Gold collapsing. Bitcoin UP.

[Peter R]

@albin: I'd be fine with Vitalik's suggestion too. As long as Qmax > Q*, the existence of the block size limit should have no significant economic impact. I think his proposal will achieve that.

@chriswilmer had a similar proposal as well (one I actually liked more than Vitalik's).

 

[albin]

I respect Garzik's intentions with BIP100, but I think his approach is fundamentally wrong. Miners should vote based on the blocks they actually create if snything, and not some meta strategy of embedding votes as messages.

Flexcap so far is super vague, but from what I understand so far it is completely unacceptable. Maxwell and Friedenbach should make an altcoin if they want to do crackpot experiments like that. The hubris behind the idea is absolutely unacceptable, that devs like this abuse their "expert" status in applied mathematics as leverage to over design complex systems of economic incentives.

The only acceptable blocksize solution is either none at all, or merely maintaining the original anti-DoS plan by making blocksize outpace empirically-observed real usage with a reasonable headroom.

 

[cypherdoc]

with any of these proposals it's important to KISS.

the fewer lines of code changed and the fewer economic concepts required, the better. simplicity and understandability is important.

 

[albin]

In my heart of hearts I can't avoid feeling like Flexcap is the real objective here, in order to create a command and control infrastructure over transaction fees.

If the resource requirements of running "economically dependent full nodes" were really the concern, then Gavin's attempts to actually discuss what spec they're designing for wouldn't be met with crickets. There is a tacit unwillingness to discuss anything whatsoever to do with defining what those resource requirements actually are, and what would be a reasonable bandwidth/memory/storage spec for a full node. This is a very concrete subject where an extremely productive (albeit boring!) debate could really achieve results.

 

[Melbustus]

...
The only acceptable blocksize solution is either none at all, or merely maintaining the original anti-DoS plan by making blocksize outpace empirically-observed real usage with a reasonable headroom.

Agreed.

I'm starting to think that Vitalik's original suggestion (which is implemented in a lot of altcoins) might be the best thing: take some kind of moving average of recent blocksize and multiply by some factor to create headroom. Say for example take twice the average of the last x number of blocks. This is functionally unlimited blocksize, but there being some kind of ceiling appeases the concern that some shocking attack block will be unleashed on the network. I like this as second best to no blocksize cap at all because it's functionally unlimited except in very pathological edge cases.

Yeah, for example, monero implements something like this, right? If this is on the table, I think it falls in the better half of potential solutions. Added benefit: investors have always said that if an alt pioneers a sufficiently useful feature, bitcoin would likely just incorporate such feature after it's been proven out in the alt-space. This would be a substantial validation of that theory (and de-validation of the long-term viability of 99+% of alts).

 

[Zarathustra]

I'm starting to think that Vitalik's original suggestion (which is implemented in a lot of altcoins) might be the best thing: take some kind of moving average of recent blocksize and multiply by some factor to create headroom. Say for example take twice the average of the last x number of blocks. This is functionally unlimited blocksize, but there being some kind of ceiling appeases the concern that some shocking attack block will be unleashed on the network. I like this as second best to no blocksize cap at all because it's functionally unlimited except in very pathological edge cases.

Yes, that kind of 'FlexCap' wouldn't block the stream on the mainchain. But I guess the Blockstream devs will propose a cap that's small enough to enforce the stream into (their) sidechains.

 

[Peter R]

@maaku7's abstract for SBHK:

The "flexcap" is a suggested improvement to Bitcoin which allows a miner to exceed current block size limits when sufficient fee pressure exists, and periodically adjusts such limits according to observed demand.

A miner may create a larger block by claiming less subsidy than they would normally be entitled to, the new limit being a function of how much subsidy was given up. It is in the miner's rational best interest to do so when the available fees exceed the subsidy foregone. The deferred subsidy may be claimed by future miners who choose to create smaller blocks than the present block size limit would have allowed. The base block size limit is periodically adjusted based on an average of the declared limits of the past N blocks. In this way miners are able to generate larger blocks in response to fee market demands in a non-gameable way, and the base limit is adjusted when faced with persistent demand.

This talk will both present the general concept of a "flexcap", and discuss in detail a specific implementation thereof, including a proposed choice of parameters appropriate for Bitcoin.

Mark Friedenbach
(Affiliation: Blockstream)

 

[Peter R]

https://www.reddit.com/r/Bitcoin/comments/3tu1tb/it_actually_doesnt_really_matter_if_blockstream/

 

[AdrianX]

Thinking about it - most of the big miners must be well below break-even and must have lost a lot of money betting on the price?

Meanwhile their very (in-)action is keeping the price down but they don't see the connection.

@awemany The Big miners who make there own chips are not well below break-even.
If they are running equipment like me that is over 1.5 years old, operating costs are still in the realm of about 60% of revenue, my costs are relatively high. Break even probably happened on the total investment a few months after deployment. If Miners have new efficient equipment the operating costs of production are a fraction of income.

The retail AntMiner S7 Asics will likely ROI for retail consumers in 6 months on an inflated capital cost of $1500. The capital investment will be revenue positive until the beginning of 2018 even if the network grows at an exponential rate of 3% wile price remains flat.

In reality the retail price for the $1500 ASIC breaks down into 5 PCB's and 2 fans and an enclosure that should have a production cost of under $50.

This is where the Core Developers get centralization it all wrong. They think it's the ROI of individual miners that is unsustainable because individuals like me don't have the benefit of low capital investment. (I buy my production equipment in an immature market that has little competition at this early stage so I pay an inflated premium). They don't understand the market and they think it's going to all centralize in big mining outfits that make the equipment, and overlook the competition between these outfits.

While that's true in part, the return on investment is huge so there will be lots of competition, and as AntMiner proves, selling equipment is a great insurance policy on that investment it allows them to get a return up front without the risk of mining themselves.

Any one wanting to compete in the industry with relatively low risk (amusing bitcoins is going to succeed) still has an opportunity for a great ROI. Competition is guaranteed when you look at the opportunities. It's my feeling the Core Developers are just risk adverse when it comes to industrial production of mining equipment. (if they cant make a 1000% return in 6 months well mining is not worth it.)

 

[albin]

@Peter R

The conflict really is about Qmax > Q* vs Qmax <= Q* (or even "whatever your Q* is lame stupid troll").

I really need to see something other than posturing and passive-aggressive shenanigans from Maxwell & Co. to justify blocksize as the necessary mechanism for supply scarcity to take what they're saying seriously. I have been so eager to be charitable about anything at all they might have to say that isn't pompously trying to lose people in a corn-maze of circumlocution. Artificially-constrained blocksize as a necessary economic parameter is just way too extraordinary a claim to accept on the unconscionable lack of evidence and argumentation the community has been publicly presented.

 

[Melbustus]

https://www.reddit.com/r/Bitcoin/comments/3tu1tb/it_actually_doesnt_really_matter_if_blockstream/

Looks like some smallblockers found that thread....sheesh.

 

[VeritasSapere]

A problem is I could re-phrase your statement above to say "We want Chinese miners to agree to mine larger blocks, so that more competent miners can become more profitable than you and out compete you".

The problem today is more than 50% of mining might be in the less competent or don't care category, because the coinbase payout enables them to be so. However at the same time we need mining agreement to adopt BIP101 or something similar.

Bitpay, coinbase and 99% of users could all decide to adopt larger blocks. But if 51% of the miners refuse and only build on 1MB or less blocks, then we are stuck on a 1MB chain. That is the way mining based security works.

In the end we need miner buy in, but I've started to become worried about gaining that buy in.

I do disagree with the idea that such a high percentage of miners are less competent. Since the theory of Bitcoin so to speak is that it does rely on this presumption of the miners doing what is best for Bitcoin, based on game theory, economics and psychology. The miners will essentially follow the money. In some ways, the miners act as a proxy for the economic majority. The way this works today is that pools are proxies for the miners and the miners are then in turn proxies for the economic majority. Since miners do need to follow the chain which has the most value, it is not necessarily that clear for me at this time what the sentiment of the miners will be in a few months from now, it could be that they are just waiting to see what Core will do before January. Since the majority of the mining power does support an increased blocksize after all.

Essentially I do think that mining works, in the sense that they will be incentivized to do what is best for Bitcoin. I am a miner myself, not a large one by any scale, but I do understand that after investing a lot of value and effort there certainly is an incentive to support what is best for Bitcoin. It is similar to owning Bitcoins, however a mining farm is not necessarily always as easy to sell especially without significant loss. Many miners I suspect might also be ideologically motivated, you need to at least believe in Bitcoin to a certain extend to invest in such an operation.

 

[Peter R]

Looks like some smallblockers found that thread....sheesh.

Wow! What a swing in the votes! I wonder if someone has a bot that has lots of accounts and logs in slowly using different IP address to sink certain threads. I've seen this behaviour several times now where a thread makes its way to the top of the first page quickly, and then slowly sinks to only ~50% upvoted.

 

[Zangelbert Bingledack]

Maybe they have their own forum somewhere where these things get linked. The same happens when we link to threads or comments from here, though to a lesser extent. It's more pronounced when it's an old one like the "MikeCoin" thread. I've been actually voting everything toward 1 when I can, because otherwise it looks like manipulation.

 

[VeritasSapere]

Some of you might already be familiar with some of my politics on bitcointalk. I have tried to express my alternative opinion on that forum for quite some time now in the face of some pretty horrendous trolling.

Much to my surprise however gmaxwell himself actually responded to one of my posts. He was not doing himself any favors however by responding to this statement of mine: "Bitcoin should not have any leadership, which is why the development needs to become more distributed."

He responded to me saying this, somewhat implying that he did disagree with me. I do not expect him to respond again since I can not imagine him doing so without it reflecting badly on him or he would at the very least need to change his position. Here is the post anyway, I thought you might all find it interesting:

https://bitcointalk.org/index.php?topic=1162684.msg13025363#msg13025363

I was rather pleased with my counter argument, I made sure it was water tight, not everyday I have a Core developer responding to me on that thread. lol

 

[Zangelbert Bingledack]

Gmax:

Do not be confused: The fact that most of the dozens of developers choose to collaborate to build a stronger and better Bitcoin implementation than we could build alone does not make development non-distributed.

The process is Bitcoin is one that amplify the independence of developers generally, even beyond the level of open source software, -- including down to the fact that multiple developers must collaborate to produce releases (so we do not just end up with only one or two person who knows how), that any user can produce the same binaries that we do, that we do not have an auto-update process, and so on. Many more complex features begin life in developers personal forks (which exist, though most don't make releases from them intended for the public (although Luke-Jr has for years)). Our software licensing enables developers to go off and do their own thing based on the codebase, even if the original authors strongly disagree with them.

When you say stuff like this, given the permissive open source software and development process what you're effectively saying is that developers should cooperate less and instead expend their efforts on more duplicated work.

Interesting. He's basically right here, because he's saying "anyone can fork it, go right ahead!" And it also points to why there is so much dev centralization as it is: it seems the natural tendency is for dev to centralize so as not to duplicate work, except when there is real controversy, which we encountering just now for the first time. This is where the forking mechanism comes into play, but Core/Blockstream want to call XT an "attack on Bitcoin." If Gmax was among those calling XT an attack, this seems hypocritical. I seem to remember him saying that, but maybe it was only the others??

EDIT: Now that I've read your (@VeritasSapere's) response, I notice you say a lot of the same things. The advent of real controversy makes forking as a governance mechanism a useful and necessary tool for the first time. This is a major paradigm shift for people to wrap their heads around and probably accounts for a lot of the confusion in the debate.

 

[cypherdoc]

The Big miners who make there own chips are not well below break-even.
If they are running equipment like me over 1.5 years old, operating costs are still in the realm of about 60% of revenue, my costs are relatively high. Break even probably happened on the total intestine a few months after implementation. If Miners have new efficient equipment the production operating costs are a fraction of income.

The retail AntMiner S7 Asics will likely ROI for retail consumers in 6 months on an inflated capital cost of $1500. The capital investment will be revenue positive until the binning of 2018 even if the network grows at an exponential rate of 3% wile price remains flat.

In reality the retail price for a $1500 miner breaks down into 5 PCB's and 2 fans and enclosure that shroud have a production cost of under $50.

This is where the Core Developers get centralization it all wrong. They think it's the ROI of individual miners that is unsustainable because individuals like me I don't have the benefit of low capital investment, (I buy my production equipment in an immature market that has little competition at this early stage). They don't understand the market and think it's going to all centralize in big mining outfits that make the equipment, and overlook the competition between these outfits.

While that's true in part, the return on investment is huge so there will be lots of competition, and as AntMiner proves, selling equipment is a great insurance policy on that investment.

but any one wanting to compete in the industry with relatively low risk still has an opportunity for a great ROI, competition is almost guaranteed when you look at the opportunities. it's my feeling the Core Developers are just risk adverse when it comes to industrial production of mining equipment. (if they cant make a 1000% return in 6 months well mining is not worth it.)

This is incredibly important info. Thanks for illuminating this for everyone.

 

[Zangelbert Bingledack]

Shower thought: what's really the difference between these two scenarios?

1. Core devs recommend an updated version of Core with a 2MB cap.
2. Core devs simply recommend a 2MB cap (and people set it themselves in BU or whatever)

 

[cypherdoc]

I do disagree with the idea that such a high percentage of miners are less competent. Since the theory of Bitcoin so to speak does rely on this presumption of the miners doing what is best for Bitcoin, based on game theory, economics and psychology. The miners will essentially follow the money. In some ways, the miners act as a proxy for the economic majority. The way this works today is that pools are proxies for the miners and the miners are then in turn proxies for the economic majority. Since miners do need to follow the chain which has the most value, it is not necessarily that clear for me at this time what the sentiment of the miners will be in a few months from now, it could be that they are just waiting to see what Core will do before January. Since the majority of the mining power does support an increased blocksize after all.

Essentially I do think that mining works, in the sense that they will be incentivized to do what is best for Bitcoin. I am a miner myself, not a large one by any scale, but I do understand that after investing a lot of value and effort there certainly is an incentive to support what is best for Bitcoin. It is similar to owning Bitcoins, however a mining farm is not necessarily always as easy to sell especially without significant loss. Many miners I suspect might also be ideologically motivated, you need to at least believe in Bitcoin to a certain extend to invest in such an operation.

yeah, this is my opinion.

initially, before 100, all miners incl the chinese were on board with 101 and 8MB blocks. then, unfortunately Garzik gave them even more power thru meta-voting. mistake. miners are making the transition to more fiscally responsible strategies i'm confident. it'll take time but we are likely to see some movement in Dec/Jan.
[doublepost=1448226485][/doublepost]

Wow! What a swing in the votes! I wonder if someone has a bot that has lots of accounts and logs in slowly using different IP address to sink certain threads. I've seen this behaviour several times now where a thread makes its way to the top of the first page quickly, and then slowly sinks to only ~50% upvoted.

miniblockists tend to be the most toxic and accusatory using allegations that project their own morals and actions. i'm sure most of us would never even consider using a sockpuppet account. it's a childish and unproductive use of time for beginners.

 

[Zangelbert Bingledack]

Btcd dev Dave Collins, sounds like he has BU-like leanings: