[Bitcoin-development] questions about bitcoin-XT code fork & non-consensus hard-fork

Lodewijk andré de la porte l at odewijk.nl
Thu Jul 2 16:24:20 PDT 2015


2015-07-02 18:38 GMT+09:00 rysiek <rysiek at hackerspace.pl>:

> Dnia niedziela, 28 czerwca 2015 20:52:43 Sean Lynch pisze:
> > Wow, that's a low blow. Arguing by authority, and then a false dichotomy:
> > "either you know more about Bitcoin than X, or you should not have a
> voice
> > at
> >  all on this"
> >
> > Might I suggest considering arguing on the merits instead, next time? :)
> >
> >
> >
> >  Perhaps if you bothered to read more than the last message in the thread
> > you would realize that I already attempted that. I think your
> expectations
> > are a bit high when there are people on the thread arguing that we should
> > really consider the opinions of those making death threats. IOW listen to
> > the terrorists.
>
> But that's exactly why I was so surprised. I filter out the people that
> make
> no sense or deat5h threats, but when I see a person that usually seems to
> make
> sense, and yet lands a straw man, I'm taken aback. :)
>

Saying Gavin knows quite a lot about Bitcoin is not an argument of
authority. It doesn't make him right, but it does make him very likely to
be right. Certainly more likely than the heapings of shitty arguments,
especially those that are technically disconnected from what bitcoin is.

When you listen to the arguments about blocksize it usually goes something
like:

Pro:
 * Cheaper transactions
 * More truly Bitcoin transactions
 * Simplest way to scale

There's some derivative arguments regarding poor people's access to Bitcoin
and a herd of applications that just require cheap access to the
blockchain. I think the most noteworthy argument is:
 * Bitcoin becomes more competitive/attractive as a result of cheaper
transactions

Con:
 * We can already fit quite some txs
 * the blockchain will be less wieldable
 * the bandwidth required to keep up with the blockchain (regardless of
storage) will increase
 * we can scale externally

The derivative arguments here are also poor people's access to Bitcoin, wrt
bandwidth, but the argument holds up much worse as SPV would work just fine
over those rare very slow connections. Then there's a lot of downright
fearmongering like "only institutions will run full nodes". Economically
that's just not true, and technologically it's not important - crypto
features allow usability immediately (simplefied (commercial) APIs, "blind"
tx broadcasting), strong guarantees *very *quickly (SPV), and certainty and
independence if you'd like it (a pruning full node), and history novelty at
only minor hassle (less than a day's work). Even a 100 fold increase in
blocksize would not radically change this.

It's pretty annoying to have an even bigger blockchain, don't get me wrong
on that, but that's just the way Bitcoin works: a blockchain that grows
with use. There's no reason for it to truly upset you, either. Running a
full node is already something you don't do for no reason at all. I can't
really make this argument as well as I like. The point is that if you have
a reason you would still do it later, and if you don't you don't already.
Some people noted that the pruning makes it possible to run a full node on
their phones. Cool! But there's no reason to. In fact, you won't because
it'd drain your battery. We'll be okay without the one silly geek that does
it anyway.

So.. these points were already hard to argue against clearly. Then there's
"we can scale externally".... The trouble is that there's so many ways,
like pinning (sidechains/mastercoin), exclusively inter-institutional
settlement, debt based moneys ("the bearer of this token is entitled
to..."), and all of them could work! In fact, we could just abandon Bitcoin
alltogether! And that's the core of my counterargument: we don't have to
cripple Bitcoin, so let's not. Let's not make it more complicated than it
has to be. If we do scale externally, let it be for exceedingly good
reasons and at exceedingly competitive prices.


Mostly I see people haggling over nothing on Reddit, and even here. There's
also confusion about roles in the Bitcoin ecosystem, and about ideals. Then
there's confusion about consent, and how to manage it. It's probably
because all those things are badly fleshed out. Hard-to-achieve consensus
has advantages; Bitcoin will be technologically stable. The roles will
shift anyway. Ideals are not inherent, only cold hard currency is. It's
implications depend on culture and math, both generally not very well
understood. Reddit has never been conducive to quality argument - at
fantastic loss to everyone but the populists.

I think, ultimately, that giving Gavin ultimate authority for 10 years then
freezing the whole ordeal would probably work out rather well. I think it's
ran afoul of his control too soon. We already have some weird bugs in
Bitcoin. Then there's some strange (unstudied) economical effects like the
halving (and why not gradual decrease?). I'm sure lots of people found
similar oddities from their perspectives. There's just not much to do about
it anymore now.

The Chinese mining pools stamping their document regarding increasing the
blocksize to 8mb was extraordinary. They ignored Gavin's 20mb or anything
proposal ("or anything" probably mostly to make 20mb seem reasonable). They
stamped - a thing completely outdated in the West, yet common in Asia (and
pretty badass). They are a group of Chinese making law for everyone.
(not-a-magnet-or-similar-document-oriented-link:
http://i.imgur.com/JUnQcue.jpg) Gavin appeased them and provided for
unattended growth with a period 8mb increase. Smart move.


So - conclusively - Gavin is a hero, the Internet's retarded, Bitcoin is in
policy jail but it likes it there. Oh, and the free-est market of the world
is already significantly run by Chinese. If you read this far, thanks.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/html
Size: 7476 bytes
Desc: not available
URL: <https://lists.cpunks.org/pipermail/cypherpunks/attachments/20150703/8c030ab3/attachment-0002.txt>


More information about the cypherpunks mailing list