Date: Mon, 08 Apr 2019 10:45:29 +0000
To: Aymeric Vitte <vitteaymeric@gmail.com>
From: ZmnSCPxj <ZmnSCPxj@protonmail.com>
Reply-To: ZmnSCPxj <ZmnSCPxj@protonmail.com>
Message-ID: <GkSgHQYXGen75-KP_N2VbK1EmY5DSDe0sncJBU77l6_2xdYhh9Yw5rQSgtPuwXJnMlnA0j195hkMfhnxhGkMERa3kXXW6KvR5qt88oSNGvY=@protonmail.com>
In-Reply-To: <392367fe-b1d7-7d47-01de-ebb4b7142ead@gmail.com>
References: <IAFPSZAn6TYt348fmmnPznQ_ApG7pa48eMjzTgrjuVAt6fS1tNieRxlcIXyTATy2vjZCUn4wVQcsyDlyb_3Ip46BstFRikB95-lKewAZBEE=@protonmail.com>
	<d1cfa2e9-69e4-ee02-4c10-23b2b1a30e00@gmail.com>
	<TF9WSGU6njZqgOyJF5-m1gYMwfgUCStjUV-IpRuX67w1Z6jL2Tdarr6PCOUO1vFb9hz_jWnbe_5Tg8E_a9iyPeXIY_iJUf9YN8u9xB4SC90=@protonmail.com>
	<392367fe-b1d7-7d47-01de-ebb4b7142ead@gmail.com>
Feedback-ID: el4j0RWPRERue64lIQeq9Y2FP-mdB86tFqjmrJyEPR9VAtMovPEo9tvgA0CrTsSHJeeyPXqnoAu6DN-R04uJUg==:Ext:ProtonMail
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: Bitcoin Protocol Discussion <bitcoin-dev@lists.linuxfoundation.org>
Subject: Re: [bitcoin-dev] Smart Contracts Unchained
Precedence: list

Good morning Aymeric,

> Hi,
>
> Apparently you are not a fan of ethereum, as far as I can tell ethereum
> sidechains look like a mess with stupid tokens/transactions flooding the
> network while they are completely centralized, but some bitcoin
> sidechains can easily compete with this too, like Tether, don't even
> understand how anyone can give some credit to that stuff the way it is
> implemented, and if bitcoin fails that would be the same as for ethereum

I prefer to be more precise in my terminology.
Colored coins are not the same as sidechains, and there are colored coins a=
nd then there are colored coins.
This mechanism does not propose some change in colored coins.
An important aspect of colored coins is that one can foist them on somebody=
 else to extract things of real value from them, but this mechanism is more=
 strongly for a fixed set of participants.

I strongly suspect that Bitcoin will outlast Ethereum, but that is rather n=
ot very related to this topic.

> Most likely everyone would agree if the escrow disappears, but not sure
> at all, let's imagine 1 to N put 10K on the table for a game, they
> update the states and at the end N wins everything, N is rich and don't
> care finally if the others cheaters have their coins locked (and to lose
> 10K), same with setting up a new escrow to resolve the conflict
>

Indeed.
Still, the option to do so exists, and sometimes all that is needed for hum=
ans to do the right thing, is to be given the option to do so.

> I think that you should highlight this (and what private key corresponds
> to E + h(E | s) * G, not sure it's trivial for everybody), probably a
> way to get this more decentralized is to reward the escrows (what is the
> interest here for people to run a smart contract platform?)

I assumed both were obvious, but I suppose a few more words about those wou=
ld not be amiss.

>
> For lightning, maybe it's a question of wording, I consider it as a
> sidechain AND methods that can be used by other sidechains, as well as
> the others you quoted, even if only two people in the world use
> lightning, it is still decentralized, because it sustains itself alone

Again, I prefer precision in my terminology.
For me, a sidechain is a blockchain of some sort.
In particular, a kind of Merklized singly-linked list containing representa=
tions of transformations of state, is how I define blockchain to be.

No such Merklized singly-linked list exists in Lightning Network, thus I do=
 not consider it, "blockchain".
And thus I do not consider it "sidechain", as a sidechain is a blockchain.
Current LN does use "shachains" by Rusty, but shachains are not Merklized s=
ingly-linked lists, but are instead a kind of inverse mountain range struct=
ure.

Still, one might consider both federated sidechains and Lightning Network t=
o have a "federated" offchain structure.
This is because the coins on the Bitcoin blockchain are locked to a multisi=
gnature and activity is not recorded on the Bitcoin blockchain.
However, in LN, each channel is a 2-member federation (you and a counterpar=
ty) and the mechanism in LN requires consensus (2-of-2) rather than a quoru=
m (m-of-n).
This greatly increases the security of LN: the owner of funding on an LN ch=
annel can always refuse to sign an update if the other member of the federa=
tion is taken over.
Compare this to the quorum that typical federations have, where takeover of=
 a sufficient quorum is enough to steal funds from the remaining federation=
.
https://zmnscpxj.github.io/offchain/safety.html

Regards,
ZmnSCPxj