SegWit

There’s an interesting thing going on right now in Bitcoin land. The core developers have released a new version of the core Bitcoin software which includes a number of updates and the “segregated witness” approach to scaling the Bitcoin technology. This release won’t be “confirmed” until 95% of nodes adopt it. That 95% is a choice that the core developers made.

Right now “SegWit” is at roughly 1/3 of nodes (based on this tweet which I have no idea if its valid). The Internet is for fake news after all 🙂

I am curious what all of you Bitcoin folks out there in the AVC community think of SegWit’s prospects and what it means for Bitcoin if it is not confirmed.

#blockchain

Comments (Archived):

  1. JimHirshfield

    SegWhat?

      1. JimHirshfield

        Rare sighting

        1. Lawrence Brass

          The only thing missing is the chainsaw… I miss JLM.

          1. JimHirshfield

            Hahahaha

          2. BU HardF$ck Clown

            https://www.reddit.com/r/Bi…Bitcoin Unlimited (BU) – The developers have realized an attacker can “disrupt the BU chain.” Fixing this has “an extremely detrimental effect on chain convergence”, according to BU’s “chief developer”. Rather than prudently advising people not to run BU, complexity is increased furtherThere is the clown’s now ? 🙂

  2. William Mougayar

    I would add that this is probably even more serious than previous activations fights.

  3. Jake Brukhman

    SegWit itself and the way SegWit is activated in the network is a reflection of high conservatism in the Bitcoin community in adopting changes. SegWit is meant to be a (relatively minor) mitigation to the network throughput problem, so its failure to adopt just means that “Bitcoin scalability” is delayed some more and that about a year was spent on instrumenting a change that will not be accepted.

    1. jason wright

      the early adopters in the bitcoin community have what they hold, which is already far too much to risk losing by allowing open innovation of the network, and i understand that mentality. unfortunately it results in stagnation. they need to be willing to take a haircut.

      1. Jake Brukhman

        Yes, that’s what it certainly seems like. While SegWit is intellectually clever, technologically speaking, it is a workaround or hack. One that took significant time (a year!) to implement. I cannot imagine a technology team at, say, Amazon advocating for such solutions when other completely obvious solutions are on the table. To me, that edges toward the conclusion that the core team is not incentivized to innovate the protocol in usual and expected ways, and that’s unfortunate. On the other side of the table, one can see how introducing changes to working codebases destabilizes them; Bitcoin is a unique project in that destabilization of its codebase directly and swiftly translates to economic losses for stakeholders.

  4. creative group

    CONTRIBUTORS:There have been questions we have followed regarding Segwit’s 2M block size and the lack of testing before implemention. The block size is a contention within the Blockchain community. (Awaiting William Mougayar to explain why). Segwit changes the structure and is said to miss testing which is important for stability.Some say the Bitcoin developers are not moving fast enough to catch up with the demand to serve regular people not understanding the technology and platform and the developers are doing their due diligence to get it right living through the scandals that can actually kill the platform before the adoption of a greater and expanded user base.

  5. Richard Ginsberg

    For Sh*ts and giggles and research, I actually invested in an ASIC miner (and run it) to better understand this industry. The ASIC hashing power is an order of magnitude higher hashing power than CPU, GPU and FPGA now. So the software running on non-ASIC miners are irrelevant.It turns out most don’t solo mine unless they have atleast 10-20 of the latest ASIC miners. This leaves everyone else with less hashing power joining pools, so as to divide the luck and profit of mining a block. *Pool miners don’t choose the bitcoin version they want to run, the pool operator does.* This is how the Chinese control so many miners even though mining hardware are scattered around the world (predominately where electricity is cheap).The Chinese have made the investment in the pools, and have recruited the miners, therefore control the adoption of Bitcoin changes by proxy. Therefore, every pool on the planet needs to switch to SegWit to get past 95% hurdle.

    1. Richard Ginsberg

      Actually I’ve seen one pool that allows miners to vote on the version they want their hashes to go. Screenshot of config: http://imgur.com/a/F5q8U

    2. Richard Ginsberg

      FYI, The recruiting of miners into a given pool is based on frequency of luck, uptime, credibility and fees deducted.

    3. SysMan

      Wrong. You doesn’t get the point.1. More higher math power – more protected bitcoin as asset. higher price can be assigned.2. CPU/GPU/FPGA/ASIC – is power optimization and trx cost optimization… going back to CPU – higher risk, 100x times higher TRX FEE will be for transaction. Simple

      1. Richard Ginsberg

        The point I was making was that the actual hash generation from miners seem indifferent on proposals. They are for the most part backing pools with the lowest fees, best luck in mining recent blocks.The pools are very political on the proposals as it affects their future business.

  6. ZekeV

    It seems like an improvement, and technically a neat trick. But the magnitude of the improvement is linear, while the network is growing exponentially. Once the network hits its limits, then we are quickly back to arguing over whether bitcoin bandwidth should be a scarce thing regulated by tx fees, or something else.

  7. awaldstein

    Fascinating.Are you saying that this is a democratic process of what gets adopted? A popular vote?If so…seems a crazy and I would say potentially unproductive way to determine what becomes the status quo honestly.Can’t wrap my head around this.

    1. pointsnfigures

      Sarcasm, feel yours, yes..

    2. William Mougayar

      There were hard politics involved, so it makes things more complicated. If it was a truly democratic, decentralized innovation process with equal voices and respect, the outcome would be different.

      1. awaldstein

        With all due respect William, why?Help me understand why this can work?

        1. William Mougayar

          It’s a long story. Bitcoin governance is what it is, and there has been a lot of fighting over it in the past year.

          1. awaldstein

            Everything with Bitcoin is a long story it seems.

          2. Adocracy

            It just takes some dedicated effort – not a passing interest – to be able to suss out the details.

          3. awaldstein

            We build visions of a changed world on details of possibility I agree.Bitcoins communications breakdown is that there is no truly articuable vision.The investment vision for USV is clear. The reason why a CMO of an enterprise SaaS company or Disney or Whole Foods should care is not.Not a problem but that is the case.

      2. LE

        Is this ‘vote’ one of those things where the large group of miners (in China) have total control over whether this gets adopted or not?

    3. jason wright

      There are more than 5000 nodes that make up the decentralised bitcoin network. Each is running a copy of the bitcoin software.+95% of them will need to update the software to include this SegWit code to enable it on the network for the upgrade.

      1. awaldstein

        thanks–gives me a sense of the complexity of this endeavor.gotto wonder how the end solution becomes simple and elegant when the process of stabilizing the core is so challenging.

        1. Alex B.

          This is incorrect. The 95% is a signaling method for miners only to demonstrate that their infrastructure is ready for the network to upgrade. See BIP9 https://github.com/bitcoin/

          1. awaldstein

            thanks.crazy how unclear this is to the normal tech outsider.

          2. Alex B.

            It may sound elitist but to be honest this field is so specialized that you should not trust any opinion without validating the information in the traditional technical venues where the experts gather. Misinformation is rampant and there are entire industries with agendas and interests in perpetuating these midunderstandings to profit from them.

          3. awaldstein

            @wmoug:disqus is my guide when i feel the need to go deep on anything blockchain.

          4. Alex B.

            I’m sorry to say William is not an expert but a consultant with low-level knowledge of the technology.And NO, the fact that he wrote a book on them does not validate his status

          5. awaldstein

            we disagree on this completely.

          6. Alex B.

            With all due respect, I fail to understand how you can rightfully evaluate this.

          7. awaldstein

            every person chooses their leaders, their experts, their heroes and their communities.that’s simply life.those who claim there are objective criteria above our own studied beliefs are simply wrong.nothing is completely objective. there is no one truth nor one authority on most anything.

          8. Alex B.

            This is science we are talking about and while there is room for subjectivity it’s simply wrong to pretend that everyone’s opinion is equally valuable.The Bitcoin protocol, for one, is completly objective.

          9. Lawrence Brass

            I perceive your statement more as emotional, than objective.It is just code, you know.cheers.

          10. Alex B.

            It’s code that some people’s lives depend on. There’s no room for half-truths.

          11. William Mougayar

            And you are doing a great job being a troll and an idiot.

          12. Geoffrey Hamilton

            Actually he’s one of the only other people on the thread who both know what is really going on and are not captured by bankster interests.

          13. William Mougayar

            both of you seem to be colluding and have your own agenda which is known, and further disguised behind pseudo-identities, which says nothing about who you really are.

          14. Geoffrey Hamilton

            I don’t know Alex at all, so I can’t speak for him, but I am using my real identity. Stop by the next BitDevs meetup at USV and maybe you’ll even meet me. We’re not colluding, we just have similar perspectives because we have similar maximalist motivations, It’s fine that you don’t share those motivations, but you can’t have a credible opinion of what Bitcoin should do if you don’t even believe in the project in the first place.The arguments against SegWit are as weak, weak enough that they just seem like obstructionism for its own sake. And that is anyone’s right, though it is more likely that those obstructionists are being bribed or otherwise coerced than it is that they actually believe in and are invested in Bitcoin yet still want to obstruct the mutually agreed upon roadmap.

          15. jason wright

            i have edited my grammar. hopefully this conveys my meaning better.

          16. Alex B.

            Your statement remains incorrect. BIP9 and the 95% threshold has nothing to do with the “5000 nodes that make up the decentralised bitcoin network”.

          17. jason wright

            who chose the 95% threshold, and what name do you give to the 95%?

          18. jason wright

            thanks. what name do you give to the 95%?

          19. Alex B.

            It is part of the state transition protocol. Its parameters are set according to the BIP being proposed. It can be referred to as the “Locked-in threshold”

          20. jason wright

            thanks. so it’s not 95% of the ‘nodes’. it’s 95% of the….?

          21. Alex B.

            95% of the blocks signaling support for SegWit over the last 2016 blocks

          22. jason wright

            thanks. blocks.so nodes and miners are mutually exclusive participants on the network?

          23. Alex B.

            Yes, the network has fragmented a long time ago. While miners are also nodes, there are also validating nodes (the 5000 you refer to) that are not involved in the soft-fork activation process. I wrote a little about this here https://medium.com/@bergeal

          24. jason wright

            thanks. now we are at the granular level, where the technical nuances become clearer.so all are nodes, but not all nodes are miners, and the miner nodes need to generate blocks that are in total 95% SegWit compliant within the year for the upgrade to be successfully implemented?how many of the +5000 nodes are mining?what is the economic incentive for setting up a node (validating) but not mine with it? where’s the return on the investment of time and capital resources? Or, are these the exchanges et.c.?

          25. Alex B.

            >need to generate blocks that are in total 95% SegWit compliant within the yearIn a 2016 block period yes.>how many of the +5000 nodes are mining?We don’t know the actual number of miners given that this is a pseudonymous network. You can see the distribution of mining power here at the bottom: https://www.blocktrail.com/BTCThe incentive to run a node is to participate in the network as a legitimate peer who validates the integrity of his own financial activity without relying on a trusted-third-party. It also helps relaying transactions around the network and creating redundancy so the network operates runs without a hitch.There is no direct ROI other than ensuring the integrity of the network and validating the work of miners to ensure that the monetary policy is respected. This distribution of trust is effectively what gives Bitcoin its value.

          26. jason wright

            thanks.if we reach the stage within the year where 95% of blocks are SegWit compliant and therefore SegWit becomes established, will all nodes (whether mining, pooling, validing – all) need to be running that SW version of the software to be able to participate in the network?

          27. Geoffrey Hamilton

            No, this is the advantage of a soft fork: old nodes can still validate the SegWit blocks even if they don’t totally understand what is going on in them.

          28. Geoffrey Hamilton

            The fact that nodes do not need to upgrade is what makes this a “soft” fork; it means that they will still be able to validate the blocks even though they don’t understand the specifics of the transaction going on in them. The nodes can still validate that sufficient proof-of-work was done to spend all of the satoshis in the transaction.One of the things that the actual “segregated witness” change (vs. the whole package of changes this blog post is referring to) does is it allows you to make future upgrades within the witness opcodes, and this will allow for future upgrades to continue happening via soft-fork.

  8. jason wright

    95% is a high bar even for ‘chinacoin’.As an aside, Circle has announced it is not continuing as a bitcoin exchange.

  9. William Mougayar

    I have this dream where Satoshi Nakamoto appears and speaks against this proposal. The 1/3 pro- votes seems to have plateaued, so maybe they were hearing Satoshi’s voices.Kidding aside, Segwit as a soft fork creates a whole new way of doing the same thing without stopping the old way, so it’s problematic from the way I see it, even if it was “tested”. You can test something and it will work, but it might still be the wrong direction. If it passes, it will crown the current team of core developers and their shrinking number of followers as the ruling governors of Bitcoin’s future, and I predict they will lose yet another slice of developers who don’t agree with this direction.

    1. Twain Twain

      How is a fork in Bitcoin different from a fork in Ethereum? That too raised questions about the consensus models and governance structures. Common problems for both camps, it seems.At the core, developers need to ask the question: “Do signatures have to accompany every transaction? How much bandwidth inefficiencies arises because of that? What’s the efficiency improvement if signatures are separated from transactions?”They just have to think in terms of the throughput metrics.

      1. William Mougayar

        The fork itself is not the issue. The nature of it is.

        1. Reinout Heeck

          This sounds like it is only a technological issue, however part of said ‘nature’ is not about technology at all but about the bait-n-switch early adopters sense because of the direction the current developer crew chose:Limiting throughput forever in order to induce higher transaction fees that wil pay for mining half a century from now. These developers chose to introduce a regulted ‘fee market’ with knobs adjusted by themselves as they see fit.So in big part it is much like Trump and Brexit, a large chunk of the ecosystem wants another crew at the top spot, SegWit is merely an opportunity for many to lob a figurative molotov cocktail.

    2. Geoffrey Hamilton

      Why is it a problem to create a new transaction type? We already have more than one way of “doing the same thing without stopping the old way” in the form of multi-sig addresses.I am all for bigger blocks but all I am reading in your comment is shilling for your bankster and regulatory overlords who fear that lightning network will make bitcoin useful as cash again by enabling microtransactions and obfuscating senders.

      1. William Mougayar

        I’m all for these new capabilities, but the current method chosen creates a management difficulty that’s not desirable for all participants.

        1. Geoffrey Hamilton

          In your original post you said that it is the doing the same thing a different way while still being able to do it the old way that is problematic, why is that problematic? Just because of node inertia? You think that if they were forced to hard fork that it would be easier to get onboard?If that is your line of reasoning I completely disagree. If it’s about the 95% I may be able to agree that is too high and possibly unachievable given that 5% of nodes haven’t even upgraded past 0.12.0.

          1. William Mougayar

            In its current form, the proposal would give a lot less influence to miners in the long term, just an fyi.

          2. Geoffrey Hamilton

            Less influence on the scripting language, but that doesn’t matter. What matters is that signatures still validate, only I can send coins from keys controlled by me, and anyone can independently validate the transactions I send.What matters is that old versions of the software can still validate the blockchain and can still send their money to new versions. What matters is that I can put a node in space, never update it, and have it still validate.Ultimately we may have to hard fork for block size but hopefully this will be a one time event that is built in a dynamic way so that it will scale going forward.

    3. awaldstein

      If you are dreaming about this you are my hero!While I do push on you as is my nature, I trust you 100% as my expert on bitcoin.I do not say this lightly but i respect the effort and understanding that you have committed to this.

      1. LE

        I trust you 100% as my expert on bitcoinI trust William myself on many things not just bitcoin or blockchain.

    4. Alex B.

      Clearly William you are out of touch with the Bitcoin ecosystem while you have your head in the blockchain clouds.SegWit adoption is on a voluntary basis by every user on the network. If it comes to activate and gets adopted it will signal that the network as a whole, individually and independently has agreed that it is the best way forward.SegWit is supported by almost every company in the ecosystem: https://bitcoincore.org/en/…The group shrinking in number is the one opposing it using nothing more than fallacies and populist arguments. The Core project has seen an increase in activity and developers so I can only infer from your comments that you are intentionally spreading FUD that is motivated by whatever bias you have.

      1. William Mougayar

        You have eronously characterized me, and I don’t take that lightly. But I don’t have time to respond and react to your bullying statements which are a reflection of how Bitcoin core modus operandi has been.

        1. Alex B.

          You have made several loaded statements which clearly shine light on your intention and bias. Every single post you have made in this thread is seemingly an attempt to discredit the incredible engineering work that is SegWit and teach the people about a controversy that does not exist, what gives?

          1. William Mougayar

            I have a right to my (informed) opinion. Just respect it, if you can; and refrain from personal attacks, if you can.

          2. Alex B.

            I also have a right to point out that your opinion or the way you present it seems more uninformed than anything. I’m just asking that you avoid deceiving people by tainting your input with some sort of political bias, if you can.

          3. William Mougayar

            giving you a warning that if you can’t see the difference in sticking to discussing issues without personal attacks, then we will consider banning you here.

          4. Alex B.

            I expected nothing less. Can you point out where I attacked your character and not your message? I have issue with the way you present the situation, is that not “discussing issues”?

      2. jason wright

        you’ve been very helpful. i thank you for that. your tech rep is presently high.fudders are duffers, and i know William isn’t a duffer.p.s. this is not bitcointalk. fudding here would probably have little impact on the crypto space 🙂

  10. Sebastian Wain

    SegWit is a revolution within the Bitcoin community and a chance to bring new innovations. I was personally involved with the implementation of a BIP to support sidechains, making it possible, for example, to extend smart-contracts capabilities.I thought 2016 was going to be the Ethereum year but ended up being the SegWit year.

    1. William Mougayar

      “I thought 2016 was going to be the Ethereum year but ended up being the SegWit year.”Why does one have to lose for the other win? Bitcoin and Ethereum are different, in more ways than one. I don’t think groups should fight with each other. The ecosystem is vast. The fight is in getting adoption beyond the 10 Million people that barely know what is going on.

      1. Sebastian Wain

        I am just saying that I expected 2016 to be a great year for Ethereum but not for Bitcoin. These personal expectations were mainly based on the different cycles of these technologies: Bitcoin traction was not growing as expected while Ethereum offered new innovations that were never tested before in other cryptocurrencies/public blockchains.I would love to see more interoperability between different actors because some communities act as if the innovation ended in Bitcoin. The Bitcoin improvement proposal I mentioned is 100% connected with this and benefits sidechains from RSK, Blockstream, and many others.

        1. Mike Cautillo

          It was a fantastic year for Ethereum, the Ethereum Classic chain was born!!!

          1. Richard Ginsberg

            Wasn’t expecting ETC. couldn’t even fathom that was even a possibility. Watching as it continues to play out.

      2. Alex Rinaldi

        Nowhere did he say ethereum lost or even allude to the idea.Furthermore, to say ethereum performed below expectations this year would be an understatement.

  11. Geoffrey Hamilton

    It’s less about node adoption and more about miner adoption. 95% of blocks mined over a difficulty period need to signal for this change. This is why ViaBTC with 8% of hashrate signaling for Unlimited can block it by themselves even if all nodes upgrade.

    1. LE

      How is it that the segwit % can go down?

      1. Doug Ellis

        Great question. My bot looks at this site: https://bitcoincore.org/en/…which tracks adoption across three time periods (or block count periods)(1) the last 144 blocks (this is the one I am looking at)(2) the last 2016 blocks(3) since the start of the last periodThe difference among these three is not that important here. What each is looking at is the % of blocks mined by miners who support SegWit. Depending on which miners are “winning” the blocks, the percentage will go up or down a bit. Over time, however, these percentages will track overall miner adoption.

    2. Alex Rinaldi

      This comment should be upvoted. It correctly points out SegWit adoption is about 95% of the hash rate signaling support. Nodes do need to signal support, but that will be a given if miners are anywhere near the 95% needed.It should also be pointed out that this BIP (Bitcoin improvement proposal) has until November 2017 to be supported before SegWit will no longer be available as a soft fork.We should also link to the direct source rather than a random twitter account: https://bitcoincore.org/en/

      1. Richard Ginsberg

        Most of the hash rate origins are not running Full Node, Mining Software, or Pool software. They are running ASIC appliances that connect to pools that offer the highest likelihood of payout. The Pool operators themselves are running the software that sets the version to 0x20000002 when they mine the block if they support SegWit.The Miners are basically giving signatory authority to the Pool operators. The bulk chose version 0x20000000 on behalf of all participating miners.

        1. freedom

          Check Slush – they offer choise & freedomvs CensoredAntPool, BW, DiscusFish / F2Pool, ViaBTC, HaoBTC – now choise, but very vocal about “censorship” …buuuuu

          1. Richard Ginsberg

            Yes, I included a screenshot of slush config on previous post giving the miner choice.

      1. Alex B.

        As a point of reference, the uptake of Segwit as a soft-fork is on pace to be the fasted activated soft-fork since the BIP9 process was introduced.

    3. Richard Ginsberg

      https://coin.dance/blocks#b…This is a nice way of looking at the latest blocks, the origin of miner (pool) and their support. More information on non-responsive desktop version though.

  12. Tony

    SegWit is a terrible approach to a simple problem, RAISE THE BLOCK SIZE LIMIT.

    1. SandwichOfEarl

      Raising the blocksize limit isn’t a simple solution though, because we know hardforks can go very wrong (look at ethereum and ethereum classic).

  13. Muneeb Ali

    If it’s not activated now, it will likely get activated later on. It’s a very carefully designed and tested improvement that has a lot of scalability benefits amongst other things.

  14. Alex B.

    There are 39 comments on this post yet NO person has mentioned that the 95% threshold has absolutely nothing to do with the nodes on the network.Please, for the love of all things, verify with someone competent before spreading misinformation. It’s nice for you to take interest but there is already enough misunderstanding as it is…

    1. fredwilson

      Thanks for the clarification. That’s what communities are for

    2. Geoffrey Hamilton

      I mentioned this in my comment yesterday.

  15. Marty Bent

    I think SegWit is a well thought out and logical scaling solution for Bitcoin right now. Big fan of KISS principles and taking your time to get something right the first time and I think the Core team has exemplified these principles while developing and launching SegWit.

  16. Bitcoin Error Log

    Exactly. It has no implications on Bitcoin’s “prospects”. It’s supposed to be difficult to change Bitcoin. Reliability and security come first.

  17. Brandon Burns

    I’m more curious to see a post and conversation around fake news. I’ll keep my hopes up… 🙂

  18. Jose Paul Martin

    Besides the name, an interesting development – will keep my eye on this.

  19. Richard Ginsberg

    I was giving this a bit more thought since many may be confused about the support for SegWit ebbing and flowing throughout the day. Bitcoin recalculates difficulty every 2016 blocks mined. The vote for SegWit is always on the block level, tallying the % with version 0x20000002 for the last 2016 blocks.In theory SegWit could be put in effect if the avg 30% of miners/pools that support it, just so happened to mine atleast 1916 of the last 2016 blocks. It’s the same theory that you could flip a coin and it lands heads repeatedly in succession N-number of times.Barring random walk and six sigma, the 3 major pools that control ~50% of hashes will need to be negotiated with.The votes could be bought as well if a pool supporting SegWit incentivized miners by offering bonus payouts for mining with the pool. So long as the proposed change also doesn’t affect the miners bottom line. It is a one line change to point your miner to another pool, akin to logging into a router and changing SSID. There is no loyalty between miners and pools; just miners looking to pay down on hardware investments and look towards ROI.

  20. Adocracy

    @william Mougayer – is there somewhere you can point me to that has your collected assessment of SegWit? Because as I see it, there is a ton of intellectual “industry” support for it, plus the promise of future expansion/experimentation being accomplished via soft fork, apart from miner approvals. Hard to see that as a lose.

  21. jason wright

    is it called Segregated Witness because under it the bitcoin nodes will divide with one group having the ability to generate SW compliant blocks and the other group only the ability to validate their legitimacy by ‘observation’?