What would you pay for the keys to the genesis block/address

[zebedee]

I doubt satoshi owns the key of this genesis bitcoin address due to the fact that the 50 BTC is unspendable. So many ppl just send dust amount to this genesis address to destroy their bitcoin. 

You're assuming a lot.  The genesis block was the only block where in some sense Satoshi was guaranteed the coins, as by construction he was competing with no one.  He may have viewed this as unethical, and therefore deliberately left out the code to add the genesis coinbase to the UTXO set - I find it hard to believe, though possible, that he wouldn't have realized that it wasn't in the UTXO set.  I suspect it was a deliberate choice.

Once he had released the code and anyone could download it, he might have ethically viewed all subsequent blocks as fair game (though no one other than Hal has stepped forward to claim a block earlier than #78) and so mined for all he could.

[thejaytiesto]

https://blockchain.info/block-index/14849


https://blockchain.info/address/1A1zP1eP5QGefi2DMPTfTL5SLmv7DivfNa


So there it is the beginning of Bitcoin  .  The original 50 coins are still there plus some nice people that wanted to donate to it.

btc value is 65.44243457 btc


But there it is.  I would buy it from Satoshi  for 100 btc.

I sure he wants more. I am also sure many people would give him quite a bit more then 500 btc for it to belong to them.

So simple question what would you pay for it?

I don't think Satoshi is in a hurry for money to sell something as iconic as the first BTC address ever. When BTC is mainstream and makes it into history books, the value of owning that first ever address will be unvaluable. Satoshi knows this and im sure he wouldn't agree to sell it, unless he totally needed to.

[teukon]

so is the total number of "usable" bitcoins, assuming none are lost to accidents, 20,999,950 instead of 21 million?

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are fully spendable today.
• some have yet to be generated (do not yet exist).
• some are "lost" because a corresponding private key was lost (8999 BTC example by Stone Man) but are spendable in principle (cryptography breakthrough, quantum computers).  Some of these will be easier to find than others: Casascius coins for example made us of the mini private key format and so are protected by 128 bits of entropy rather than the usual 160 bits.
  
• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address; cryptograffiti.info/)
  
• some are unspendable due to protocol bugs (the 50 BTC in question and the 100 BTC mentioned by zebedee above)
• some are locked by scripts which provably return false on any input.  Some of these are so badly lost they don't even have addresses! (367.75849319 BTC example).
  
• 23.1 mills (23.1 millibitcoins = 0.0231 BTC) may well never be mined into existence in the first place due to rounding all subsidies down to the nearest satoshi.
• some have already lost the chance to ever be mined into existence (example by midnightmagic: block #124724 mints only 49.999 999 99 BTC).
  

Note: I use "exist" here for any amounts of BTC in an up-to-date unpruned Bitcoin UTXO database.  There are of course different interpretations of what it means for a bitcoin to exist.  Even this seemingly simple definition is not water-tight as two different full nodes may disagree on the UTXO and in particular it allows one to claim that at one point, there existed more than 180 billion bitcoins.

[futureofbitcoin]

so is the total number of "usable" bitcoins, assuming none are lost to accidents, 20,999,950 instead of 21 million?

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are fully spendable today.
• some have yet to be generated (do not yet exist).
• some are "lost" because a corresponding private key was lost (8999 BTC example by Stone Man) but are spendable in principle (cryptography breakthrough, quantum computers).  Some of these will be easier to find than others: Casascius coins for example made us of the mini private key format and so are protected by 128 bits of entropy rather than the usual 160 bits.
  
• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address; cryptograffiti.info/)
  
• some are unspendable due to protocol bugs (the 50 BTC in question and the 100 BTC mentioned by zebedee above)
• some are locked by scripts which provably return false on any input.  Some of these are so badly lost they don't even have addresses! (367.75849319 BTC example).
  
• 23.1 mills (23.1 millibitcoins = 0.0231 BTC) may well never be mined into existence in the first place due to rounding all subsidies down to the nearest satoshi.
• some have already lost the chance to ever be mined into existence (example by midnightmagic: block #124724 mints only 49.999 999 99 BTC).
  

Note: I use "exist" here for any amounts of BTC in an up-to-date unpruned Bitcoin UTXO database.  There are of course different interpretations of what it means for a bitcoin to exist.  Even this seemingly simple definition is not water-tight as two different full nodes may disagree on the UTXO and in particular it allows one to claim that at one point, there existed more than 180 billion bitcoins.

And this is why I said ASSUMING NONE ARE LOST. From the very beginning, there was never a possibility of having 21,000,000 usable bitcoins, or, is there an extra block to account for the unusable first block, and the amount from the first block is usually left out of the total?

Quite a simple question, and while I appreciate the two attempts to answer my question, and it seems you guys put a lot of time and effort into explaining, my question was not really answered.

[lucullus]

Three hundred dollars!

[ensurance982]

so is the total number of "usable" bitcoins, assuming none are lost to accidents, 20,999,950 instead of 21 million?

There were 100 BTC lost in two duplicated coinbase transactions (i.e. two pairs of blocks with the same coinbase txn, and as the protocol assumes txnID and output index are sufficient to designate an unspent output, means the later coinbase "overwrote" the earlier as it hadn't been spent by overwrite time), which is why version 2 blocks were created to eliminate that loophole.  So no, there are 150 BTC at first glance "granted" in blocks that are actually irretrievable forever.

There have been cases of miners claiming less than the reward + fees, "just for the hell of it", though I believe those are fractional BTC.

Then there was the Karpeles gox fuckup where he sent coins to provably unspendable scripts, because testnet was too much effort.  That was around 2600 BTC IIRC.

So almost 3 thousand BTC simply don't exist any more, no matter what random numbers some lucky bastard generates.

Very very very interesting information, didn't know about a lot of those things. The Karpeles incident doesn't even surprise me, to be honest. But pushing code to a live system is scary already, not testing it when it comes to literal money... Glad Gox is gone now...

[jacktheking]

Hmm.. I wont be buying it for more than 65 BTC - not gonna buy it. I will only be buying it if more and more people is getting crazy for first 50 Bitcoin. At that time.. I dont think I stand a chance to get it.

[ensurance982]

Three hundred dollars!

I guess it is obvious that you would always be willing to pay less than the block/address' actual net worth at the time. But I think this address does have memorabilia-value and people could still send many coins directed at it. It's sort of an investment

[teukon]

so is the total number of "usable" bitcoins, assuming none are lost to accidents, 20,999,950 instead of 21 million?

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are fully spendable today.
• some have yet to be generated (do not yet exist).
• some are "lost" because a corresponding private key was lost (8999 BTC example by Stone Man) but are spendable in principle (cryptography breakthrough, quantum computers).  Some of these will be easier to find than others: Casascius coins for example made us of the mini private key format and so are protected by 128 bits of entropy rather than the usual 160 bits.
  
• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address; cryptograffiti.info/)
  
• some are unspendable due to protocol bugs (the 50 BTC in question and the 100 BTC mentioned by zebedee above)
• some are locked by scripts which provably return false on any input.  Some of these are so badly lost they don't even have addresses! (367.75849319 BTC example).
  
• 23.1 mills (23.1 millibitcoins = 0.0231 BTC) may well never be mined into existence in the first place due to rounding all subsidies down to the nearest satoshi.
• some have already lost the chance to ever be mined into existence (example by midnightmagic: block #124724 mints only 49.999 999 99 BTC).
  

Note: I use "exist" here for any amounts of BTC in an up-to-date unpruned Bitcoin UTXO database.  There are of course different interpretations of what it means for a bitcoin to exist.  Even this seemingly simple definition is not water-tight as two different full nodes may disagree on the UTXO and in particular it allows one to claim that at one point, there existed more than 180 billion bitcoins.

And this is why I said ASSUMING NONE ARE LOST. From the very beginning, there was never a possibility of having 21,000,000 usable bitcoins, or, is there an extra block to account for the unusable first block, and the amount from the first block is usually left out of the total?

Quite a simple question, and while I appreciate the two attempts to answer my question, and it seems you guys put a lot of time and effort into explaining, my question was not really answered.

Thanks for the compliment.  I do pride myself on trying to actually answer questions here and not second-guess people.

I did read the question and took "assuming none are lost" on board.  Unfortunately, although this question seems simple, it is actually rather muddy and depends acutely on what is meant by "lost".

I can confirm that there is no "extra block" to make up for this 50 BTC output.  The fifth point of my cursory categorisation above claims this 50 BTC output to be part of the 21 million total.

Even before genesis, Satoshi did not intend for 21 million usable bitcoins to even be theoretically achievable.  The subsidy rounding code, for the sake of simplicity, sacrifices 23.1 mills (23.1 millibitcoins = 0.0231 BTC), leaving a theoretical maximum valid UTXO total of 20 999 999.976 900 00 BTC.  (Aside: At one point the intention was for a maximum of just 20 million BTC).

[futureofbitcoin]

Thank you. I always wondered about how there could be exactly 21 million coins despite bitcoins not being infinitely divisible (at least not currently). Now it makes total sense. Thank you for taking the time to actually read and answer. So many people nowadays don't even read the OP.

[Bizmark13]

How about satoshi takes out the dust coins, and make the genesis key public?

I don't see why this couldn't be done. After all, NXT pretty much did the exact same thing when its founder purposely made the private key of the genesis account easy to guess.

collector items are usually something you can show to others. For example, you can't really collect a secret that Einstein shared with your great grandfather and passed down to you. There's no verifiability and you can't really show it to anyone.

Pretty much the same with these addresses. I mean I guess you can "show" you have access to it by sending some bitcoins to someone's wallet from that address, but that's not really showing the item itself.

You could also sign a message from it as well.

Simple! The account holds 65.44243457 BTC! I would send 65.44233457 (0.0001 tx fees)!
No more! No Less 
#Like_A_BosS

I wouldn't pay more than 65BTC or the equivalent in fiat. It's doesn't have any special value to me and I wouldn't treat it as a collectible.
I'm sure it's worth a lot more to some people, just like art. For one person it's just a pretty picture worth $50. For another, who's a fan of the artist and collects his works it can be worth a million.

yeah I would think to some it has a lot of value. to others it is worth just 65 btc in change.

I've noticed many other people saying that they would buy it for ~65 BTC as well.

To be clear, the initial 50 BTC in that account is unspendable. It would really only be worth ~15 BTC + the value of the coins that might be sent to it in the future (which probably isn't significant now that the price of BTC has risen so much) + the collectible/coolness value of having the private keys to the genesis block. The value of the second one is speculative and the value of the last one is entirely subjective.

so is the total number of "usable" bitcoins, assuming none are lost to accidents, 20,999,950 instead of 21 million?

There were 100 BTC lost in two duplicated coinbase transactions (i.e. two pairs of blocks with the same coinbase txn, and as the protocol assumes txnID and output index are sufficient to designate an unspent output, means the later coinbase "overwrote" the earlier as it hadn't been spent by overwrite time), which is why version 2 blocks were created to eliminate that loophole.  So no, there are 150 BTC at first glance "granted" in blocks that are actually irretrievable forever...

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:...

I understand futureofbitcoin wasn't referring to coins that were lost after the creation of the genesis block, but to expand on the above two replies, this link might be of interest:

http://bitcointalk.org/index.php?topic=7253

if it can't be payout, then it worth nothing.
The only use of the genesis  block/address key is used to prove the identity of satoshi.

It is no worth more than 65.44243457 BTC if the amount is spendable otherwise it is worthless.

The 50 BTC is unspendable. The ~15 BTC that others have sent are spendable. At the bare minimum, it would still be worth ~15 BTC.

I doubt satoshi owns the key of this genesis bitcoin address due to the fact that the 50 BTC is unspendable. So many ppl just send dust amount to this genesis address to destroy their bitcoin. 

You're assuming a lot.  The genesis block was the only block where in some sense Satoshi was guaranteed the coins, as by construction he was competing with no one.

Only by construction. We know that he was mining pretty much alone for an entire week. And even after that, it was mostly just him and Hal Finney mining. A single 50 BTC block really wouldn't have changed things at all. A conscious decision to make the genesis block coins unspendable by design would probably have been done out of principle rather than for any practical reasons.

He may have viewed this as unethical, and therefore deliberately left out the code to add the genesis coinbase to the UTXO set - I find it hard to believe, though possible, that he wouldn't have realized that it wasn't in the UTXO set.  I suspect it was a deliberate choice.

Once he had released the code and anyone could download it, he might have ethically viewed all subsequent blocks as fair game (though no one other than Hal has stepped forward to claim a block earlier than #78) and so mined for all he could.

The fact that the genesis block coins are unspendable was a much later discovery although I guess it's possible that Satoshi knew of this beforehand. I believe the assumption at the time was that they were in fact spendable and to the best of my knowledge, Satoshi didn't say anything that would indicate otherwise.

Anyway, I'll see if I can find the exact quote...

EDIT: Found it:

This is how the Bitcoin client worked when Satoshi created it: The client maintains a block database and a transaction database. When it finds that its block database is empty, it inserts the genesis block into its block database to get things started. This block includes a transaction sending 50 BTC to 1A1z..., but the client does not insert this transaction into its transaction database. So even though this transaction is part of the block chain, if the client sees a transaction spending this 50 BTC, it won't be able to find the 50 BTC transaction in its transaction database, and the spending transaction will be rejected. In other words, the genesis block's transaction isn't considered to be a "real transaction" by the original Bitcoin client.

(Current versions of Bitcoin-Qt handle the block/transaction databases in a totally different way, so the genesis block's transaction is now just a weird special case in the code.)

This could easily have been an oversight, or Satoshi could have intended this transaction to be unspendable. This quirk wasn't discovered until after Satoshi disappeared, so no one had a chance to ask him.

Link: http://www.reddit.com/r/Bitcoin/comments/1nc13r/the_first_50btc_block_reward_cant_be_spend_why/

Thank you. I always wondered about how there could be exactly 21 million coins despite bitcoins not being infinitely divisible (at least not currently). Now it makes total sense. Thank you for taking the time to actually read and answer. So many people nowadays don't even read the OP.

It might be possible to asymptotically reach 21 million coins if Bitcoin adopts sub-satoshi mining rewards since the reason why the total supply is 0.0021 BTC less than 21 million BTC in the first place seems to be because there is no 0.000000005 BTC mining reward after the 6,929,999th block:

http://en.bitcoin.it/wiki/Controlled_supply#Projected_Bitcoins_Long_Term

[zebedee]

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address;

I guarantee you trillions of private keys corresponding to that address exist.  Mathematically around 2^96 should exist by the pigeon-hole principle.  So the probability that none do is so crushingly humungously vanishingly small that it makes creating a SHA256 preimage look like child's play.

[teukon]

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address;

I guarantee you trillions of private keys corresponding to that address exist.  Mathematically around 2^96 should exist by the pigeon-hole principle.  So the probability that none do is so crushingly humungously vanishingly small that it makes creating a SHA256 preimage look like child's play.

It is true that there are approximately 2⁹⁶ distinct private keys per address on average but I don't know how much can be said about the "lumpiness" of the distribution at this scale.  It seems plausible to me that something subtle in the algorithm causes some addresses to have many more than the average number of solutions while others have far fewer.  I don't see why there cannot be gaps of unobtainability scattered throughout the distribution, gaps too narrow to be discovered even with extensive testing.

If the distribution is seemingly random and uniform at the scale of individual addresses then I completely agree with your assessment of the probabilities.  Indeed, I would go further and claim that RIPEMD-160 is likely surjective.

[ensurance982]

This is a thorny topic and really depends on what you mean by "usable" and "lost to accidents".  Of the oft-cited 21 million BTC total:

• some are at addresses for which it is not known that a corresponding private key exists (1111111111111111111114oLvT2; The Bitcoin Eater address;

I guarantee you trillions of private keys corresponding to that address exist.  Mathematically around 2^96 should exist by the pigeon-hole principle.  So the probability that none do is so crushingly humungously vanishingly small that it makes creating a SHA256 preimage look like child's play.

True, but it doesn't matter that there is an infinite amount of private keys to any given address. The only thing important is the fact that the probability of generating a matching private key for a given address is 1/[numberOfBitcoinAddresses]