I still don't really understand how a channel works under dispute, e.g. both party claim different truth. I think that is not possible without a judge-like third party involved

And the traditional approach is also trustless: Bitpay have a 1000 btc account at Coinbase, Coinbase have a 1000 btc account at Bitpay, if Bitpay run away, his account at Coinbase will belong to Coinbase, vice versa

First, this is a terrible analogy. Changes to the Bitcoin protocol should not be made in any way similar to the way presidents are elected. In any case, presidents can be (and often are) elected by a plurality, not a majority. A candidate with less than 50% of the vote won in 1992, 1996 and 2000.


I don't have an interest in BlockStream at all. I think the threat of a hard fork with only approximately 75% of the mining power and with a significant portion of the community against it is very dangerous. I started to write that I have a "deep interest" in Bitcoin, but I'm not sure that's true anymore. What I see happening now makes me tend to think the Bitcoin "community" isn't so different from other communities. People split into tribes, shout at each other, call each other names, and then celebrate their victories or console themselves with their losses. I suppose I was too optimistic to think a cryptocurrency community would be different. One reason I thought it was different was that I had the impression that the fundamental rules were fixed and no longer in the hands of fallible humans. It turns out this isn't true.

Regarding BlockStream, it's clear that much of the Bitcoin community finds them controversial. It's also clear that they're doing some groundbreaking research, and that this research would apply to other cryptocurrencies than Bitcoin. Perhaps BlockStream should just implement their ideas for Litecoin and leave it to those in control of XT to develop what Bitcoin is to become. Would you find that preferable to the current situation?


I'm not sure what you would consider a "logical argument." As a logician, I'm inclined to interpret it literally, but I suspect it's not what you intended. A logical argument is a deduction starting from some axioms and leading to a conclusion. In your post you actually referenced two different possible conclusions: XT should be opposed. vs. The block size should not be increased. Now, it should be easier to give a logical argument for why XT should be opposed, since if someone already shows the block size should not be increased, then it logically follows that XT should be opposed.

Of course, it's impossible to conclude that XT should either be opposed, supported or even ignored unless we start from some axioms. This gets to the root of the issue. Different people have different fundamental beliefs about what Bitcoin is and what it should be.

Often I've seen the argument that Bitcoin should be censorship-resistant way for individuals to control their finances free from government control. We could take this as an axiom. Another axiom could be that for a cryptocurrency to remain censorship-resistant it is vital that it can be safely run behind Tor. Finally, we could add an axiom that states that some of the new code in Bitcoin XT makes it difficult to run Bitcoin XT safely behind Tor. With axioms like these, and possibly some more, we could chain together a logical argument ending with "XT should be opposed." I'll flesh out the details of the argument upon demand.

Now, of course, you could say it isn't a logical argument because you don't accept one or more of the axioms, but this is not a criticism of the argument. It's a criticism of the axioms. I could give many logical arguments (and for a reasonable donation I'd be willing to formalize them), but you could always reject the conclusion by rejecting some axioms used. That's just how logic works.

You have certain axioms of your own that I've seen expressed in many places. An assertion that is often made by supporters of XT is that BlockStream wants to keep the block size limit in 1MB so they can make more profit. Using this as an axiom, and probably a few other axioms, one could probably logically argue that BlockStream has nefarious motives for opposing XT. However, even this wouldn't logically rule out the possibility that BlockStream (or, more precisely, the employees of BlockStream) have both nefarious and intellectually pure motives for opposing XT. The possibilities aren't exclusive. Maybe one could argue about whether their primary motive is nefarious or pure, as presumably there is only one primary motive. Logic forces one to be annoyingly precise.

It's natural when people get into these kind of tribal arguments that statements such as "XT should be opposed" or "XT should be supported" themselves rise to the level of axioms, or, fundamental beliefs. Once that happens, any statements that would contradict the fundamental belief are immediately rejected. At that point it's impossible to have a consistent set of beliefs/axioms that would imply the opposing statement. In other words, eventually people are so sure of their position that it's impossible to convince them of anything contrary to it. If that's what's happened here, then every logical argument someone could give which concludes "XT should be opposed" would necessarily have at least one axiom you would reject. Your only other choices would be to change your mind about the conclusion or ignore the argument. Realistically, ignoring arguments is typically what people do in these situations.

My prediction for Thursday....

let's list a few of the tactics employed by the Blockstream devs:

1.  appeal to authority- "the entire dev/technical community is against this"-which they're not if you look outside core dev.
2.  scare tactics- "if Bitcoin forks, it will fail"
3.  character assassination- "Gavin hasn't coded for over a year", "we are shocked by his behavior", "he's out courting merchants, exchanges, & miners behind our backs".

i honestly ask you, which of the 3 tactics has Gavin employed on Reddit or here?  my answer is NONE.  and here we yet another example of #2 tonite by gmax:

https://www.reddit.com/r/Bitcoin/comments/3awomg/how_the_bitcoin_experiment_might_fail/

he's getting pounded too.  and these are the reasons Blockstream will lose in the end.  no one trusts them.

[–]nullcGreg Maxwell - Bitcoin Expert 14 points 1 month ago*

To give an external perspective from Luke's on this;

I now agree that BIP17 was better; not just slightly better but very clearly better in important, meaningful ways.

At the time I preferred BIP16, in hindsight I made an error in reasoning-- part of this was that I incorrectly believed that the difference was smaller than it actually was: There were substantial differences in terms of limitations of P2SH that none of us (probably not even Luke) understood at the time. (I say not even Luke because Luke's primary argument was about the "aesthetics" of the implementation, which were not-- by themselves-- that persuasive to anyone.)

A lack of care in this deployment resulted in substantial non-trivial (3-5 blocks) network forks for a period of about two months. BIP16's irregularity resulted in alternative implementers being somewhat slower to implement and more error prone. Luke just gave a concrete example (the idiotic 520 byte limit; which I think we didn't even realize existed in BIP16 at the time, spec was updated later to mention it); another is that you can't combine multiple P2SH scripts, e.g. to get an OR where you can spend a coin with either a small script or a big one-- that one we knew existed but didn't give adequate thought to, it's turned out to be a rather annoying limitation.

At the time the fact that the BIP16 approach had more testing and work on it, especially after the embarrassing design reboot after OP_EVAL turned out to be vulnerability introducing, and Gavin's strong preference for it, combined with the assessment of "little difference" made me prefer BIP16; as anything else would have meant additional delays. In reality, it was basically years before there was widespread P2SH use, an additional delay would have been better. It's worth noting that the person (roconnor) who found the OP_EVAL flaw also preferred BIP17-- which was probably a sign we should have paid more attention to. It's not a big regret but I do consider it a lesson.

It didn't help that some people (mostly not Luke himself) tried to construct a public drama with non-technical community members over what was still just a kind of boring technical argument. Unfortunately; trying to make drama can have exactly the opposite effect of embedding people in their positions and making them immune to reason. It's almost universally a bad move if you care about achieving a high quality result. I keep trying to remind myself of the importance of deciding to be immune to political drama in order to avoid the failure modes it creates if you let it influence you.

You can see a chart of opinions at the time: https://en.bitcoin.it/wiki/P2SH_Votes BIP16 was a condorcet winner; but most people considered either approach acceptable.

Security is definitely a large roadblock for those considering to buy bitcoins, but I don't agree that the bitcoin price is dictated by people who don't spend or sell any bitcoins. It's a part of a much more complex equation, because if nobody would spend or sell any coins the market would be extremely illiquid which also deters potential buyers (no sellers, no buyers). So it's not that easy.