Bitcoin Community is Withering—It has Lost the Excellent Circle
Background
Though China has long been puzzled by its brain drain in the form of elite emigration, it is still out of expectation that the rising industry of Bitcoin is also losing its elite.
Circle, a then sought-after start-up that received funding of 60 million USD with China’s Baidu among its investors, has announced it will cut off its business with bitcoin system and shift to blockchain.
Last week a friend asked me why his account balances in his wallet of Bitkan were negative numbers and doubted there might be bugs. It turned out that it was because the transaction fees for transferring bitcoin are too high and the balances of users of small amount transaction can’t offset the transaction fees. Now wallet begin to subsidize these users to ensure a quick transaction confirmation for them.The transaction fee of bitcoin has been risen to 0.5 mbtc from 0.1 mbtc of June, the time when the bitcoin block became full.The even worse experience is that any dynamic fees recommended by any wallet now cannot guarantee a confirmation within 10 minutes while transactions before the block was full can all be confirmed within 10 minutes. This uncertainty is indeed awful.
Satoshi has predicted the need to scale the block and has given the scaling plan.
In 2013 when block was 300kb full, developer Gavin began to promote scaling. The senior developers including Gavin and Jeff expected that block could full after 3 years and began to lead Core team to promote hard-fork scaling. But Core team has squeezed out the visionary developers including Gavin and Jeff. Gavin was compelled to set up another team in an attempt to resume the influence of bitcoin.
In 2015, the Chinese miner-based entrepreneur team began to negotiate with Core and no results have gained in the near 2-year-long process.Core is so shameless that they use censorship to face the opinions from the community.Bitcoin has missed the best timing to scale and has operated at full capacity for half a year. Bitcoin has lost its far-reaching impact for half a year.Core team confine the block size to a small one, making bitcoin expensive, inconvenient and hard to attract new users. Core’s choices are far from supportive of the development of bitcoin companies. In the context, to cut off business with bitcoin system becomes a natural choice. Circle, though a company that has gained huge investment, is hardly an exception.
Core highly promotes anonymity, which is against the mainstream of the human civilization development.
It’s really hard to understand why Core is so obstinate when it comes to scaling, a plan that is in the interest of the whole community. The mystery of Core’s motives has been unveiled in the scaling conference in Milan.
The people with influence behind the Core are anti-governmentalists. Their main goal is to build bitcoin into a type of currency that can resist government’s check and inspection. If bitcoin block is too big, then the ability to resist government’s check will be weakened. Particularly, Core members want to run full nodes through Tor. On Milan conference, Bitcoin Core clearly exposed their thoughts to the public. They controlled the agenda only discussing anonymity instead of scaling. They want make blockchain ledger a complete anonymous one. As for lighting, Core members actually don’t care. Anyone discerning will know lightning can’t be done in a 1MB size block.
Core’s idea to change the transparent ledger of bitcoin into a black and anonymous one will make bitcoin a tool for criminals. This is not only against the will of Satoshi, but also against the of mainstream entrepreneurs and the majority of the community. It is even a deviation from the trend of human civilization.The majority of the people want to see bitcoin a world-class application instead of a tool for money laundry, gambling and drug trafficking.
The public ledger of Bitcoin has already caused misunderstandings. The Silk Road case is still simmering and Coinbase is now negotiating with American government. If Core successfully make bitcoin utter anonymous, Bitcoin will be in a state of full hostility to people and governments. The law-abiding and professional enterprises of bitcoin will find it hard to provide services to the users.
Any business starter with dream and mission will not pin its future on an anonymous ledger. The leave of Circle is a natural reaction.
Core’s high-handedness to the bitcoin community is sterilizing the bitcoin’s soil for start-ups.
I tried to promote bitcoin to a lot of strangers in the first half of this year and found out that more than 99% of the people thought bitcoin is something negative.
Some of them liken bitcoin to heresy, which I think is misunderstanding. But in the past 6 months, the Core’s core members, the most powerful people in the community, have used the means of heresy to manage the communication. One out of their ten sentence is deceitful. They will censor, delete and block posts. They will hire sockpuppets to squabble and even openly coerce miners.
Isn’t this a kind of heresy? Among the “congregation” are users, miners, business owners. They beg Core to scale the block even just to scale to 2M block. But Core member show no interests in negotiation and do whatever they want to.
When Gavin and Mike released Bitcoin XT, Core maligned them as criminals;
When Coinbase was ready to test Bitcoin XT, Core said Coinbase was a company of alt coin;
When miners released new version of Bitcoin Classic, Core said Classic’s developer were making the community fall apart;
When Bitcoin Unlimited released Xthin, Core released the Compact Block, a copy of BU’s Xthin Block, and cursed BU devs for not following BIP development norms
When Viabtc used BU to mine, Core said BU haven’t gone through thorough tests.
…
Core cannot withstand any project that runs counter to their lust. Bitcoin becomes Core’s private garden granting no permission to anyone who wants to sow grass in it. Core will say you are alt coin.
Users and business owners are all outrageous and are criticizing Core. But Core hasn’t made any change. Circle’s leave is inevitable.
*This is the translation of article written by Huang Shiliang, famous opinion leader. The article dose not represent the opinions of 8btc. see: The original text.
COMMENTS(218)
Here is the link to the original comment thread. Or you can comment here to start a discussion. Author: LightningHuang
In 2015, the Chinese miner-based entrepreneur team began to negotiate with Core and no results have gained in the near 2-year-long process.
~ Insert skeleton here ~ “Waiting for 2Mb blocks”
Particularly, Core members want to run full nodes through Tor. On Milan conference, Bitcoin Core clearly exposed their thoughts to the public. They controlled the agenda only discussing anonymity instead of scaling. They want make blockchain ledger a complete anonymous one.
Interesting opinion, I wonder if everyone has been paying their taxes.
As for lighting, Core members actually don’t care. Anyone discerning will know lightning can’t be done in a 1MB size block.
Personally think lighting is being lined up to replace bitcoin in the long run but that’s just me.
FUD, FUD, & more FUD, don’t bother!
A shitty Chinese op-ed, bashing Core, translated into broken English with terrible punctuation.
Thanks for the legit post.
I wept tears of laughter, a genuinely funny attempt to manipulate. Thanks OP!
High street banks are scalable second layer solutions to the central backend, yet they create the majority of the money of the world.
If the majority of bitcoin users are forced onto 2nd-layer only then we’ll end up with a few LN nodes controlling the network. With only a few points of failure it will be much easier to target for shut down and censorship. The number of actual nodes carrying the blockchain will not have any influence in decentralization and will leave the network as fully centralized. This seems to be the real goal of Lightning.
A damning report on the current state of the bitcoin comunity.
They beg Core to scale the block even just to scale to 2M block. But Core member show no interests in negotiation and do whatever they want to.
When Gavin and Mike released Bitcoin XT, Core maligned them as criminals;
When Coinbase was ready to test Bitcoin XT, Core said Coinbase was a company of alt coin;
When miners released new version of Bitcoin Classic, Core said Classic’s developer were making the community fall apart;
When Bitcoin Unlimited released Xthin, Core released the Compact Block, a copy of BU’s Xthin Block, and cursed BU devs for not following BIP development norms
When Viabtc used BU to mine, Core said BU haven’t gone through thorough tests.
…
Core cannot withstand any project that runs counter to their lust. Bitcoin becomes Core’s private garden granting no permission to anyone who wants to sow grass in it. Core will say you are alt coin.
Users and business owners are all outrageous and are criticizing Core. But Core hasn’t made any change. Circle’s leave is inevitable.
among its investors, has announced it will cut off its business with bitcoin system and shift to blockchain.
lol
shift to blockchain
among its investors, has announced it will cut off its business with bitcoin system and shift to blockchain.
lol
shift to blockchain
Core/BS want sidechains. LN is not their baby.
The Chinese miners need to take the courageous step of adopting BU.
The Core team “layer 2” dogma is never going to change.
Until they do, we will go round and round, and could be having the same conversation in three years time.
I suppose it’s the same thing,drive up cost of layer 1,people have to use layer 2 (LN or sidechain) - sidechains will probably be easier to use eventually (I still can’t see LN working in practice, sidechains are just like bitcoin, proven to work) , BC wins.
Yes. That old comic is entirely right. They saw an opportunity to block the stream by keeping the 1MB dam (or meager 2MB segwit) and syphoning off profits for themselves, to generate at least a billion dollars in profits (by the basic logic of venture capital investing; $75 million in investment on a longshot won’t be gunning for anything less than that).
Unfortunately for them, I don’t think 2WP sidechains are really going to be a thing.
This ‘article’ is full of shit.
There will be no weeping.
Interesting how comments here are being reflected back onto the site?!
I think that they could be a thing fairly easily. Was it drivechain that recently proposed a working method that core refused due to ‘patent issues’?
Chinese miners are just as much to blame as blockstream. It’s time we hold them accountable just like we are trying to do with blockstream. If all the minerals revolted against core and ran an alternate client this discussion would have ended a long time ago.
Drivechains seem a lot more promising to me. Paul Sztorc at least understand quite a bit of economics, unlike the main Core devs - although he has some big blind spots as well.
“Then they fight you.”
It’s easy. Just some computer settings. This is good for the interaction for reddit and 8btc. You know 8btc is the largest news/forum platform about Chinese bitcoin community
Oh ffs, can we stop with this attacking conversation of segwit vs BU? Either it’s a constructing criticism or just leave it out of both subs!
Users are getting sick and tired of the attacks on both sides.
The twats that think it’s one vs the other are dumb as a nail. The politics is destroying the technology and deviating from science. Proof to that is the only thing they can do is personal attacks and avoid narrowing the conversation to criticize either solution.
Of course for people shorting the currency it’s working like a charm, so maybe that’s why we see a surge of these posts.
i’d be more inclined to guess that a greater than a minority part of the ecosystem feels disenfranchised because they think that with the status quo leadership there will be no block size increase ever despite actual technological and software advances. For instance… the arbitrary argument that some use that a hard fork must take a year of planning… if a hard fork was ever actually meant to occur they could have included the code with segwit and had it activate 1 year after segwit and it would have satisfied that imaginary argument. Except they didn’t. There seems to be a clear intention to not increase the blocksize ever. That means the market has to go around them.
My own opinion is that we should err on the side of smaller(but obviously larger than we have now) and the best way to do that would actually be to offer some sort of compromise (which they seem incapable of).
This is outright and utterly false. Segwit is a stepping stone enabling many more upgrades that will have a cumulative benefit to scaling, both on chain and off chain. Almost no one wants the status quo, but that has always been a potential outcome if consensus isn’t found on the nature of changes. Thats always been the case, and denying that is lying to yourself.
The argument of long lead time in a hardfork is not arbitrary, it is rooted in the need for security. A hardfork is an incompatible upgrade, you need to socially organize the upgrade outside of the system, and people screwing up fragments the system and can lose people money. As well, hard forking without consensus will guarantee people will refuse to upgrade and stay on the original network, which will cause unbelievable chaos for the ecosystem at large. Advocating this is reckless and stupid.
Segwit is the compromise. Denying this is gaslighting. People making arguments along the line of yours have the same fatal flaw, you immaturely and in a very childlike manner abstract the market away as some amorphous thing that always agrees with your wants and needs. You are dead wrong. The market is composed of individuals, you, me, everyone out there participating in Bitcoin. And cumulatively it looks like the market isn’t bothered by the status quo now, supports Core by a vast majority, and isn’t buying these bullshit delusional attempts to rewrite and white wash history.
There seems to be a clear intention to not increase the blocksize ever.
Obviously wrong, because SegWit already contains a block size increase.
Hardly anybody says there should never be any further block size increase, even on the Core side.
i’d be more inclined to guess that a greater than a minority part of the ecosystem feels disenfranchised because they think that with the status quo leadership there will be no block size increase ever
Segwit is a blocksize increase; it could have been active on the network for 2 months if it wasn’t being blocked by miners. So that makes no sense.
Well it’s obviously one vs the other since BU’s reason for existence is to block segwit
So there you have it: The only reason for Bitcoin Unlimited existence is to block segwit.
Sure things. Has nothing to do with their vision of market-based blocksize-limits.
Yea ive seen the Chinese exchanges/miners especially saying SegWit will make Bitcoin ‘too complicated.’
Personally I support a maxblocksize increase greater than the current SWSF proposal but why do you think Bitcoin Unlimited is not packaging a stripped down version of Segwit in a Hard fork when it’s clear it would help them sway consensus ? They don’t want a malleability fix because they don’t want lightning, while some of the mining pools themselves don’t want neither lightning nor any privacy enhancement.
Then why don’t they include segwit with BU ? I’m all for increasing the blocksize personally but not without a malleability fix.
You just believe everything you read? Did you stop for one second and think how they intend to block anything? It’s open source, even you can pick it up and change it to whatever you want. Not to mention it’s decentralized.
i’d be interested in Segwit support inside of BU since that means that with the recent 40% hash behind BU and the 30% hash behind segwit we would be close to actually getting something done.
I’ve heard it several times that they “don’t want to be forced” on to lightning. Nothing against lightning itself. I have no qualms with lightning, but I do somewhat doubt how much it will actually help scale the amount of users (rather than amount of transactions). I actually have higher hopes for sidechains than lightning network for scaling.
There is a separate argument that segwit is “ugly” code and flexible transactions is nicer, but there is only 1 dev behind flexible transactions and BU and flexible transactions are separate. I could personally go for either since I like the idea of flexible transactions and think if it got more developer input it might amount to something, and I think the fear against segwit being ugly soft fork code to be overblown.
With hashrate ? As they are doing now with Segwit ?
It’s not that they don’t want to be forced in lightning, they don’t want any functional (need malleability fix) lightning at all. Malleability fix will be a vaporware with Bitcoin Unlimited becasuse there is no will to fix it. Rather a will not to fix it!
If BU goes through, it forks and there will be two solutions. Most of people on the core side should signal segwit pretty easily and there you have it. No one blocked shit. Bitcoin is free and open source, yet people love to paint doom and war. Stop believing the Drama queens.
I don’t understand the concept of your question. That they don’t implement SegWit does not mean that they are solely here to block it. There is some room between …
I’m tired of this discussion. As if it was just malice or supid to not like everything in SegWit. In short:
SegWit does only fix malleability of SegWit-UTXO. If this is the condition for increasing the blocksize we probably will never do. Also malleabiliy and capacity are not really connected.
Designed as it is SegWit is simply not usefull for increasing blocksize for all transactions. As it increases the size of average tx and creates an spam-attack-vector two times bigger than the transaction capacity it is more of an obstacle.
Also, many people don’t think it is helpful to introduce more magic numbers and change the whole concept of blocksize to technically regulate an economic variable like UTXO.
There are plans with BU to implement SegWit as a hardfork, as this eliminates most downsides but keeps the advantages, when - if ever - the moment comes and BU introduces a hardfork. But this are just plans, and I guess the BU-devs would be pretty happy if Core SegWit architects cooperate for such a venture.
BU also cooperates with Classic, which plans to fix malleability with flexibleTransactions. I don’t think this will ever be introduced, but it is still an interesting concept that could help Bitcoin to greatly evolve.
After all it is a thing of priority, nothing more. You think malleability is more important than capacity, while BU thinks that capacity is the most urgent problem and that malleability can be fixed later.
a chinese guy writes anti-privacy articles? what a surprise 😛
of course they hate everything what will make Bitcoin more private and useful. ChinaBU is trying hard to hinder bitcoins growth.
Not a dev only user, I wanted to write something but Cbergmann beat me to it lol.
Many reasons, imho main ones, segwit includes a discount that is not much liked and another fixed limit (blockweight).
Also another reason is that BU is based on core 0.12 codebase, it is not so easy to merge this with the 0.13 codebase (I tried myself but gave up after an hour of fixing merge conflicts).
Core could also merge BU’s blocksize setting, maybe with a very sane default value EB1/AD999999, but probably not going to happen.
Well the BU threat is putting a damper on the bitcoin rally. This should be a signal that miners should stop taking matters into their own hands. Let the protocol evolve organically and recognize the tremendous support for segwit and get ready. Otherwise they will just run bitcoin into the ground im afraid.
Honestly, not signalling BU is better than signalling it, if you dont like SegWit. Because the market gets worried of hardfork and split. Its an irrational move to signal BU from a miner point of view.
I cant believe people are still pushing that narrative. I took a quick look at BU and i understood immediately that the blocksize will not be market based. Either it will never change, or it will change when a conglomerate of miners start mining larger blocks. Has nothing to do with a market based blocksize. Come on.
I’m pretty sure this means doom and war.
That’s it people, pack it up, this guy has all figured out. We can all believe you and just follow blindly.
This is outright and utterly false. Segwit is a stepping stone enabling many more upgrades that will have a cumulative benefit to scaling, both on chain and off chain. Almost no one wants the status quo, but that has always been a potential outcome if consensus isn’t found on the nature of changes. Thats always been the case, and denying that is lying to yourself.
So a part of the ecosystem doesn’t feel disenfranchised by no blocksize increase? if that is the case how do you explain the rapid rise in bitcoin unlimited hashrate? Oh right that is just one person signaling. /s
The argument of long lead time in a hardfork is not arbitrary, it is rooted in the need for security. A hardfork is an incompatible upgrade, you need to socially organize the upgrade outside of the system, and people screwing up fragments the system and can lose people money. As well, hard forking without consensus will guarantee people will refuse to upgrade and stay on the original network, which will cause unbelievable chaos for the ecosystem at large. Advocating this is reckless and stupid.
This is pretty much proven wrong by recent events. Look at how the ethereum hard fork happened in a matter of hours over contentious reasons and how its market cap is now larger than before. Ethereum has hard forked several times and this time it was actually contentious. If you don’t call that success…
Segwit is the compromise. Denying this is gaslighting. People making arguments along the line of yours have the same fatal flaw, you immaturely and in a very childlike manner abstract the market away as some amorphous thing that always agrees with your wants and needs. You are dead wrong. The market is composed of individuals, you, me, everyone out there participating in Bitcoin. And cumulatively it looks like the market isn’t bothered by the status quo now, supports Core by a vast majority, and isn’t buying these bullshit delusional attempts to rewrite and white wash history.
I’m simply reiterating what Steven Pair said in his recent article:
Markets and economics will eventually overwhelm any ideological stance community members hold.
Maybe he’s trying to educate people? Bitcoin Unlimited would kill Bitcoin decentralization by letting most powerful miners to centralize. No bitcoiner would ever want this. Yet many non-technical people are being fooled by money, emotions etc.
Bitcoin Unlimited would kill Bitcoin decentralization by letting most powerful miners to centralize
How do you see this unfolding?
Because to me, bigger blocks doesn’t centralize shit. Hard drives, Bandwith and processing power has increased magnitudes in the last few years and are very very cheap.
Correct me if I’m wrong but I believe a hard fork would be pretty bad for Bitcoin:
Security is reduced.
Investors might go insane.
Each network only has half the nodes.
Let me explain to you. This has been repeated here many times, but it people tend to still not know it. Yet they talk about BU as a possible solution to anything. BU-idiots oversimplify stuff that can’t be simplified.
Check out https://www.reddit.com/r/Bitcoin/comments/5g1x84/bitcoin_unlimited_bu_median_value_of_miner_eb/
Bitcoin Unlimited would allow miners to choose how big blocks they produce and support. According to them, these settings should be dynamic and vary between each individual miners. So those values are whatever individual miners want them to be. Consensus should somehow magically form. Magically, because it’s not known to a man. 🙂 Essentially, as explained in the link, the network could be split into two by making one specific-sized block.
On a fundamental level, miners would choose how big blocks they can make. Remember that mining blocks effectively requires you to download and validate others blocks as soon as possible. Better resourced miners could make blocks that are just fine for them, but would drop say 5% of the existing miners off as those 5% can’t keep up with the better resourced miners. More resourced miners would produce blocks that are too resource-heavy for smaller miners to handle. This would force those 5% out because for those 5% mining wouldn’t be worth it if they can’t get the blocks in time. Now repeat this scenario over and over again, always killing 5% of the existing miners. 95% turns into ~90% etc.. Soon only the most well-resourced miners exist and rule the blockchain.
Maybe you should invest your time in more than a quick look?
Either it will never change, or it will change when a conglomerate of miners start mining larger blocks
You nearly get it. There are clear incentives in Bitcoin itself that the blocksize will only change if a mining cartell is behind the change that has the support of the economic majority and a mojority of developers.
It’s like:
AntPool: Hey, we need more space. Who is with us?
OtherPools: Yes, we are.
Miners: Hey, exchanges, what do you think of 2 MB?
Exchanges: OK.
Miners+Exchanges: Hey, developers, do you think we can handle 2 MB?
Developers: Yes, maybe even 4.
Miners+Exchanges+Developers: Ok, so lets do this thing …
The current setup enables a small part of developers to block such an increase. B-U just eliminates this possibility and gives the ecosystem the option to attempt to do this upgrade by their own.
Every attempt of miners to enforce a new limit will fail. With Core, with Klassik, with Unl1mited. This doesn’t work with a thing with bitcoin. Never.
Question: If the blocksize limit is raised can it subsequently be lowered if need be?
If the blocksize limit is raised can it subsequently be lowered if need be?
Interesting questions. You seem to assume that the miners will say: “Hey, let’s make a new limit of 10 mb”. I don’t think this to be true.
Imho the blocksize will increase in the way that is least controversial: by baby-stepps which are needed to clear the mempool (except of the case of spam-attacks). So it will be more like: “Hey, let’s try 1.5 MB and see how long it lasts and what does it do to the network.”
A decrease of the Blockchain should not be needed this way. But if, you surely can softfork a rule that blocks after height X shall not be bigger than Y MB. With assumevalid you maybe don’t even need a softfork.
But this is just a guess of me.
Also, I had to shut down my full node a couple days ago as SSD storage is just wayyyy too expensive. Bandwidth on a Gigabit line is also not cheap.
Try buying a dedicated server and running a full node on it. It will not be cheap at all.
I can’t imagine the costs of running a full node with even bigger block sizes.
Fun fact: Jihan is an investor at 8BTC.com
The twats that think it’s one vs the other are dumb as a nail.
Thanks for clearing that up.
The saddest part is imagining anyone listening to ver.
Only the dominant coin matters. I want people to make informed choices. That if they support BU, they won’t get smartcontracts, lightning, enhanced privacy.
Perhaps you provide the content of these conversations, or at least a name of the person you were talking to?
Isn’t most of this true today already, just on a different scale? Why don’t we reduce the blocksize so more people can afford to secure the network?
of course they hate everything what will make Bitcoin more private and useful.
/tinfoil hat on
PBoC already did try hinder all the privacy enhancement through Bitmain. The reason the exchange’s withdrawal was extended was due to their failure to affect the fork.
/tinfoil hat off
No, the BU fork would have 2 nodes and 40 proxies
The core fork has hundreds of known physically located in private home full nodes. Plus thousands more on VPSes from many companies internationally.
My main concern would be if there was significant decrease in the node count attributed to an increase in blocksize there would be a corrective mechanism to reverse course and preserve Bitcoin’s resiliency against malicious actors.
Also, I suspect that no block size limit is, from a purely free market economics point of view, correct but politically unlikely at this point in time. People seem to view Bitcoin as working ‘good enough’. Perhaps down the road when a greater proportion of the global economy’s monetary foundation rests on Bitcoin it will be non-contentious to remove the limit. I think the BU guys may have a solution to a long term problem but the appetite for implementation now is too early.
Lol, it wont work that way.
Miners: Blocks are full, lets increase them. Whos with us?
Developers: Its not safe.
Exchanges: We are indifferent
Miners will just start making larger blocks anyway because why not include a few more transactions, you know for those extra fees? Some miners only find a single block per day. Why not stuff a huge chunk of mempool inside it for 0.5btc extra?
They said it many times in the other sub that the real benefit that they all want to see is Bitcoin with a new dev team. So if we just fire Core and Blockstream, give commit back to Gavin, they would be happy to adopt segwit and LN. Oh wait, BU dev team probably cannot maintain the code.
Good job the nodes will punish those blocks then isn’t it.
Bitcoin is built on every participant acting in a way they think is best for them, which leads to the natural evolution you are talking about. Even if you don’t like the way this evolution might take.
Well that’s not true - there’s 2 solutions on the table from the big block side already (at least). It’s just seen as a lower priority than the blocksize increase. Just like in the other side it was seen as a higher priority.
Yeah, I made the picture sharper than it is. Your version is also interesting:
Miners: Let’s hardfork to 2 MB.
Developers: It’s not safe.
Exchanges: We don’t care.
Miners: Hey, exchanges, let’s fork against the developers.
Exchanges: Wait, let’s ask the developers again.
It is exactly like it is for years, just with the option to try it out if you think you have a majority. The entities, which have the most weight during a fork - miners and exchanges - have at the same time most at stake. They will be very very careful to not do something wrong …
How do you know they will? Some nodes may reject it, others wont. Eventually the nodes that didnt accept it would have to, because a portion of the network already did and whoops 5-10 minutes later new block found, its built on top, no going back now.
this post was actually fairly neutrally written man.
I quite like the idea of SegWit combined with Lightning Network. With Lightning Network (if the malleability issue can be rectified), SegWit isn’t really necessary, but it will make the protocol more robust for later. Lightning Network is the way forward.
95% devs support SegWit.
Better question: why were you using an SSD to store the blockchain?
I personally run a full node on SSD storage (along with quite a few other services). You can do so pretty cheaply ($20) on digital ocean also on SSD storage if you like. Things certainly get a lot cheaper when not on SSD.
BU has 25% network support. they could hard fork, and theoretically at BEST, BTC loses 25% of its value. BTC-BU will trade at a 1:4 rate of BTC-Core.
Then financial markets must decide between BTC-BU and BTC-Core, and obviously the combined market cap of both would drop due to this uncertainty.
At this point, an ambiguity is created that I honestly imagine will be not as bad as people think, but still much worse than the ETC vs ETH mess (which has ironed out pretty well)
Yes understand: too complicated to track transactions to please the Big brother chinese government.
They’re not larger blocks if the majority of nodes accept them, they’re normal sized blocks.
I’m not sure you understand how AD works. It’s certainly not how you described
SSD storage is just wayyyy too expensive
You can easily get a 1TB HDD for less than $50. You’ll have plenty of space for a loooooooong time. Why the hell would you use an SSD to store the blockchain?!?!
Except it won’t be based just on opinions but on large scale tests.
There we go with the personal attacks again.
Yet, you haven’t said shit about how BU is supposed to block segwit. You just swallow everything that is said against BU without even questioning.
Laptop
PoW cost is the real limit, not bandwidth
Well kind of. It’s just drama bullshit against BU like they are talking about a war, yet none of them even explain how they are supposed to block segwit. I’m neutral, I just want affordable transactions and don’t care about solution. But I can see false accusations here because blocking segwit is technically impossible.
Dude, mining power doesn’t dictate value. Mining follows the market, not the other way around. Any sub-50% fork can effectively be ignored if the majority of users stayed on the original chain. The miners would then be forced to abandon the fork due to lack of value.
Only the dominant coin matters.
NO, just no! It’s not black vs white. It’s not them vs us. It’s not a war. It’s a democracy, people looking for affordable transactions and you can still use the other chain after they fork.
Nobody knows how many nodes would be on each side until it actually happens.
Segwit has no effect on that at all
None of these pools are actually running BU except Roger.
Is there a way to tell for sure? I suspect the same as well.
Why run an inefficient client that is 3500 commits behind core master when you can just modify the coinbase text.
ur right about that and i do apologize. i edited my comment to remove the attacks but guess you had already read them.
it does nothing but make u think im an asshole.
but nah, were on the same team for sure. we just want bitcoin to grow.
im just concerned about the value impact a HF could have from BU. if Core does a HF to bigger blocks (not unlimited), thats fine. i just dont want fragmented community
There has been many hard forks. In Bitcoin and in other currencies. It’s not good, but it’s not pretty bad.
Yes. For years the default Bitcoin blocksize was a few hundred kilobytes, below the 1 MB cap that was originally set to reduce the image of spam.
i just dont want fragmented community
I don’t want it either, but humans are humans and this is how democracy works. Yet some people are trying to turn this into a war because it can bring them profit.
Major players are fragmenting it, not the users. The users can chose to see this and tell them to shut up, or they can incentivise it by making even more posts attacking people and solutions.
That’s why I started responding to this post because most of it is smoke drama bullshit to blind and enrage people. When you start asking for technical facts to prove their claims, none come forward.
HF might happen and if so there might be two solutions and people can then chose their sides, or just keep on both. It’s never a us vs them situation, which some players in this sub like to make.
You are completely ignoring my points
Yes, definitely! A mechanism like emergent consensus is a big incentive for everybody to engage in and support tests and science around the capacity of Bitcoin.
If you look at scientific papers, this is a highly underrepresented topic. There are hundreds of papers about privacy, value and offchain-networks, but not a dozen of papers about the capacity to scale onchain … (and those are old and calculate with outdated bitcoin-technology)
Yet I see that in most threads. There’s always somebody saying hardforks must never happen, that decentralization means that grandma’s computer always will be able to run a full node, that all the transactions that won’t fit into the block just was spam anyway (because can’t afford the fees = spam), that paying for coffee wasn’t the point of Bitcoin (and ignoring all the important stuff that becomes impossible), etc…
I’m assuming you aren’t a developer just like me.
So we’re in a frustrating position.
What role do you think users (who arent devs, or market whales or industry players) have in contributing to this conversation, if any?
If I understand u correctly u prefer ppl only contribute if they have technical facts backing their opinion right? Keep shit civil
USB to SATA enclosure
Segwit fix malleability only for segwit transaction ! What’s your problem with that ? There is no other way to fix malleability in a backward compatible way. Are you trying to mislead people ? All that matters is that it’s fixed so we can have non maleable transactions.
3,4 Then get rid of the magic numbers! There are “plans”, a.k.a vaporware. Considering the risk involved in every HF, we have all interest in grouping them together. Moreoever, BU would have way more chance to reach consensus if was integrating Segwit. Not doing it now proves there is no desire to implement it.
Why do you even mention it if won’t be implemented and is not even past design stage
I think Malleability and capacity are both equally important especially because fixing malleability brings more capacity. Segwit itself brings a blocksize increase as well. So no, the real answer is that I want both more direct onchain capacity through a blocksize increase and fixing maleability but that BU pushers don’t want a malleability fix at all (and are so undermanned they can’t provide it anyway according to you) for reasons explained in my post.
Yes it will stay on the table so that people don’t realize until it’s too late (after a Bitcoin unlimited fork)
I’d bet you how many Bitcoins you want that they won’t package a malleability fix with their BU hardfork, when 1/ It would give them more chance at reaching consensus 2/ It makes sense to package hard fork to diminish split risks. Because guess what ? The mining pools don’t want to fix maleability! And Roger and Olivier might actually agree with them anyway.
A very good collection of the main arguments. OK, you exaggerated each one for best effect, but it’s still good.
Maybe “there’s always somebody” because they are right.
Ofc and he is directly paying Huang Shiliang to write his propaganda.
A mechanism like emergent consensus is a big incentive for everybody to engage in and support tests and science around the capacity of Bitcoin.
Are you high or something? What do you think has been going on the last few years?
It is just as likely that nobody will do any reasearch on blocksize and just accept whatever the miners feel like mining. Path of least resistance. I mean do you expect miners to care? Why would they? They will just include more and more transactions. They wont be doing research or any of that stuff. They will just fill blockspace more and more because they can. OR they wont do it at all, a case can be made for that as well i guess. But if you give the miners power and expect them to not abuse it you are going to have a bad time.
Or there’s always somebody because humans are bad at risk evaluation
I don’t think it would be polite to do so. But you can ask Roger and Olivier about it and see their answers.
Actually Segwit is necessary to have a useful Lightning network. Lightning devs explained that to me in person (and to anyone who want to listen).
Contribute if they have technical facts backing their opinion right? Keep shit civil
Yeah, users should keep with facts and question everything. Drama, politics and war talk is for soap operas and movies. Usually people who start with personal attacks are the same who want to divide people even more.
Everyone wants bitcoin to be worth more. Miners, users, devs, etc. Everyone has a common goal. Segwit and BU are not mutually exclusive. I’m ok with both solutions, but I do see a lot of attacks and posts that disappear in this sub towards BU.
Bitcoin started focused on science and math and should be kept that way. It’s what made it strong.
Yes it has by allowing future privacy enhancements worked on by core devs and making things like tumblebit more efficient. Did you read the links I provided were Huang Shiliang is arguing against Segwit specifically because It will improve privacy ? I’m not inventing it.
because blocking segwit is technically impossible.
how so? you only need 6-10% hashpower to block segwit. it’s completely free as well, since you get to keep the block reward.
That’s because they had overwhelming consensus. And I don’t remember Bitcoin forking on purpose. I think they forked it once by mistake though but the update got rolled back.
Either way, avoiding a hard fork is better better than not.
Some of us have been saying this for months. Welcome to the party! 😉
Segwit hasn’t been blocked by anything. Even with BU’s rise in hashing support to 40% as of today… it isn’t actually blocking Segwit. Segwit grew to about 23% support and never got higher than that in about four months of existence. It needs 95% as per its design. For a long time Segwit should have needed to climb to 75% support for BU to actually be blocking it since BU only had about 20%. Segwit simply doesn’t have network support. Today with BUs new hashing power support… Segwit would need to be at 55%ish support to have BU blocking it. Bitcoin was designed for 1 hash = 1 vote… and that is it… that is the consensus system. BU is playing by the fundamental rules of the system since the start. Segwit just hasn’t achieved the required network support.
Fortunately the BU devs and most of their supporters seem happy to have Segwit or a similar innovation applied to BU after the fork. So BU+Segwit+Lightning is their goal also.
I don’t think you understand how markets or money work. Money has huge network effects and there will be one dominant coin.
No it’s not otherwise they would do it at the same time.
Segwit has wide support in the community, much more than BU. Most bitcoin companies are segwit ready. There’s only a small group around Roger Ver and the mining pools blocking it.
The 1 hash = 1 vote consensus system is one thing, but there’s economic incentives for Miners to anticipate consensus otherwise they might end up mining a worthless coin. I’m just worried miners might misjudge the support for BU or the strengh of their position (influencing the market with a hashrate display of power).
In the end the markets will decide which fork is valuable if there is a fork. It won’t be necessary the one that starts with the most hashrate. Otherwise Bitcoin mining centralisation would be very worrying. That’s game theory that prevents this from being too big of a problem.
$20 a month is not cheap. I could get another dedicated server for $20 and that would be a lot more useful to me than a full node.
I’m just a high school student but I want to contribute to the network as well.
Well, if you don’t want to spend 20, don’t get it backed by SSD.
Initially I was thinking they were just using BU as leverage to force Core handsthen I realized it’s not that when I saw how serious they were in pushing BU and that every time I called for a malleability fix to be included or some form of middleground fork, it was met with “crickets”.
Btw feel free to retweet my tweet linking to this article if you want, I think most people are still not aware of this. It’s important if they try to push a fork that
1/ people know what they are getting in BU
2/they realize they have the true power through buying and selling and that hashrate will follow.
How can BU be blocking anything if a large share of the hashrate isn’t signalling for BU or Segwit?
What is the economic argument for scaling on-chain vs. scaling off-chain with something like LN?
But coinjoin already exists, so what’s the problem?
Bitcoin is far from being perfectly anonymous, there’s a lot of progress to be made. Chain analysis is a thing. Do you know that Chinese exchanges were tracking every transactions even before PBOC?
Another coin will simply not work. There are tons of altcoins out there but none have garnered the same success as Bitcoin. It has been proven again and again that it simply doesn’t work.
Also, that’s not how democracy works. Let’s take the US for example. If all the states just kept splitting off and becoming independent, we wouldn’t have the US anymore.
Because guess what? The people pushing BU are THE EXACT SAME ONE blocking Segwit!
You know what? Some states wouldn’t mind that, especially after the last election.
Yeah sure, but it would decrease the influence and power of the US.
That’s not what we are trying to do here. We are trying to increase the influence of Bitcoin.
I just don’t want to spend extra money on a full node. I want to use my existing infrastructure to help.
So you’re saying the people “pushing BU” aren’t signalling BU and are instead signalling for neither?
“buy things to store the blockchain” doesn’t really help your argument.
why do you ask?
What I meant: I searched for papers examining the capacity of the current system, explaining, what happens, when you increase blocksize, with different factors and scenarios.
There are really few papers, some presentations, but none with current technology, various growth scenarios and a full set of factors (CPU, memory, HD, bandwidth, UTXO, IBD and so on). The set of data to make any reliable estimation about the capacity of the system is extremely poor.
I would speculate closer to 99%. Remember that Garzik and Gavin both support SW. Who are you left with that actually writes code and does not just copypasta from core?
95% of all statistics is made up.
Why would anyone believe what you are saying?
It seems to me that scaling on-chain necessarily leads to more centralization thru increased resources needed to run a full node. Off-chain scaling with LN does not have this tradeoff. Of course it is not entirely clear how many nodes are needed to ensure that BTC is resilient to attack by nefarious actors who would like to see it destroyed. The question then becomes; does this potential centralization risk of on-chain scaling outweigh the benefits of off-chain scaling? I just don’t see significant benefits of on-chain scaling at this particular time in BTC’s evolution but can see a role for it in the future.
Compact blocks (xthin blocks) already solved the problem of “can’t get blocks fast enough” for very large blocks though, that’s an old argument.
The fact is: BU code base doesn’t include SegWit. Even though it is trivial to use the recent version and add BU patch on top of it, they deliberately made it hard to merge SegWit. So they deliberately removed SegWit.
Anybody who denies this is dumb as a nail.
I’m sick of BU apologists. “Let’s be friendly/play nice/compromise”. The thing is, BU is very divisive by itself, we need to understand what’s going on.
I think that the market would value the chain it can actually make transactions on, not the one that costs $1+ to use and MAYBE your tx will make it into a block in the next day.
But maybe that’s just me…
Man what a reasonable conversation that is happening here, it’s good to see.
Well the BU threat is putting a damper on the bitcoin rally.
Says who?
I take a different viewpoint - the unreliability and slowness of transactions right now due to limited blockspace is putting a damper on the bitcoin rally. In fact, I’ve even seen the market bump up in the past on pro-BU news being released. But you know, that’s probably nothing, right? 😉
And replay attacks were abound, tons of people lots money, the ecosystem was in chaos, Coinbase ended up losing money fucking up the chain split, BTC-E didn’t give any ETC because they pocketed it or fucked up too, 51% attacks were threatened and attempted, and at the time is caused a massive drop in price. You are a completely reckless and immature goon if you seriously think it is wise or safe to repeat that with Bitcoin, something people actually use for real economic activity.
All the more reason that we should hardfork ONCE to remove the blocksize limit entirely so that we won’t have to go through this again in the future.
Seriously, remove the blocksize limit and I will fully support and applaud other scaling solutions being built. I know many other BU-supporters feel the same way. I just don’t like the way that SW/LN is forced on us as the ONLY possible scaling solution. The market should decide which scaling solutions are best and to what capacity they should be used, not a dictatorship built from a corporation.
well the market will return to normal as soon as it realises that the miners are bluffing. But when the news broke about Antpool alot of people were questioning bitcoins future. But the market is starting to realise that the miners probably wont contentious hardfork and that 1mb blocks will probably be ok.
Segwit has wide support in the community, much more than BU.
Do you have proof of that? I’ve seen the opposite to be true. Segwit-ready doesn’t mean endorsement of segwit, and more and more people and businesses are recognizing the need for larger blocks today, not years down the road.
The 1 hash = 1 vote consensus system is one thing, but there’s economic incentives for Miners to anticipate consensus otherwise they might end up mining a worthless coin. I’m just worried miners might misjudge the support for BU or the strengh of their position (influencing the market with a hashrate display of power).
In the end the markets will decide which fork is valuable if there is a fork. It won’t be necessary the one that starts with the most hashrate. Otherwise Bitcoin mining centralisation would be very worrying. That’s game theory that prevents this from being too big of a problem.
Completely agree. It’ll be interesting to see how this ends up playing out, but the incentives will keep the miners from doing anything stupid.
All the more reason that we should hardfork ONCE to remove the blocksize limit entirely so that we won’t have to go through this again in the future.
Absolutely not. Especially not with BU, which comes with the guarantee of large reorgs whenever the blocksize is raised. Its possible to create 2 MB blocks that take 15 minutes to validate. If a miner has no blocksize restriction whatsoever, not only is there no predictability to the cost of running a node, without further optimizations of Bitcoin miners could create attack blocks would take hours to validate. Essentially stalling the entire chain.
There are so many complex facets of things intertwined with this issue, the paramount one being the security and integrity of the network at large, and naive or misinformed(or malicious) dismissal of this is unacceptable. It is not as simple as smaller blocks = bad, bigger blocks = good.
So a part of the ecosystem doesn’t feel disenfranchised by no blocksize increase?
Segwit is a blocksize increase. BU is entirely political, and a bluff. None of these pools are actually running BU except Roger.
This is pretty much proven wrong by recent events. Look at how the ethereum hard fork happened in a matter of hours over contentious reasons and how its market cap is now larger than before.
And replay attacks were abound, tons of people lost money, the ecosystem was in chaos, Coinbase ended up losing money fucking up the chain split, BTC-E didn’t give any ETC because they pocketed it or fucked up too, 51% attacks were threatened and attempted, and at the time is caused a massive drop in price. You are a completely reckless and immature goon if you seriously think it is wise or safe to repeat that with Bitcoin, something people actually use for real economic activity.
I’m simply reiterating what Steven Pair said in his recent article:
No you are not, all the market participants supporting Core are not supporting a refusal to ever increase the blocksize, they are refusing to support that at the moment for a much more intelligent upgrade ready right now. You are intentionally misrepresenting what other people are doing and supporting, completely misrepresenting the situation, and completely misrepresenting what the market is and what it is doing. You are spreading a mash of lies and misrepresentations, and using deceptive language to attempt to sound reasonable.
I’m questioning bitcoin’s future BECAUSE we’re stuck at 1MB blocks. And I know I’m not the only one.
Price dropped because of China PBOC news, not because of antpool news.
Which part you don’t believe ?
1) Large reorgs with blocksize increase won’t happen. A miner wouldn’t be stupid enough to mine a block larger than less than 50% of the mining community would support.
2) So what if someone can create a block that takes 15 minutes to validate? It doesn’t matter, no one will accept it, it’ll simply be orphaned away. If someone actually tries that “attack”, they’ll simply be wasting a block reward.
3) Again, no miners are going to create attack blocks that would take hours to validate because no other miners would accept such a block. You couldn’t stall the entire chain this way.
I welcome you to detail any other issues you have with eliminating the blocksize. All of them I’ve been presented with so far are eliminated with simple economic theory.
Why would anyone believe any of it?
Well I haven’t asked businesses about it, so you have more information than I do. Thanks for sharing what you know.
Compact blocks (xthin blocks) already solved the problem of “can’t get blocks fast enough” for very large blocks though, that’s an old argument.
Nope. Malicious (or profit-seeking) miners would make blocks that do not work with compact blocks. E.g. they make a block full of earlier-unseen transactions.
The business I have talked to are pro-segwit. Most people want both a blocksize increase and segwit. Few people support BU. Mainly a subset of holders not happy because some core devs are blocking a hf blocksize increase. I understand the reasons for blocking a HF (it’s “dangerous” so better package them and do just 1 or as few as necessary).
But personally I would support it. (2mb + segwit)
This is just false. Almost everyone in rbtc wants to see layer2 scaling, they just think a blocksize increase is more pressing. It’s about priorities.
it’s completely free as well, since you get to keep the block reward.
Really? Then why don’t you get 6-10% hashpower, if it’s so easy and free
The argument (from the second 8btc.com link) that LN will devalue Bitcoin because it will become a metaprotocol connecting blockchains - so you will primarily use LN rather than Bitcoin, and that LN will include many other coins, really threw me. It made sense to me and I hadn’t thought of it before.
Effectively this seems to remove scarcity, because if LN becomes the protocol people are using for the majority of transactions, and it includes an arbitrary number of coins from every blockchain which joins the network, then there is no scarcity and little inherent value to any of the coins.
What are the specific arguments against this being true?
The sheeple are not the one get to decide. They are just useful idiots. If it’s not packaged, there is no guarantee it will be done.
I think Satoshi not including the lifting of the limit in the code should have at least taught you that.
i think its pretty safe to say that bitcoins future wont be bright if we get stuck at 1mb blocks. But with segwit the blocks can be a little bigger and segwit is really easy to activate!
maybe I have. doesn’t help me to activate segwit though. blocking is easier than activating.
Of course they won’t package a Malleability fix with the blocksize fork. It’s more risky to do multiple changes at once. That comes later,one big step at a time.
risking a split each time ? No it’s much safe to package changes. Especially when as I said fixing malleability would rally much more pople to BU.
1) If we can’t trust the majority of miners to do what is in their own economic interest then we can’t trust bitcoin at all.
2) No, but that is something that is easily coded in. If validation takes longer than 15 seconds then it is deemed an attack block, so dump it. Miners could signal other miners what their acceptable validation time limits in the block coinbase, and a miner would not mine a block greater than the median of those signaled limits.
3) Again, that is something that is easily coded in. The chain cannot be stalled in this way.
We need more than a little bigger.
There’s nothing to stop Core from taking back power from BU by offering SegWit after the hard fork.
Larger blocks take longer to relay. If a miner produces a block which takes longer to relay than another block found at around the same time, they run the risk of having their block orphaned. This hits them in their back pocket, by quite a significant amount at today’s exchange rates.
Miners need to balance block relay times, with block sizes. The bigger the block, the more transactions they can include within it. More transactions, more users, more fees. However, there is a natural limitation which is network propagation times.
With this natural economic incentive in place, and traditional market forces, the network will naturally converge to the optimal blocksize that the network can support.
We don’t need to have this centrally controlled.
You are jumping to conclusions based on anecdotal evidence. You pointed to the conversations of a few to represent the motivations and thoughts of the many. Very poor scientific reasoning.
They are both risks, but different risks.
The risk of packaging is extra complexity leads to higher bug possibility. For our first real hard fork change, I’d definitely keep it simple. Block size only.
If a Malleability fix is as non contentious as you think then there will be no split when a 2nd hardfork happens.
In the last 24 hours, 34% of blocks did not signal for either Segwit or BU.
As you need 95% for Segwit, you could just as easily say that the 34% of miners who are not signalling are blocking Segwit.
For those interested, 33% signalled for BU and 22% signalled for Segwit.
Im pretty sure a malleability fixed is contentious to some extend (opposed by some miners). Its going very hard to have non contentious fork around these issues so better multiply the contentious forks and splits
I’ve created a question on Bitcoin.SE to collect this frequent question you’re answering here. Would you please consider posting your comment there?
→ Why do people expect Bitcoin Unlimited to cause miner centralization?
It’s plausible that multiple coins implementing an LN-alike payment channel mechanism could exist in parallel, and this could reduce friction for exchanging those coins (among other things). But any LN transaction is an on-chain transaction, just suspended (and thus not broadcast). LN cannot make any altcoin freely interconvertible with, or indistinguishable from, Bitcoin; LN transactions moving bitcoin will only ever move bitcoin, and if you only want to accept bitcoin via LN, that’s easy (and in fact the default). So I can’t see how this is actually a problem at all.
Larger blocks take longer to relay. If a miner produces a block which takes longer to relay than another block found at around the same time, they run the risk of having their block orphaned. This hits them in their back pocket, by quite a significant amount at today’s exchange rates.
Miners need to balance block relay times, with block sizes. The bigger the block, the more transactions they can include within it. More transactions, more users, more fees. However, there is a natural limitation which is network propagation times.
What happens when e.g. 99% of the network is OK with the larger block? The remaining 1% is not and their voice is not heard because for 99% it’s OK.
What happens when the 1% gets dropped, say 100 times? When it happens relatively slowly, it’s very hard to see happening and it centralizes mining and Bitcoin network.
With this natural economic incentive in place, and traditional market forces, the network will naturally converge to the optimal blocksize that the network can support.
This would be free market where miners can kill each other and network would blindly follow. 1% getting kicked out of network at time is not visible to the majority. Bitcoin is for 100%, so this kind of free market enabling people to be kicked out of network should not be supported by any Bitcoiners.
We don’t need to have this centrally controlled.
We need consensus and not control by miners. Nothing is centrally controlled. If you want to update the hard protocol rules, you need real consensus. Consensus is very hard to measure. Current consensus is 1 MB maximum block sizes.
BU exists to scal bitcoin because Core hasn’t proposed any workable solutions to do that. Bitcoin is not some smart contract, two layer block chain controlled by blockstream (hello centralization). That’s what the market (i.e. Hash power) is telling you right now and you need to listen
There are other ways to solve transaction malleability besides SegWit that are less complex and more maintainable. But the more pressing issue is the block size, and that is what BU focuses on fixing right now.
And what prevents that if SegWit were the preferred option?
If you’re talking about flexible transactions, it’s much more complex and not backward-compatible.
Sure. Only the maximum blocksize is raised, so you can always create smaller blocks.
If the maximum needs to be lowered it would be a softfork (just like the introduction of the 1mb limit).
Its not an on-chain transaction, its a potential on-chain transaction, which is not quite the same, although I see your point.
But to play devil’s advocate let’s say the majority of shops, services etc accept LN payments, and LN supports multiple blockchains. They are not actually accepting Bitcoin specifically. So if non-lightning transactions become uncommon as people are using LN instead then Bitcoin loses the value from being money - because you can use any coin not just one of 21 million for the majority of things Bitcoin is used for.
If enough people leave all their coin in LN channels (and considering how many people leave their coins on exchanges I’d say there’s a fair chance a lot would) then Bitcoin becomes like the gold in banks which is never redeemed, only rather than issuing more notes than they have gold they just issue notes for silver, copper and tins of mackerell, which all work just as well - they can effectively allow the creation of notes for anything.
I would worry that you then end up in a situation where only notes which are ‘backed’ by something have value because Bitcoin, for example, can no longer claim to have value for its use as a payment system with a network of users.
The impression that a lot of people have is that Core and Unlimited will split as two separate coins competing over hashing power and market value, the way Ethereum and ETC split. If that were to happen, it probably wouldn’t be the end of the world. Unfortunately, that’s not what’s going on. If it were, BU could split today, anytime they wanted, and have their own coin.
What’s going on is that Unlimited and Core are competing over the same coin. The reason BU won’t activate without at least 75% miner support is because it can’t survive otherwise and would likely destroy Bitcoin. Why? Because without overwhelming support, there is always the possibility that the network reverts back to Core due to Core receiving more hashing power, and if it were to do that, all the bitcoin transactions between the time BU was activated and the time the network reverted back to Core would be considered invalid due to invalid blocks caused by excessive size. (Remember the recent incident where bitcoin.com mined a block that was too large and it was rejected by the network?) This would be a catastrophe for Bitcoin, people’s bitcoin would simply vanish! The exchanges would go insolvent. The reputation of Bitcoin would be ruined forever. Bitcoin would die.
And BU simply doesn’t have enough support to guarantee overwhelming consensus, and likely never will. Which means that if BU were to activate prematurely, with only a slight hashing majority but always the possibility of Core regaining control, no one would accept bitcoin transactions anymore as long as BU was in control, because as long as this was true, no bitcoin transaction would be guaranteed and might eventually just vanish.
The only safe way to hard fork, the only way that wouldn’t risk destroying Bitcoin entirely, is for BU to do exactly what ETC did, which is to split off as an altcoin. But they don’t want to do that, they want control of BTC.
So I think many people are confused about this, and they look at the ETH/ETC split and say, “well, that wasn’t so bad.” But the problem is that BU doesn’t want to do a clean split like that, they want control of Bitcoin, but it is impossible for them to do so as long as Core is still in play, because as long as it is, there not only exists the possibility that BU could destroy Bitcoin by being out of consensus with Core, but this possibility will make people stop using bitcoin altogether.
If the limit is removed I wonder what the chances are that miners create smaller blocks on average than currently.
Can’t we just go back to the roller coaster gifs?
There’s always somebody saying hardforks must never happen
What you mostly see is people saying (a) hardforks are not a rational choice when a softfork is available and (b) hardforks should never happen *without overwhelming consensus (obvious case being an out-and-out bug).
I think both are entirely sensible.
Mostly, yes. But I’ve seen at least a dozen people over the years arguing that it MUST NOT happen, ever.
Many companies are afraid of voicing support for anything but Core because they will get treated like Coinbase (when they merely indicated that they were going to test big blocks).
No one wants to come out and say it. That doesn’t mean they don’t support it.
One could quite easily say “Those pushing SegWit are blocking BU”
You haven’t made your point.
There are different degrees of positions …
One group only wants to see hard forks for bugs or protocol failures = perhaps a ~1k btc users of 10+ million
Second group never wants to see a HF ever and still runs patched 0.5.4 nodes - http://thebitcoin.foundation ~ users
Core devs are conservative when it comes to HF’s but most of them are ok with them occurring if needed or consensus is found.
Both positions are extremely rare but the interesting aspect is these conservative groups tend to be early adopters who have many bitcoins and treat bitcoin as digital gold so their opinion matters in the sense of the economic vote.
No malleability fix can solve it for existing transactions without also invalidating existing scripts and time locked transactions. The safest way to do that is with a new transaction version, so old ones work the old way.
Segwit doesn’t increase capacity recklessly. Malicious miner can still attack the network of course, but not by making very large blocks.
This is why luke-jr’s +17%/year HF proposal actually proposes a blocksize decrease if activated before April 2017. It’s a valid argument about decentralization, but not very tasty to most people since blocks are already full.
ahh, well i see this as good and bad news.
bad news is that fork is more contentious.
good news is there is no way BU is able to do it.
Well, Bitcoin and all the various alts would technically each have their own LN. If a business wants to accept “bitcoin via LN”, they need to set up a Bitcoin payment channel; if they want to accept “litecoin via LN”, they need a Litecoin payment channel (and wallet, possibly node, &c.); and so forth for every different alt.
These would not be the same network; they’d be separate networks, with overlapping participants. Exchanging bitcoin for any alt, or vice versa, is always a deliberate operation; this is no conceptually different from the current status quo, where you can send someone bitcoin on-blockchain in exchange for an altcoin on-blockchain. So if some other alt has permanent inflation, it still wouldn’t erode the purchasing power of bitcoin on LN, because there would be a market between that alt and bitcoin similar to those nowadays, in which the permanently-inflating alt would presumably permanently lose value. Businesses that accepted hypothetical-inflation-alt would have to do so at an exchange rate.
There are other second-layer technologies, e.g. Interledger, which claim to have native multiple-currency operation. But I don’t understand these well enough to explain them, and they don’t seem to be anyone’s preferred path to scaling anyway.
Bitpay says they get business users making medium-large transactions (tons of coffee, not cups) and that the fee is no problem to these users. I’m quite happy with bitcoin becoming used for international clearing, rather than “liberty!!! on-chain!!!” and stuff like instant 0.1BTC gambling payouts, and happily await level 2 solutions for everything else
coinjoin isn’t used by a big enough portion of users, and it’s reportedly not working very well. Increased usage of coinjoin would just bloat the blockchain. But we need an interplanetary solution before space colonization can start.
Yep, and that’s why we have parallel validation.
Indeed. The possibility for individuals to do the validation is one of the key parts in decentralization of Bitcoin.
merge BU’s blocksize setting
If it’s easy and not much work to integrate that into newest core, then create a pull request for it already 🙂
There’s likely more free help and resources for automatic testing to get like that, than isolating away from core. Knots (luke-jr) might have some advice on this. And BU could maybe get help fixing the EB/AD consensus fail for free from somebody that considers it a nice intellectual challenge.
bitcoin for any alt, or vice versa, is always a deliberate operation
Whilst that is true, who is the deliberator?
I guess here is the core of my concern and I’m actually not sure if it is valid: It seems to me that LN raises the possibility of decentralized exchanges between all LN supported coins, which can be manipulated via smart contracts and therefore automated. So someone requesting money in one currency may receive it from somebody sending in another currency.
In addition, many merchants and service providers are (at least at the moment) likely to denominate their prices in fiat rather than BTC, and receive the equivalent value in any coin. So that doesn’t look like separate networks with overlapping participants to me. It looks like one network with multiple points of entry - none of which is preferable to a fiat linked coin.
This is definitely something that could happen. But any LN node which will receive payments in one coin and forward them in another is just equivalent to a market maker in a market between those two coins. These can and should be automated, and the algorithm they’ll run is just one which sets the exchange rate based on supply and demand. So any coin which inflates itself will lose value in those exchange rates; if a fiat-linked coin is accepted, it’ll presumably fluctuate along with the price of fiat relative to crypto.
Merchants, under such a scheme, could accept whatever coins they wanted. The decision of which coin to accept would just be equivalent to holding your assets in that coin, which increases the price relative to the amount of your assets. Anyone who doesn’t trust a given coin would refuse to accept it; payers could pay with that coin, which would be equivalent to selling the untrusted coin in exchange for the trusted coin. The merchant who only accepted a given coin is at no risk from fluctuations in any other coin.
Meanwhile, a node that forwards inter-coin payments at a rate different from the market-determined exchange rate is just acting irrationally, and they’ll get all their liquidity arbitraged away.
So any coin which inflates itself will lose value in those exchange rates; if a fiat-linked coin is accepted, it’ll presumably fluctuate along with the price of fiat relative to crypto.
Hopefully. But you lose nothing in the exchange rates if the recipient denominates their prices in fiat and you pay in a fiat backed coin; but you do lose in the exchange rate if the recipient actually needs fiat and you pay with Bitcoin.
This argument has been debunked so many times. Can’t they wrap their head around the concept of the network effect?
less complex and more maintainable.
Than SegWit? Are you crazy? SegWit is not only a relatively straightforward solution, it’s elegant.
why would you use a SATA these days for anything?
I think that BU will be valued more if they change their name to BJ Unlimited (stands for Bitcoin Judas Unlimited) But the market will flock to the free BJs and the BJU coin will sore in value.
BU has 25% network support.
BU has 25% of miner support. It has almost zero percent of network support.
Which is why there is lightning.
1 MB cap that was originally set to reduce the image of spam.
Prove it.
No it doesnt overlap the same. Especially because main backing from BU and against segwit comes from miners.
Well i talked to companies privately so no it’s not the pb
Most mining pools are acting in concert.
Bitcoin suffers from a governance flaw. It’s too easy to block upgrades the stakeholders want. This means another coin will take the lead eventually. The only option is to diversify.
It’s good that bitcoin is conservative. Money should be. If stakeholders REALLY wanted the same thing it would be done. Problem is there’s a few different groups pushing for different things for now.
We don’t know what the stakeholders want because there is no system to find out. It might be there is consensus but competitor coins are stalling bitcoin to capture marketshare.
I don’t subscribe to this theory becasue I see no basis in it.
Seems like most I can find right now is this;
https://www.reddit.com/r/Bitcoin/comments/449f57/the_circuit_breaker_and_satoshi_or_why_the_one/
He only really talked about it as something that could changed if needed, and in the context of preventing attacks. At that time the limit was 32 MB, with blocks at around 500KB in size.
Why would he say it can be changed so easily if he meant it to be permanent or thought it was absolutely necessary? He talked about SPV mode and how full nodes eventually would run in server halls, so from context it seems clear he did it as a temporary protection against spam by miners.
1) Thats why you run a node.
2) You’ve just imposed arbitrary limits on what people can and can’t spend dunce.
3) Easilly? Code it up buddy.
I’m done engaging with you. You cannot actually present a coherent statement or argument to stand on its own, you simply re-actively reply in a contrarian manner to me with no substance.
If these things are so easy, go do them yourself instead of whining in such an entitled manner on the internet that someone not getting paid should just do what you want because you want it.
Help solve problems instead of bitching about them.
1) Miners have demonstrated a long series of stupid decisions recently. Bad assumption.
2) So what? Do you think there is some Chinese Foxconn worker sitting at a computer manually deciding what blocks to accept or not? Thats not how this works.
3) Bad assumption. You very well can stall the chain this way.
What you just said is by no means whatsoever economic theory. Its you just making completely baseless statements, and presenting them as fact. They are founded in no reason whatsoever, make blind assumptions, and run completely contrary to how the software and network works. You are clueless.
Seems like most I can find right now is this;
So you just making shit up, aren’t you? You said :
1 MB cap that was originally set to reduce the image of spam.
So prove it. You can’t, can you?
How about you try to prove it was meant to be permanent? Or whatever else you think it is for.
No, you made a statement :
1 MB cap that was originally set to reduce the image of spam.
Prove it, or you’re just making shit up. You are, aren’t you?
Repeat after me :
I have no idea why the 1MB cap was introduced by Satoshi.
That statement better reflects what you actually know, and what you can actually prove, doesn’t it?
Except lightning doesn’t exist…
Yes it does.
Don’t you ever tire of being wrong?
Because it’s cheap
I know there are prototypes and alphas out there, I mean it doesn’t exist as something we can actually use right now.
Please sign in first