[Cryptography] threat models, was Krugman blockchain currency skepticism

[Patrick Chkoreff]

John Levine wrote on 08/05/2018 11:01 AM:

> If you're OK with a centralized system with tethered assets, why don't
> you use Paypal?  Internal transactions are fast and free, value is
> firmly tethered and exchangable into banks all over the world.  Before
> arguing that they're not trustworthy, please compare the number of
> people who trust Paypal to do their transactions (on the order of 200
> million) with those who use any sort of cryptocurrency.  Why are they
> all wrong?

I never said they were all wrong.  Paypal is a perfectly fine
alternative as long as they don't cut you off.  Then you can use Bitcoin
instead.  Recall that my original point was to answer Krugman's question
"what problem does it [bitcoin] solve?"


>> I tend to think the future is in individually issued real bills of
>> exchange, redeemable for assets, with invoices signed by the payee and
>> those signed invoices further signed by the payer, and exchanges cleared
>> by the issuer.  Bills of exchange by different issuers could be traded
>> on a market floor to allow price discovery based on issuer reputation
>> and asset value.
> 
> That worked OK in the 1700s but there's probably reasons we've moved
> beyond it.  But again, what's your threat model?

The threat model is that a payment processor cuts off your service
because somebody doesn't like you.


>   Do you assume these
> are recorded on some sort of distributed consensus blockchain?

No.  Some issuers will use double-entry bookkeeping just like Paypal.
Others may use blinded tokens.


>   If so, who runs it, and why do you trust them?

I will trust some issuers and payment platforms more than others.  Other
people will make a different assessment.  That's what makes a market.

I am less concerned about accepting an issue which I trust less if I can
trade it for an issue which I trust more.  It may well be that the
person on the other side of the trade has exactly the opposite position
of trust with respect to the two issues -- or at least has a more
pressing need for the one I have versus the one he has.


>   (The answer isn't "nobody",
 someone's doing the mining or equivalent.)  If an issuer doesn't
> redeem a bill, what recourse do you have other than to denounce him?

It depends on the contract and the jurisdiction.


> Can anyone be an issuer?

Yes, though not everyone will find traction in a wide market.  Some may
only trade within cliques, or networks of cliques.


> If so, why can't I issue bills each good for
> a kilo of unobtainium and run away with the money?

You could.  However there are many examples of issuers/payment
processors with a long track record of not doing such things.


The concerns you raise regarding trust in central servers/issuers and
trust in redeemable assets serve as excellent points in favor of Bitcoin.

With Bitcoin, you don't have to ask "Why should I trust Issuer X to
clear my payments, and why should I trust Issuer X to redeem my digital
tokens for real assets like gold or $100 bills?"  The answers are "there
is no Issuer X" and "there are no real assets."  Real assets are so 18th
Century anyway.


-- Patrick