[Cryptography] Proof of Prophecy (POP) Re: Hashgraph

Ersin Taskin hersintaskin at gmail.com
Wed Jan 3 10:57:24 EST 2018


Message: 4
Date: Mon, 1 Jan 2018 12:44:07 -0500
From: Phillip Hallam-Baker <phill at hallambaker.com>
To: Ismail Kizir <ikizir at gmail.com>
Cc: Cryptography Mailing List <cryptography at metzdowd.com>
Subject: Re: [Cryptography] Hashgraph
Message-ID:
        <CAMm+LwhVBsfhHrNCZFEFs2=ghiGOn5vciqsPNGs-2-tnKMsZ2w at mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

On Sat, Dec 30, 2017 at 2:18 PM, Ismail Kizir <ikizir at gmail.com> wrote:

>> I wish a happy new year for everyone.
>> And I want to ask:
>> I haven't read anyone mentioning about Hashgraph on the list:
>> It's a very fast and promising technology. Better than blockchain, it
>> seems.
>> The only weakness, as far as  I could understand so far, is that they
>> need the consensus of 2/3 of the nodes.
>>
>> https://hashgraph.com/

>It is not clear what SwirlID are claiming. They certainly cannot claim the
>idea of cross linking multiple independent Harber-Stornetta notary logs.
>People have been discussing approaches to doing that for years. The
>original CT specs had in them the notion of chattering between the logs to
>bind them into one.
>
>I am not sure why consensus is desirable other than trying to reconstruct
>Blockchain ideology without mining.
>
>Let us say BofA, Chase and Barclays are running a notary chain and they
>cross notarize every hour. All that means is that when I make a
>transaction, on the BofA chain, then BofA could in theory repudiate it in a
>60 minute (max) window and for that 60 minutes I have to trust BofA. So
>what? I already trust my banks with my money.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/
attachments/20180101/420bb30b/attachment-0001.html>


>I think that, Blockchain was a good start, but I consider it as a
>"Trojan horse of the banking system".
>Because of its need of very high resources(both computational power,
>storage and network), Blockchain will condemn us to central
>authorities and this is contradictory to its reason of creation.
>That's why, I am excited about "more democratic", "more decentralized"
>technologies.
>As far as I understand, at least for the moment, Hashgraph can't give
>us what it is promising to give. (Patent issue and (maybe also)
>reliability issue).

Regards
Ismail Kizir


Dear Phillip and Ismail,

I agree with you on trust and Bitcoin. I want to add that the trust
anathema is one of the key issues of the cryptocurrency eco-system today.
And consensus can be used to form the next block for fault
tolerance (including Byzantine). I am talking about a few seconds standard
Quorum consensus. So once we are saved from the trustlessness myth we can
go back to powerful textbook solutions to be applied to cryptocurrencies
and use our creativity for efficiency, security, privacy, functionality,
and not for and because of unnecessary issues due to anhcorlessness.

In fact, I am working on a paper to propose a new protocol to Bitcoin. It
is called Proof of Prophecy (POP). POP is a special case of Proof of Trust
(POT, also defined in the paper). The basic idea is that you cannot have
Anchorlessness and Robustness at the same time (sort of CAP Theorem in the
cryptocurrency context). Consider Bitcoin for instance. It is slow,
cumbersome, expensive, environmentally hostile, etc. You don't even have
determinism. You are not sure if your valid transaction will make it to the
ledger because some miners may not like your fee or whatever or those who
mine your transaction cannot win the lottery. The so claimed consistency is
very eventual. Imagine you need AT LEAST one hour to be ALMOST sure. A
Byzantine
tolerant digital cash system can achieve a few seconds absolute transaction
determinism (a valid transaction sent to the system must be registered to
the ledger and an invalid transaction must be rejected at all times) and be
used in POS scenarios.

A kid from the cryptocurrency kingdom should say the king is naked for the
sake of the kingdom.

POW has done a good job in the birth phase of Bitcoin when mining could be
done by individuals via PC's. Its survival at the beginning owes a lot to
the POW based solution to the double spend problem. To the philosophy of
decentralization. But now POW mining is a business where you need special
and big investment. The vast majority of power is in the hands of a few
pool managers and mining companies. Bitcoin is far from complete
decentralization. It is environmentally hostile. Furthermore, it is
vulnerable to super-rational attack (defined in the paper). The problem
with Proof of Stake (POS), etc. is also well mentioned in many papers so I
don't want to repeat here. Replacing POW with another form of trustlessness
does not resolve the fundamental issue of anhcorlessness. They are
extremely vulnerable to super-rational attacks and doomed to evolve to
power consolidation.

And we have natural anchors that we amazingly ignore. One such anchor has
already been created by Bitcoin itself through POW. POW has already
achieved its mission and must hand the flag to POP.  I plan to share the
draft of my paper here on this list and wish to take feedback, this
quarter. I can share pre-draft earlier with those interested.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20180103/a9e9bd1e/attachment.html>


More information about the cryptography mailing list