Gold collapsing. Bitcoin UP.

Peter R

Well-Known Member
Aug 28, 2015
1,398
5,595
Are you at least prepared to assume good faith for the purposes of productive conversation? Out of interest why do you think Gavin is wrong when he says Blockstream acts in good faith?

I assumed good faith for a long time, but I've changed my mind. Given the censorship, the attacks, the inconsistencies, and the overall weirdness, it's too much for me to look the other way.

I now believe that either they are desperate to build a business around LN/sidechains/etc. and they believe that crippling the main chain is necessary, they are working to undermine Bitcoin for reasons that I don't understand, or their leadership is mentally unstable.
 

molecular

Active Member
Aug 31, 2015
372
1,391
Jonny1000 said he solo mined and never found a block.
The pool will also tell you immediately if your shares are valid thus helping the network.

In my view saying that you solo mined but never found a block does not give you any credibility, in fact it hurts your credibility.
point taken and I agree about the credibility.

Jonny1000 said he solo mined and never found a block.
I remember when the electric company called to enquire about my large bill and make sure that I wasn't doing anything illegal with the power.
I actually called mine beforehand and increased the monthly pay (the do yearly accounting) so I'd have an easier time decreasing it again when time came... it even worked.
 

Aquent

Active Member
Aug 19, 2015
252
667
What exactly is strong consensus? Plus your argument is a bit circular. You seem to be saying you agree with hardforking to 2mb, but because it was proposed by classic you disagree, and because you disagree it is contentious and because it is contentious it should not happen.

If it is correct that you agree with hard-forking to 2mb, why does it matter who or what proposes it? As Bitcoin is decentralized, anyone is free to propose anything they wish. Why would who or what proposes something you agree with suddenly make you disagree with what you agree with?

As for your main premise that change should not happen when there is no consensus, even though somehow the individuals who are making it lack consensus seem to agree with the change, but for some reason disagree at the same time, thus making it have no consensus, I find it contradictory to - presumably - your position that softforks are fine without consensus.

As we know, a softfork can change the limit as segwit - soonish - plans and a soft fork can change the 21 million cap as well. Therefore it likewise should require "strong consensus", but presumably you think changing parameters through a soft fork doesn't require such consensus. Perhaps you can expand on why, because from a logical point of view your argument seems incoherent and contradictory.

The truth of the matter is, isn't it, that certain individuals do not wish to increase maxblocksize at all, but do not want to state so, leading them to make contradictory arguments of you need consensus, but, although I agree, I will still disagree or you need strong consensus for a hard-fork but at the same time you don't need strong consensus for a soft-fork even though anything can be changed with a soft-fork.

The same mental dissonance can be found in regards to fees. The plan is, presumably, to increase fees, but if it is shown that fees are increasing, then all the sudden this fact becomes FUD.

Or, the plan is for bitcoin to operate under full blocks, but when it is shown blocks are full, the fact is inconvenient and argued away as spam.

Supporters of big blocks have been straight with people by using rational and logical arguments. Perhaps you should try it @jonny1000 , rather than using emotional and irrelevant arguments about attacks or subjective appeal to an ill-defined "strong consensus" or the classic I agree with 2mb, but I disagree because it is contentious, and I am making it contentious because I disagree and I disagree because suddenly making a proposal is an attack, although I do agree with the proposal, I just disagree with... what, who made it? Or is it because it did not follow some "process" which doesn't exist and has clearly failed as we are seeing Maaku and Gmaxwell disagreeing on what you agreed to in the roundtable "consensus" even though a timeline of 1 year is to be proposed with presumably a threshold of 95%.

Maybe, @jonny1000 , all these emotional arguments have nothing to do with the only logical argument from small blockers made by gmax in 2013 - bitcoin works "only" because of the 1mb limit - otherwise it would not be decentralized.

Now, perhaps the small blockers can prove a negative and maybe they have a crystal ball as well, but at least show us the respect of not saying you agree while saying you disagree in the same sentence and actually argue the logical and rational point rather than making emotional arguments about "attacks" .

Truth of the matter is 75% or locking out 25% or classic or core or 1 year lead time or 28 days or a million of other arguments have nothing whatever to do with us still having no capacity increase and the only thing that has anything to do with it is the 1mb forever mindset (despite vocally saying otherwise) as is very clear from gmax's and maaku's statement in regards to your hard-fork proposal.

By the way, didn't Jonathan Toomim pull request a 2mb increase? I can't recall, I think at the time he said he was going to. If he did, and - apparently according to what you are saying core developers agree with a 2mb increase - then why not merge that pull request or if parameters were different why not change them and merge the changed pull request?

Now, if you are saying that would have been under "influence" and it needed to be shown that others could not just change the parameters without "consensus", how could you be influenced to agree to something you already agree with? An attack suggests a forced change in regards to something you disagree with, but if you are saying core developers actually agree with 2mb then what part is the attack? That a proposal was made? Or is the attack in fact the opposite, that despite seemingly everyone agreeing as you are suggesting nonetheless it was decided to remain at 1mb forever, presumably because they agree with 2mb, which is why they stood at 1mb and made no hardfork pull request. It is, of course, because they agree with 2mb that they consider a proposal of 2mb an attack.

Maybe you can enlighten us on all these contradictions Jonny, and, more importantly, if you have been keeping up with segwit, give us an update on whether it is moving at all after 6 months, despite being tested for a year or so on a sidechain and therefore not being complex, but ready any day now.
 
Last edited:

Zangelbert Bingledack

Well-Known Member
Aug 29, 2015
1,485
5,585
@jonny1000

I think I at least understand your view now, and part of the view of Core and its supporters. It does actually make sense if you assume every *current participant* in Bitcoin must consent to all changes to "Bitcoin." I think the key piece you are missing is that "participant" and "Bitcoin" are fluid concepts, and thus it is quite possible to have consensus among all "participants" in "Bitcoin" while not having one single united direction that must be adhered to for Bitcoin to work. I'll elaborate via example.

Let's take the future attempt to change the 21M coin limit. To put some rough numbers on it, suppose at a given time 20% of current participants (stakeholders, investors, users) representing 40% of the economic power of the community are clamoring to double the inflation rate.

I say, oh well, price will drop by 40% at most, though we anti-inflationists can sell off our coins in the inflationary fork to recover the loss so the only people who are really hurt by this are the inflationists. Unless their fork does succeed, in which case we who sold our coins in that fork are the ones who lose money (nothing happens to the purchasing power of those who slept through the fork).

Anyway, as the forked ledger situation above shows, "participants" and "Bitcoin" are fluid when the Bitcoin ledger can be copied with participants going their separate ways. All participants in what *they* think is Bitcoin stay in consensus at all times. Or put more carefully, for any given participant in a given version of Bitcoin, they find at all times that there is perfect consensus among them. Different participants will want to call a fork "Bitcoin" and another fork "not Bitcoin," and likewise they will call some people "participants in Bitcoin" and other people "no longer participants in Bitcoin," so within their viewpoint they always have "consensus among all Bitcoin participants."

Yet from the third-party perspective of someone who was never involved in Bitcoin, the "Bitcoin community" has split and there is no consensus, because the two groups disagree *with each other* about what Bitcoin is. That's irrelevant to each set of participants, though, other than that they may be annoyed by the potential confusion for new investors and that new investors may be drawn into the "wrong ledger" (nevertheless, these concerns can be addressed just by not selling all their coins in the other ledger after the fork happens; and presumably if there is such a huge split, the market was demanding it so the total value of the two ledgers combined will be greater than it ever was as a single ledger - not a bad situation at all).

Needless to say, it very much concerns me when I see arguments about consensus with no mention of investors and ledger copying dynamics. It makes me think the analyst doesn't even consider such things. I am hard pressed to recall any mention of either investment aspects of the consensus process or ledger copies on the Core dev mailing list or by Core devs or their supporters on reddit. It seems to be assumed to be exclusively a compsci issue.

By the way, I don't have any inclination to think you are Adam Back or are affiliated with Core/BS. I also doubt Blockstream is pursuing *nefarious* self interest in trying to keep the blocksize cap centrally controlled and low, but I do think they are pursuing just plain self interest. They believed Bitcoin couldn't work, and they never actually understood it, so they created a company to profit off what they considered to be saving it.

I say "considered," because now that they have attracted all that venture capital, it would be quite difficult personally, professionally, and financially to change course even if a new argument was raised that made them realize they are wrong. That's the reason the self-interest angle cannot be dismissed just because they already held the views before starting the company. (However, given we can peacefully fork as illustrated above, I don't really find BS to be a long-term concern. And I am fine assuming good faith, because for most things I discuss it doesn't matter whether they are trying to destroy Bitcoin deliberately or are just misguided and trying to help.)
 
Last edited:

Inca

Moderator
Staff member
Aug 28, 2015
517
1,679
@lunar @Norway

I tried that formula from wolfram and it I get the same output which is good! For those interested the practical relevance was of trying to generate a mathematical model based around proximity of laparoscopic trocar sites and likelihood of collisions to try and optimise surgical port placement.

Code at: github.com/surg0r/intersect

--

Any thoughts on the price action cypherdoc or predictions for the Halving? Are any of you holding ethereum?
 
  • Like
Reactions: lunar

satoshis_sockpuppet

Active Member
Feb 22, 2016
776
3,312
As I have explained many times, Core has not had a opportunity to do a responsible HF, due to the attacks.
That is the problem. If you see another implementation as an "attack" you shouldn't be leading the standard implementation. I've never seen Unlimited, Classic or btcd devs accuse each other of "attacking bitcoin". And I think the way SW is deployed is as much a reason to abandon core asap as the limit crippling.
Had Gavin come forward, at this point, with a 4MB proposal, to come into effect in May 2016, it would have been resoundingly accepted.
Bullshit. Complete bullshit. You know as well as anybody that Core would have "discussed" that proposal as much as any other.
You see what you are doing? Core "promised" a 2MB HF this summer and you're already collecting reasons ("attacks") why they can't do a HF now as well.
[doublepost=1463915678][/doublepost]
What exactly is strong consensus?
It's manipulation. Core and BS knowingly took the technical consensus (longest chain) and used the word to manipulate the people into thinking that some kind of undefined "community consensus" is necessary to do anything.
It's perfect breeding ground for psychopaths like Maxwell.
[doublepost=1463916344,1463915286][/doublepost]
Trouble in paradise?
 

jonny1000

Active Member
Nov 11, 2015
380
101
By the way, I don't have any inclination to think you are Adam Back or are affiliated with Core/BS. I also doubt Blockstream is pursuing *nefarious* self interest in trying to keep the blocksize cap centrally controlled and low, but I do think they are pursuing just plain self interest. They believed Bitcoin couldn't work, and they never actually understood it, so they created a company to profit off what they considered to be saving it.

I say "considered," because now that they have attracted all that venture capital, it would be quite difficult personally, professionally, and financially to change course even if a new argument was raised that made them realize they are wrong. That's the reason the self-interest angle cannot be dismissed just because they already held the views before starting the company.
I agree that there is a large amount of subconscious bias and unhelpful influences on both sides of this argument, driven by stubbornness and ego, in particular. I do not think unsubstantiated accusations of deliberate nefarious influences such as for personal financial gain or being a government agent, are helpful or true.
[doublepost=1463919501][/doublepost]
That is the problem. If you see another implementation as an "attack" you shouldn't be leading the standard implementation. I've never seen Unlimited, Classic or btcd devs accuse each other of "attacking bitcoin".
I have repeated this many times, nobody sees other implementations as an attack. Classic is an attack because it aims to eliminate a consensus rule, locking in significant opposition at the time of activation, as part of a deliberate and stated aim of changing the rules on a significant part of the network, against their will. Please can you stop mischaracterizing this as if people regard any alternative implementation as an attack. This mischaracterisation causes unhelpful confusion and animosity. Why do you insists on making this false and destructive claim?

I am not saying launching this attack is immoral, I just think it is vital that Bitcoin has the incentives in place to ensure the attack is defeated. The attack is being defeated by participants acting freely.
  • 95% of miners are not participating in Classic
  • 82% of nodes are not participating in Classic
  • 85% of developers are not participating in Classic
  • 85% of investors, in the only polls which exists, do not want Classic.
I therefore do not see what you are complaining about. If you want a coin with the existing Bitcoin holders, launch a new coin with the Bitcoin UTXO (or hash of it) in the new genesis block. If you want 2MB or more as the blocksize limit in Bitcoin, give strong consensus a chance. I am very confident that once you give strong consensus a chance you will get overwhelming support from miners, who only avoided supporting Classic by a minuscule whisker, after being put under enormous pressure, because they didn't want to lock in 25% opposition. The stronger the consensus you require the higher the chance of success. I estimate 95% is a victory maximization threshold. If you do withdraw Classic and come out with new HF attempt, please do not try and push it and choose 85% and a 3 month grace period, just choose 95% and 6 months and get it done.

[doublepost=1463919954,1463919254][/doublepost]
I find it contradictory to - presumably - your position that softforks are fine without consensus.
I support the Core team doing softforks with a 95% threshold, just like I would support a shift to 2MB with a 95% threshold. However I recognize that softforks can be done more easily, strictly speaking they can be done by 51% of miners. If the Core team put forward a 51% softfork I would oppose it and advocate a higher threshold, however I recognize that miners could win this battle.

There seems to be some confusion on this thread, that my desire for strong consensus before a hardfork is purely an ideological position. My reasoning for this requirement is driven by how the network actually works. A hardfork pushes existing clients onto another chain, therefore strong consensus is needed to ensure only a tiny minority remain on that chain. To clarity further, this is a fact, because it is how nodes operate, it is not an opinion or ideology.
 
Last edited:

Aquent

Active Member
Aug 19, 2015
252
667
Actually, I think now it is a pretty good time to pull request in Bitcoin Core a 2mb increase. The code is mostly written. The 75% threshold can be changed to 95%, just change the number, or better, just make it a flag day with a one year lead time from now. Should take maybe an hour or so to get it all done. Although not sure if it needs modifications due to segwit.

Obviously it is far too little and far too late, but at least it is some increase when it seems that there might be none except for maybe segwit sometime in autum if this year at all.

Not sure why it hasn't been pull requested yet. Maybe you can do so @jonny1000 ?

"strictly speaking [softforks] can be done by 51% of miners"

Strictly speaking, hardforks can be done by 51% of miners too. Bringing up a node isn't hard. It would create two chains, but I think at least two softforks have created two chains, in 2013 and one during summer 2015. The monetary incentives ensure that two chains don't co-exist. Moreover, the human incentives can make sure that the minority chain can not operate by directly forking their blocks.

Perhaps for you it isn't ideological, but gmaxwell calling you and others a dipshit for agreeing to a 95% or whatever threshold hardfork with a one year lead time shows clearly that it has nothing to do with irrelevant 75% or 95% or 99% threshold or a flagday, but with 1mb forever.

Did you see who Luke went to consult first on the HK hardfork? Popescu. And popescu is very much ideologically 1mb forever as I suspect is Todd and Gmax and maybe some others. By your definition of "strong concensus" as long as Mr Popescu disagrees and/or gmax (who clearly does according to his statement) then it is contentious.

Under that view of 1mb forever, then the emotional and divisive argument of "attack" makes more sense with the threshold nonsense just some cover for the real argument - 1 mb forever. While your argument that I agree with 2mb, but 2mb is an attack doesn't make any sense.

In any event, it would make sense to see a pull request regarding 2mb this month. Not sure why it hasn't been done already considering that most of the code is written. If any other changes that are contained in the wishlist are to be added, then they can have their own pull requests. We can then see whether there is any "strong consensus".

I very much doubt it, just as I doubt we'll see any segwit any time soon. And while I was hopping we'll see LN by June as Adam Back stated and Poon, I obviously very much doubt that will happen too.

But hey, we've been talking about it for more than a year now. Miners made their choice. Just sit back and enjoy the show I guess, but people are very worried, complaining about tx delays, etc., while there is no indication whatever of any support for capacity increase.

In those circumstances, perhaps you should be persuading your own 1mb forevers that an increase is really desirable and soon, rather than telling us here that our attempt to avoid the PR disaster we are experiencing were or are an attack.
 
Last edited:

AdrianX

Well-Known Member
Aug 28, 2015
2,097
5,797
bitco.in
@jonny1000
I support the Core team doing softforks with a 95% threshold, just like I would support a shift to 2MB with a 95% threshold. However I recognize that softforks can be done more easily, strictly speaking they can be done by 51% of miners. If the Core team put forward a 51% softfork I would oppose it and advocate a higher threshold, however I recognize that miners could win this battle.
You're accepting the notion that the bitcoin protocol should be decided by a centralised committee consisting of just the miners and the developers who write their code. You outright ignore this as controversial.

That 95% threshold includes only the consensus of a handful of people. It excludes the nodes who by design are supposed to protect maintain and validate the protocol.

You're advocating for a radical change to how Bitcoin was designed (and has been working for most of it's existence) you're unequivocally supporting centralized control of the protocol and somehow think that a 95% or variation thereof consensus among the group of central planners is OK.

Keep in mind miners may have no say other than to agree in this 95% if they have a consenting opinion they are potentially at risk of being expelled from the Relay Network under the control of the central planners.

It looks like the irony of your position is lost on your earlier statement that the nodes play a functional role in the preservation of the protocol.
 
Last edited:

jonny1000

Active Member
Nov 11, 2015
380
101
Did you see who Luke went to consult first on the HK hardfork? Popescu. And popescu is very much ideologically 1mb forever as I suspect is Todd and Gmax and maybe some others. By your definition of "strong concensus" as long as Mr Popescu disagrees and/or gmax (who clearly does according to his statement) then it is contentious.

Under that view of 1mb forever, then the emotional and divisive argument of "attack" makes more sense with the threshold nonsense just some cover for the real argument - 1 mb forever. While your argument that I agree with 2mb, but 2mb is an attack doesn't make any sense.
With all due respect, I think you are focusing on the wrong people. I will try to be as careful as I can, to explain a difficult topic. In my view there are extremists on both sides of this debate. On the large blockist side, which remember I am on, as I want larger blocks, we need to recognize we have been totally outclassed and outplayed by small block extremists. This is because the large block side made terrible strategic mistakes like a proposal locking in 8GB blocks now or locking in 25% miner opposition when activating 2MB blocks.

The best analogy I could think, hopefully now not controversial, is the anti terrorism policy of the US government post 9/11. There was too much focus on the radical Islamic extremist, who hated American freedoms and wanted to murder as many civilians as possible. The US government incorrectly overestimated the size and significance of this group. Therefore the policy was not to negotiate and be ruthlessly violent against these opponents. The mistake was best described by president George Bush himself, in his “you are with us or with the enemy” statement. With all due respect to American policy makers in this difficult time, they simply did not recognize winning the war was about getting a billion Muslims on their side and the number of people that actually only wanted to murder as many American’s as possible was pathetically small.

Source:
The 1MB forever crowd is ridiculously tiny and maybe consists of 4 or 5 people. Please stop making the mistake of focusing too much on them and try to get the 95% of people in the middle, on your side. You are overestimating how large the number of 1MB forever people are, they can never get to 5% of the hashrate, nowhere near.

(Please note I am not conflating the 1MB forever people with terrorists, its just an example)
 
Last edited:

Aquent

Active Member
Aug 19, 2015
252
667
On what basis do you make the statement that 1mb forevers are a tiny sub-group of small blockers?

Gmaxwell just called you a dipshit for agreeing to capacity increase with a 95% threshold. According to theymos "consensus" definition, that makes it contentious. Theymos too, btw, disagrees with your HK hardfork. I'm not sure if I recall Todd making likewise statements disagreeing.

According to most polls some 80% agree on an increase, with almost all businesses etc, but their opinion doesn't matter. It is only one person, Wladimir, who merges the code and he has stated if it is contentious then he won't merge it.

Gmax's statement makes it contentious. As does maaku's, as probably would Popescus, or Theymos. So, it may be 4 or 5 people, but they happen to be the ones who can persuade the maintainer to not merge the code.

So, maybe you can try and persuade the guy who called you a dipshit to support your roundtable "consensus" rather than telling us we are incompetent or attacking, because under your logic it was us who created an unreliable network while bitcoin related alternatives not only do not exist, but are months away and in the process infuriating users, including in DNM subreddits.

That said, you may have a point about the 75% threshold in the context of that 90% 1992 agreement, not sure why it wasn't changed at the time, but had it not been the threshold it would have been something else not least because bitcoin core has been using a 75% activation threshold since forever, making that argument disingenuous .

Really, it should just be a flagday. That's how Nakamoto suggested it should happen and that's how Ethereum did it back in Spring with, surprise surprise, zero problems.

But the chances of it happening seem minimal and nothing to do with Classic, especially when miners have made it clear they want the change to come from core, but everything to do with the 1mb forever mindset.
 

Melbustus

Active Member
Aug 28, 2015
237
884
...
Needless to say, it very much concerns me when I see arguments about consensus with no mention of investors and ledger copying dynamics. It makes me think the analyst doesn't even consider such things. I am hard pressed to recall any mention of either investment aspects of the consensus process or ledger copies on the Core dev mailing list or by Core devs or their supporters on reddit. It seems to be assumed to be exclusively a compsci issue.
Yes, exactly. I've noticed that most in the small-block camp are careful to refer to the bitcoin community specifically as "the technical community" (or similar language), especially when discussing consensus. This is unfortunate as, at its heart, Bitcoin is an economic incentive system that *must* be driven by as-unencumbered-as-possible market-forces in order for it to work long term while maintaining key properties (supply transparency, uncensorability, etc).

But fundamentally, on a technical level, I think many small-blockists do not actually think PoW mining works:

...I also doubt Blockstream is pursuing *nefarious* self interest in trying to keep the blocksize cap centrally controlled and low, but I do think they are pursuing just plain self interest. They believed Bitcoin couldn't work, and they never actually understood it, so they created a company to profit off what they considered to be saving it.
...
Yes, I fully agree with this. Adam's (and others') comments have been clear that they think PoW mining is a tragedy of the commons scenario, and that they think getting on-chain fees high relatively soon is the only way to potentially make it work. This obviously goes back to the above point: this viewpoint looks very linearly/narrowly/myopically at mining dynamics without appreciating real-world market-forces, and the ability of the market to come up with solutions to what appear to be "problems". (And FWIW, if one doesn't think PoW mining can work, why care about Bitcoin at all? If permissionless verification doesn't *fundamentally* work, it should be obvious that no amount of meddling in the natural market dynamics can both make it work and keep it decentralized, but I digress...)

So I agree that Blockstream's motivations are not malicious, but I do think the business model centered on off-chain scaling probably serves to at-least sub-consciously re-enforce their core belief that PoW mining doesn't actually work (which, btw, is one of Tim Swanson's core beliefs at R3) without herculean, carefully considered, and elaborate technical feats of centrally-engineered crypto-economics.

I think many small-blockers would do well to read the following in full, as they offer a market-based understanding of how Bitcoin works and adapts, which is really the only way Bitcoin continues to function long-term while retaining its key properties:

https://news.bitcoin.com/konrad-graf-bitcoin-block-size-economy/

https://bitcoinism.liberty.me/economic-fallacies-and-the-block-size-limit-part-2-price-discovery/
 

cypherdoc

Well-Known Member
Aug 26, 2015
5,257
12,994
Well done

What exactly is strong consensus? Plus your argument is a bit circular. You seem to be saying you agree with hardforking to 2mb, but because it was proposed by classic you disagree, and because you disagree it is contentious and because it is contentious it should not happen.

If it is correct that you agree with hard-forking to 2mb, why does it matter who or what proposes it? As Bitcoin is decentralized, anyone is free to propose anything they wish. Why would who or what proposes something you agree with suddenly make you disagree with what you agree with?

As for your main premise that change should not happen when there is no consensus, even though somehow the individuals who are making it lack consensus seem to agree with the change, but for some reason disagree at the same time, thus making it have no consensus, I find it contradictory to - presumably - your position that softforks are fine without consensus.

As we know, a softfork can change the limit as segwit - soonish - plans and a soft fork can change the 21 million cap as well. Therefore it likewise should require "strong consensus", but presumably you think changing parameters through a soft fork doesn't require such consensus. Perhaps you can expand on why, because from a logical point of view your argument seems incoherent and contradictory.

The truth of the matter is, isn't it, that certain individuals do not wish to increase maxblocksize at all, but do not want to state so, leading them to make contradictory arguments of you need consensus, but, although I agree, I will still disagree or you need strong consensus for a hard-fork but at the same time you don't need strong consensus for a soft-fork even though anything can be changed with a soft-fork.

The same mental dissonance can be found in regards to fees. The plan is, presumably, to increase fees, but if it is shown that fees are increasing, then all the sudden this fact becomes FUD.

Or, the plan is for bitcoin to operate under full blocks, but when it is shown blocks are full, the fact is inconvenient and argued away as spam.

Supporters of big blocks have been straight with people by using rational and logical arguments. Perhaps you should try it @jonny1000 , rather than using emotional and irrelevant arguments about attacks or subjective appeal to an ill-defined "strong consensus" or the classic I agree with 2mb, but I disagree because it is contentious, and I am making it contentious because I disagree and I disagree because suddenly making a proposal is an attack, although I do agree with the proposal, I just disagree with... what, who made it? Or is it because it did not follow some "process" which doesn't exist and has clearly failed as we are seeing Maaku and Gmaxwell disagreeing on what you agreed to in the roundtable "consensus" even though a timeline of 1 year is to be proposed with presumably a threshold of 95%.

Maybe, @jonny1000 , all these emotional arguments have nothing to do with the only logical argument from small blockers made by gmax in 2013 - bitcoin works "only" because of the 1mb limit - otherwise it would not be decentralized.

Now, perhaps the small blockers can prove a negative and maybe they have a crystal ball as well, but at least show us the respect of not saying you agree while saying you disagree in the same sentence and actually argue the logical and rational point rather than making emotional arguments about "attacks" .

Truth of the matter is 75% or locking out 25% or classic or core or 1 year lead time or 28 days or a million of other arguments have nothing whatever to do with us still having no capacity increase and the only thing that has anything to do with it is the 1mb forever mindset (despite vocally saying otherwise) as is very clear from gmax's and maaku's statement in regards to your hard-fork proposal.

By the way, didn't Jonathan Toomim pull request a 2mb increase? I can't recall, I think at the time he said he was going to. If he did, and - apparently according to what you are saying core developers agree with a 2mb increase - then why not merge that pull request or if parameters were different why not change them and merge the changed pull request?

Now, if you are saying that would have been under "influence" and it needed to be shown that others could not just change the parameters without "consensus", how could you be influenced to agree to something you already agree with? An attack suggests a forced change in regards to something you disagree with, but if you are saying core developers actually agree with 2mb then what part is the attack? That a proposal was made? Or is the attack in fact the opposite, that despite seemingly everyone agreeing as you are suggesting nonetheless it was decided to remain at 1mb forever, presumably because they agree with 2mb, which is why they stood at 1mb and made no hardfork pull request. It is, of course, because they agree with 2mb that they consider a proposal of 2mb an attack.

Maybe you can enlighten us on all these contradictions Jonny, and, more importantly, if you have been keeping up with segwit, give us an update on whether it is moving at all after 6 months, despite being tested for a year or so on a sidechain and therefore not being complex, but ready any day now.
 
  • Like
Reactions: bluemoon

Inca

Moderator
Staff member
Aug 28, 2015
517
1,679
@jonny1000 unfortunately Gregory Maxwell is a 1mb forever person. A lone voice against the crowd perhaps.

Sadly that lone voice has taken control of the open source development of Core. A company he founded with Luke-jr currently employs several Core developers - their livelihoods rest literally in his hands. All senior developers disagreeing with the direction the project has taken are now no longer involved or have been expunged ('expert developer consensus achieved!').

He is surrounded by lickspittles and yes-men such as btcdrak (who is he exactly?) and has closed off normal lines of developer communication whilst supporting active and outrageous censorship of fora with the highest bitcoin traffic 'moderated' by Theymos.

These things are necessary because what he has set out to do is flatly against the wishes of the wider community, the userbase and the major companies in the space who are actively diversifying to other blockchains such as ethereum in protest. Companies such as Coinbase which are in large part responsible for on boarding large numbers of users in the US.

He may be one lone voice. But he is a particularly destructive and divisive influence upon the community because he will do what he feels is right against the wishes of everyone else - and use any means to do so - especially if he stands to personally benefit - damn the consequences. Reading the Wikimedia details or how he misattributed numerous github commits to himself is a highly illuminating insight into his character.

Ethereum supporters are cheering him on with glee. He is the strongest advert to date for ethereum over bitcoin going forward. The only thing he is lacking from an outsiders perspective is a 'sponsored by JP Morgan badge' on the hood of his car.
 
Last edited:

Aquent

Active Member
Aug 19, 2015
252
667
@jonny1000 I don't want to start a never ending debate about minute points almost no one cares about. We have been doing so for more than a year now. On that specific point, almost all softforks, before BIP9, were activated at 75% and then enforced at 95%, so using a threshold that has always been there is not an "attack". There's a difference between activation and enforcement, but the point still stands and prevents you, on a logical basis, from rationally viewing it as an "attack" on that point alone.

I don't really know about your own specific views, although I think we have exchanged debates before, not sure if you are Johnson Lou/Lau (sorry can't recall the name) btw? I think you were happy with Bitcoin Unlimited from what I can recall?

Your argument about Classic being an attack, in my view, is fully wrong, because if the question was some 75% or 95% consensus, then bitcoin core could have just merged a hardfork code with a 95% consensus, thus making any "attack" vanish and as I stated earlier one can't argue they did not want to appear influenced because you can't influence someone to agree to what they already agree. However, they didn't merge any such code and I'd be very surprised if they do.

In my view, these arguments are all a sideshow and a distraction because they can not make the real argument which is, in my opinion, 1mb forever. They clearly see that as a fully loosing proposition, so they don't argue it, although they did do so in 2013. Instead focusing on irrelevant and potentially divisive aspects, such as we are the rocket engineers, they have done no coding, if you merge 2MB I'll quit, etc.

They speak about hard-forks of some code which is specifically designed to not have two chains, while at the same time creating a hard-fork in humans and splitting the community, while looking at users complaining about transactions stuck for hours and instead of responding calling people "dipshits".

Whatever, not my choice. The miners have made theirs. They waiting for this HK hardfork code to be merged in just over a month now. Gmax has made it clear that's not going to happen so, shit-show continues I guess.