[Bitcoin-segwit2x] PROPOSAL: B0RG (Bitcoin zero replay, guarantee) - Ensuring a smooth 2X upgrade without a chain split

Wed Oct 25 01:38:04 UTC 2017

Given that the futures markets believe that 1X will be far more valuable
than 2X, there's only a few means by which 2X proponents and miners are
likely to be able to shift the expected values more in favour of 2X.

The least insidious and most natural is simply to mine only 2X - eg if 99%
or even 95% of miners only mined 2X, it's conceivable that the very long
time to find blocks and to adjust difficulty on 1X might make it
sufficiently unusable that its value in expectation goes down.

However I find this somewhat doubtful, as there is now a fairly strong
precedent of coins retaining significant value even when their chain is
fraught (Bitcoin Cash in the early days, and arguably even now with its
schizophrenic EDA), or even where no chain exists! (observe Bitcoin Gold,
in all its ignominy).

Given this, it seems pretty likely that the market can stay 'irrational'
(may not actually be irrational if you look at the longer-run incentives)
longer than miners can stay solvent, so defections become more likely the
longer the price continues to ascribe much greater value to 1X than 2X,
despite problems with actually using 1X to transact.

That means that if miners are really very determined that 2X be the only
surviving chain, which seems possible - observe current favoured framing of
it as an 'upgrade' amongst proponents, then they may employ more insidious
tactics. These could include selfish mining on the 1X, with or without
transactions. If they chose to include transactions in their selfish-mined
blocks, they can also double-spend costlessly.

I say costlessly, but that's only true in Bitcoin terms - it could crash
the real value of those coins though. Since double spending against
exchanges screws the entire ecosystem, it seems unlikely they'd do that, I
agree.

But in any case, more explicit attacks like any of these would be much more
likely to move the expected value.

This is where something like B0rg is potentially very powerful, because it
creates the possibility of forcing an all-or-nothing equilibrium, so a
simple majority of support quickly converts itself into consensus at one of
the two extrema (in a way futures prices do the same - as there is positive
feedback from the price to miner support, and to the extent that you
believe miner support affects viability/expected value, back to price. So
we should expect futures price to be 'bipolar', just as we'd expect miner
suppport with something like B0rg to be bipolar)

For the record I actually think B0rg is evil, and I'm also not sure that in
its current form it's found a way to achieve what it's trying to do - force
consensus as the stable equilibrium (for a start it seems that there is a
problem with withholding blocks). But I think I agree that something like
it could, when combined with an initial simple majority of miners
supporting it, move incentives and expected values very far very quickly.

The existence of futures also makes these attacks cost less, because miners
could buy 2X at a large discoutn just before the fork, knowing their
intentions. Of course, this fact, combined with the low current value of
the 2X future, is actually pretty reassuring, from a Hansonian perspective,
as it's a signal that miners too believe that 2X won't be worth more.

*Ben Kloester*

On 25 October 2017 at 12:00, Jacob Eliosoff via Bitcoin-segwit2x <
bitcoin-segwit2x at lists.linuxfoundation.org> wrote:

> I wouldn't assume that "clearly very detrimental to the future utility of
> Bitcoin" is much of a disincentive to double-spend attacks, or much
> reassurance to an exchange risking one, particularly where "Bitcoin" =
> "one, minority-hash branch of Bitcoin".  The persistent attacks on Ethereum
> last year (replay, then gas pricing exploits) show that when attacks are
> feasible, it's fairly likely some schmuck will employ them.  Eg in the
> extreme case where hashrate is split 99%/1%, we can probably agree
> accepting payments on the 1% chain is unwise.
>
> That said - 99/1 seems implausible to me, and as long as both chains have
> nontrivial (>5%?) support, my *guess* is Alex is right that double-spends
> (or, more likely, empty-/bad-block attacks) are unlikely.
>
>
> On Oct 24, 2017 8:42 PM, "Alex Morcos via Bitcoin-segwit2x" <
> bitcoin-segwit2x at lists.linuxfoundation.org> wrote:
>
> If we start seeing 6-block deep reorgs because miners are attempting to
> pull off a double-spend attack then I think the fundamental economic
> incentives of Bitcoin will be shown to have failed and we might as well all
> go home.  It's somewhat reasonable to come to differing opinions on what's
> long term economically rational for miners as to which fork to mine, but
> it's clearly very detrimental to the future utility of Bitcoin to try
> double spend attacks.  I'm quite confident we will not see that.
>
>
>
> On Tue, Oct 24, 2017 at 8:11 PM, Vinny Lingham via Bitcoin-segwit2x <
> bitcoin-segwit2x at lists.linuxfoundation.org> wrote:
>
>> I’d love to get a sense of what the hash rate threshold will be for
>> exchanges to stop accepting 1x coin, even with 6 confirmations? With the
>> risk of a selfish mining attack increasing the lower the 1x hashrate is,
>> why would an exchange accept the coins knowing that someone could be double
>> spending across multiple exchanges?
>>
>> https://bitcointalk.org/index.php?topic=2127697.0
>>
>> I’m just trying to understand this attack vector better... I haven’t
>> heard a good response to this question yet - would really appreciate some
>> insight here. This is largely why I believe hashrate secures the network
>> and why I’ve been against the UASF.
>>
>> Sent from my iPad
>>
>> On Oct 24, 2017, at 13:26, Erik Aronesty via Bitcoin-segwit2x <
>> bitcoin-segwit2x at lists.linuxfoundation.org> wrote:
>>
>> At this point *either* chain can be the "legacy" chain.   Most exchanges
>> are going to support both coins.   Coinbase's announcement that 2x will be
>> initially called B2X was big (they reserve the right to rename it later).
>>
>> Users will be thus be free to dump either coin according to their
>> personal feelings on the matter.
>>
>> Hashpower is probably not going to be a consideration for the majority of
>> holders.   At the end of the day... if users dump 1x, miners will jump on
>> 2x, and if users dump 2x, miners will jump on 1x.
>>
>> Trying to predict what investors and users will do is really not
>> "politics".   It's pure speculation.   And basing any real decisions off
>> "hashpower signaling" is patently absurd.
>>
>> In other words:
>>
>>  - This is a risky proposition either way
>>  - There is no reasonable way to predict the outcome
>>
>> Anyone who claims otherwise is probably entrenched in a particular
>> "camp", and isn't thinking rationally about the current situation.
>>
>> (Personally, I'm simply going to dump on the first day... regardless of
>> the price and based on my preference.  I suspect I'm in the minority here.)
>>
>>
>> On Tue, Oct 24, 2017 at 4:16 PM, Phillip Katete via Bitcoin-segwit2x <
>> bitcoin-segwit2x at lists.linuxfoundation.org> wrote:
>>
>>> Unfortunately, it is a double whammy for the legacy chain.
>>>
>>>
>>>
>>> Despite there being a general assumption that the legacy chain will have
>>> at least 20% of the current network hash, it’s going to be less than that
>>> and more likely the majority of that hash will mine coinbase only blocks in
>>> order to stall the chain. The closer the attack happens to the fork time
>>> (and therefore unknown block times & mempool), the more potent the attack
>>> as the fees will still be low relative to when block times are more
>>> accurately predicted and mempool settled.
>>>
>>>
>>>
>>> Once confirmation times on the legacy chain become unusually long (and
>>> the price begins to tend to reflect that), then attackers can abandon the
>>> chain and the ardent legacy miners will also start divesting/splitting
>>> their hash to the 2x chain.
>>>
>>>
>>>
>>> *From: *Peter BitcoinReminder.com via Bitcoin-segwit2x
>>> <bitcoin-segwit2x at lists.linuxfoundation.org>
>>> *Sent: *24 October 2017 20:41
>>> *To: *Erik Voorhees <erik at shapeshift.io>
>>> *Cc: *Melvin Carvalho via Bitcoin-segwit2x
>>> <bitcoin-segwit2x at lists.linuxfoundation.org>
>>> *Subject: *Re: [Bitcoin-segwit2x] PROPOSAL: B0RG (Bitcoin zero replay,
>>> guarantee) - Ensuring a smooth 2X upgrade without a chain split
>>>
>>>
>>>
>>> You are also missing the point that many people dont use bitcoin for
>>> payments - but as store of value - so they don’t care if there are long
>>> confirmation times (or even none in one week!) - you only see it from the
>>> payment/transaction side, but this is not the case for many people which
>>> are motivated to stay and fight for bitcoin.
>>>
>>>
>>>
>>> Open your mind - there are many people which give a shit about
>>> confirmation times and fees.
>>>
>>>
>>>
>>> Am 24.10.2017 um 21:04 schrieb Erik Voorhees via Bitcoin-segwit2x <
>>> bitcoin-segwit2x at lists.linuxfoundation.org>:
>>>
>>>
>>>
>>> Peter - I disagree that 1x won’t have a market price. Exchanges can turn
>>> off all deposits and withdraws completely for a week and you’ll still see
>>> markets of both coins working, because there is already 1x and 2x in the
>>> exchanges (in large quantities).
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Kind regards,
>>>
>>> -Erik Voorhees
>>>
>>> CEO ShapeShift AG
>>>
>>>
>>>
>>> On October 24, 2017 at 12:59:28 PM, Peter (dizzle at pointbiz.com) wrote:
>>>
>>> I agree the minority chain will quickly grind to a halt. Pure SPV users
>>> will just want Bitcoin to work as Moe indicated. They won't know if they
>>> are on the 1x or 2x chain but what they will know is Bitcoin (BTC) still
>>> works and in 1 day 2x (BTC) will pass 100 blocks and have a market price
>>> and be usable by exchanges globally. The 1x chain will not have a market
>>> price for 5 days because it didn't reach the 100 block milestone to enable
>>> deposits and coin splitting. In that time it can die out. Also, it's likely
>>> to take an extra week or so for exchanges to get all their coin splitting
>>> transactions confirmed by mixing with 1x coinbases. Honeybadger won't care
>>> much and people won't want to say 1x is BTC because then they have to admit
>>> BTC had a 1 or 2 week service interruption.
>>>
>>>
>>>
>>> Sorry if this is not technical enough for the list.
>>>
>>>
>>>
>>> Regards
>>>
>>> Peter
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Bitcoin-segwit2x mailing list
>>> Bitcoin-segwit2x at lists.linuxfoundation.org
>>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-segwit2x
>>>
>>>
>> _______________________________________________
>> Bitcoin-segwit2x mailing list
>> Bitcoin-segwit2x at lists.linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-segwit2x
>>
>>
>> _______________________________________________
>> Bitcoin-segwit2x mailing list
>> Bitcoin-segwit2x at lists.linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-segwit2x
>>
>>
>
> _______________________________________________
> Bitcoin-segwit2x mailing list
> Bitcoin-segwit2x at lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-segwit2x
>
>
>
> _______________________________________________
> Bitcoin-segwit2x mailing list
> Bitcoin-segwit2x at lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-segwit2x
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-segwit2x/attachments/20171025/5cf5adf5/attachment-0001.html>