Sending coins to trash -- probably the biggest threat to bitcoin

[solex]

The issue has been clarified until it is as crystal clear as a glacial melt-water stream. Please, please re-read all the responses above.

I didn't yet see anyone confirming whether I had correctly interpreted point 2 in https://bitcointalk.org/index.php?topic=199542.msg2079488#msg2079488

Are you saying that as soon as that one bitcoin is seeded into circulation, and another chunk of it is eventually trashed, the process can continue to repeat itself indefinitely?

No. We are saying that it is not feasible to trash a significant number of bitcoins, as the value of the remaining ones (or portions) increases, so there is always an adequate quantity of bitcoins for monetary use.

Basically, the bitcoin network would not have to undergo some major and costly tuning in order to accommodate such repeat cycles of coin trashing by rogue miners, am I right?

Right. There is no feedback mechanism  for the network to respond to coin trashing and no feedback process is required. The network, and indeed, whole ecosystem behaves perfectly well whether the number of bitcoins increases (through mining) or decreases (through trashing exceeding mining). 11 million are mined already, so "rogue miners" won't get them.

Check out this address: http://blockchain.info/address/1BitcoinEaterAddressDontSendf59kuE

No private key exists, so 0.41196588 BTC has been trashed and permanently unspendable. Yet, this has no effect on the network.

[1715097050]

1715097050

[1715097050]

1715097050

[1715097050]

1715097050

[1715097050]

1715097050

[1715097050]

1715097050

[niniyo]

This does not threaten bitcoin any more than gold would be threatened by the prospect of people buying it and dumping it into the ocean.  If you do it, then you've just stupidly wasted money!

[Zaih]

Yeh, numbers mean nothing. 1 bitcoin or 21 million.

[wumpus]

Yes it'd take quite some nihilism to dump sizable amounts of coins in the trash just to make a statement at this point   And if so you're only making other people's coins worth more.

[BTConomist]

Yes it'd take quite a some nihilism to dump sizable amounts of coins in the trash just to make a statement at this point   And if so you're only making other people's coins worth more.

I'm afraid it's not quite that simple... You see, a single bitcoin that gets seeded into the BTC-based economy has the potential to grow BTC money supply (not to be confused with BTC monetary base of 21 million bitcoins) by 100,000,000 BTC (i.e. your average satoshi, but in a consumer-friendly notation). So, assuming that the BTC-based economy will eventually flourish, can you imaging the ripple effect an act of trashing a single bitcoin could have on that economy? This is precisely what I was hoping everyone here would look into, from a technical perspective (i.e. the resilience of bitcoin protocol to withstand such threats, if there were ever to be one), rather than focus their attention on bashing my concern as being invalid. My concern is just as valid as any other concern before it, even if it came about as part of diligent economic hacking, not code hacking. So, could some of you please take a look at the technical side of this -- is there an infinite loop (as perhaps was suggested by our very own SgtSpike) that a rouge miner would have to contend with when trying to carry out this type of an attack? Because if there is an infinite loop, then there's nothing to worry about -- bitcoin is poised to make history!

[kjj]

I'm afraid it's not quite so simple... You see, a single bitcoin that gets seeded into the BTC-based economy has the potential to grow BTC money supply (not to be confused with BTC monetary base of 21 million bitcoins) by 100,000,000 BTC (i.e. your average satoshi, but in a consumer-friendly notation). So, assuming that the BTC-based economy will eventually flourish, can you imaging the ripple effect an act of trashing a single bitcoin could have on that economy? This is precisely what I was hoping everyone here would look into, from a technical perspective (i.e. the resilience of bitcoin protocol to withstand such threats, if there were ever to be one), rather than focus their attention on bashing my concern as being invalid. My concern is just as valid as any other concern before it, even if it came about as part of diligent economic hacking, not code hacking. So, could some of you please take a look at the technical side of this -- is there an infinite loop (as perhaps was suggested by our very own SgtSpike) that a rouge miner would have to contend with when trying to carry out this type of an attack? Because if there is an infinite loop, then there's nothing to worry about -- bitcoin is poised to make history!

I have said many times over the years that we would very likely switch to a 128 bit integer for the value field for aesthetic reasons (native word size of 128 bit CPUs) long before it became economically necessary.  I doubt that even a determined attack could deplete the bitcoin supply enough to change that.

The software change would be easy.  Getting the network to go along with it would be hard right now because there is no need, but would be (relatively) easy if the need ever develops.

P.S.  Your thinking seems clouded.  1 BTC is the fundamental unit, not the satoshi.  The network is currently using a representation that goes to 8 decimal places because 8 decimal places is more than enough for the foreseeable future.  We call the current minimum protocol unit the satoshi because it was the first such unit, and we are sentimental creatures.  That unit has no other special features.  The value of 1 BTC is infinitely scalable; the software and network can be adjusted as necessary.  One satoshi would be just as good for commerce as 21,000,000 BTC, it would just require a different scaling factor.

[DeathAndTaxes]

It isn't a threat.  Fees for Bitcoin are likely to remain low.  Any attacker with >51% of hashpower can do a lot more than destroy coins at high cost.  Any attacker with less can't materially change the amount of Bitcoins.

Bitcoin fees have averaged 0.03% of value transferred.  Velocity (annual transaction volume / money supply) is ~10 and unlikely to be higher.  Let say Bitcoin stabilizes with fees at ~0.1% of volume transferred.  Then an attacker with 50% of hashpower could destroy 10*50%*0.1% = 0.5% of the money supply each year.  An attacker with less hashing power would have even less effect.

Now a small annual reduction in money supply has no detrimental effect but even if you think it does it would take 20 years to destroy just 10% of the money supply and 60 years to destroy 25%.  It is a complete non-issue.  Only the stupidest attacker in the world (who also has millions of dollars a year to burn) would engage in an attack like this and it would have absolutely no effect.

If 10% of the money supply was destroyed the exchange rate would rise ~10%.  With 8 digits of precision it would take centuries to raise the valuation of Bitcoin so high as to make transaction unweildly and if/when you did the protocol could be forked to add another 8 digits of precision.  In the process the "attacker" would massively increase the wealth of Bitcoin holders who don't destroy their coins.

[gmaxwell]

The software change would be easy.  Getting the network to go along with it would be hard right now because there is no need, but would be (relatively) easy if the need ever develops.

Right. Given a suitable deployment horizon I would expect the only opposition would be Luke (or some disciple of his order, N generations removed) because the rescaling would break tonal bitcoin.

You can even get four more places while fitting in 63 bits (but not while fitting in a Decimal64).
 

[2112]

I'm afraid it's not quite that simple... So, assuming that the BTC-based economy will eventually flourish, can you imaging the ripple effect an act of trashing a single bitcoin could have on that economy?

You have an imagination of some sort techno-taliban: like if dynamiting the Buddha statues would damage buddism.

[odolvlobo]

... You see, a single bitcoin that gets seeded into the BTC-based economy has the potential to grow BTC money supply ... by 100,000,000 BTC...

According to you, then burning a $100 bill would have a big effect on the U.S. economy, right? It would lower the GDP by $10 billion!

The first problem is your number.

If you are talking about fractional reserve banking, then the multiplier has traditionally been about 10 (though Goldman Sachs and its ilk managed to leverage it up to 40 for themselves, becoming a contributing factor to the financial crisis).

If you are talking about a fiscal multiplier, then the number is probably somewhere between 0.5 and 10. Read this: http://en.wikipedia.org/wiki/Fiscal_multiplier

The second problem is that the total wealth in the world is not equal to the total value of all the bitcoins. The portion of wealth in the world destroyed as a result of destroying a bitcoin would be imperceptible, even with the multiplier.

[Stephen Gornick]

I'm afraid it's not quite that simple... So, assuming that the BTC-based economy will eventually flourish, can you imaging the ripple effect an act of trashing a single bitcoin could have on that economy?

You have an imagination of some sort techno-taliban: like if dynamiting the Buddha statues would damage buddism.

This recurring and overblown worry about lost coins has me wondering of the source.  

I'm presuming this has little to do with truly protecting bitcoin, but instead is all about getting a foot in the door with some method of demurrage (call it reclaiming of "lost" coins or whatever).   Get that implmented and then a protocol change for currency inflation or some type of taxation is just an incremental step further.

[ctrlphreak]

Hasn't this topic already been discussed to death?

[jaywaka2713]

Miners can burn their money, but it is way to expensive to run such hardware to do so. As previously discussed, Bitcoin is infinitely divisible. I could see us breaking down Bitcoin 8 more decimal places when we have calculated that only about 10,000 BTC are left in existence. However this "trashing coins" process is one of the cons of having a decentralized currency. No central authority issues wallets, so there is no verification such wallets exist. Clients should check if the address coins are being sent to is in fact a valid address, but other than that there is almost nothing we can do about it.

[DeathAndTaxes]

Even if a client doesn't check, the protocol requires that a tx have valid addresses otherwise the tx is invalid.

Note there is a difference between invalid address and a valid one in which the private key is lost/unknown (or was never known).

[Tirapon]

The decimal point is arbitrary. If we got down to 21 BTC we'd just have to divide the satoshi into 1 million smaller units. Problem solved.

Also as you throw away money, the result would be that the value of what's left must increase proportionately as the supply drops. This would happen gradually, so if someone dumped a load of coins it would take a while for equilibrium to be reached again. But essentially anyone doing this would be giving all of their money to everybody else. So let them go for it.

[2112]

This recurring and overblown worry about lost coins has me wondering of the source.  

I'm presuming this has little to do with truly protecting bitcoin, but instead is all about getting a foot in the door with some method of demurrage (call it reclaiming of "lost" coins or whatever).   Get that implmented and then a protocol change for currency inflation or some type of taxation is just an incremental step further.

I understand your concern above, but my thinking is more skewed to applying the Hanlon's razor: a combination of two phenomenons:

a) intense misinformation about Bitcoin's properties both amongst its supporters and its detractors;

b) influx of extremely literally-minded people who have difficulty with abstract or transcendental thinking;

makes people worry that the bean-counting professions will either run out of beans to count or bean-counting will be supplanted by e.g. pea-counting.

I don't see any cunning or even any long-term planning: it is just a pre-conditioned response to an unfamiliar stymuli.

[jaywaka2713]

Even if a client doesn't check, the protocol requires that a tx have valid addresses otherwise the tx is invalid.

Note there is a difference between invalid address and a valid one in which the private key is lost/unknown (or was never known).

Thats why I was talking about one of the cons of decentralization. There is no way to verify if an address has been generated or not because nobody distributes them. It's just something we have to deal with, but isnt a problem.

[BTConomist]

I could see us breaking down Bitcoin 8 more decimal places when we have calculated that only about 10,000 BTC are left in existence.

Thanks for confirming that I have correctly interpreted the second point in SgtSpike's answer. Unfortunately, this now means that I have to go back to the drawing board and figure out at what point it would make sense to add more decimal places in order to avert an inflationary spiral that will eventually result from lost coins (if we ever succeed at building a flourishing BTC-based economy around the bitcoin protocol).

I know that the issues I'm trying to bring to the miners' attention may appear somewhat odd at this point, but please take note of the inferences that I've made throughout this thread, as understanding them would prove essential in a miner's ability to take maximum advantage of the GOLD 2.0 era we're passing through right now. I don't think that it will be years before I can finally share (in layman's terms) what had clicked in my head after I came across one of Gavin's old tweets, back in 2011 (see a hard copy below). It's just I haven't yet found a professional writer (with background in bitcoin and economic theory) to help me put the right words on paper, so to speak. Why would I need a professional writer? Well, as some of you may have already noticed, any attempt that I take to construct a structural argument around BTC money supply (not monetary base units, a.k.a. raw bitcoins) it only seems to take everyone's attention away from the main points of that argument... It appear that I'm just not cut out for explaining intricate things in prose.

[solex]

Unfortunately, this now means that I have to go back to the drawing board and figure out at what point it would make sense to add more decimal places in order to avert an inflationary spiral that will eventually result from lost coins ...

These are the fundamental errors in your logic:
(a) Lost coins are deflationary not inflationary.
(b) There is no spiral. Just because one coin is lost there is no causative connection so that more and more are lost because of the earlier losses.

[odolvlobo]

It appear that I'm just not cut out for explaining intricate things in prose.

It is more likely that you just don't have a clue. haha sorry couldn't resist.

Seriously, the problem is that you haven't stated what you think the problem is. Also, you keep bringing up Gold 2.0 like we know what you are talking about and like it is a real concept.

You don't have to write prose. Just write a series of steps -- cause and effect, and we can help fill in the details.