Is this idea to counter lost bitcoins possible?

[scrybe]

What I mean is that Bitcoin, as it stands now, is in heavy favour of those that possess Bitcoins. What you are essentially saying by "feeing up" "lost" coins, is you want to swing it the other way BACK towards those that don't have Bitcoins yet, you want to ease some supposed deflationary spiral that will work against those that don't hold Bitcoins.

Oh here we go again with the profit argument. Yeah lets just ignore this idea of stabilizing the money supply just because you ELITE BITCOIN HOLDERS with a crap load of coins can make a profit from it.

And your logic is completely flawed anyway, because people not holding coins are hardly going to give a shit about something which will happen 100 years from now. I do in fact hold bitcoins, and quite a few of them, so this move would not be in my best interest.

What I know is this, right now, my Bitcoins are MINE, not the networks, not Satoshi's or the developer's, not some government's, not the UN's, MINE.

No one is saying they aren't yours. You can easily keep them if you take the effort every 100 years to show the network that those coins are still active. I mean what is so blasphemous about this if it can help stabilize the money supply. Oh the horror...

You are proposing stealing from my pocket, or my childrens, or my grandchildrens, or...

It DOES NOT MATTER WHEN IT HAPPENS this is still theft, destabilization, and a very, very bad idea.

Please explain, as you have been asked multiple times, why a STABLE NUMBER OF COINS is required in any way at all. Your root premise is flawed as multiple people have pointed out.

[1714885081]

1714885081

[1714885081]

1714885081

[1714885081]

1714885081

[1714885081]

1714885081

[bitfreak!]

Please explain why Bitcoin must continue indefinitely and never become a scarce commodity?

It already is a scarce commodity because the supply is limited. I'm suggesting that we make it a stable scarce commodity, a bit like how gold doesn't just disappear into thin air, the supply on Earth remains consistently stable.

There is also the point that you appear to be willfully missing. BitCoin was created with a fixed set of rules that a lot of folks are vested in, changing those requires a consensus, so your proposal is unlikely to ever get acceptance for "BitCoin."

I am not missing that point, I am well aware of it and well aware that the chance of this change ever being implemented is extremely remote. But I am still going to discuss my reasoning and opinion for why it does need to be changed.

[Peter Todd]

Please provide a single example.  It took roughly 15 years for MD5 to go from theoretical attack to "fully broken" and even today it requires a huge amount of computing power to cause a MD5 collision.  At each improvement of the attack it still required a huge amount of computing power but the speed improvement over brute force grew by magnitudes.  Almost like mining but in reverse ("difficulty" declined over time).   Note this doesn't mean MD5 should be used, there simply are better safer algorithms but the idea that it went from academic to "instahack" with negligible computing power is simply innacurate.

Please name a single mainstream cryptographic algorithm (not some security through obscurity or proprietary solution created by an novice) which was broken wide open in a single step.

Name a single example of a broken public-key algorithm previously in wide use in the first place. Frankly we just don't know because we've yet to see a public-key algorithm broken in recent history.

FWIW MD5 collisions can be generated in 2^20 time right now, trivial. More importantly though look at Bitcoin difficulty, which is of course a partial pre-image: the security of the network is just a matter of a few "bits" of difficulty, yet even just a single paper in crptography is likely to remove at least a bit or two of security from an algorithm. It's easy to see why an algorithm can go from secure to insecure overnight if that last step happens at the wrong place, while at the same time having warning is likely.

Lets suppose for instance RIPEMD160 pre-image collisions went from 2^160 difficulty to 2^120: it'd be worrying, but still Bitcoin would be ok. Now another paper or six are published, and you're down to 2^70: rather worrying, but Bitcoin is still ok for now, if just barely. (the whole Bitcoin network has done about 2^66 SHA256's in total remember) One more paper though, another 10 bits, and suddenly stealing funds sent to address hashes is quite practical with just a few ten thousand in equipment, and each additional bit of security lost halves that difficulty.

There's a reason the difficulty target is a number, rather than just a number of bits.

[bitfreak!]

You are proposing stealing from my pocket, or my childrens, or my grandchildrens, or...

Ok let me give you an analogy. You place some gold in a locked box and then 100 years later I find the box just sitting in some remote location covered in dust. I am going to take that box and bust it open and take the gold, because clearly some one has lost it and left it there. Now if it weren't lost, I wouldn't have the chance to find a 100 year old box covered in dust because someone would have come in, cleaned the dust off and relocated it to a more secure location, giving clear indications that the box is certainly not lost or abandoned. The fact is that anyone would have a very large chance to stop their coins from being re-mined, and if they fail to do so then the coins are fairly considered to be lost, because there are real lost coins out there which need to be found. If you happen to get caught up in that process it's you're own fault for not taking action to secure ownership of the coins within a reasonable time frame.

Please explain, as you have been asked multiple times, why a STABLE NUMBER OF COINS is required in any way at all.

I have already explained throughout this thread why a stable money supply is more economically sound than infinite deflation. Infinite anything is absurd. Of course it's not necessarily an absolute necessity, but it is more economically sound and therefore superior and more desirable.

[TangibleCryptography]

Meh.  2^56 operations would still be roughly the equivelent of difficulty 16 million.  The idea that is "quite practical" is kinda a weak argument.  You are talking building a network say 10% of the size of the bitcoin network would allow you to steal a couple dozen addresses per day.   Of course that assumes that security can be reduced to 2^56 operations.  I pointed out that even SHA-1 which has been known compromised since 2005 still requires > 2^61 operations. 

Will coins be stolen.  Sure in time but it won't be these "insta-hack" nonsense you espouse.   The period of time between improved attacks is usually measured in years.  So when security is reduced to 2^56 you will see the largest dormant addresses "remined" and then it really won't be worth it.  Maybe in 4-5 years someone will improve the attack further and the medium sized addresses will be compromised.   While this brings coins into the network it will hardly be this "instantly every RIPEMD-160 address is owned by a single person in the span of a few seconds.

It will take time and honestly won't be much different than mining now.  There will be risk involved.  Do I bought a $10K RIPEMD-160 cracker out of FPGAs?  What if others get the best addresses first?  What if BFL releases a RIPEMD-160 ASIC before I pay off my "rig".   I would also point out that it would be decentralized.   It would be the FIAT command from on high deeming RIPEMD-160 addresses unfit for use and therefore nulled.  I would never, ever, ever support it.  I would spend the last of my coins fighting it and if it still did happen I would have no reason to use Bitcoin.    Bitcoin where wealth can be confiscated at will because the elites deem it necessary?  I will go back to Gold.    Many Goldbugs BTW fear this very idea that Bitcoin isn't truly Tangible.   If it can be confiscated by decree then they are right.

[scrybe]

You are proposing stealing from my pocket, or my childrens, or my grandchildrens, or...

Ok let me give you an analogy. You place some gold in a locked box and then 100 years later I find the box just sitting in some remote location covered in dust. I am going to take that box and bust it open and take the gold, because clearly some one has lost it and left it there. Now if it weren't lost, I wouldn't have the chance to find a 100 year old box covered in dust because someone would have come in, cleaned the dust of and relocated it to a more secure location, giving clear indications that the box is certainly not lost or abandoned. The fact is that anyone would have a very large chance to stop their coins from being re-mined, and if they fail to do so then the coins are fairly considered to be lost, because there are real lost coins out there which need to be found. If you happen to get caught up in that process it's you're own fault for not taking action to secure ownership of the coins within a reasonable time frame.

This argument falls on it's face. With BitCoins you CANNOT "clearly" know that they have been "lost." There is no dusty box or other indication besides the last time the address was used. If I have placed some gold in a safe I expect it to stay there. If my great-great grandchildren re-discover and open by old safe with paper bitcoins in it, you want the gold to already be gone.

Please explain, as you have been asked multiple times, why a STABLE NUMBER OF COINS is required in any way at all.

I have already explained throughout this thread why a stable money supply is more economically sound than infinite deflation. Infinite anything is absurd. Of course it's not necessarily an absolute necessity, but it is more economically sound and therefore superior and more desirable.

No, you have not explained, you have claimed this repeatedly (and without attribution) but you have not justified these statements in any significant way. No mentions of the possible consequences or scenarios that you are testing against. You are just saying it over and over again.

Since you are arguing that something which already exists should be changed, and everyone else is arguing for it to stay the same, you have the burden of proof. I'm offering you a way to convince some folks rather than just saying "should be" over and over again.

There have been a large number of substantial objections raised to your idea including practicality, social contracts, existing code, design intentions, market forces, future impacts, and most importantly the immutability of transactions themselves. You are questioning the very foundations of bitcoin by proposing that transactions can EVER be undone, your burden of proof is enormous.

[Rudd-O]

The beauty of Bitcoin is that it is voluntary. You can fork it and implement your idea, see how it goes.

I don't want to create a new fork of bitcoin, I have no intention to do so, nor the skills to do so. I want to make bitcoin more economically sound and stable over the very long term. Sure, the network could continue operating with only one BTC, but that is not healthy or sound in the long term. The fact this argument has been turned into some vendetta against a cartel of elite bitcoiners is absolutely ridiculous.

You don't want to create an alt coin? Live with the one you got cos you ain't changing the existing one. Or don't use Bitcoin. It bears repeating: unlike government money (which has the properties you seem to want), nobody is forcing you to use Bitcoin. You already have what you want available to use today. Use dollars and stop trying to bring its fucked-up disadvantages to Bitcoin.

At any rate, you ain't fooling anyone. This whole 'making Bitcoin more healthy' and 'stable money supply' is statist code speak for imposing your opinion on stealing from existing wealth owners. Perhaps that explains why you annoyingly keep repeating the same nonsensical and absurd conclusions of 'stealing old money is good' like a broken record without actually ever substantiating them.

[Peter Todd]

Meh.  2^56 operations would still be roughly the equivelent of difficulty 16 million.  The idea that is "quite practical" is kinda a weak argument.  You are talking building a network say 10% of the size of the bitcoin network would allow you to steal a couple dozen addresses per day.   Of course that assumes that security can be reduced to 2^56 operations.  I pointed out that even SHA-1 which has been known compromised since 2005 still requires > 2^61 operations.

Stealing a couple dozen addresses a day can give you serious amounts of money. There are addresses that have historically had amounts in the low millions assigned to them, and I'm sure the unspent-tx-out set has plenty of addresses with at least tens of thousands attached to them.

Will coins be stolen.  Sure in time but it won't be these "insta-hack" nonsense you espouse.   The period of time between improved attacks is usually measured in years.  So when security is reduced to 2^56 you will see the largest dormant addresses "remined" and then it really won't be worth it.  Maybe in 4-5 years someone will improve the attack further and the medium sized addresses will be compromised.   While this brings coins into the network it will hardly be this "instantly every RIPEMD-160 address is owned by a single person in the span of a few seconds.

It will take time and honestly won't be much different than mining now.  There will be risk involved.  Do I bought a $10K RIPEMD-160 cracker out of FPGAs?  What if others get the best addresses first?  What if BFL releases a RIPEMD-160 ASIC before I pay off my "rig".   I would also point out that it would be decentralized.   It would be the FIAT command from on high deeming RIPEMD-160 addresses unfit for use and therefore nulled.  I would never, ever, ever support it.  I would spend the last of my coins fighting it and if it still did happen I would have no reason to use Bitcoin.    Bitcoin where wealth can be confiscated at will because the elites deem it necessary?  I will go back to Gold.    Many Goldbugs BTW fear this very idea that Bitcoin isn't truly Tangible.   If it can be confiscated by decree then they are right.

That's all well and good, but when the difference to that stealing happening over a matter of years, or a matter of days or weeks is one paper, it'll cause the value of Bitcoin to tank all the same. Mining as we know it is carefully controlled and highly predicable, "mining" by attacks on cryptography isn't.

Also, if a RIPEMD160 attack does happen, we have just two backup options: move to SHA256 for address hashes, or move to pure public keys, both of which are backwards compatible changes. If an ECDSA break happens we have no choice but to move to a new PK algorithm. Doing that is a hard-fork change if you want to be able to spend your coins, and thus this discussion will come up. Given how many people consider protection against inflation as one of the core principles of Bitcoin protection against the risk of inflation by releasing old lost coins will be discussed. You're hung up on the idea of it being confiscated by "decree", but plenty more are just going to worry about their wealth being confiscated however it happens.

Of course all this depends on how many coins are lost; maybe it won't be an issue. Note that one way this all could be handled is to have the new rules in this hard-fork scenario only allow old coins to be spent on a schedule, with limits on the volume of such spending. You could say that after block n only transactions from weak to strong are allowed, and only xBTC of those transactions per block. (with the obvious tx fee competition happening) Or the rules could state that making the coins undependable only happens if a certain percentage of coins in circulation remain under weak keys beyond some date, thus limiting the possible inflation. We may find in such a scenario that that percentage is low enough that people aren't worried about the inflationary wealth confiscation. Of course coming to consensus in any case will be hard. Even just moving to a new PK algorithm will be bad enough.

I really, really, really want to see the look on those Goldbugs faces when someone mines the first meteorite, or someone makes a breakthrough in gold nucleosynthesis... In the latter I suspect they'll start arguing that gold isn't gold unless it's totally free of trace radioactive contamination.

[bitfreak!]

This argument falls on it's face. With BitCoins you CANNOT "clearly" know that they have been "lost." There is no dusty box or other indication besides the last time the address was used. If I have placed some gold in a safe I expect it to stay there. If my great-great grandchildren re-discover and open by old safe with paper bitcoins in it, you want the gold to already be gone.

Of course you can never truly know if they are lost, however if people are given a truly fair window of time to show the network that they aren't lost I fail to see the problem. You could tell your grandchildren to make sure they they enforce their right to the coins before that window expires. And if they also want to put coins in storage for a long period of time and hand them down, they can do the same thing. It's not rocket science, nor some tedious task which needs to be repeated every year. We are talking extremely long periods of time here.

[Rudd-O]

Of course you can never truly know if they are lost, however if people are given a truly fair window of time to show the network that they aren't lost I fail to see the problem.

Translation from statist codespeak to English:

I want to impose on Bitcoin a policy of "Prove to me that you own what is yours, or else I'll pretend it's not yours, steal it from you, and call that 'fair'."

The whole point of this is, of course, to impose Keynesian policy by discouraging long-term saving.

This is why I just love Bitcoin-the-existing-blockchain: it makes it impossible for misinformed and propagandized fools to impose their terrible cached ideas on everyone else.

Adding to ignore list... done.

[scrybe]

This argument falls on it's face. With BitCoins you CANNOT "clearly" know that they have been "lost." There is no dusty box or other indication besides the last time the address was used. If I have placed some gold in a safe I expect it to stay there. If my great-great grandchildren re-discover and open by old safe with paper bitcoins in it, you want the gold to already be gone.

Of course you can never truly know if they are lost, however if people are given a truly fair window of time to show the network that they aren't lost I fail to see the problem. You could tell your grandchildren to make sure they they enforce their right to the coins before that window expires. And if they also want to put coins in storage for a long period of time and hand them down, they can do the same thing. It's not rocket science, nor some tedious task which needs to be repeated every year. We are talking extremely long periods of time here.

...and you ignored all the rest of my points and replied flippantly to this one.

Ignored.

[Rudd-O]

This argument falls on it's face. With BitCoins you CANNOT "clearly" know that they have been "lost." There is no dusty box or other indication besides the last time the address was used. If I have placed some gold in a safe I expect it to stay there. If my great-great grandchildren re-discover and open by old safe with paper bitcoins in it, you want the gold to already be gone.

Of course you can never truly know if they are lost, however if people are given a truly fair window of time to show the network that they aren't lost I fail to see the problem. You could tell your grandchildren to make sure they they enforce their right to the coins before that window expires. And if they also want to put coins in storage for a long period of time and hand them down, they can do the same thing. It's not rocket science, nor some tedious task which needs to be repeated every year. We are talking extremely long periods of time here.

...and you ignored all the rest of my points and replied flippantly to this one.

Ignored.

But, but, but TRULY!  TRULY!  TRULY!  OUTRAGEOUS!

Haha :-) 

He's clearly not engaged in a conversation with anyone here -- the very second anyone questions his terrible ideas, he immediately regresses to unsubstantiated suppositions, opinions and theories, making his participation here useless, bad and boring.  I added him to my ignore list too.

[Meni Rosenfeld]

I would like to mention in this context something which does not get the attention it deserves.

It is plausible that the currently used version of ECDSA will be broken eventually. When weaknesses start to be found people will start moving their coins to more secure addresses. But the lost coins will remain where they are, and when it is finally broken, whoever does it will find himself with a huge treasure of coins. This isn't stable and is not how Bitcoin is supposed to work.

I think we need to consider agreeing that some time after the signature algorithm shows weaknesses, we will delete all old coins to which the keys were lost. It's either that or have them all suddenly move to one party.

I don't think we need to agree to that at all.  Who makes "you" (to mean not just you but anyone who feels they have the authority to control the wealth of others) to decide when wealth should be confiscated in order to protect others.  You are the bitcoin "elite" you need to protect others by confiscating their wealth?  You are talking about the role of a state.  The elite protecting the masses by confiscating wealth through force.  Bitcoin is dead if/when the "self proclaimed elites" out of fear decide they need to protect it by confiscating wealth.

I think of it more as a community consensus to make sure Bitcoin works the way it was designed to - that lost coins remain lost. I think when the time comes we'll figure out what is a fair solution.

That said I understand the issues with this deletion, and hence I will not press this further at this point.

[Rudd-O]

I would like to mention in this context something which does not get the attention it deserves.

It is plausible that the currently used version of ECDSA will be broken eventually. When weaknesses start to be found people will start moving their coins to more secure addresses. But the lost coins will remain where they are, and when it is finally broken, whoever does it will find himself with a huge treasure of coins. This isn't stable and is not how Bitcoin is supposed to work.

I think we need to consider agreeing that some time after the signature algorithm shows weaknesses, we will delete all old coins to which the keys were lost. It's either that or have them all suddenly move to one party.

I don't think we need to agree to that at all.  Who makes "you" (to mean not just you but anyone who feels they have the authority to control the wealth of others) to decide when wealth should be confiscated in order to protect others.  You are the bitcoin "elite" you need to protect others by confiscating their wealth?  You are talking about the role of a state.  The elite protecting the masses by confiscating wealth through force.  Bitcoin is dead if/when the "self proclaimed elites" out of fear decide they need to protect it by confiscating wealth.

I think of it more as a community consensus to make sure Bitcoin works the way it was designed to - that lost coins remain lost. I think when the time comes we'll figure out what is a fair solution.

That said I understand the issues with this deletion, and hence I will not press this further at this point.

I am positive of one thing: if at some point any "self-appointed elite" with commit access manages to sneak in the source code a proviso to steal existing coins -- regardless of the form of this proviso -- I will immediately fork the code, revert the change, and start running nodes to keep the original non-organized-theft code alive.  Then we'll see if people choose to stay with a currency paradigm that unilaterally decided to steal from them, or continue using the original paradigm.

[MoonShadow]

Keep in mind I know very little about the complex inner workings and technical details of bitcoin so keep your answers relatively simple.

It would work, but it's unnecessary.  It's extremely improbable that the crypto algo currently in use for creating address keypairs will remain secure for that entire time period.  Bitcoin includes an upgrade path to a more secure algo, without exposing any security issues in the meantime, should it start to look like the algo is at risk.  The leading 1 at the beginning of current bitcoin addresses is what denotes the version of bitcoin address.  Bitcoin doesn't presently recognize any other, but it eventually shall.  At some point, it's going to become apparent that the current algo is at risk of breakdown, for whatever reason, and the community is going to transition to another algo.  Any unclaimed transactions that have not been moved to addresses using this new algo will ultimately become salvage to whomever can brute force a collision first.  Thus, lost bitcoins are not really lost permanently, just not likely to be recovered within the lifetime of anyone here.

EDIT: Actually, Bitcoin does recognize a special address version for the "testnet", but those are not usable addresses in the live bitcoin network.

[bitfreak!]

I want to impose on Bitcoin a policy of "Prove to me that you own what is yours, or else I'll pretend it's not yours, steal it from you, and call that 'fair'."

In the real world if you leave something untouched for over 100 years and then fail to prove it's yours when some one else tries to take it, they have every right to take it. What reason do they have to actually believe it's not lost if you can't prove it's not lost? Of course you might say "it's sitting in my house", but unfortunately there is no such easy way to prove that bitcoin is sitting in an address which proves you own it. But there are still easy ways to prove you own it... but that's not the problem here, the problem here is that you simply don't want to have the network discern any difference between lost coins and active coins, because you only want to make a profit from infinite deflation, regardless of the economic holes in such a system.

I could argue infinite inflation really isn't bad... you know it just benefits a few people, but in reality the market will always adjust even if we have a quadrillion trillion trillion dollars in circulation. No problem folks, that's entirely economically healthy and I see no reason why we should try to avoid that if the system can still work with that much money in circulation.  

But I don't think you will even be able to read this since you've ignored me and don't see anything you don't want to see. Well done.

[xxjs]

Oh here we go again with the profit argument. Yeah lets just ignore this idea of stabilizing the money supply just because you ELITE BITCOIN HOLDERS with a crap load of coins can make a profit from it.

Why sulk, just jump in. You can be an early adopter.

[bitfreak!]

Oh here we go again with the profit argument. Yeah lets just ignore this idea of stabilizing the money supply just because you ELITE BITCOIN HOLDERS with a crap load of coins can make a profit from it.

Why sulk, just jump in. You can be an early adopter.

I already have jumped in and I already do own a fair amount of BTC. That's not the point here, I'm not trying to profit from infinite deflation, my motive is to create a more economically sound currency which will be more stable over the very long term. My main motivation is not greed or irrational fear of having my coins stolen 100 years from now.

[MoonShadow]

Oh here we go again with the profit argument. Yeah lets just ignore this idea of stabilizing the money supply just because you ELITE BITCOIN HOLDERS with a crap load of coins can make a profit from it.

Why sulk, just jump in. You can be an early adopter.

I already have jumped in and I already do own a fair amount of BTC. That's not the point here, I'm not trying to profit from infinite deflation, my motive is to create a more economically sound currency which will be more stable over the very long term. My main motivation is not greed or irrational fear of having my coins stolen 100 years from now.

You cannot hope to do this, for that is what Bitcoin already is.  Bitcoin's current exchange instability has zero to do with it's deflationary nature, because that won't occur until about 2130.  Bitcoin is still hugely inflationary, and will be more inflationary than the US $ and the Euro for another four years or more; unless something dramatic happens to one of those other currencies.  Bitcoin's current instability is most related to the very tiny size of the market for which it serves.  Even in 2130, Bitcoin will not be particularly deflationary.  The monetary base will simply be very stable, which is not the same as deflationary.  Those who assume that Bitcoin will prove to be deflationary because of lost addresses taking bitcoins out of circulation are still assuming that to be so.  Bitcoin is based upon Austrian Economic Theory, and there is no more 'sound' basis for a currency.  If you want to start a perpetually inflating version, go right ahead, but it's already been done several times, and every time it has faded away in mere months.  Bitcoin was designed to mimic gold, and it does this very well.

[bitfreak!]

Bitcoin is based upon Austrian Economic Theory, and there is no more 'sound' basis for a currency.  If you want to start a perpetually inflating version, go right ahead, but it's already been done several times, and every time it has faded away in mere months.  Bitcoin was designed to mimic gold, and it does this very well.

Clearly you haven't understood much of what I've said in this thread. I'm not talking about anything related to infinite inflation. Let me repost some things I've said in this thread and the other thread about a design contract so that you can see what I'm describing is CLOSER to gold and MORE aligned with the Austrian mind set.

If 500 tons of gold were lost at the bottom of the ocean for many years and the loss of this gold therefore gets factored into the gold price, I think that gold would be a better currency if the lost gold is never found because the price would then be more stable.

No the price would not be more stable. It would shift in one direction constantly for as long as we used it. But we don't live in some irrational world where lost gold can never be found again, so the opposite thing also applies; when gold is found the market adjusts and factors that into the gold price. In a world where all the gold has been mined, we would only have some gold being lost and some gold being found. Overall, this would result in the value of gold remaining at a stable pace, and not shifting infinitely in any one direction. The main factor applying a change to the value of gold will be natural market forces such as supply and demand in accordance with natural economic growth. It doesn't take a genius to see that this economic model is by far superior to a model which shifts infinitely in one direction, whether that direction is inflation or deflation. Both options taken to infinite extremes are not economically sound.

This is a sarcastic remark which helps illustrate my point further:

I could argue infinite inflation really isn't bad... you know it just benefits a few people, but in reality the market will always adjust even if we have a quadrillion trillion trillion dollars in circulation. No problem folks, that's entirely economically healthy and I see no reason why we should try to avoid that if the system can still work with that much money in circulation.

And this was my first analogy using gold:

To further illustrate the point I'm trying to make here, let me draw a comparison to gold, as people often like to compare bitcoin to gold in the way they function.

Gold can be lost, but not in the same way bitcoins are lost. When a person loses some gold, it doesn't just vanish into to thin air with virtually no chance of ever being recovered again, in the way bitcoins are lost. It will be some where waiting for someone to find it again at some point, even if they have to mine it out of the ground again. Would any of you really like a world where gold could vanish into thin air, and the supply of gold would slowly get smaller and smaller? Well of course some of you probably would judging by this thread so far.

But my point is that's not a healthy economic system, anymore than infinite inflation is a healthy economic system. Both are fundamentally flawed at a very basic level. Like gold, we need some system which will allow us to rediscover the lost bitcoins and bring them back into the system. The idea of a limited supply is great and fantastic, but it's not just a limited supply, it's an ever decreasing limited supply. It's easy to see that a stable money supply is by far the most superior currency structure, and bitcoin should be designed in the most superior way.