Bitcoin: More ideology than trustworthy currency | ZDNet

Exposed: How Bankers are trying to centralize and highjack Bitcoin by buying "supporters" and promoters (like OpenBazaar team) for the B2X (S2X/NYA) attack on Bitcoin.

*Open Bazaar was crossed-out after their S2X support retraction, see edit at bottom.
These guys have deep pockets, but as you will see below, they are funded by even deeper pockets.
We can't leave this to chance or "the markets to decide" when there is such a malicious intent to manipulate the markets by those powerful players. So that's why all the people saying: "Don't worry, S2X won't happen" or "S2X is DOA" need to stop, we are at a 'make-or-break' moment for Bitcoin. It's very dumb to underestimate them. If you don't know yet who those malicious players are, read below:
We need to keep exposing them everywhere. Using Garzik as a pawn now, after they failed when they bought Hearn and Andresen (Here are the corrupted former 'good guys'), they are using the old and effective 'Problem-Reaction-Solution' combined with the 'Divide & Conquer' strategies to try to hijack Bitcoin. Well, effective before the current social media era, in which hidden motives can be brought to the light of day to be exposed.
Public pressure works when your profits depend on your reputation. The social media criticism worked for companies like Open Bazaar, which after weeks of calling them out on their S2X support, they finally withdrew it.
Please contact the companies on these lists if you have any type of relationship with them, we have just a few days left until the fork:
Regarding OpenBazaar:
* openbazaar (OB1) developer appears to be spreading pro s2x fud. someone needs to fork their project
* PSA : Open Bazaars latest investment round was for 200K from Barry Silberts DCG (Digital Currency Group)
(See edit at the bottom)
B2X (S2X/NYA) is nothing more than an open attack on Bitcoin, not an "upgrade" as they want to sell it. This attack has no 'consensus', at all. It was "agreed" by a bunch of miners and corporations behind closed doors, with no community nor developers support. Only miners and a few millionaires that stand to profit from the B2X attack support it. The vast majority of the Bitcoin community is totally against this attack on Bitcoin. Most of those companies are under DCG group:
Every bitcoiner should know about what DCG (Digital Currency Group) is, and call out publicly these crooks and the people they bribed that are working for the Corporations/Bankers against Bitcoin:
Brian Armstrong, Winklevoss brothers, Bobby Lee, Peter Smith, Nic Cary, Haipo Yang, Rick Falkvinge, Jon Matonis, Wences Casares, Tony Gallippi, Mike Belshe, Ryan X Charles, Brian Hoffman/Sam Patterson/Chris Pacia (and all OB1 team)(see edit at the bottom), Gavin Andresen, Jeff Garzik, Mike Hearn, Roger Ver, Jihan Wu, John Mcaffe, Craig Wright, Barry Silbert, Larry Summers, Blythe Masters, Stephen Pair, Erik Voorhees, Vinny Lingham, Olivier Janssens, Jeremy Allaire, Peter Vessenes, Bruce Wagner, Brock Pierce, Aaron Voisine/Adam Traidman/Aaron Lasher (Breadwallet team), Glenn Hutchins (Federal Reserve Board of Directors), Bill Barhydt and Jiang Zhuoer.
Once people are informed, they won't be fooled (like all the poor guys at btc) and will follow Bitcoin instead of the S2X or Bcash or any other centralized altcoin they come up with disguised as Bitcoin.
DCG (Digital Currency Group) is the company spearheading the Segwit2x movement. The CEO of DCG is Barry Silbert, a former investment banker, and Mastercard is an investor in DCG.
Let's have a look at the people that control DCG:
http://dcg.co/who-we-are/
Three board members are listed, and one Board "Advisor." Three of the four Members/advisors are particularly interesting:
Glenn Hutchins: Former Advisor to President Clinton. Hutchins sits on the board of The Federal Reserve Bank of New York, where he was reelected as a Class B director for a three-year term ending December 31, 2018. Yes, you read that correctly, currently sitting board member of the Federal Reserve Bank of New York.
Barry Silbert: CEO of DCG (Digital Currency Group, funded by Mastercard) who is also an Ex investment Banker at (Houlihan Lokey)
And then there's the "Board Advisor,"
Lawrence H. Summers:
"Chief Economist at the World Bank from 1991 to 1993. In 1993, Summers was appointed Undersecretary for International Affairs of the United States Department of the Treasury under the Clinton Administration. In 1995, he was promoted to Deputy Secretary of the Treasury under his long-time political mentor Robert Rubin. In 1999, he succeeded Rubin as Secretary of the Treasury. While working for the Clinton administration Summers played a leading role in the American response to the 1994 economic crisis in Mexico, the 1997 Asian financial crisis, and the Russian financial crisis. He was also influential in the American advised privatization of the economies of the post-Soviet states, and in the deregulation of the U.S financial system, including the repeal of the Glass-Steagall Act."
https://en.wikipedia.org/wiki/Lawrence_Summers
Blythe Masters:
Former executive at JPMorgan Chase.[1] She is currently the CEO of Digital Asset Holdings,[2] a financial technology firm developing distributed ledger technology for wholesale financial services.[3] Masters is widely credited as the creator of the credit default swap as a financial instrument. She is also Chairman of the Governing Board of the Linux Foundation’s open source Hyperledger Project, member of the International Advisory Board of Santander Group, and Advisory Board Member of the US Chamber of Digital Commerce.
https://en.wikipedia.org/wiki/Blythe_Masters
Seriously....The segwit2x deal is being pushed through by a Company funded by Mastercard, Whose CEO Barry Silbert is ex investment banker, and the Board Members of DCG include a currently sitting member of the Board of the Federal Reserve Bank of New York, and the Ex chief Economist for the World Bank and a guy responsible for the removal of Glass Steagall.
It's fair to call these guys "bankers" right?
So that's the Board of DCG. They're spearheading the Segwit2x movement. As far as who is responsible for development, my research led me to "Bitgo". I checked the "Money Map" https://i.redd.it/15auzwkq3hiz.png And sure enough, DCG is an investor in Bitgo.
(BTW, make sure you take a good look take a look at the money map and bookmark it for reference later, ^ it is really helpful.)
"Currently, development is being overseen by bitcoin security startup BitGo, with help from other developers including Bloq co-founder Jeff Garzik."
https://www.coindesk.com/bitcoins-segwit2x-scaling-proposal-miners-offer-optimistic-outlook/
So Bitgo is overseeing development of Segwit2x with Jeff Garzick. Bitgo has a product/service that basically facilitates transactions and supposedly prevents double spending. It seems like their main selling point is that they insert themselves as middlemen to ensure Double spending doesn't happen, and if it does, they take the hit, of course for a fee, so it sounds sort of like the buyer protection paypal gives you:
"Using the above multi-signature security model, BitGo can guarantee that transactions cannot be double spent. When BitGo co-signs a BitGo Instant transaction, BitGo takes on a financial obligation and issues a cryptographically signed guarantee on the transaction. The recipient of a BitGo Instant transaction can rest assured that in any event where the transaction is not ultimately confirmed in the blockchain, and loses money as a result, they can file a claim and will be compensated in full by BitGo."
Source: https://www.bitgo.com/solutions
So basically, they insert themselves as middlemen, guarantee your transaction gets confirmed and take a fee. What do we need this for though when we have a working blockchain that confirms payments in the next block already? 0-conf is safe when blocks aren't full and one confirmation should really be good enough for almost anyone on the most POW chain. So if we have a fully functional blockchain, there isn't much of a need for this service is there? They're selling protection against "The transaction not being confirmed in the Blockchain" but why wouldn't the transaction be getting confirmed in the blockchain? Every transaction should be getting confirmed, that's how Bitcoin works. So in what situation does "protection against the transaction not being confirmed in the blockchain" have value?
Is it possible that the Central Bankers that control development of Segwit2x plan to restrict block size to benefit their business model just like our good friends over at Blockstream attempted to do, although unsuccessfully as they were not able to deliver a working L2 in time?
It looks like Blockstream was an attempted corporate takeover to restrict block size and push people onto their L2, essentially stealing business away from miners. They seem to have failed, but now it almost seems like the Segwit2x might be a culmination of a very similar problem.
Also worth noting these two things, pointed out by Adrian-x:
  1. MasterCard made this statement before investing in DCG and Blockstream. (Very evident at 2:50 - enemy of digital cash watch the whole thing.) https://www.youtube.com/watch?v=Tu2mofrhw58
  2. Blockstream is part of the DCG portfolio and the day after the the NYA Barry personal thanked Adam Back for his assistant in putting the agreement together. https://twitter.com/barrysilbert/status/867706595102388224
So segwit2x takes power away from core, but then gives it to guess who...Mastercard and central bankers.
So, to recap:
Did we just spend so much time fighting and bickering with core that we totally missed the REAL takeover of Bitcoin, happening right before our eyes, by the likes of currently serving Federal Reserve Bank of New York Board Members?
And before you dismiss all those hard and documented facts as just a 'conspiracy theory', think about this:
Of course, who thought that the ones holding the centralized financial power today (famous for back-door shady plots to consolidate even more power and control), would sit on their hands and let Bitcoin just stroll in and easily take that power away from them?
So, it is not a crazy conspiracy theory, but more like the logical and expected thing to happen. Don't let it happen.
Edit: Formatting.
Edit 2: Brian Armstrong taken out of the 'bad guys' list.
Edit 3: Welp, Brian Armstrong back on the blacklist for this flip-flop. And added Winklevoss Brothers for this, and Bobby Lee for this.
Edit 4: Due to Brian Hoffman just issuing this excellent and explicit S2X/NYA support retraction, I created this post to apologize for my previous posts (calling them out for the S2X support) and I will be editing my posts to reflect this positive change. I'm gladly back to being a supporter of the great and promising project that OpenBazaar has proven to be.
Edit 5: Added Blythe Masters (How could we leave her out?).
Edit 6: Added links to lists of companies supporting S2X/NYA.
submitted by readish to Bitcoin [link] [comments]

Bitcoin Cash Merchant Adoption Blitzkrieg - Day 1 - 4 Dec 2017

As many of you know, the Accept Bitcoin Cash initiative site has launched thanks to the hard working Accept Bitcoin Cash team.
However it's not enough to rest on our laurels and post Vegeta memes all day hoping for the price to go up. We've got some major catching up to do. What we need to do is focus on massive merchant adoption. Bitcoin Cash is only useful as a store of value if you can actually buy things with it. The only way that happens is by letting merchants know that we want to pay them with Bitcoin Cash and that we're prepared to do so.
Bitcoin Cash actually benefits from being a functional network, unlike Bitcoin Legacy with its high fees, congestion and Tab payment mechanism which is always 18 months away. Bitcoin Cash has a lot of good selling points, even more so than most other cryptocurrencies if you actually research their fundamentals. Any merchants currently supporting BTC should be able to support Bitcoin Cash with a few tweaks to their software.
So the plan is each day we have a new merchant/service/exchange we want to target together in a coordinated campaign. We can start with some of the biggest global ones first so the network effect grows faster. All we need to do is send one tweet each asking them to support Bitcoin Cash. It will only take a few seconds of your time. We need everyone involved, some have more Twitter outreach and influence than others. Don't just sit on the sideline and watch, get involved too. You want your Bitcoin Cash to grow in value too, right?
Our critics call us "coffee coin" so let's try getting a major cafe chain onboard. To kick things off, I thought we could start off with Starbucks. I like a nice ice cold frappuccino as much as the next guy (and my waistline shows it). Next time we can do Amazon, Netflix, Uber, Lamborghini, Newegg, airlines, online services etc, whatever you like. Feel free to post suggestions for next time in the comments. I don't even need to be running the show here. With Bitcoin Cash everyone's the CEO. If I haven't posted the daily thread by 00:00 UTC due to work, travel etc feel free to copy the template and post it yourself.
Here's our plan for today:
In the future Starbucks or whoever else add support for Bitcoin Cash, go buy something with Bitcoin Cash to show your support. I know Bitcoin Legacy has a "HODL at all costs" culture, but if we want Bitcoin Cash to be accepted everywhere and deprecate the banking system then we all need to actually use it for payments. For me personally I will do a bit of long term saving and spending. I'll try save up a certain amount long term for retirement, save up another amount and cash out that amount for a house one day, maybe another portion for a car and the rest I'll set aside for spending. You can come up with your own strategy like buying a small amount each month with your paycheck.
Once more merchants start coming on board we can also make an issue or Pull Request to keep the site up to date. This will be the go-to site to know who is currently accepting Bitcoin Cash and where you can go to spend it. If you don't see a merchant or service on there feel free to add it too. The more the merrier.
submitted by n9jd34x04l151ho4 to btc [link] [comments]

Bitcoin, huh? WTF is going on? Should we scale you on-chain or off-chain? Will you stay decentralized, distributed, immutable?

0. Shit, this is long, TLWR please! Too long, won't read.
EDIT: TLDR TLWR for clarity.
1. Bitcoin, huh? Brief introduction.
There are 3 sections to this overview. The first section is a brief introduction to bitcoin. The second section looks at recent developments in the bitcoin world, through the analogy of email attachments, and the third section discusses what could be next, through the perspective of resilience and network security.
This is just a continuation of a long, long, possibly never-ending debate that started with the release of the bitcoin whitepaper in 2008 (see https://bitcoin.org/bitcoin.pdf). The recent mess during the past few years boils down to the controversy with the block size limit and how to appropriately scale bitcoin, the keyword appropriately. Scaling bitcoin is a controversial debate with valid arguments from all sides (see https://en.bitcoin.it/wiki/Block_size_limit_controversy).
I have researched, studied, and written this overview as objectively and as impartially as possible. By all means, this is still an opinion and everyone is advised to draw their own conclusions. My efforts are to make at least a few readers aware that ultimately there is only one team, and that team is the team bitcoin. Yes, currently though, there are factions within the team bitcoin. I hope that we can get beyond partisan fights and work together for the best bitcoin. I support all scaling proposals as long as they are the best for the given moment in time. Personally, I hate propaganda and love free speech as long as it is not derogatory and as long as it allows for constructive discussions.
The goal of this overview is to explain to a novice how bitcoin network works, what has been keeping many bitcoin enthusiasts concerned, and if we can keep the bitcoin network with three main properties described as decentralized, distributed, immutable. Immutable means censorship resistant. For the distinction between decentralized and distributed, refer to Figure 1: Centralized, decentralized and distributed network models by Paul Baran (1964), which is a RAND Institute study to create a robust and nonlinear military communication network (see https://www.rand.org/content/dam/rand/pubs/research_memoranda/2006/RM3420.pdf). Note that for the overall network resilience and security, distributed is more desirable than decentralized, and the goal is to get as far away from central models as possible. Of course, nothing is strictly decentralized or strictly distributed and all network elements are at different levels of this spectrum.
For those unaware how bitcoin works, I recommend the Bitcoin Wikipedia (see https://en.bitcoin.it/wiki/Main_Page). In short, the bitcoin network includes users which make bitcoin transactions and send them to the network memory pool called mempool, nodes which store the public and pseudonymous ledger called blockchain and which help with receiving pending transactions and updating processed transactions, thus securing the overall network, and miners which also secure the bitcoin network by mining. Mining is the process of confirming pending bitcoin transactions, clearing them from the mempool, and adding them to blocks which build up the consecutive chain of blocks on the blockchain. The blockchain is therefore a decentralized and distributed ledger built on top of bitcoin transactions, therefore impossible to exist without bitcoin. If someone claims to be working on their own blockchain without bitcoin, by the definition of the bitcoin network however, they are not talking about the actual blockchain. Instead, they intend to own a different kind of a private database made to look like the public and pseudonymous blockchain ledger.
There are roughly a couple of dozen mining pools, each possibly with hundreds or thousands of miners participating in them, to several thousand nodes (see https://blockchain.info/pools and https://coin.dance/nodes). Therefore, the bitcoin network has at worst decentralized miners and at best distributed nodes. The miner and node design makes the blockchain resilient and immune to reversible changes, making it censorship resistant, thus immutable. The bitcoin blockchain avoids the previous need for a third party to trust. This is a very elegant solution to peer-to-peer financial exchange via a network that is all: decentralized, distributed, immutable. Extra features (escrow, reversibility via time-locks, and other features desirable in specific instances) can be integrated within the network or added on top of this network, however, they have not been implemented yet.
Miners who participate receive mining reward consisting of newly mined bitcoins at a predetermined deflationary rate and also transaction fees from actual bitcoin transactions being processed. It is estimated that in 2022, miners will have mined more than 90% of all 21 million bitcoins ever to be mined (see https://en.bitcoin.it/wiki/Controlled_supply). As the mining reward from newly mined blocks diminishes to absolute zero in 2140, the network eventually needs the transaction fees to become the main component of the reward. This can happen either via high-volume-low-cost transaction fees or low-volume-high-cost transaction fees. Obviously, there is the need to address the question of fees when dealing with the dilemma how to scale bitcoin. Which type of fees would you prefer and under which circumstances?
2. WTF is going on? Recent developments.
There are multiple sides to the scaling debate but to simplify it, first consider the 2 main poles. In particular, to scale bitcoin on blockchain or to scale it off it, that is the question!
The first side likes the idea of bitcoin as it has been until now. It prefers on-chain scaling envisioned by the bitcoin creator or a group of creators who chose the pseudonym Satoshi Nakamoto. It is now called Bitcoin Cash and somewhat religiously follows Satoshi’s vision from the 2008 whitepaper and their later public forum discussions (see https://bitcointalk.org/index.php?topic=1347.msg15366#msg15366). Creators’ vision is good to follow but it should not be followed blindly and dogmatically when better advancements are possible, the keyword when. To alleviate concerning backlog of transactions and rising fees, Bitcoin Cash proponents implemented a simple one-line code update which increased the block size limit for blockhain blocks from 1MB block size limit to a new, larger 8MB limit. This was done through a fork on August 1, 2017, which created Bitcoin Cash, and which kept the bitcoin transaction history until then. Bitcoin Cash has observed significant increase in support, from 3% of all bitcoin miners at first to over 44% of all bitcoin miners after 3 weeks on August 22, 2017 (see http://fork.lol/pow/hashrate and http://fork.lol/pow/hashrateabs).
An appropriate scaling analogy is to recall email attachments early on. They too were limited to a few MB at first, then 10MB, 20MB, up until 25MB on Gmail. But even then, Gmail eventually started using Google Drive internally. Note that Google Drive is a third party to Gmail, although yes, it is managed by the same entity.
The second side argues that bitcoin cannot work with such a scaling approach of pre-meditated MB increases. Arguments against block size increases include miner and node centralization, and bandwidth limitations. These are discussed in more detail in the third section of this overview. As an example of an alternative scaling approach, proponents of off-chain scaling want to jump to the internally integrated third party right away, without any MB increase and, sadly, without any discussion. Some of these proponents called one particular implementation method SegWit, which stands for Segregated Witness, and they argue that SegWit is the only way that we can ever scale up add the extra features to the bitcoin network. This is not necessarily true because other scaling solutions are feasible, such as already functioning Bitcoin Cash, and SegWit’s proposed solution will not use internally integrated third party as shown next. Note that although not as elegant as SegWit is today, there are other possibilities to integrate some extra features without SegWit (see /Bitcoin/comments/5dt8tz/confused_is_segwit_needed_for_lightning_network).
Due to the scaling controversy and the current backlog of transactions and already high fees, a third side hastily proposed a compromise to a 2MB increase in addition to the proposed SegWit implementation. They called it SegWit2x, which stands for Segregated Witness with 2MB block size limit increase. But the on-chain scaling and Bitcoin Cash proponents did not accept it due to SegWit’s design redundancy and hub centralization which are discussed next and revisited in the third section of this overview. After a few years of deadlock, that is why the first side broke free and created the Bitcoin Cash fork.
The second side stuck with bitcoin as it was. In a way, they inherited the bitcoin network without any major change to public eye. This is crucial because major changes are about to happen and the original bitcoin vision, as we have known it, is truly reflected only in what some media refer to as a forked clone, Bitcoin Cash. Note that to avoid confusion, this second side is referred to as Bitcoin Core by some or Legacy Bitcoin by others, although mainstream media still refers to it simply as Bitcoin. The core of Bitcoin Core is quite hardcore though. They too rejected the proposed compromise for SegWit2x and there are clear indications that they will push to keep SegWit only, forcing the third side with SegWit2x proponents to create another fork in November 2017 or to join Bitcoin Cash. Note that to certain degree, already implemented and working Bitcoin Cash is technically superior to SegWit2x which is yet to be deployed (see /Bitcoin/comments/6v0gll/why_segwit2x_b2x_is_technically_inferior_to).
Interestingly enough, those who agreed to SegWit2x have been in overwhelming majority, nearly 87% of all bitcoin miners on July 31, 2017 prior to the fork, and a little over 90% of remaining Bitcoin Core miners to date after the fork (see https://coin.dance/blocks). Despite such staggering support, another Bitcoin Core fork is anticipated later in November (see https://cointelegraph.com/news/bitcoin-is-splitting-once-again-are-you-ready) and the "Outcome #2: Segwit2x reneges on 2x or does not prioritize on-chain scaling" seems to be on track from the perspective of Bitcoin Core SegWit, publicly seen as the original Bitcoin (see https://blog.bridge21.io/before-and-after-the-great-bitcoin-fork-17d2aad5d512). The sad part is that although in their overwhelming majority, the miners who support SegWit2x would be the ones creating another Bitcoin Core SegWit2x fork or parting ways from the original Bitcoin.
In a way, this is an ironic example how bitcoin’s built-in resiliency to veto changes causes majority to part away when a small minority has status quo and holds off fully-consented progress. Ultimately, this will give the minority Bitcoin Core SegWit proponents the original Bitcoin branding, perhaps to lure in large institutional investors and monetize on bitcoin’s success as we have it seen it during the past 9 years since its inception. Recall that bitcoin of today is already a decentralized, distributed, immutable network by its definition. The bitcoin network was designed to be an alternative to centralized and mutable institutions, so prevalent in modern capitalist societies.
Bitcoin Core SegWit group wants to change the existing bitcoin network to a network with dominant third parties which, unlike Google Drive to Gmail, are not internal. In particular, they intend to do so via the lightning network, which is a second layer solution (2L). This particular 2L as currently designed relies on an artificial block size limit cap which creates a bottleneck in order to provide high incentives for miners to participate. It monetizes on backlog of transaction and high fees, which are allocated to miners, not any group in particular. Cheaper and more instantaneous transactions are shifted to the lightning network which is operated by hubs also earning revenue. Note that some of these hubs may choose to monitor transactions and can possibly censor who is allowed to participate in this no longer strictly peer-to-peer network.
We lose the immutability and instead we have a peer-to-hub-to-peer network that is mutable and at best decentralized, and certainly not distributed (see https://medium.com/@jonaldfyookball/mathematical-proof-that-the-lightning-network-cannot-be-a-decentralized-bitcoin-scaling-solution-1b8147650800). For regular day-to-day and recurring transactions, it is not a considerable risk or inconvenience. And one could choose to use the main chain any time to bypass the lightning network and truly transact peer-to-peer. But since the main chain has an entry barrier in the form of artificially instilled high transaction fees, common people are not able to use bitcoin as we have known it until now. Peer-to-peer bitcoin becomes institution-to-institution bitcoin with peer-to-hub-to-peer 2L.
To reiterate and stress, note the following lightning network design flaw again. Yes, activating SegWit and allowing 2L such as lightning allows for lower transaction fees to coexist side by side with more costly on-chain transactions. For those using this particularly prescribed 2L, the fees remain low. But since these 2L are managed by hubs, we introduce another element to trust, which is contrary to what the bitcoin network was designed to do at the first place. Over time, by the nature of the lightning network in its current design, these third party hubs grow to be centralized, just like Visa, Mastercard, Amex, Discover, etc. There is nothing wrong with that in general because it works just fine. But recall that bitcoin set out to create a different kind of a network. Instead of decentralized, distributed, immutable network with miners and nodes, with the lightning network we end up with at best decentralized but mutable network with hubs.
Note that Bitcoin Core SegWit has a US-based organization backing it with millions of dollars (see https://en.wikipedia.org/wiki/Blockstream and https://steemit.com/bitcoin/@adambalm/the-truth-about-who-is-behind-blockstream-and-segwit-as-the-saying-goes-follow-the-money). Their proponents are quite political and some even imply $1000 fees on the main bitcoin blockchain (see https://cointelegraph.com/news/ari-paul-tuur-demeester-look-forward-to-up-to-1k-bitcoin-fees). Contrary to them, Bitcoin Cash proponents intend to keep small fees on a scale of a few cents, which in large volume in larger blockchain blocks provide sufficient incentive for miners to participate.
On the one hand, sticking to the original vision of peer-to-peer network scaled on-chain has merit and holds potential for future value. On the other hand, 2L have potential to carry leaps forward from current financial infrastructure. As mentioned earlier, 2L will allow for extra features to be integrated off-chain (e.g. escrow, reversibility via time-locks), including entirely new features such as smart contracts, decentralized applications, some of which have been pioneered and tested on another cryptocurrency network called Ethereum. But such features could be one day implemented directly on the main bitcoin blockchain without the lightning network as currently designed, or perhaps with a truly integrated 2L proposed in the third section of this overview.
What makes the whole discussion even more confusing is that there are some proposals for specific 2L that would in fact increase privacy and make bitcoin transactions less pseudonymous than those on the current bitcoin blockchain now. Keep in mind that 2L are not necessarily undesirable. If they add features and keep the main network characteristics (decentralized, distributed, immutable), they should be embraced with open arms. But the lightning network as currently designed gives up immutability and hub centralization moves the network characteristic towards a decentralized rather than a distributed network.
In a sense, back to the initial email attachment analogy, even Gmail stopped with attachment limit increases and started hosting large files on Google Drive internally, with an embedded link in a Gmail email to download anything larger than 25MB from Google Drive. Anticipating the same scaling decisions, the question then becomes not if but when and how such 2L should be implemented, keeping the overall network security and network characteristics in mind. If you have not gotten it yet, repeat, repeat, repeat: decentralized, distributed, immutable. Is it the right time now and is SegWit (one way, my way or highway) truly the best solution?
Those siding away from Bitcoin Core SegWit also dislike that corporate entities behind Blockstream, the one publicly known corporate entity directly supporting SegWit, have allegedly applied for SegWit patents which may further restrict who may and who may not participate in the creation of future hubs, or how these hubs are controlled (see the alleged patent revelations, https://falkvinge.net/2017/05/01/blockstream-patents-segwit-makes-pieces-fall-place, the subsequent Twitter rebuttal Blockstream CEO, http://bitcoinist.com/adam-back-no-patents-segwit, and the subsequent legal threats to SegWit2x proponents /btc/comments/6vadfi/blockstream_threatening_legal_action_against). Regardless if the patent claims are precise or not, the fact remains that there is a corporate entity dictating and vetoing bitcoin developments. Objectively speaking, Bitcoin Core SegWit developers paid by Blockstream is a corporate takeover of the bitcoin network as we have known it.
And on the topic of patents and permissionless technological innovations, what makes all of this even more complicated is that a mining improvement technology called ASICboost is allowed on Bitcoin Cash. The main entities who forked from Bitcoin Core to form Bitcoin Cash had taken advantage of patents to the ASICboost technology on the original bitcoin network prior to the fork (see https://bitcoinmagazine.com/articles/breaking-down-bitcoins-asicboost-scandal). This boost saved estimated 20% electricity for miners on 1MB blocks and created unfair economic advantage for this one particular party. SegWit is one way that this boost is being eliminated, through the code. Larger blocks are another way to reduce the boost advantage, via decreased rate of collisions which made this boost happen at the first place (see https://bitcoinmagazine.com/articles/breaking-down-bitcoins-asicboost-scandal-solutions and https://bitslog.wordpress.com/2017/04/10/the-relation-between-segwit-and-asicboost-covert-and-overt). Therefore, the initial Bitcoin Cash proponents argue that eliminating ASICboost through the code is no longer needed or necessary.
Of course, saving any amount electricity between 0% and 20% is good for all on our planet but in reality any energy saved in a mining operation is used by the same mining operation to increase their mining capacity. In reality, there are no savings, there is just capacity redistribution. The question then becomes if it is okay that only one party currently and already holds onto this advantage, which they covertly hid for relatively long time, and which they could be using covertly on Bitcoin Cash if they desired to do so, even though it would an advantage to a smaller degree. To be fair to them, they are mining manufacturers and operators, they researched and developed the advantage from own resources, so perhaps they do indeed have the right to reap ASICboost benefits while they can. But perhaps it should happen in publicly know way, not behind closed doors, and should be temporary, with agreed patent release date.
In conclusion, there is no good and no bad actor, each side is its own shade of grey. All parties have their own truth (and villainy) to certain degree.
Bitcoin Cash's vision is for bitcoin to be an electronic cash platform and daily payment processor whereas Bitcoin Core SegWit seems to be drawn more to the ideas of bitcoin as an investment vehicle and a larger settlement layer with the payment processor function managed via at best decentralized third party hubs. Both can coexist, or either one can eventually prove more useful and digest the other one by taking over all use-cases.
Additionally, the most popular communication channel on /bitcoin with roughly 300k subscribers censors any alternative non-Bitcoin-Core-SegWit opinions and bans people from posting their ideas to discussions (see https://medium.com/@johnblocke/a-brief-and-incomplete-history-of-censorship-in-r-bitcoin-c85a290fe43). This is because their moderators are also supported by Blockstream. Note that the author of this overview has not gotten banned from this particular subreddit (yet), but has experienced shadow-banning first hand. Shadow-banning is a form of censorship. In this particular case, their moderator robot managed by people moderators, collaboratively with the people moderators, do the following:
  • (1) look for "Bitcoin Cash" and other undesirable keywords,
  • (2) warn authors that “Bitcoin Cash” is not true bitcoin (which objectively speaking it is, and which is by no means “BCash” that Bitcoin Core SegWit proponents refer to, in a coordinated effort to further confuse public, especially since some of them have published plans to officially release another cryptocurrency called “BCash” in 2018, see https://medium.com/@freetrade68/announcing-bcash-8b938329eaeb),
  • (3) further warn authors that if they try to post such opinions again, they could banned permanently,
  • (4) tell authors to delete their already posted posts or comments,
  • (5) hide their post from publicly seen boards with all other posts, thus preventing it from being seeing by the other participants in this roughly 300k public forum,
  • (6) and in extreme cases actually “remove” their valid opinions if they slip by uncensored, gain traction, and are often times raise to popularity as comments to other uncensored posts (see /btc/comments/6v3ee8/on_a_reply_i_made_in_rbitcoin_that_had_over_350 and /btc/comments/6vbyv0/in_case_we_needed_more_evidence_500_upvotes).
This effectively silences objective opinions and creates a dangerous echo-chamber. Suppressing free speech and artificially blowing up transaction fees on Bitcoin Core SegWit is against bitcoin’s fundamental values. Therefore, instead of the original Reddit communication channel, many bitcoin enthusiasts migrated to /btc which has roughly 60k subscribers as of now, up from 20k subscribers a year ago in August 2016 (see http://redditmetrics.com/btc). Moderators there do not censor opinions and allow all polite and civil discussions about scaling, including all opinions on Bitcoin Cash, Bitcoin Core, etc.
Looking beyond their respective leaderships and communication channels, let us review a few network fundamentals and recent developments in Bitcoin Core and Bitcoin Cash networks. Consequently, for now, these present Bitcoin Cash with more favorable long-term prospects.
  • (1) The stress-test and/or attack on the Bitcoin Cash mempool earlier on August 16, 2017 showed that 8MB blocks do work as intended, without catastrophic complications that Bitcoin Core proponents anticipated and from which they attempted to discourage others (see https://jochen-hoenicke.de/queue/uahf/#2w for the Bitcoin Cash mempool and https://core.jochen-hoenicke.de/queue/#2w for the Bitcoin Core mempool). Note that when compared to the Bitcoin Core mempool on their respective 2 week views, one can observe how each network handles backlogs. On the most recent 2 week graphs, the Y-scale for Bitcoin Core is 110k vs. 90k on Bitcoin Cash. In other words, at the moment, Bitcoin Cash works better than Bitcoin Core even though there is clearly not as big demand for Bitcoin Cash as there is for Bitcoin Core. The lack of demand for Bitcoin Cash is partly because Bitcoin Cash is only 3 weeks old and not many merchants have started accepting it, and only a limited number of software applications to use Bitcoin Cash has been released so far. By all means, the Bitcoin Cash stress-test and/or attack from August 16, 2017 reveals that the supply will handle the increased demand, more affordably, and at a much quicker rate.
  • (2) Bitcoin Cash “BCH” mining has become temporarily more profitable than mining Bitcoin Core “BTC” (see http://fork.lol). Besides temporary loss of miners, this puts Bitcoin Core in danger of permanently fleeing miners. Subsequently, mempool backlog and transaction fees are anticipated to increase further.
  • (3) When compared to Bitcoin Cash transaction fees at roughly $0.02, transaction fees per kB are over 800 times as expensive on Bitcoin Core, currently at over $16 (see https://cashvscore.com).
  • (4) Tipping service that used to work on Bitcoin Core's /Bitcoin a few years back has been revived by a new tipping service piloted on the more neutral /btc with the integration of Bitcoin Cash (see /cashtipperbot).
3. Should we scale you on-chain or off-chain? Scaling bitcoin.
Let us start with the notion that we are impartial to both Bitcoin Core (small blocks, off-chain scaling only) and Bitcoin Cash (big blocks, on-chain scaling only) schools of thought. We will support any or all ideas, as long as they allow for bitcoin to grow organically and eventually succeed as a peer-to-peer network that remains decentralized, distributed, immutable. Should we have a preference in either of the proposed scaling solutions?
First, let us briefly address Bitcoin Core and small blocks again. From the second section of this overview, we understand that there are proposed off-chain scaling methods via second layer solutions (2L), most notably soon-to-be implemented lightning via SegWit on Bitcoin Core. Unfortunately, the lightning network diminishes distributed and immutable network properties by replacing bitcoin’s peer-to-peer network with a two-layer institution-to-institution network and peer-to-hub-to-peer 2L. Do we need this particular 2L right now? Is its complexity truly needed? Is it not at best somewhat cumbersome (if not very redundant)? In addition to ridiculously high on-chain transaction fees illustrated in the earlier section, the lightning network code is perhaps more robust than it needs to be now, with thousands of lines of code, thus possibly opening up to new vectors for bugs or attacks (see https://en.bitcoin.it/wiki/Lightning_Network and https://github.com/lightningnetwork/lnd). Additionally, this particular 2L as currently designed unnecessarily introduces third parties, hubs, that are expected to centralize. We already have a working code that has been tested and proven to handle 8MB blocks, as seen with Bitcoin Cash on August 16, 2017 (see https://www.cryptocoinsnews.com/first-8mb-bitcoin-cash-block-just-mined). At best, these third party hubs would be decentralized but they would not be distributed. And these hubs would be by no means integral to the original bitcoin network with users, nodes, and miners.
To paraphrase Ocam’s razor problem solving principle, the simplest solution with the most desirable features will prevail (see https://en.wikipedia.org/wiki/Occam%27s_razor). The simplest scalability solution today is Bitcoin Cash because it updates only one line of code, which instantly increases the block size limit. This also allows other companies building on Bitcoin Cash to reduce their codes when compared to Bitcoin Core SegWit’s longer code, some even claiming ten-fold reductions (see /btc/comments/6vdm7y/ryan_x_charles_reveals_bcc_plan). The bitcoin ecosystem not only includes the network but it also includes companies building services on top of it. When these companies can reduce their vectors for bugs or attacks, the entire ecosystem is healthier and more resilient to hacking disasters. Obviously, changes to the bitcoin network code are desirable to be as few and as elegant as possible.
But what are the long-term implications of doing the one-line update repeatedly? Eventually, blocks would have to reach over 500MB size if they were to process Visa-level capacity (see https://en.bitcoin.it/wiki/Scalability). With decreasing costs of IT infrastructure, bandwidth and storage could accommodate it, but the overhead costs would increase significantly, implying miner and/or full node centralization