Poll Finds 70% of Bitcoin Users Prefer Segwit for Scaling(coinjournal.net) |
Poll Finds 70% of Bitcoin Users Prefer Segwit for Scaling(coinjournal.net) |
Though I'd like to try and provide a summary for people who don't know what's going on.
Bitcoin desperately needs to scale to handle higher transaction demand. It can take multiple hours for transactions to be confirmed, and due to the demand, fees are high ($2-3 per transaction for fast confirmation.) A technical approach (SegWit) was proposed and implemented by the Bitcoin developers in order to improve transaction capacity a couple years ago. This also brings along a number of innovations which can help implement what is termed 'Layer 2' solutions to happen inside Bitcoin, instead of using an altcoin like Litecoin or Ethereum. Think of this enabling a Visa-like entity to provide instant settlement services denominated in BTC, and secured by the Bitcoin blockchain.
However, this upgrade is very contentious with the miners who are entrusted with securing the blockchain against attackers. Bitcoin mining now is a professional enterprise where major players need to be located in places with cheap ASIC production and cheap electricity to have any hope of making a profit. In some cases, they see these new layer 2 solutions as reducing their transaction fees, which are becoming a large percentage of their profits. Instead, they have been proposing just a larger base block size limit, which will solve the transaction backlog and send the fees to the miners, but then not make progress moving forward with what users and developers want, which is increased innovation. In addition, the larger block size is seen as increasing centralization of mining since only larger providers will be able to deal with the extra data rates (since there is an advantage to having quick transfers of newly created blocks when mining.)
Since Bitcoin is decentralized, anyone can run any software. There are mechanisms in place for miners to 'vote' with their hashpower/mining rate to agree on software changes, then after enough support, the mechanism activates. This is the easiest way to make sure global upgrades happen smoothly and in a coordinated fashion. However, users are becoming desperate and are looking for other ways to make these innovations happen without needing support from the miners.
Unfortunately the Internet and echo chambers have destroyed the hope of meaningful discussion about this topic. Please, HN commenters, do not bring in Reddit-level comments here... Provide sources and a balanced view, and please be respectful.
However, one critical issue is omitted in the above: The suggested approach (SegWit), albeit probably more by coincidence than by intention, also corrects a shortcoming of the existing transaction protocol that lets you reuse parts of the computation when trying to find a suitable block, leading to a net gain of ~20% hash power over other miners if you exploit it.
Critically, said approach ("AsicBoost") is patented! Therefore, not all miners can use the approach, leading to a centralization of hash power in contrast to the original intention of Bitcoin.
Miners who currently benefit from this oversight in the protocol have no short-term interest in removing it via a change of protocol headers that SegWit entails, in addition to the other reason you mention, i.e., higher transaction fees if there is a backlog due to small blacks.
Of course I support any efforts to get rid of shortcuts in the POW function ASAP, and by any means necessary. The whole reason behind a POW function is to ensure a fair competition.
Satoshi Nakamoto himself believed that bitcoin block size would have to increase rapidly, and that on-chain scaling is good enough for nearly all transactions. He wrote that blocks could scale up from the current 1MB to over 700MB as bandwidth and disks got cheaper while Bitcoin got more popular. [1] He foresaw the day when a "full node" would not be someone's laptop, but rather something the size of a data center. There would only be maybe a hundred full nodes in the world, run by universities or major bitcoin stakeholders, and everyone else would use various kinds of thin clients to interact with the network.
Gavin Andresen, the developer that Satoshi hand-picked to lead Bitcoin after his departure, shared this vision. He believed that on-chain scaling via bigger blocks is technically simpler (and simple is good!), and the resulting loss in full node count is nothing to worry about.
I'd also like to point out that one of the most modern coins, Ethereum, actually has both. They let miners vote on block size (big blocks), and they support off-chain scaling (like SegWit). They have both, and Satoshi thought Bitcoin should someday have both.
So let's get both! =)
[1] http://satoshinakamoto.me/2008/11/02/re-bitcoin-p2p-e-cash-p...
I agree that bigger blocks are needed to ease even non-witness/standard transaction backlogs, and to ease up on wallet and other providers that want a faster network in-place without needing updates that work in concert SegWit like the Lightning Network. The 1 MB was originally an anti-spam limit. In a dream world, we would have a dynamic anti-spam block size. But I don't see consensus happening any time soon. :(
The idea that Bitcoin's high transaction demand makes transactions slow is a serious misconception.
Bitcoin transaction fees are a supply and demand system, and each transaction you send is a bid for the limited supply of blockchain space. If you place a sufficiently high bid, you'll get into the next block almost every time. The price of that bid is getting a lot higher than it used to be, but that's a case of Bitcoin becoming expensive to use, not unreliable.
OTOH, let's suppose you're bid is lower. I think where part of the misconception comes from is that both supply and demand are variable, the latter because of fluctuations in how many transactions people send, and the former because blocks are found randomly in a Poisson process. So if you want, you might get lucky and pay a lower price. But that's the same with any commodity with a free market - you wouldn't complain about gas becoming unreliable because you have to wait a few days for the price to dip, you'd still say it was becoming expensive. And like gas, as you lower your bid you very quickly find yourself waiting forever.
The other root cause of this misconception, is many of the people pushing the "Bitcoin is becoming unreliable" meme have business interests where they want Bitcoin to be a payment system first and foremost. That demands low fees. Unfortunately for them, Bitcoin's underlying technology is fundamentally ill-suited to that need and always will be - you need layers on top of Bitcoin such as Lightning to suit those needs.
Even worse, for some of those people they want Bitcoin to be a transparent payment system, with every payment easily visible, e.g. due to promises made to government regulators, or because their very business model is to deanonymise Bitcoin transactions. That subset's needs are very incompatible with scaling Bitcoin, as every scalability improvement will inevitably improve privacy against some attackers when compared to the current "everything public" Bitcoin protocol. For example, even Paypal offers improved privacy for myself in Canada, against the government of Iran snooping on my payments. Equally, a common complaint in the fintech world is the poor privacy properties of blockchains...
But I definitely agree that it is working 'as designed' at the moment, and actually handling the situation of high demand reasonably well. And I can't wait for new technologies to become available on top of BTC for innovation at a global scale.
BTW thanks for all the work you have done with Bitcoin, I hope you have been able to stay above the fray on this issue for the most part.
Over the last 6 months, all that has been replaced with suspicion and hate. Both sides now have their own heavily-moderated forums that mostly don't tolerate dissent from the party line. There are elaborate theories getting posted about why those who disagree with you are really all part of some huge conspiracy or astroturfing campaign. Naturally, there is not a shred of evidence for any of this.
I miss the old days.
Last weekend I found out I had a handful of old mBTCs lying around, and while transferring them I made the mistake of only paying a $.35 fee instead of the currently recommended $2. The transaction [1] still hasn't gone through and I expect it will stay in limbo for at least several weeks.
Considering what Bitcoin was supposed to be, this a pretty big step back and the situation doesn't compare very favorably to the big payment networks, to say the least. No matter which side of the User Activated Soft Fork divide you are on, I think everyone already agrees that something needs to be done real soon. If the SegWit switch doesn't go as planned and the community fails to reach consensus once again, things are going to look pretty bleak.
[0] https://blockchain.info/charts/mempool-size?timespan=60days [1] https://blockchain.info/tx/e2456f994910fb40d938d5ebe39b40df6...
Activating Segwit as a soft fork would prevent the ability to use AsicBoost and therefore a lot of people suspect that most of this debate is actually just a delay tactic by Bitmain to maintain their mining dominance with covert AsicBoost. They likely have no intention to support Bitcoin Unlimited (they don't seem to be running the software).
Segwit recently activated on Litecoin only after the threat of a User Activated Soft Fork (much easier to execute on Litecoin). The Segwit activation threshold on Litecoin was 75% of the hashpower (compared to 95% on Bitcoin). At one point Segwit had achieved 75% of the hashpower until suddenly 3 mining pools increased their hashpower by about 300% in 1 day to block Segwit (the same day that Jihan Wu, CEO of Bitmain declared that the Litecoin miners to be shipped needed "additional firmware testing"). These pools were Antpool, LTC.top, and LTC1BTC [4], all of which controlled a much smaller fraction pre-segwit. For this reason, it's likely that Jihan Wu controls both Antpool and BTC.top and therefore about 30% of the hashpower on Bitcoin.
[1] https://cve.mitre.org/cgi-bin/cvename.cgi?name=+CVE-2017-923... [2] https://bitcoinmagazine.com/articles/bitmain-may-be-infringi... [3] https://blog.bitmain.com/en/tag/asicboost/ [4] https://www.litecoinpool.org/pools
It's quite possible that they're funding both sides of the debate to make sure none of them generates a super-majority.
It's also quite possible that Bitcoin developers who were once heavily invested in Bitcoin have already switched to Ethereum, and now are actively sabotaging it.
In short: extraordinary claims require extraordinary evidence.
If you want to claim an altcoin conspiracy to destroy Bitcoin, prove it. Otherwise take your baseless gossip elsewhere, please.
The whole value proposition of mining is that rewards would shift over time from newly-issued coins to transaction fees. Any solution which offers an alternative for transaction settlement and reduces the demand for normal on-chain transactions and the fees from them is, then, an attack on the basic value proposition of Bitcoin mining.
Personally, I think that mining only should not be a for-profit enterprise. Only parties that have goals aligned with the long-term economic future of Bitcoin should be involved in mining, like major exchanges, merchant providers, etc. The value proposition was only ever to secure the network and get some financial help with doing so.
Otherwise, you have to have constant innovation and investment otherwise you are pointlessly fighting the difficulty level adjustment.
And 40% for Bitcoin Unlimited (Emergent Consensus)
Because each side thinks they are acting selflessly in the best interests of Bitcoin with sound technical reasoning, while the other side has sinister ulterior motives. Bitcoin scaling debates (like most political debates) are riddled with assumptions of bad faith.
What really makes it complicated is that, to some extent, both sides are right: there really are people on each side who stand to profit greatly from their proposal winning and thus have their finger on the scale. Of course, each side only notices these people when they belong to the other camp.
They will often assert strongly that certain claims are undeniably true from a technical perspective, without stating the large number of assumptions inherent to the conceptual framework they use to make such claims.
Many would argue that their assumptions are not all obviously reasonable, yet rarely does the discussion even manage to broach these underlying aspects.
So far Bitcoin has proven to me over and over again that it just doesn't work.
What the fuck is "Emergent Consensus" or "Segregated Witness" or any of this? From the perspective of a regular person, I just don't care. I don't see why anyone else should, either. Why is one Bitcoin worth more than thousands of units of the strongest currencies in the world?
One bitcoin is worth more than the current world reserve currency because the market demands it. Even if it is mostly speculators with cheap money that flows from all of the central banks printing it and offering negative interest rates.
Alternative options such as EC (which uses Xthin blocks) have already proven to have multiple vulnerabilities and there are many instances of large scale crashing of nodes. Of course the miners that are signaling for this option aren't actually using the software (it's not at all stable, they don't actually want EC and are just using it to stall for higher fees) so they were not affected by the crash, if they were 40% of the hashpower would have gone offline which is a pretty big deal.
https://www.reddit.com/r/btc/comments/6eul2q/only_61_days_to...
From what I read segwit sounds like a reasonable solution to what seems like a very real practical issue with bitcoin today. So far all the people I've seen arguing against this change seem more interested in fueling conspiracy theories than actually explaining why it would be a bad move (and I get this vibe from your post as well).
So what's your position exactly? That the current issues with transfer fees and the long term usability of bitcoin are overstated? That they're real problems but it's too early for this push? That the solution is not the right one? But if not, which one?
* Realistically, can not be activated (95% support level required, way more than achievable). Means we'll stuck with current 1MB forever waiting for segwit
* Upon activation it does not add any benefits. Real increase in throughput will be achieved later, then all major players updated their client code, started crafting segwit transactions and moved coins to new segwit addresses. Means it delays throughput increase even more
* After full adoption, throughput will be increased from 1MB/10min up to ~1.7-2MB/10min while worst-case scenario (useless, specially crafted transactions) is 4MB - miners should be prepared for it. Means "2MB forever" and "hard to increase block size because miners should be prepared for 4x load"
There are other points as well, i'll stop here. Simple block size increase don't have such cons
This is not an argument against segwit, but against the method of it's activation. The rationale behind this high activation treshold is well summarized by greg maxwell [1]
> Upon activation it does not add any benefits. Real increase in throughput will be achieved later, then all major players updated their client code, started crafting segwit transactions and moved coins to new segwit addresses. Means it delays throughput increase even more
If only merchants or services who need to do lots transactions enable segwit, everyone benefits from the compressed output as there is more space on the chain for non-segwit wallets even if they don't upgrade or are aware of what segwit is
> After full adoption, throughput will be increased from 1MB/10min up to ~1.7-2MB/10min while worst-case scenario (useless, specially crafted transactions) is 4MB - miners should be prepared for it. Means "2MB forever" and "hard to increase block size because miners should be prepared for 4x load"
Miners should be prepared to give the service users want and pay for with fees. "Specially crafted transactions" included.
[1] https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2017...
* It fixes transaction malleability and quadratic hashing problem. Transaction malleability is definitely a benefit upon activation. Oh... and it's also a block size increase so there is that benefit. Oh... and lightning network is basically ready. Oh... and rootstock. If the network NEEDS NEEDS NEEDS more capacity so badly, then throughput increases should happen very very quickly, because why would people wait to upgrade if the need is so incredibly dire.
* Don't even know what you're arguing here but "2MB forever" is not grounded in any sort of reality except one that people have made up.
Oh definitely not that - it's a huge problem! Places that accepted bitcoin no longer do, and bitcoin's market share is plummeting - there's no reason it every should have happened. But instead of a simple fix to the problem (upping a temporary limit put in place in 2010 - that's eight year old tech), now Blockstream is trying to push a massively complex revamping. It won't even help with the scaling problem - it only discounts Segwit enabled transactions (since Litecoin enabled Segwit, almost nothing has used it). It's primary purpose to enable second layer solutions Blockstream can consult on. Even if something like Segwit was a solution, FlexTrans is a much better way to go. Even if Segwit activated tomorrow, there'd be little-to-no backlog relieft, and Bitcoin would be forever saddled with inescapable tech debt.
to me it sounds like Segwit is trying to #1 fix a problem and #2 add new features.
Wouldn't it be a lot faster and less risky to first fix the problem (#1) and wait with implementing new features (#2) until all the problems are solved?
if (blocknumber > 500000) maxblocksize = 8000000;
A more longer term problem is the governance problems it entails. Large economic interest will want to adjust that value to their liking, so it should preferrably be a little more non-arbitrary.
So a more flexible approach is necessary, preferrably one that is backwards compatible for old nodes. Extension blocks, and a special form of extension blocks called segregated witness, can do exactly that. They have other problems, but most are pretty well understood by now, and most developers see segregated witness blocks as a way to more than double block size whilst minimizing risks.
(It also brought along the possibility to fix other long standing problems with the transaction format, so in the end several other things such as script versioning, uxto defragmentation and non-malleability was stuffed in there.)
I also see that some people worry that using bigger and bigger block sizes could end up "concentrating the mining power in the hands of a few miners" but since that already seems to be the case I'm not sure it's a very good counter argument.
It's pretty obvious to me that linearly scaling solutions like simply increasing the block size are not sufficient to achieve global mainstream adoption (at least without further centralizing mining), much less the sort of machine-to-machine / micropayment revolution that a scalable cryptocurrency could unlock.
Yes - the former now (or last year!) to resolve the massive backlog as new adopters come in. We later need something to enable greater scale than blocksize alone, but not yet; other cryptos like Ethererum are able to handle a much higher transaction throughput. When the time comes when something like SegWit IS needed, FlexTrans is a much better solution.
But we HAD years to do this. Why didn't we do it?
Why did it take so long to develop? I think it's a combination. Many people wanted to get it right, a bad solution could be worse than no solution. There was also concern about miners blocking it, as many had been very skeptical about lowered fees.
Why does it take so long to active? Yeah, this is the tough one. It was deployed the same way as other soft forks. But this turned out to be quite contentious as it played right into an ongoing governance conflict. Worst case, the current deployment has to time out later this year and then it can be safely re-done.
I would rather the core developers focus on the current capacity problems and deal with use cases like instant transactions at another time. Make bitcoin work as it's supposed to and let the community decide on adding features later - don't try to push them both together.