LearnBitcoin

Bitcoin Dev Archive 2017

BIP - Dead Man's Switch

9 messages, 11 to 12 Dec 2017. Started by Teweldemedhin Aberra.

Teweldemedhin Aberra (2), Douglas Roark, Nick Pudar, Radoslaw Biernacki, Pieter Wuille, Chris Riley, Luke Dashjr, Ricardo Filipe

Permalink Original thread on gnusha.org

Teweldemedhin Aberra permalink original

It is estimated that about 4 million of the about 16.4 Bitcoins ever mined are lost forever because no one knows the private keys of some Bitcoin addresses. This effectively mean there are actually only 14.4 million Bitcoins in circulation even though 16.4 million are mined. There is no way of eliminating the human errors that cause these losses of Bitcoin from circulation, while the number of Bitcoin that will ever be mined is capped at 21 million. This means the total number of Bitcoins that are in circulation will eventually become zero, bringing the network to an end.

The solution this BIP proposes is to implementing a dead man's switch to Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to dormant addresses to automatically expire. A Bitcoin address is deemed dormant if it is not used in transactions for some fixed length of time, say ten years.

The calculation of the miner's reward should take into account the Bitcoins that has expired. This means there is a possibility that miner's reward can increase if sufficient number of Bitcoins expire.

Ref:

http://fortune.com/2017/11/25/lost-bitcoins/

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free. www.avast.com <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Douglas Roark in reply to Teweldemedhin Aberra permalink original

With all due respect, this isn't a BIP. It's idle speculation regarding what one person considers to be a problem and others may not. Please read https://github.com/bitcoin/bips/blob/master/bip-0001.mediawiki and try again. Among other things:

- Convince us this is a real issue, and that your data is accurate. Who's to say which coins are truly lost? Maybe the people controlling the keys are just laying low, wish to use their coins in 2112, etc. - Propose formulas for how the miners will take everything into account, and explain why they're acceptable. - Write code that implements your ideas.

Good luck!

Nick Pudar in reply to Teweldemedhin Aberra permalink original

This topic has come up several times in recent years. While it is well intentioned, it can have devastating outcomes for people that want to save long term. If such a system were implemented, it would force people to move funds around in order to not get nullified. In that process, it introduces multiple opportunities for errors. Cold storage should be able to stay cold. I personally would be apprehensive about implementing this kind of a system.

...via Android

From: Teweldemedhin Aberra via bitcoin-dev Sent: Monday, December 11, 1:04 PM Subject: [bitcoin-dev] BIP - Dead Man's Switch To: bitcoin-dev@lists•linuxfoundation•org

It is estimated that about 4 million of the about 16.4 Bitcoins ever mined are lost forever because no one knows the private keys of some Bitcoin addresses. This effectively mean there are actually only 14.4 million Bitcoins in circulation even though 16.4 million are mined. There is no way of eliminating the human errors that cause these losses of Bitcoin from circulation, while the number of Bitcoin that will ever be mined is capped at 21 million. This means the total number of Bitcoins that are in circulation will eventually become zero, bringing the network to an end. The solution this BIP proposes is to implementing a dead man's switch to Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to dormant addresses to automatically expire. A Bitcoin address is deemed dormant if it is not used in transactions for some fixed length of time, say ten years. The calculation of the miner's reward should take into account the Bitcoins that has expired. This means there is a possibility that miner's reward can increase if sufficient number of Bitcoins expire.

Ref: http://fortune.com/2017/11/25/lost-bitcoins/

Virus-free.
        <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
www.avast.com
Radoslaw Biernacki in reply to Teweldemedhin Aberra permalink original

Aside from that such change would require a hard fork it also violates one of basic rules of bitcoin, which has long term consequences for miners and for whole Bitcoin economy. In short, after altering the supply limit it would not be "bitcoin" anymore.

On Mon, Dec 11, 2017 at 6:30 PM, Teweldemedhin Aberra via bitcoin-dev < bitcoin-dev@lists•linuxfoundation•org> wrote:

Quoted text (36 lines)
> It is estimated that about 4 million of the about 16.4 Bitcoins ever mined
> are lost forever because no one knows the private keys of some Bitcoin
> addresses. This effectively mean there are actually only 14.4 million
> Bitcoins in circulation even though 16.4 million are mined. There is no way
> of eliminating the human errors that cause these losses of Bitcoin from
> circulation, while the number of Bitcoin that will ever be mined is capped
> at 21 million. This means the total number of Bitcoins that are in
> circulation will eventually become zero, bringing the network to an end.
>
> The solution this BIP proposes is to implementing a dead man's switch to
> Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to
> dormant addresses to automatically expire. A Bitcoin address is deemed
> dormant if it is not used in transactions for some fixed length of time,
> say ten years.
>
> The calculation of the miner's reward should take into account the
> Bitcoins that has expired. This means there is a possibility that miner's
> reward can increase if sufficient number of Bitcoins expire.
>
> Ref:
>
> http://fortune.com/2017/11/25/lost-bitcoins/
>
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
> <#m_-612306741899295358_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists•linuxfoundation•org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>
>
Pieter Wuille in reply to Nick Pudar permalink original

On Dec 11, 2017 10:23, "Nick Pudar via bitcoin-dev" < bitcoin-dev@lists•linuxfoundation•org> wrote:

This topic has come up several times in recent years. While it is well intentioned, it can have devastating outcomes for people that want to save long term. If such a system were implemented, it would force people to move funds around in order to not get nullified. In that process, it introduces multiple opportunities for errors. Cold storage should be able to stay cold. I personally would be apprehensive about implementing this kind of a system.

Furthermore, if it rewards miners with funds that are expired, it creates terrible incentives. Miners in their best interest could choose to censor transactions that move funds close to their expiration time, to increase their own future rewards.

Cheers,

Chris Riley in reply to Teweldemedhin Aberra permalink original

Hi, 1. If there are 16.4 million mined and 4 million are lost, that results in 12.4 million in circulation vs 14.4 million. 2. Satoshi addressed this as have numerous other people ( https://bitcointalk.org/index.php?topic=198.msg1647#msg1647 ) - lost coins decrease supply, increasing value of the remaining coins. 3. This assumes this is a problem. Bitcoin is divisible, 100 million, potentially more if necessary. ( https://en.bitcoin.it/wiki/Help:FAQ#How_divisible_are_bitcoins.3F) 4. Why is it okay to steal bitcoins from people who's bitcoins have been "dormant" for a fixed period, 10 years in your example? 5. What happens to bitcoins that, say, Hal Finney still had (if any) and he put in cold storage while he is in ultimate cold storage ( https://en.wikipedia.org/wiki/Hal_Finney_(computer_scientist)#Death) ? Ditto for someone, say, in a coma for 11 years, in jail for 11 years or any other similar event? Or a 20 year old sets aside coins for retirement. The following year, the system is changed, and when he looks again after not paying attention for a decade or two, they are gone. 6. This encourages censorship by miners for people attempting to move coins. 7. This has been discussed many times before and everyone is welcome to fork bitcoin code and the block chain and convince people to follow this chain and code. Then you can see if you can get many people to agree that this is a good idea.

On Mon, Dec 11, 2017 at 12:30 PM, Teweldemedhin Aberra via bitcoin-dev < bitcoin-dev@lists•linuxfoundation•org> wrote:

Quoted text (36 lines)
> It is estimated that about 4 million of the about 16.4 Bitcoins ever mined
> are lost forever because no one knows the private keys of some Bitcoin
> addresses. This effectively mean there are actually only 14.4 million
> Bitcoins in circulation even though 16.4 million are mined. There is no way
> of eliminating the human errors that cause these losses of Bitcoin from
> circulation, while the number of Bitcoin that will ever be mined is capped
> at 21 million. This means the total number of Bitcoins that are in
> circulation will eventually become zero, bringing the network to an end.
>
> The solution this BIP proposes is to implementing a dead man's switch to
> Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to
> dormant addresses to automatically expire. A Bitcoin address is deemed
> dormant if it is not used in transactions for some fixed length of time,
> say ten years.
>
> The calculation of the miner's reward should take into account the
> Bitcoins that has expired. This means there is a possibility that miner's
> reward can increase if sufficient number of Bitcoins expire.
>
> Ref:
>
> http://fortune.com/2017/11/25/lost-bitcoins/
>
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link>
> <#m_2495265241835744459_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists•linuxfoundation•org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>
>
Luke Dashjr in reply to Teweldemedhin Aberra permalink original

You can implement this already, but only for ~1 year expirations.

IF <normal script> ELSE <1 year> CHECKSEQUENCEVERIFY ENDIF

Perhaps it would make sense to propose a flag extending the range of relative lock-times so you can do several years?

Luke

On Monday 11 December 2017 5:30:37 PM Teweldemedhin Aberra via bitcoin-dev wrote:

Quoted text (30 lines)
> It is estimated that about 4 million of the about 16.4 Bitcoins ever mined
> are lost forever because no one knows the private keys of some Bitcoin
> addresses. This effectively mean there are actually only 14.4 million
> Bitcoins in circulation even though 16.4 million are mined. There is no way
> of eliminating the human errors that cause these losses of Bitcoin from
> circulation, while the number of Bitcoin that will ever be mined is capped
> at 21 million. This means the total number of Bitcoins that are in
> circulation will eventually become zero, bringing the network to an end.
>
> The solution this BIP proposes is to implementing a dead man's switch to
> Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to
> dormant addresses to automatically expire. A Bitcoin address is deemed
> dormant if it is not used in transactions for some fixed length of time,
> say ten years.
>
> The calculation of the miner's reward should take into account the Bitcoins
> that has expired. This means there is a possibility that miner's reward can
> increase if sufficient number of Bitcoins expire.
>
> Ref:
>
> http://fortune.com/2017/11/25/lost-bitcoins/
>
>
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campa
> ign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> www.avast.com
> <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campa
> ign=sig-email&utm_content=webmail&utm_term=link>
> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
Teweldemedhin Aberra in reply to Luke Dashjr permalink original

Hi, The only solution other than Dead Man's Switch to avoid gradual loss Bitcoins in transaction is increasing the divisibiliy of Bitcoins. Then Bitcoin values will need integer of more than 64 bits. Could that be done with soft fork?

On Dec 11, 2017 9:42 PM, <bitcoin-dev-request@lists•linuxfoundation•org> wrote:

Quoted text (51 lines)
> You can implement this already, but only for ~1 year expirations.
>
> IF <normal script> ELSE <1 year> CHECKSEQUENCEVERIFY ENDIF
>
> Perhaps it would make sense to propose a flag extending the range of
> relative
> lock-times so you can do several years?
>
> Luke
>
>
> On Monday 11 December 2017 5:30:37 PM Teweldemedhin Aberra via bitcoin-dev
> wrote:
> > It is estimated that about 4 million of the about 16.4 Bitcoins ever
> mined
> > are lost forever because no one knows the private keys of some Bitcoin
> > addresses. This effectively mean there are actually only 14.4 million
> > Bitcoins in circulation even though 16.4 million are mined. There is no
> way
> > of eliminating the human errors that cause these losses of Bitcoin from
> > circulation, while the number of Bitcoin that will ever be mined is
> capped
> > at 21 million. This means the total number of Bitcoins that are in
> > circulation will eventually become zero, bringing the network to an end.
> >
> > The solution this BIP proposes is to implementing a dead man's switch to
> > Bitcoin addresses. The dead man's switch causes the Bitcoins assigned to
> > dormant addresses to automatically expire. A Bitcoin address is deemed
> > dormant if it is not used in transactions for some fixed length of time,
> > say ten years.
> >
> > The calculation of the miner's reward should take into account the
> Bitcoins
> > that has expired. This means there is a possibility that miner's reward
> can
> > increase if sufficient number of Bitcoins expire.
> >
> > Ref:
> >
> > http://fortune.com/2017/11/25/lost-bitcoins/
> >
> >
> > <https://www.avast.com/sig-email?utm_medium=email&utm_
> source=link&utm_campa
> > ign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
> > www.avast.com
> > <https://www.avast.com/sig-email?utm_medium=email&utm_
> source=link&utm_campa
> > ign=sig-email&utm_content=webmail&utm_term=link>
> > <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
Ricardo Filipe in reply to Teweldemedhin Aberra permalink original

You can do it on 2nd layer solutions such as the lightning network, with their own format. On the base layer you cannot do it without a hard fork, or it would undermine the invariants of bitcoin.

2017-12-12 8:24 GMT+00:00 Teweldemedhin Aberra <teweldemat@gmail•com>:

Quoted text (159 lines)
> How?
>
>
>
> From: Ricardo Filipe
> Sent: Tuesday, December 12, 2017 4:44 AM
> To: Teweldemedhin Aberra; Bitcoin Protocol Discussion
> Subject: Re: [bitcoin-dev] BIP - Dead Man's Switch
>
>
>
> yes
>
>
>
> 2017-12-12 1:10 GMT+00:00 Teweldemedhin Aberra via bitcoin-dev
>
> <bitcoin-dev@lists•linuxfoundation•org>:
>
>> Hi,
>
>> The only solution other than Dead Man's Switch to avoid gradual loss
>
>> Bitcoins in transaction is increasing the divisibiliy of Bitcoins. Then
>
>> Bitcoin values will need integer of more than 64 bits. Could that be done
>
>> with soft fork?
>
>>
>
>> On Dec 11, 2017 9:42 PM, <bitcoin-dev-request@lists•linuxfoundation•org>
>
>> wrote:
>
>>>
>
>>> You can implement this already, but only for ~1 year expirations.
>
>>>
>
>>> IF <normal script> ELSE <1 year> CHECKSEQUENCEVERIFY ENDIF
>
>>>
>
>>> Perhaps it would make sense to propose a flag extending the range of
>
>>> relative
>
>>> lock-times so you can do several years?
>
>>>
>
>>> Luke
>
>>>
>
>>>
>
>>> On Monday 11 December 2017 5:30:37 PM Teweldemedhin Aberra via
>>> bitcoin-dev
>
>>> wrote:
>
>>> > It is estimated that about 4 million of the about 16.4 Bitcoins ever
>
>>> > mined
>
>>> > are lost forever because no one knows the private keys of some Bitcoin
>
>>> > addresses. This effectively mean there are actually only 14.4 million
>
>>> > Bitcoins in circulation even though 16.4 million are mined. There is no
>
>>> > way
>
>>> > of eliminating the human errors that cause these losses of Bitcoin from
>
>>> > circulation, while the number of Bitcoin that will ever be mined is
>
>>> > capped
>
>>> > at 21 million. This means the total number of Bitcoins that are in
>
>>> > circulation will eventually become zero, bringing the network to an
>>> > end.
>
>>> >
>
>>> > The solution this BIP proposes is to implementing a dead man's switch
>>> > to
>
>>> > Bitcoin addresses. The dead man's switch causes the Bitcoins assigned
>>> > to
>
>>> > dormant addresses to automatically expire. A Bitcoin address is deemed
>
>>> > dormant if it is not used in transactions for some fixed length of
>>> > time,
>
>>> > say ten years.
>
>>> >
>
>>> > The calculation of the miner's reward should take into account the
>
>>> > Bitcoins
>
>>> > that has expired. This means there is a possibility that miner's reward
>
>>> > can
>
>>> > increase if sufficient number of Bitcoins expire.
>
>>> >
>
>>> > Ref:
>
>>> >
>
>>> > http://fortune.com/2017/11/25/lost-bitcoins/
>
>>> >
>
>>> >
>
>>> >
>
>>> >
>>> > <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campa
>
>>> > ign=sig-email&utm_content=webmail&utm_term=icon> Virus-free.
>
>>> > www.avast.com
>
>>> >
>
>>> >
>>> > <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campa
>
>>> > ign=sig-email&utm_content=webmail&utm_term=link>
>
>>> > <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>>
>
>>
>
>> _______________________________________________
>
>> bitcoin-dev mailing list
>
>> bitcoin-dev@lists•linuxfoundation•org
>
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>
>>
>
>