Gold collapsing. Bitcoin UP.

[BldSwtTrs]

Do you remember when Bitcoin was an experiment which could change the world?

Now the new narrative is that Bitcoin is a system responsible for a lot of money so we need to be careful for whatever we do with it. And we also need to think of it as a little fragile and precious backbone of everything else, because success is assumed.

Bitcoin has a cultural problem. A very serious one. I think the narrative "Bitcoin is an experiment" stopped when Gavin lost its leadership role. The economic illeterates geeks took over and imposed a new one.

 

[Mengerian]

Good interview with Daniel Krawisz.

Talks about:

• Problems with Segwit - Jaqen Hash’ghar article
• Quadratic hashing issue
  
• Chain splits versus altcoins
• Idea of client that can handle multiple chains gracefully
• Fork future markets
• Whether verifying-only nodes add value to the network
• The roles of investors and coders

 

[Roger_Murdock]

Activating SegWit for its tiny, one-time throughput increase strikes me as a bit like drinking your own urine when you're lost in the woods and dying of thirst: it might provide some temporary relief, but it's not exactly a viable long-term solution to the actual problem, and you really shouldn't have allowed yourself to get into a situation where the proposition would start to seem like a tempting one.

What's really sad is that Bitcoin isn't actually lost in the woods. The situation is really more like that of a dumb kid who's camping in his own backyard. He could just go inside and open the fridge to get a drink, but he's too scared to leave the tent after his "friend" told him a--he swears, "100% true"--campfire story about "The Hard Fork Man." (It's basically a knock-off of this one, but instead of a hook for a hand... well, you can probably figure it out.)

Hmm, I can't decide if that's a great analogy or one of the dumbest things I've ever written, but I'm gonna go ahead and hit "Post."

 

[lunar]

@Mengerian We are all faceless men, disciples of the many faced god, - All men must serve, Faceless men most of all, and Mr Bitcoin is a ruthless master.

@DanielKrawisz nails this interview, a great watch. The discussion on chain forks and alt-coins is the crux of the issue. Perhaps a simpler way of phrasing the point, would be to say:

'Chain forks are 100% value preserving, they allow participating investors to assign value to the fork they believe will increase the usefulness of the bitcoin economy, without risk to non-participating bitcoin holders. Whereas alt-coins are value diluting, they re-assign value to new coins at the expense of all non participating bitcoin holders.'

It's great to see the idea of multiple competing chains, which share a common ancestor ledger, getting more airtime. This is without a doubt the future. The concept of a client that will track divergent ledgers is well worth exploring. I'm not yet convinced, it's something that will definitively become a reality - in that the economic incentives are so strong, I believe the weaker chain will almost immediately die.

However a client designed specifically to handle such a divergent ledger scenario will be a massive improvement on the current generation. Parallel validation seems like a logical starting point here, leading into two isolated databases.

wrt. fork futures. I think it's been said, @Zangelbert Bingledack or @Mengerian but there appears to be a key problems here, that I've not seen a solution to. Namely they need a fixed reference point. I can't see how you can have a true futures market without knowing in advance that a fork is going to happen and what each side of that of the fork represents? In the example of a hard fork to >1MB at say block 500,000 it's easy we have the reference point, but if you try to apply similar logic to the current state of the market, everything becomes intangible.

Are we going to need a prediction market for a future fork market? Messy.

Anyway great talk Daniel we need more of these. Once people understand the points you are making it becomes obvious that the Core roadmap, is either grossly deluded or just blatantly malicious.

 

[albin]

The Krawisz interview really got me thinking, if segwit activates and transactions actually happen in mainnet, we're tied to supporting it forever. Given Sergio's recent work, and the fact that segwit was decided upon on a high-level literally immediately after the Hong Kong presentation with zero peer review or feedback from anyone in the community outside the wizards clique, are we ready to commit to supporting this scheme indefinitely moving forward for as long as Bitcoin exists?

 

[AdrianX]

Given what we've learned about the ramifications of non contentious soft forks like the 1MB limit something like segwit will be irreversible.

Is there still a threat it will get activated?

I can only reflect what someone else said, centralized planing always results in the misallocation of resources.

 

[Zangelbert Bingledack]

@lunar

The solution is for miners to act unilaterally (or at least credibly front that they will by scheduling a hard fork flag day). In other words, the very thing Core doesn't want them to do.

 

[jbreher]

if segwit activates and transactions actually happen in mainnet, we're tied to supporting it forever.

OTOH, with the legacy anyone-can-spend nature of SegWit transactions, the more such transactions get made, the greater the incentive for miners to collude to award all those anyone-can-spend transactions to themselves. As the stored value reaches into the $trillions, is this really a resistable honeypot?

 

[freetrader]

BTCfork has the technology for a flag day / soft-fork triggered minority hard fork.

I'm sure we are not alone. If SegWit gets pushed through, there *will* be spin-offs.

 

[Zangelbert Bingledack]

The other thing I always wonder about Segwit is, what happens if the 95% mining super-majority later becomes a minority, as has been the normal historical progression (Ghash.io had 50% and is now gone, so if it had been part of the 95% in favor of Segwit, the Segwit-supporting miners could now be in the minority).

One could argue that any SW tx should be immediately "stolen" by miners in order to convince people not to do something so dumb ever again.

Then again, some say anyone-can-spend doesn't work that way, so maybe I am missing something technical that adds more safety to it even if later miners disrespect Segwit.

 

[molecular]

Activating SegWit for its tiny, one-time throughput increase strikes me as a bit like drinking your own urine when you're lost in the woods and dying of thirst: it might provide some temporary relief, but it's not exactly a viable long-term solution to the actual problem, and you really shouldn't have allowed yourself to get into a situation where the proposition would start to seem like a tempting one.

Even worse: In an attempt to avoid losing all self-respect while trying the direct approach, you're drinking it from a pot you found laying around, adding risk of poisoning and/or infection.

 

[molecular]

Good interview with Daniel Krawisz.

Talks about:

• Problems with Segwit - Jaqen Hash’ghar article
• Quadratic hashing issue
  
• Chain splits versus altcoins
• Idea of client that can handle multiple chains gracefully
• Fork future markets
• Whether verifying-only nodes add value to the network
• The roles of investors and coders

A couple of remarks:

I value Krawisz very highly for his economic viewpoints and expertise. In the technical field he seems a bit lacking. It shows in his attempts to explain the quadratic hashing problem (which is as we know largely a non-issue rather easily solved/contained)

Hist distinction of altcoins vs. spinoff are valid. Also his view of "aggregate bitcoin market cap" shared by all spinoffs and his conclusion of there not developing a world with a multitude of bitcoins is accurate.

I also share his insistence to 'put the investor first' by offering choice (by split in the case of the blocksize issue) and letting the market "figure it out". I also agree that *in the long run* one chain will dominate and therefore the issue of which chain gets to name itself "bitcoin" will solve itself. However his view on how merchants / users would want to react regarding payments in case of a split is naive at best. It seems to me he's porting over his investor view of not having to care about a split and being "automatically diversified" (valid). Doesn't work for payments: transactions are valid on both chains... problem is: they wont get mined equally and utxo sets will quickly diverge (intentionally and unintentionally). Just closing your eyes to that problem and assuming that "wallet software will try to get transactions mined on both chains" is not adequate imo. As a merchant you're going to get screwed left and right by users knowing their shit. Payment processors can definitely help here, but shit needs to be done. If we split, we're going to have to go through a rather long dark valley regarding Bitcoins path of adoption and growth. I'm personally willing to accept that and pull through.

I'd be happy to see if an exchange would take him up on the offer to work on future market definitions. As we can see with bitfinex tokens, good definitions are crucial (I still think the bfx token definitions are intentionally biased)

It makes me very sad everytime I see a figure I respect fall for the core narrative (slush being the most recent example, although I'm not sure at this point) and my knee-jerk reaction is always to suspect coercion or hidden incentives.

It's good to see Daniel hasn't toppled over.

I'll read (I noticed I haven't done so completely) that "a fork too far" article next...

 

[AdrianX]

OTOH, with the legacy anyone-can-spend nature of SegWit transactions, the more such transactions get made, the greater the incentive for miners to collude to award all those anyone-can-spend transactions to themselves. As the stored value reaches into the $trillions, is this really a resistable honeypot?

It seems to me miners will destroy confidence if they ever do mess with those anyone-can-spend transactions. It reflects badly on the developers who pushed it miners pay the price - I believe miners will be incentivized to keep the confidence in bitcoin even if it entails doubling down on a flawed rule and colluding to keep it in place.

 

[Zangelbert Bingledack]

@molecular

The "porting over" of the investors' neutral view of chain splits to merchant/consumer transactions is something I've worried about in the multi-chain Bitcoin concept, but I'm not seeing any real dealbreakers yet.

The problem of simultaneous tx propagated over both chains after a split taking different amounts of time to confirm is annoying, but it seems worst case to only increase the confirmation time marginally. I.e., wallets might consider a transaction to have 6 confirmations only once it has 6 in both chains, so there are two chances for it to take a long-ass time rather than just one. Longer but not that long. Even if there were 8 chains, the average time for all 8 to get a certain number of confirmations is not wildly longer than for one chain, as far as I can tell.

Besides that, since UTXOs will diverge, payments by default in wallet software will have to happen in all chains in the right proportion to be considered "paid," so it seems the only people who will have coins in only certain chains or in odd proportions will be investors. Consumers and merchants who don't wish to take a position never need to mess with it. If anyone tries to pay them without paying in every chain, their wallet will not accept it. So within that sphere the ledger proportions held by each user should not really vary (if they have one BTC-a, they will always also have one BTC-b, etc. - no more and no less).

This will require wallets to be a lot more active in updating and tracking market prices, overall making them a lot more integrated in users' lives. It's a good opportunity for wallet software devs as a business, I think.

And of course the fallback position if this really can't be made to work for some reason is that this very same fact prevents a second coin from ever having much value in the first place (the ETC/ETH persistent survival only being thanks to no commerce happening in Ethereum anyway).

 

[go1111111]

Activating SegWit for its tiny, one-time throughput increase strikes me as a bit like drinking your own urine when you're lost in the woods and dying of thirst: it might provide some temporary relief, but it's not exactly a viable long-term solution to the actual problem,

It seems that many people believe that if we activate SegWit, we somehow become "locked in" to Core's scaling plan, and will have a much harder time hard forking in the future. I'm skeptical of this. Part of the difficulty of convincing people to hard fork now is that SegWit is available and is an actual short term scaling alternative to a HF. If we activate SegWit and fees drop, but then in 6 months fees are back to where they are now, Core has just lost their best argument against doing a HF. A HF becomes the clear reasonable path forward by virtue of it being the only real option.

BU supporters being anti-segwit comes off as spiteful and politics driven, and IMO is bad PR for the big-block cause.

 

[molecular]

@molecular
The "porting over" of the investors' neutral view of chain splits to merchant/consumer transactions is something I've worried about in the multi-chain Bitcoin concept, but I'm not seeing any real dealbreakers yet.

The problem of simultaneous tx propagated over both chains after a split taking different amounts of time to confirm is annoying, but it seems worst case to only increase the confirmation time marginally. I.e., wallets might consider a transaction to have 6 confirmations only once it has 6 in both chains, so there are two chances for it to take a long-ass time rather than just one. Longer but not that long. Even if there were 8 chains, the average time for all 8 to get a certain number of confirmations is not wildly longer than for one chain, as far as I can tell.

If it was just different amounts of time, I'd tend to agree things might be manageable by smart enough wallets. However I'm expecting one of the chains to have significantly higher capacity and also that it will be used (if this doesn't happen naturally, there will probably be "spam attacks" (real ones this time) on the lower-capacity (or both) chains). The lower-capacity chain could fall back permanently with transactions dropping off the network or being in an unknown state for very long periods of time. You'd end up with an increasing number of transactions *never* confirming on all chains and nothing definite being decided in any way. An unbearable situation.

Even if that happened just on occasion, payment processors / merchants / any other recipients of funds would have to be prepared to somehow handle refunds or crediting of partially transferred funds using huge timeouts and whatnot. A nightmare. More than just an annoyance you'll have to admit.

I honestly can't imagine it working well enough at all. I think merchants will just have to choose a chain or offer to be payed in any one of x chains or at least otherwise be clear about the fact that they expect coins from all chains and transparent about their processes and policies regarding this. Again: huge pain.

Don't get me wrong: I'm all *for* the market deciding this. I just hope it will be quick.

 

[Erdogan]

With the current plan, only smallblockers can split the chain.

Let's say largeblocks comes with the 75% hashpower majority, and we have a split, the largeblock chain will quickly take over, because people will see that bitcoin is now viable for more people, more people will be using it and aquiring a cash balance, supporting the value and the mining.

If this does not spur new interest, or if there were some miscalculation or lying in the signalling, the smallblock chain will take the lead, but in that case the largeblock chain will stop. It is not useful for anybody to have a largeblocks chain that is shorter. The largeblocks chain will stop, and we will reorg back to the smallblocks chain, with the opportunity to try again later.

Therefore (again with the current plan of waiting for majority support) I am not worried about a chain split, and I think most wallet suppliers and vendors can do alright by just following the longest chain, disregarding blocksize (consider largeblocks valid). Maybe a short pause in business while the dust settles. I think the only actors needing to support two chains will be the exchanges, if anyone needs to.

 

[AdrianX]

It seems that many people believe that if we activate SegWit, we somehow become "locked in" to Core's scaling plan, and will have a much harder time hard forking in the future. I'm skeptical of this.

look at how a non contentious soft fork like the 1MB block limit is hard to revere where there is no money at risk, no try imagine revering segwit where there are coins on unsupported formats. - Its a total lock in, with an 75% discount should you adopt the new format. once activated there is no turning back that is the only way forward.

If we activate SegWit and fees drop, but then in 6 months fees are back to where they are now, Core has just lost their best argument against doing a HF.

Core are not in this to win arguments they are in it to introduce new incentive rules bundled with Segwit. But make no mistake limiting transaction volume is supported by all the influential Core developers, they see it as necessary, and are working on layer 2 transaction networks to save users from those high fees.

The fork in the future becomes more contentious, and the new block weight makes it more expensive in terms of bandwidth cost, compared to a simple capacity upgrade.

 

[79b79aa8]

in case it has not been noticed, bitmex is offering futures contracts for big block activation and for segwit activation, to expire at the end of 2017.

https://blog.bitmex.com/site_announcement/announcing-new-binary-futures-segwit-big-blocks/

right now traders are betting harder on SW than on big blocks. trading is thin.

 

[albin]

look at how a non contentious soft fork like the 1MB block limit is hard to revere where there is no money at risk, no try imagine revering segwit where there are coins on unsupported formats. - Its a total lock in, with an 75% discount should you adopt the new format. once activated there is no turning back that is the only way forward.

The only way to reverse segwit should it turn out to be significantly undesirable is for a different cryptocurrency to usurp Bitcoin in the market, because once segwit goes live, it has to be supported for all eternity, because as long as some segwit UTXO still exists out there, no longer supporting it is equivalent to burning somebody's money.