address someone else or no-one has control over. If you simply take care when dealing with them then yes, they are indestructible as long as you use common sense.

[DeathAndTaxes]

As they are digital I am going to assume you do not mean indestructible in the literal sense of the word. They cannot practically be bruteforced, nor can they be sent to an invalid address. So the only way they can be lost is via human error. One has to forget where they left them, or forget the key to access them, or send them to a random address someone else or no-one has control over. If you simply take care when dealing with them then yes, they are indestructible as long as you use common sense.

They can be destroyed in the block creation process (mining).  They can be destroyed so they are not indestructible.

[Dabs]

It's going to take awhile for bitcoins to be destroyed by mining. Which miner in their right mind would do that? Which pool? And before anything major happens, only a few bitcoins will be destroyed that way, maybe at most a day's worth, before everyone knows about it, then all the miners and pools will fix it.

As far as everyone is concerned, we will still see about 11.8 million coins by the end of this year (2013) and about 20 million coins before the year 2025.

The few that will be destroyed by mining less than the block reward and subsidy and fees... Well, no one is going to do them on purpose, and even if someone intended to do them on purpose they would only waste their time and money.

[Phrenico]

Whether or not miners make the coinbase = Subsidy + Fees, I think it's fair to say that bitcoins sent as transaction fees are destroyed.

[DeathAndTaxes]

Whether or not miners make the coinbase = Subsidy + Fees, I think it's fair to say that bitcoins sent as transaction fees are destroyed.

Well by that logic all coins are destroyed in any tx (and an equal number of new ones created).  However under normal circumstances there is no NET coins destroyed.   Tx inputs = Tx Outputs + Miner fee.   As long as miner pays out those fees in coinbase then no net coins are destroyed so making a distinction that tx fee outputs are destroyed but not normal tx outputs is without merit.