Re: [bitcoin-dev] On the scalability issues of onboarding millions of LN mobile clients

[Chris Belcher <belcher@riseup.net>]

On 05/05/2020 16:16, Lloyd Fournier via bitcoin-dev wrote:
> On Tue, May 5, 2020 at 9:01 PM Luke Dashjr via bitcoin-dev <
> bitcoin-dev@lists.linuxfoundation.org> wrote:
> 
>> On Tuesday 05 May 2020 10:17:37 Antoine Riard via bitcoin-dev wrote:
>>> Trust-minimization of Bitcoin security model has always relied first and
>>> above on running a full-node. This current paradigm may be shifted by LN
>>> where fast, affordable, confidential, censorship-resistant payment
>> services
>>> may attract a lot of adoption without users running a full-node.
>>
>> No, it cannot be shifted. This would compromise Bitcoin itself, which for
>> security depends on the assumption that a supermajority of the economy is
>> verifying their incoming transactions using their own full node.
>>
> 
> Hi Luke,
> 
> I have heard this claim made several times but have never understood the
> argument behind it. The question I always have is: If I get scammed by not
> verifying my incoming transactions properly how can this affect anyone
> else? It's very unintuative.  I've been scammed several times in my life in
> fiat currency transactions but as far as I could tell it never negatively
> affected the currency overall!
> 
> The links you point and from what I've seen you say before refer to "miner
> control" as the culprit. My only thought is that this is because a light
> client could follow a dishonest majority of hash power chain. But this just
> brings me back to the question. If, instead of BTC, I get a payment in some
> miner scamcoin on their dishonest fork (but I think it's BTC because I'm
> running a light client) that still seems to only to damage me. Where does
> the side effect onto others on the network come from?
> 
> Cheers,
> 
> LL
> 

Hello Lloyd,

The problem comes when a large part of the ecosystem gets scammed at
once, which is how such an attack would happen in practice.

For example, consider if bitcoin had 10000 users. 10 of them use a full
node wallet while the other 9990 use an SPV wallet. If a miner attacked
the system by printing infinite bitcoins and spending coins without a
valid signature, then the 9990 SPV wallets would accept those fake coins
as payment, and trade the coins amongst themselves. After a time those
coins would likely be the ancestors of most active coins in the
9990-SPV-wallet ecosystem. Bitcoin would split into two currencies:
full-node-coin and SPV-coin.

Now the fraud miners may become well known, perhaps being published on
bitcoin news portals, but the 9990-SPV-wallet ecosystem has a strong
incentive to be against any rollback. Their recent transactions would
disappear and they'd lose money. They would argue that they've already
been using the coin for a while, and it works perfectly fine, and anyway
a coin that can be spent in 9990 places is more useful than one that can
be spent in just 10 places. The SPV-wallet community might even decide
to use something like `invalidateblock` to make sure their SPV-coin
doesn't get reorg'd out of existence. There'd also likely be a social
attack, with every bitcoin community portal being flooded with bots and
shills advocating the merits of SPV-coin. This is not a hypothetical
because we already saw the same thing during the scalability conflict
2015-2017.

Before you know it, "Bitcoin" would become SPV-coin with inflation and
arbitrary seizure. Any normal user could download software called
"Bitcoin wallet" which they trust and have used before, but instead of
using Bitcoin they'd be using SPV-coin. You may be one of the 10 wallets
backed by a full node, but that won't do much good to you when 9990
users happily use another coin as their medium of exchange.

Regards
CB