Bitcoin cash mining. How to mine BCH explained.

Decred Journal — June 2018

Note: You can read this on GitHub, Medium or old Reddit to see the 207 links.

Development

The biggest announcement of the month was the new kind of decentralized exchange proposed by @jy-p of Company 0. The Community Discussions section considers the stakeholders' response.
dcrd: Peer management and connectivity improvements. Some work for improved sighash algo. A new optimization that gives 3-4x faster serving of headers, which is great for SPV. This was another step towards multipeer parallel downloads – check this issue for a clear overview of progress and planned work for next months (and some engineering delight). As usual, codebase cleanup, improvements to error handling, test infrastructure and test coverage.
Decrediton: work towards watching only wallets, lots of bugfixes and visual design improvements. Preliminary work to integrate SPV has begun.
Politeia is live on testnet! Useful links: announcement, introduction, command line voting example, example proposal with some votes, mini-guide how to compose a proposal.
Trezor: Decred appeared in the firmware update and on Trezor website, currently for testnet only. Next steps are mainnet support and integration in wallets. For the progress of Decrediton support you can track this meta issue.
dcrdata: Continued work on Insight API support, see this meta issue for progress overview. It is important for integrations due to its popularity. Ongoing work to add charts. A big database change to improve sorting on the Address page was merged and bumped version to 3.0. Work to visualize agenda voting continues.
Ticket splitting: 11-way ticket split from last month has voted (transaction).
Ethereum support in atomicswap is progressing and welcomes more eyeballs.
decred.org: revamped Press page with dozens of added articles, and a shiny new Roadmap page.
decredinfo.com: a new Decred dashboard by lte13. Reddit announcement here.
Dev activity stats for June: 245 active PRs, 184 master commits, 25,973 added and 13,575 deleted lines spread across 8 repositories. Contributions came from 2 to 10 developers per repository. (chart)

Network

Hashrate: growth continues, the month started at 15 and ended at 44 PH/s with some wild 30% swings on the way. The peak was 53.9 PH/s.
F2Pool was the leader varying between 36% and 59% hashrate, followed by coinmine.pl holding between 18% and 29%. In response to concerns about its hashrate share, F2Pool made a statement that they will consider measures like rising the fees to prevent growing to 51%.
Staking: 30-day average ticket price is 94.7 DCR (+3.4). The price was steadily rising from 90.7 to 95.8 peaking at 98.1. Locked DCR grew from 3.68 to 3.81 million DCR, the highest value was 3.83 million corresponding to 47.87% of supply (+0.7% from previous peak).
Nodes: there are 240 public listening and 115 normal nodes per dcred.eu. Version distribution: 57% on v1.2.0 (+12%), 25% on v1.1.2 (-13%), 14% on v1.1.0 (-1%). Note: the reported count of non-listening nodes has dropped significantly due to data reset at decred.eu. It will take some time before the crawler collects more data. On top of that, there is no way to exactly count non-listening nodes. To illustrate, an alternative data source, charts.dcr.farm showed 690 reachable nodes on Jul 1.
Extraordinary event: 247361 and 247362 were two nearly full blocks. Normally blocks are 10-20 KiB, but these blocks were 374 KiB (max is 384 KiB).

ASICs

Update from Obelisk: shipping is expected in first half of July and there is non-zero chance to meet hashrate target.
Another Chinese ASIC spotted on the web: Flying Fish D18 with 340 GH/s at 180 W costing 2,200 CNY (~340 USD). (asicok.comtranslated, also on asicminervalue)
dcrASIC team posted a farewell letter. Despite having an awesome 16 nm chip design, they decided to stop the project citing the saturated mining ecosystem and low profitability for their potential customers.

Integrations

bepool.org is a new mining pool spotted on dcred.eu.
Exchange integrations:
Two OTC trading desks are now shown on decred.org exchanges page.
BitPro payment gateway added Decred and posted on Reddit. Notably, it is fully functional without javascript or cookies and does not ask for name or email, among other features.
Guarda Wallet integrated Decred. Currently only in their web wallet, but more may come in future. Notable feature is "DCR purchase with a bank card". See more details in their post or ask their representative on Reddit. Important: do your best to understand the security model before using any wallet software.

Adoption

Merchants:
BlueYard Capital announced investment in Decred and the intent to be long term supporters and to actively participate in the network's governance. In an overview post they stressed core values of the project:
There are a few other remarkable characteristics that are a testament to the DNA of the team behind Decred: there was no sale of DCR to investors, no venture funding, and no payment to exchanges to be listed – underscoring that the Decred team and contributors are all about doing the right thing for long term (as manifested in their constitution for the project).
The most encouraging thing we can see is both the quality and quantity of high calibre developers flocking to the project, in addition to a vibrant community attaching their identity to the project.
The company will be hosting an event in Berlin, see Events below.
Arbitrade is now mining Decred.

Events

Attended:
Upcoming:

Media

stakey.club: a new website by @mm:
Hey guys! I'd like to share with you my latest adventure: Stakey Club, hosted at stakey.club, is a website dedicated to Decred. I posted a few articles in Brazilian Portuguese and in English. I also translated to Portuguese some posts from the Decred Blog. I hope you like it! (slack)
@morphymore translated Placeholder's Decred Investment Thesis and Richard Red's write-up on Politeia to Chinese, while @DZ translated Decred Roadmap 2018 to Italian and Russian, and A New Kind of DEX to Italian and Russian.
Second iteration of Chinese ratings released. Compared to the first issue, Decred dropped from 26 to 29 while Bitcoin fell from 13 to 17. We (the authors) restrain ourselves commenting on this one.
Videos:
Audio:
Featured articles:
Articles:

Community Discussions

Community stats: Twitter followers 40,209 (+1,091), Reddit subscribers 8,410 (+243), Slack users 5,830 (+172), GitHub 392 stars and 918 forks of dcrd repository.
An update on our communication systems:
Jake Yocom-Piatt did an AMA on CryptoTechnology, a forum for serious crypto tech discussion. Some topics covered were Decred attack cost and resistance, voting policies, smart contracts, SPV security, DAO and DPoS.
A new kind of DEX was the subject of an extensive discussion in #general, #random, #trading channels as well as Reddit. New channel #thedex was created and attracted more than 100 people.
A frequent and fair question is how the DEX would benefit Decred. @lukebp has put it well:
Projects like these help Decred attract talent. Typically, the people that are the best at what they do aren’t driven solely by money. They want to work on interesting projects that they believe in with other talented individuals. Launching a DEX that has no trading fees, no requirement to buy a 3rd party token (including Decred), and that cuts out all middlemen is a clear demonstration of the ethos that Decred was founded on. It helps us get our name out there and attract the type of people that believe in the same mission that we do. (slack)
Another concern that it will slow down other projects was addressed by @davecgh:
The intent is for an external team to take up the mantle and build it, so it won't have any bearing on the current c0 roadmap. The important thing to keep in mind is that the goal of Decred is to have a bunch of independent teams on working on different things. (slack)
A chat about Decred fork resistance started on Twitter and continued in #trading. Community members continue to discuss the finer points of Decred's hybrid system, bringing new users up to speed and answering their questions. The key takeaway from this chat is that the Decred chain is impossible to advance without votes, and to get around that the forker needs to change the protocol in a way that would make it clearly not Decred.
"Against community governance" article was discussed on Reddit and #governance.
"The Downside of Democracy (and What it Means for Blockchain Governance)" was another article arguing against on-chain governance, discussed here.
Reddit recap: mining rig shops discussion; how centralized is Politeia; controversial debate on photos of models that yielded useful discussion on our marketing approach; analysis of a drop in number of transactions; concerns regarding project bus factor, removing central authorities, advertising and full node count – received detailed responses; an argument by insette for maximizing aggregate tx fees; coordinating network upgrades; a new "Why Decred?" thread; a question about quantum resistance with a detailed answer and a recap of current status of quantum resistant algorithms.
Chats recap: Programmatic Proof-of-Work (ProgPoW) discussion; possible hashrate of Blake-256 miners is at least ~30% higher than SHA-256d; how Decred is not vulnerable to SPV leaf/node attack.

Markets

DCR opened the month at ~$93, reached monthly high of $110, gradually dropped to the low of $58 and closed at $67. In BTC terms it was 0.0125 -> 0.0150 -> 0.0098 -> 0.0105. The downturn coincided with a global decline across the whole crypto market.
In the middle of the month Decred was noticed to be #1 in onchainfx "% down from ATH" chart and on this chart by @CoinzTrader. Towards the end of the month it dropped to #3.

Relevant External

Obelisk announced Launchpad service. The idea is to work with coin developers to design a custom, ASIC-friendly PoW algorithm together with a first batch of ASICs and distribute them among the community.
Equihash-based ZenCash was hit by a double spend attack that led to a loss of $450,000 by the exchange which was targeted.
Almost one year after collecting funds, Tezos announced a surprise identification procedure to claim tokens (non-javascript version).
A hacker broke into Syscoin's GitHub account and implanted malware stealing passwords and private keys into Windows binaries. This is a painful reminder for everybody to verify binaries after download.
Circle announced new asset listing framework for Poloniex. Relevant to recent discussions of exchange listing bribery:
Please note: we will not accept any kind of payment to list an asset.
Bithumb got hacked with a $30 m loss.
Zcash organized Zcon0, an event in Canada that focused on privacy tech and governance. An interesting insight from Keynote Panel on governance: "There is no such thing as on-chain governance".
Microsoft acquired GitHub. There was some debate about whether it is a reason to look into alternative solutions like GitLab right now. It is always a good idea to have a local copy of Decred source code, just in case.
Status update from @sumiflow on correcting DCR supply on various sites:
To begin with, none of the below sites were showing the correct supply or market cap for Decred but we've made some progress. coingecko.com, coinlib.io, cryptocompare.com, livecoinwatch.com, worldcoinindex.com - corrected! cryptoindex.co, onchainfx.com - awaiting fix coinmarketcap.com - refused to fix because devs have coins too? (slack)

About This Issue

This is the third issue of Decred Journal after April and May.
Most information from third parties is relayed directly from source after a minimal sanity check. The authors of Decred Journal have no ability to verify all claims. Please beware of scams and do your own research.
The new public Matrix logs look promising and we hope to transition from Slack links to Matrix links. In the meantime, the way to read Slack links is explained in the previous issue.
As usual, any feedback is appreciated: please comment on Reddit, GitHub or #writers_room. Contributions are welcome too, anything from initial collection to final review to translations.
Credits (Slack names, alphabetical order): bee and Richard-Red. Special thanks to @Haon for bringing May 2018 issue to medium.
submitted by jet_user to decred [link] [comments]

Read: An open letter to bitcoin miner

https://keepingstock.net/an-open-letter-to-bitcoin-miners-c260467e1f0
Dear Bitcoin Miner,
My name is Jonald, and I am a Bitcoin investor.
I bought my first Bitcoins in 2013 and have been active on the Bitcointalk forum since March, 2014. I’m also a small business owner that actually uses Bitcoin for
Since Bitcoin investors and miners need each other to succeed, I wanted to take a minute to reach out to you, and send a sincere message from a “real Bitcoiner”. I’ll cut right to the chase:
I’m concerned. I believe we urgently need to find a scaling solution, and I believe the best solution is to increase the blocksize.
At least, hear me out.
Why Should You Listen to Me?
There’s a huge amount of misinformation, dishonesty, and political agendas attached to the Great Scaling Debate. The situation is serious and there’s a lot at stake here.
I am not beholden to any special interests. No one is paying me to write this. I am not a contributor to any Bitcoin projects, but I am quite familiar with the scaling topic because I’ve been following it for some time now, and I am knowledgeable enough to clearly understand the technical details.
I’ve heard all the arguments from every side of the debate, and I want to give you my honest, unbiased, unfiltered understanding of the situation.
Let’s Start At the Beginning
In 2008, Satoshi Nakamoto published a paper titled Bitcoin: A Peer-to-Peer Electronic Cash System. Everybody knows this, but the exact title needs to be repeated because today, even the most basic facets of Bitcoin are being challenged. Should Bitcoin really be “cash” or instead “digital gold”? And if we follow Satoshi’s plan, is it really peer to peer?
These questions come not so much from open-minded inquiry, but rather from a biased agenda. This would have been inconceivable a few years ago, but now things have become so political, that certain people even want to re-write the Bitcoin whitepaper.
(Attempting to re-write history has always been a favorite tactic of tyrannical elites.)
Satoshi’s Vision to Scale Bitcoin
Regardless of “which side” of the scaling debate you are on, it should not be contested that Satoshi always planned for and advocated for simple, on-chain scaling.
When asked how Bitcoin would scale to Visa-like levels, he said:
Long before the network gets anywhere near as large as that, it would be safe for users to use Simplified Payment Verification (section 8) to check for double spending, which only requires having the chain of block headers, or about 12KB per day. Only people trying to create new coins would need to run network nodes. At first, most users would run network nodes, but as the network grows beyond a certain point, it would be left more and more to specialists with server farms of specialized hardware. A server farm would only need to have one node on the network and the rest of the LAN connects with that one node.
The bandwidth might not be as prohibitive as you think. A typical transaction would be about 400 bytes (ECC is nicely compact). Each transaction has to be broadcast twice, so lets say 1KB per transaction. Visa processed 37 billion transactions in FY2008, or an average of 100 million transactions per day. That many transactions would take 100GB of bandwidth, or the size of 12 DVD or 2 HD quality movies, or about $18 worth of bandwidth at current prices.
If the network were to get that big, it would take several years, and by then, sending 2 HD movies over the Internet would probably not seem like a big deal.
Satoshi Nakamoto
Source
Disturbingly, this simple quote from Satoshi was moderated (deleted) from the bitcoin reddit page. I’ll revisit the censorship issue in a moment.
Another important fact is that the current blocksize limit of 1mb was intended to be a temporary measure. This was something ‘everyone’ knew before the debate became politicized.
One of the earliest code reviewers, Ray Dillinger, explained that he, Hal Finey, and Satoshi all agreed the limit was to be temporary.
Satoshi also provided the means to raise the limit with his famous quote:
It can be phased in, like:
if (blocknumber > 115000) maxblocksize = largerlimit
Here is one more explanation from Satoshi, in an email to Mike Hearn, about why Bitcoin never hits a scaling ceiling.
Sure, Satoshi isn’t God. The point isn’t to appeal to his authority, but simply to remember that Bitcoin always had a scaling plan in place from the beginning.
…But the “Core Devs” Had Other Ideas.
The history of the current crop of Bitcoin Core developers has been already summarized and described elsewhere.
Explanations have been given for the unproductive scaling conferences, the broken Hong Kong agreements, and so on, but it should be extremely clear to everyone, based on years of their behavior (and even their own words), that the Core group does not want to scale Bitcoin with a simple blocksize increase.
In fact, they (and their supporters) have done everything in their power to prevent this, including engaging in massive censorship.
Their primary arguments are as follows:
It is problematic to raise the limit because it requires a hard fork, which is difficult to coordinate.Bitcoin nodes should be as inexpensive to run as possible, otherwise the decentralization of Bitcoin will be threatened.Without a constraint on the blocksize, Bitcoin won’t be secure once subsidies (block rewards) decline.
None of These Arguments Have Sufficient Merit to Forestall a Blocksize Increase
I am not saying the arguments are entirely without merit. Few things in life are ever 100% black-and-white. But we have to weigh the merits of these positions against the alternatives, and against other factors in the Bitcoin ecosystem.
Let’s take one at a time:
The “Hard Forks Are Dangerous” Myth
This was a prominent talking point in 2014–2015. However, the truth is that hard forks (HF) are not necessarily dangerous, especially if they occur with a clear majority of hashing power supporting the upgraded consensus rules.
The previous group of developers, including Gavin Andresen, Jeff Garzik, and Mike Hearn, all supported upgrading Bitcoin with hard forks.
Initially, the discussion was whether the new maximum blocksize would be 2MB, 4MB, or 8MB. What begin as a minor difference of opinions between the miners somehow snowballed into a potent meme that consensus over scaling was going to be difficult.
The developers starting adding their own opinions about hard forks, creating additional friction. Yes, it is easy to claim there is contention when you are among those being contentious!
Core has no official leadership positions or governance structure. Because of this, it has been easy to justify inaction by simply concluding that “there’s no consensus”. And since they control the reference code repository, their refusal to raise the limit effects everyone else.
In practice, Core does have leaders. How else can it be explained that segwit was merged into the code (even if not activated) with practically no public debate whatsoever?
On a side note, prominent Core developers have denied that Core decides what code is published, and have denied there is any leadership. This is an example of the kind of constant misinformation that is being generated on a daily basis.
Back to the HF issue:
Many altcoins like Monero have regular hard forks. Coordination between major players in an ecosystem is not a big challenge if everyone is on the same page.
So far, I have not heard of a single problem that an altcoin had in performing a network upgrade via hard fork. So, there is evidence that they can be done safely.
In addition, if Core admits in their roadmap that eventually the blocksize will need to be increased, then why not do it now when it is badly needed? There is no logical reason why it would be more risky now rather than later.
Decentralization Myths
There are actually several myths surrounding the issue of decentralization. Let’s address the obvious ones:
The most ludicrous is the “all users should be running full nodes” idea.
As others have explained, there is no security provided to the network by non-mining ‘full nodes’. Only mining nodes secure and extend Bitcon’s distributed ledger.
The white paper explains why most users do not need to run full nodes:
It is possible to verify payments without running a full network node. A user only needs to keep a copy of the block headers of the longest proof-of-work chain, which he can get by querying network nodes until he’s convinced he has the longest chain, and obtain the Merkle branch linking the transaction to the block it’s timestamped in. He can’t check the transaction for himself, but by linking it to a place in the chain, he can see that a network node has accepted it, and blocks added after it further confirm the network has accepted it… …Businesses that receive frequent payments will probably still want to run their own nodes for more independent security and quicker verification.
The idea that a lot of non-mining full nodes will make the network more decentralized (because they can make sure the miners are behaving) is erroneous, because an SPV client can already query the network’s nodes. Generally, there would only be a problem if a majority mining of nodes were colluding dishonestly, in which case Bitcoin would be already broken.
A more valid concern is that as nodes become more expensive, eventually only large corporations will run nodes. It is true that node costs will increase over time as the network grows. However, storage, bandwidth, and processing capabilities are also constantly increasing.
Just as important: By the time that capacity increases — lets say from 3 TPS (transactions per second) to 30 TPS — the network will be so large that it likely won’t be any less decentralized, even if it costs more to run a node.
At 3000 TPS, Bitcoin would be highly dominant globally, and making use of the millions of datacenters and servers available worldwide. This was always the plan.
The Alternative Vision of Bitcoin Holds Decentralization Risks That Are Worse
Many users are not aware of the decentralization risks that come with the small-node/small-block vision of Bitcoin. Core’s vision for Bitcoin is to transform the peer-to-peer cash system into some kind of settlement network.
While this would be a way to keep node costs minimal, most users would be economically forced off the main chain because they cannot compete with institutions for fees. They would then need to get permission from trusted third parties to transact.
In my opinion, this represents a much more dangerous form of centralization than bigger blocks and expensive nodes.
The Fee-Market Failure Myth
The third primary argument of the small-block philosophy is that eventually, block rewards will run out, and mining fees will be the sole source of funding security. They then claim that without limiting the supply of transaction space, miners will be hopelessly caught in a tragedy-of-the-commons price war, with the users paying rock bottom fees, leading to a collapse of commercial mining.
There’s a few problems with this argument.
First of all, there is a natural market for every good and service in the world. There have been many price wars, but nothing with high demand ever stops being produced.
The concern that the network hashrate will become too low is based on several assumptions and variables, including the number of daily transactions, the willingness of the users to wait for confirmations, the willingness of the users to pay small amounts, the behavior of the miners, the fee policies set by various wallets, the emergent consensus on acceptable fees by the mining community, and other factors, including what actually is “too low” of a network hashrate in the first place.
The hypothetical failure of the natural fee market depends on all these assumptions combining into an unfavorable outcome, as well as the inability of the system to adjust itself favorably using any of these factors.
But, by far the biggest reason that this argument is bunk, is that it will be decades before the majority of the subsidies actually disappear.
Pure Foolishness: Overplanning the Future While Ignoring Urgent Issues Today
Why implement a plan that might help Bitcoin in 20–30 years, if it requires you to damage the user experience and erode the adoption and network effect of Bitcoin, today?
In the case of Bitcoin, it’s completely unnecessary to plan ahead that far, and the destructive consequences are already being seen.
This is the biggest reason why Core’s position should be considered indefensible. Even if their arguments have merit, it is more important to keep Bitcoin healthy right now, stay competitive, and keep the user base growing than to prevent the problems that may or may not happen later.
Even worse, those prevention plans work in direct opposition to the short term goals!
It is no less insane than demanding a bedridden hospital patient, badly in need of rest, to immediately go outside and start running laps because “exercise will help you live longer”.
What About Segwit?
It is my understanding that at “the Hong Kong meeting”, the miners agreed to Segwit PLUS a hardfork blocksize increase because they didn’t trust the Core team enough to offer satisfactory scaling in a timely manner.
I think their decision was smart. Core cannot be trusted. However, if Core changed their mind today, and agreed to the 2MB+Segwit, I would support that as a compromise to break the impasse.
They seem to be unwilling to do this.
Since miners are unwilling to accept segwit on its own, and since Core will not compromise, the only logical alternative is bigger blocks, which is the best option regardless.
What Core Wants
You may be wondering: How is it possible for people as intelligent as the Bitcoin Core developers to fail to see the obvious mistakes in their thinking?
American author Upton Sinclair’s famous quote comes to mind:
It is difficult to get a man to understand something, when his salary depends upon his not understanding it!
The Core team and their supporters want to change Bitcoin into a settlement network. They will deny this, but in my opinion, all of their actions point to this logical conclusion.
This is why they are against on chain scaling, and why segwit offers as little of it as possible while supporting their “HF are bad” narrative.
Additionally, I believe they also want to control public opinion by employing key individuals, by their associates and moderation policies on various platforms, and with an army of trolls.
They also intimidate and punish businesses that don’t fall in line. For example, coinbase.com was delisted from bitcoin.org for supporting Bitcoin XT instead of the Core client.
Despite these shenanigans, companies do support bigger blocks and on-chain scaling.
Most importantly, they want to scare you, the miner, into believing that the community doesn’t really want big blocks and if that if you mine big blocks, you’ll be forked off to a worthless coin and left with worthless ASICs.
Do not let them intimidate you.
What the Users Want
Most users just want a Bitcoin that works. They do not want slow confirmation and high fees. Most Bitcoiners that use bitcoin frequently understand the issues and support bigger blocks.
Despite all the trolling and propaganda, users controlling actual coins vote overwhelmingly in favor of Satoshi’s scaling plan.
The “Healthy Fee Market” is Already Unhealthy
Even IF a centrally planned fee market was a good idea right now, it is being managed poorly. A “healthy” fee market should strive to provide adequate fee revenue while at the same time provide a good user experience and promote growth of the network and user base.
While miner revenue is certainly adequate, the user experience is severely degraded because of slow confirmations and high fees, and this is definitely not attractive or conducive to growing the user base.
If keeping the blocksize at 1mb was an experiment to see how the fee market would develop, it has already played out its usefulness. To keep fees at a level competitive with other coins, supply must catch up with demand (we must raise the blocksize) . But these developers seem to have no interest in doing so. They would rather carry on with their agenda than serve the users.
What About Bitcoin As a Store-of-Value or as “Digital Gold”?
The great thing about Bitcoin is that it can be both a cash-like payment system and a gold-like store of value. These two aspects enhance each other.
Exposed to the propaganda that Bitcoin can’t scale as electronic cash, some users have said “that’s ok. I’m fine with Bitcoin being digital gold only”. The problem with this thinking is that Bitcoin has competition.
If another coin is useful to store value AND to transact cheaply with, it severely undermines Bitcon’s appeal to investors. At the same time, it greatly dampens demand for actual usage.
Sure, its possible that Bitcoin could survive in some form as digital gold, but it would be at a huge disadvantage.
Small Blocks Destroy Miner Revenue
At first glance, the idea that smaller blocks are bad for mining revenue may appear incorrect, since fee rates have recently exploded based on the demand of Bitcoin transactions outpacing the supply of space in the blocks.
However, this trend cannot continue for long, since users will only pay so much. At the same time, new users and new demand are being shut out from the ecosystem.
To use an analogy: Who makes more money — the farmer in town “A” selling milk from one cow? Or the farmer in town “B” selling milk from 8 cows? Townspeople in “A” might pay more per bottle, but they’ll only pay so much for it. They will start drinking something else, drink milk less often, or import their milk from another town.
Bitcoin miners simply cannot meet the demands of users at fees they are willing to reasonably pay if blocks are restricted to 1mb… and users will find satisfactory alternatives which are quickly becoming abundant.
The situation will become even worse in the long run if Core is allowed to create “second layer solutions”, because those solutions will probably not be free, and they will further absorb the money that users are willing to spend in order to transact.
This will be bad for miners, and bad for network security. It will make bitcoin even less competitive, and money will leave the ecosystem.
Price Always Lags Behind Fundamentals
It is easy to look at a high Bitcoin price and think that everything is fine. If things were going so badly, why isn’t the price dropping?
But, price doesn’t always reflect the underlying fundamentals of a market in the short term.
In the long run, fundamentals always dictate the direction of the market. Daytraders are flat at the end of the day. Speculators come and go. In the end, it’s only the long term investors and the non-speculative demand that determines the price.
The fundamental value of Bitcoin primarily comes from its usefulness as a payment system. If that system ceases to be useful, Bitcoin will cease to be valuable.
Time To Act. Let’s Help Bitcoin Grow Again.
It’s always better to fix a problem BEFORE it gets too big. As they say, “an ounce of prevention is worth a pound of cure”.
If we wait until the Bitcoin price crashes because Bitcoin is unusable as a currency, it will be too late. We would have already lost serious momentum, marketshare, users, reputation, and merchants.
This is already happening, but there is still time to act.
I urge you: don’t be complacent.
You are the miner. You have the power. Start signaling for bigger blocks today, and let’s make sure Bitcoin stays #1.
Help Spread the Word
If you’re not a miner, but a concerned investor like myself, then please spread this message far and wide, and ask the miners and pools that you know for bigger blocks.
Addendum
This article is available in several foreign language formats:
Chinese Spanish Japanese German Russian
https://keepingstock.net/an-open-letter-to-bitcoin-miners-c260467e1f0
submitted by german_bitcoiner to btc [link] [comments]

On the new batch of comments to the SEC about the SolidX ETF, some honorable mentions, and some negative comments

The SEC just posted a new batch of 286 comments on the SolidX ETF, bringing the total to 1147. I am skimming through them and posted some of the best already to this sub.
The vast majority are short comments, obviously submitted in response to some mail-in campaign. The names sound very much like the invented ones of spam emails that I have been receiving for years. A telling detail is the lack of a middle initial.
They also mostly repeat the same arguments, and many are obviously written by people who don't understand what is the ETF, only that if that SEC thing approves it then the bitcoin price will go to the moon. I have just seen a dozen that start with the same phrase "I hearby[sic] state my acceptance and full support..."
Some are so sloppy that they submit with one name but sign with a different name.
Here are some honorable mentions:
A few negative comments:
submitted by jstolfi to Buttcoin [link] [comments]

Decred Journal – September 2018

Note: you can read this on GitHub (link), Medium (link) or old Reddit (link).

Development

Final version 1.3.0 of the core software was released bringing all the enhancements reported last month to the rest of the community. The groundwork for SPV (simplified payment verification) is complete, another reduction of fees is being deployed, and performance stepped up once again with a 50% reduction in startup time, 20% increased sync speed and more than 3x faster peer delivery of block headers (a key update for SPV). Decrediton's integrations of SPV and Politeia are open for testing by experienced users. Read the full release notes and get the downloads on GitHub. As always, don't forget to verify signatures.
dcrd: completed several steps towards multipeer downloads, improved introduction to the software in the main README, continued porting cleanups and refactoring from upstream btcd.
Currently in review are initial release of smart fee estimator and a change to UTXO set semantics. The latter is a large and important change that provides simpler handling, and resolves various issues with the previous approach. A lot of testing and careful review is needed so help is welcome.
Educational series for new Decred developers by @matheusd added two episodes: 02 Simnet Setup shows how to automate simnet management with tmux and 03 Miner Reward Invalidation explains block validity rules.
Finally, a pull request template with a list of checks was added to help guide the contributors to dcrd.
dcrwallet: bugfixes and RPC improvements to support desktop and mobile wallets.
Developers are welcome to comment on this idea to derive stakepool keys from the HD wallet seed. This would eliminate the need to backup and restore redeem scripts, thus greatly improving wallet UX. (missed in July issue)
Decrediton: bugfixes, refactoring to make the sync process more robust, new loading animations, design polishing.
Politeia: multiple improvements to the CLI client (security conscious users with more funds at risk might prefer CLI) and security hardening. A feature to deprecate or timeout proposals was identified as necessary for initial release and the work started. A privacy enhancement to not leak metadata of ticket holders was merged.
Android: update from @collins: "Second test release for dcrandroid is out. Major bugs have been fixed since last test. Latest code from SPV sync has been integrated. Once again, bug reports are welcome and issues can be opened on GitHub". Ask in #dev room for the APK to join testing.
A new security page was added that allows one to validate addresses and to sign/verify messages, similar to Decrediton's Security Center. Work on translations is beginning.
Overall the app is quite stable and accepting more testers. Next milestone is getting the test app on the app store.
iOS: the app started accepting testers last week. @macsleven: "the test version of Decred Wallet for iOS is available, we have a link for installing the app but the builds currently require your UDID. Contact either @macsleven or @raedah with your UDID if you would like to help test.".
Nearest goal is to make the app crash free.
Both mobile apps received new design themes.
dcrdata: v3.0 was released for mainnet! Highlights: charts, "merged debits" view, agendas page, Insight API support, side chain tracking, Go 1.11 support with module builds, numerous backend improvements. Full release notes here. This release featured 9 contributors and development lead @chappjc noted: "This collaboration with @raedahgroup on our own block explorer and web API for @decredproject has been super productive.".
Up next is supporting dynamic page widths site wide and deploying new visual blocks home page.
Trezor: proof of concept implementation for Trezor Model T firmware is in the works (previous work was for Model One).
Ticket splitting: updated to use Go modules and added simnet support, several fixes.
docs: beginner's guide overhaul, multiple fixes and cleanups.
decred.org: added 3rd party wallets, removed inactive PoW pools and removed web wallet.
@Richard-Red is building a curated list of Decred-related GitHub repositories.
Welcome to new people contributing for the first time: @klebe, @s_ben, @victorguedes, and PrimeDominus!
Dev activity stats for September: 219 active PRs, 197 commits, 28.7k added and 18.8k deleted lines spread across 6 repositories. Contributions came from 4-10 developers per repository. (chart)

Network

Hashrate: started and ended the month around 75 PH/s, hitting a low of 60.5 and a new high of 110 PH/s. BeePool is again the leader with their share varying between 23-54%, followed by F2Pool 13-30%, Coinmine 4-6% and Luxor 3-5%. As in previous months, there were multiple spikes of unidentified hashrate.
Staking: 30-day average ticket price is 98 DCR (+2.4). The price varied between 95.7 and 101.9 DCR. Locked DCR amount was 3.86-3.96 million DCR, or 45.7-46.5% of the supply.
Nodes: there are 201 public listening nodes and 325 normal nodes per dcred.eu. Version distribution: 5% are v1.4.0(pre) dev builds (+3%), 30% on v1.3.0 (+25%), 42% on v1.2.0 (-20%), 15% on v1.1.2 (-7%), 6% on v1.1.0. More than 76% of nodes run v1.2.0 and higher and therefore support client filters. Data as of Oct 1.

ASICs

Obelisk posted two updates on their mailing list. 70% of Batch 1 units are shipped, an extensive user guide is available, Obelisk Scanner application was released that allows one to automatically update firmware. First firmware update was released and bumped SC1 hashrate by 10-20%, added new pools and fixed multiple bugs. Next update will focus on DCR1. It is worth a special mention that the firmware source code is now open! Let us hope more manufacturers will follow this example.
A few details about Whatsminer surfaced this month. The manufacturer is MicroBT, also known as Bitwei and commonly misspelled as Bitewei. Pangolinminer is a reseller, and the model name is Whatsminer D1.
Bitmain has finally entered Decred ASIC space with their Antminer DR3. Hash rate is 7.8 TH/s while pulling 1410 W, at the price of $673. These specs mean it has the best GH/W and GH/USD of currently sold miners until the Whatsminer or others come out, although its GH/USD of 11.6 already competes with Whatsminer's 10.5. Discussed on Reddit and bitcointalk, unboxing video here.

Integrations

Meet our 17th voting service provider: decredvoting.com. It is operated by @david, has 2% fee and supports ticket splitting. Reddit thread is here.
For a historical note, the first VSP to support ticket splitting was decredbrasil.com:
@matheusd started tests on testnet several months ago. I contacted him so we could integrate with the pool in June this year. We set up the machine in July and bought the first split ticket on mainnet, using the decredbrasil pool, on July 19. It was voted on July 30. After this first vote on mainnet, we opened the tests to selected users (with more technical background) on the pool. In August we opened the tests to everyone, and would call people who want to join to the #ticket_splitting channel, or to our own Slack (in Portuguese, so mostly Brazilian users). We have 28 split tickets already voted, and 16 are live. So little more than 40 split tickets total were bought on decredbrasil pool. (@girino in #pos-voting)
KuCoin exchange listed DCBTC and DCETH pairs. To celebrate their anniversary they had a 99% trading fees discount on DCR pairs for 2 weeks.
Three more wallets integrated Decred in September:
ChangeNow announced Decred addition to their Android app that allows accountless swaps between 150+ assets.
Coinbase launched informational asset pages for top 50 coins by market cap, including Decred. First the pages started showing in the Coinbase app for a small group of testers, and later the web price dashboard went live.

Adoption

The birth of a Brazilian girl was registered on the Decred blockchain using OriginalMy, a blockchain proof of authenticity services provider. Read the full story in Portuguese and in English.

Marketing

Advertising report for September is ready. Next month the graphics for all the ads will be changing.
Marketing might seem quiet right now, but a ton is actually going on behind the scenes to put the right foundation in place for the future. Discovery data are being analyzed to generate a positioning strategy, as well as a messaging hierarchy that can guide how to talk about Decred. This will all be agreed upon via consensus of the community in the work channels, and materials will be distributed.
Next, work is being done to identify the right PR partner to help with media relations, media training, and coordination at events. While all of this is coming up to speed, we believe the website needs a refresher reflecting the soon to be agreed upon messaging, plus a more intuitive architecture to make it easier to navigate. (@Dustorf)

Events

Attended:
Upcoming:
We'll begin shortly reviewing conferences and events planned for the first half of 2019. Highlights are sure to include The North American Bitcoin Conference in Miami (Jan 16-18) and Consensus in NYC (May 14-16). If you have suggestions of events or conferences Decred should attend, please share them in #event_planning. In 2019, we would like to expand our presence in Europe, Asia, and South America, and we're looking for community members to help identify and staff those events. (@Dustorf)

Media

August issue of Decred Journal was translated to Russian. Many thanks to @DZ!
Rency cryptocurrency ratings published a report on Decred and incorporated a lot of feedback from the community on Reddit.
September issue of Chinese CCID ratings was published (snapshot), Decred is still at the bottom.
Videos:
Featured articles:
Articles:

Community Discussions

Community stats:
Comm systems news: Several work channels were migrated to Matrix, #writers_room is finally bridged.
Highlights:
Twitter: why decentralized governance and funding are necessary for network survival and the power of controlling the narrative; learning about governance more broadly by watching its evolution in cryptocurrency space, importance of community consensus and communications infrastructure.
Reddit: yet another strong pitch by @solar; question about buyer protections; dcrtime internals; a proposal to sponsor hoodies in the University of Cape Town; Lightning Network support for altcoins.
Chats: skills to operate a stakepool; voting details: 2 of 3 votes can approve a block, what votes really approve are regular tx, etc; scriptless script atomic swaps using Schnorr adaptor signatures; dev dashboard, choosing work, people do best when working on what interests them most; opportunities for governments and enterprise for anchoring legal data to blockchain; terminology: DAO vs DAE; human-friendly payments, sharing xpub vs payment protocols; funding btcsuite development; Politeia vote types: approval vote, sentiment vote and a defund vote, also linking proposals and financial statements; algo trading and programming languages (yes, on #trading!); alternative implementation, C/C++/Go/Rust; HFTs, algo trading, fake volume and slippage; offline wallets, usb/write-only media/optical scanners vs auditing traffic between dcrd and dcrwallet; Proof of Activity did not inspire Decred but spurred Decred to get moving, Wikipedia page hurdles; how stakeholders could veto blocks; how many votes are needed to approve a proposal; why Decrediton uses Electron; CVE-2018-17144 and over-dependence on single Bitcoin implementation, btcsuite, fuzz testing; tracking proposal progress after voting and funding; why the wallet does not store the seed at all; power connectors, electricity, wiring and fire safety; reasonable spendings from project fund; ways to measure sync progress better than block height; using Politeia without email address; concurrency in Go, locks vs channels.
#support is not often mentioned, but it must be noted that every day on this channel people get high quality support. (@bee: To my surprise, even those poor souls running Windows 10. My greatest respect to the support team!)

Markets

In September DCR was trading in the range of USD 34-45 / BTC 0.0054-0.0063. On Sep 6, DCR revisited the bottom of USD 34 / BTC 0.0054 when BTC quickly dropped from USD 7,300 to 6,400. On Sep 14, a small price rise coincided with both the start of KuCoin trading and hashrate spike to 104 PH/s. Looking at coinmarketcap charts, the trading volume is a bit lower than in July and August.
As of Oct 4, Decred is #18 by the number of daily transactions with 3,200 tx, and #9 by the USD value of daily issuance with $230k. (source: onchainfx)
Interesting observation by @ImacallyouJawdy: while we sit at 2018 price lows the amount locked in tickets is testing 2018 high.

Relevant External

ASIC for Lyra2REv2 was spotted on the web. Vertcoin team is preparing a new PoW algorithm. This would be the 3rd fork after two previous forks to change the algorithm in 2014 and 2015.
A report titled The Positive Externalities of Bitcoin Mining discusses the benefits of PoW mining that are often overlooked by the critics of its energy use.
A Brief Study of Cryptonetwork Forks by Alex Evans of Placeholder studies the behavior of users, developers and miners after the fork, and makes the cases that it is hard for child chains to attract users and developers from their parent chains.
New research on private atomic swaps: the paper "Anonymous Atomic Swaps Using Homomorphic Hashing" attempts to break the public link between two transactions. (bitcointalk, decred)
On Sep 18 Poloniex announced delisting of 8 more assets. That day they took a 12-80% dive showing their dependence on this one exchange.
Circle introduced USDC markets on Poloniex: "USDC is a fully collateralized US dollar stablecoin using the ERC-20 standard that provides detailed financial and operational transparency, operates within the regulated framework of US money transmission laws, and is reinforced by established banking partners and auditors.".
Coinbase announced new asset listing process and is accepting submissions on their listing portal. (decred)
The New York State Office of the Attorney General posted a study of 13 exchanges that contains many insights.
A critical vulnerability was discovered and fixed in Bitcoin Core. Few days later a full disclosure was posted revealing the severity of the bug. In a bitcointalk thread btcd was called 'amateur' despite not being vulnerable, and some Core developers voiced their concerns about multiple implementations. The Bitcoin Unlimited developer who found the bug shared his perspective in a blog post. Decred's vision so far is that more full node implementations is a strength, just like for any Internet protocol.

About This Issue

This is the 6th issue of Decred Journal. It is mirrored on GitHub, Medium and Reddit. Past issues are available here.
Most information from third parties is relayed directly from source after a minimal sanity check. The authors of Decred Journal have no ability to verify all claims. Please beware of scams and do your own research.
Feedback is appreciated: please comment on Reddit, GitHub or #writers_room on Matrix or Slack.
Contributions are also welcome: some areas are adding content, pre-release review or translations to other languages.
Credits (Slack names, alphabetical order): bee, Dustorf, jz, Haon, oregonisaac, raedah and Richard-Red.
submitted by jet_user to decred [link] [comments]

Read: An open letter to bitcoin miner

https://keepingstock.net/an-open-letter-to-bitcoin-miners-c260467e1f0
Dear Bitcoin Miner,
My name is Jonald, and I am a Bitcoin investor.
I bought my first Bitcoins in 2013 and have been active on the Bitcointalk forum since March, 2014. I’m also a small business owner that actually uses Bitcoin for
Since Bitcoin investors and miners need each other to succeed, I wanted to take a minute to reach out to you, and send a sincere message from a “real Bitcoiner”. I’ll cut right to the chase:
I’m concerned. I believe we urgently need to find a scaling solution, and I believe the best solution is to increase the blocksize.
At least, hear me out.
Why Should You Listen to Me?
There’s a huge amount of misinformation, dishonesty, and political agendas attached to the Great Scaling Debate. The situation is serious and there’s a lot at stake here.
I am not beholden to any special interests. No one is paying me to write this. I am not a contributor to any Bitcoin projects, but I am quite familiar with the scaling topic because I’ve been following it for some time now, and I am knowledgeable enough to clearly understand the technical details.
I’ve heard all the arguments from every side of the debate, and I want to give you my honest, unbiased, unfiltered understanding of the situation.
Let’s Start At the Beginning
In 2008, Satoshi Nakamoto published a paper titled Bitcoin: A Peer-to-Peer Electronic Cash System. Everybody knows this, but the exact title needs to be repeated because today, even the most basic facets of Bitcoin are being challenged. Should Bitcoin really be “cash” or instead “digital gold”? And if we follow Satoshi’s plan, is it really peer to peer?
These questions come not so much from open-minded inquiry, but rather from a biased agenda. This would have been inconceivable a few years ago, but now things have become so political, that certain people even want to re-write the Bitcoin whitepaper.
(Attempting to re-write history has always been a favorite tactic of tyrannical elites.)
Satoshi’s Vision to Scale Bitcoin
Regardless of “which side” of the scaling debate you are on, it should not be contested that Satoshi always planned for and advocated for simple, on-chain scaling.
When asked how Bitcoin would scale to Visa-like levels, he said:
Long before the network gets anywhere near as large as that, it would be safe for users to use Simplified Payment Verification (section 8) to check for double spending, which only requires having the chain of block headers, or about 12KB per day. Only people trying to create new coins would need to run network nodes. At first, most users would run network nodes, but as the network grows beyond a certain point, it would be left more and more to specialists with server farms of specialized hardware. A server farm would only need to have one node on the network and the rest of the LAN connects with that one node.
The bandwidth might not be as prohibitive as you think. A typical transaction would be about 400 bytes (ECC is nicely compact). Each transaction has to be broadcast twice, so lets say 1KB per transaction. Visa processed 37 billion transactions in FY2008, or an average of 100 million transactions per day. That many transactions would take 100GB of bandwidth, or the size of 12 DVD or 2 HD quality movies, or about $18 worth of bandwidth at current prices.
If the network were to get that big, it would take several years, and by then, sending 2 HD movies over the Internet would probably not seem like a big deal.
Satoshi Nakamoto
Source
Disturbingly, this simple quote from Satoshi was moderated (deleted) from the bitcoin reddit page. I’ll revisit the censorship issue in a moment.
Another important fact is that the current blocksize limit of 1mb was intended to be a temporary measure. This was something ‘everyone’ knew before the debate became politicized.
One of the earliest code reviewers, Ray Dillinger, explained that he, Hal Finey, and Satoshi all agreed the limit was to be temporary.
Satoshi also provided the means to raise the limit with his famous quote:
It can be phased in, like:
if (blocknumber > 115000) maxblocksize = largerlimit
Here is one more explanation from Satoshi, in an email to Mike Hearn, about why Bitcoin never hits a scaling ceiling.
Sure, Satoshi isn’t God. The point isn’t to appeal to his authority, but simply to remember that Bitcoin always had a scaling plan in place from the beginning.
…But the “Core Devs” Had Other Ideas.
The history of the current crop of Bitcoin Core developers has been already summarized and described elsewhere.
Explanations have been given for the unproductive scaling conferences, the broken Hong Kong agreements, and so on, but it should be extremely clear to everyone, based on years of their behavior (and even their own words), that the Core group does not want to scale Bitcoin with a simple blocksize increase.
In fact, they (and their supporters) have done everything in their power to prevent this, including engaging in massive censorship.
Their primary arguments are as follows:
It is problematic to raise the limit because it requires a hard fork, which is difficult to coordinate.Bitcoin nodes should be as inexpensive to run as possible, otherwise the decentralization of Bitcoin will be threatened.Without a constraint on the blocksize, Bitcoin won’t be secure once subsidies (block rewards) decline.
None of These Arguments Have Sufficient Merit to Forestall a Blocksize Increase
I am not saying the arguments are entirely without merit. Few things in life are ever 100% black-and-white. But we have to weigh the merits of these positions against the alternatives, and against other factors in the Bitcoin ecosystem.
Let’s take one at a time:
The “Hard Forks Are Dangerous” Myth
This was a prominent talking point in 2014–2015. However, the truth is that hard forks (HF) are not necessarily dangerous, especially if they occur with a clear majority of hashing power supporting the upgraded consensus rules.
The previous group of developers, including Gavin Andresen, Jeff Garzik, and Mike Hearn, all supported upgrading Bitcoin with hard forks.
Initially, the discussion was whether the new maximum blocksize would be 2MB, 4MB, or 8MB. What begin as a minor difference of opinions between the miners somehow snowballed into a potent meme that consensus over scaling was going to be difficult.
The developers starting adding their own opinions about hard forks, creating additional friction. Yes, it is easy to claim there is contention when you are among those being contentious!
Core has no official leadership positions or governance structure. Because of this, it has been easy to justify inaction by simply concluding that “there’s no consensus”. And since they control the reference code repository, their refusal to raise the limit effects everyone else.
In practice, Core does have leaders. How else can it be explained that segwit was merged into the code (even if not activated) with practically no public debate whatsoever?
On a side note, prominent Core developers have denied that Core decides what code is published, and have denied there is any leadership. This is an example of the kind of constant misinformation that is being generated on a daily basis.
Back to the HF issue:
Many altcoins like Monero have regular hard forks. Coordination between major players in an ecosystem is not a big challenge if everyone is on the same page.
So far, I have not heard of a single problem that an altcoin had in performing a network upgrade via hard fork. So, there is evidence that they can be done safely.
In addition, if Core admits in their roadmap that eventually the blocksize will need to be increased, then why not do it now when it is badly needed? There is no logical reason why it would be more risky now rather than later.
Decentralization Myths
There are actually several myths surrounding the issue of decentralization. Let’s address the obvious ones:
The most ludicrous is the “all users should be running full nodes” idea.
As others have explained, there is no security provided to the network by non-mining ‘full nodes’. Only mining nodes secure and extend Bitcon’s distributed ledger.
The white paper explains why most users do not need to run full nodes:
It is possible to verify payments without running a full network node. A user only needs to keep a copy of the block headers of the longest proof-of-work chain, which he can get by querying network nodes until he’s convinced he has the longest chain, and obtain the Merkle branch linking the transaction to the block it’s timestamped in. He can’t check the transaction for himself, but by linking it to a place in the chain, he can see that a network node has accepted it, and blocks added after it further confirm the network has accepted it… …Businesses that receive frequent payments will probably still want to run their own nodes for more independent security and quicker verification.
The idea that a lot of non-mining full nodes will make the network more decentralized (because they can make sure the miners are behaving) is erroneous, because an SPV client can already query the network’s nodes. Generally, there would only be a problem if a majority mining of nodes were colluding dishonestly, in which case Bitcoin would be already broken.
A more valid concern is that as nodes become more expensive, eventually only large corporations will run nodes. It is true that node costs will increase over time as the network grows. However, storage, bandwidth, and processing capabilities are also constantly increasing.
Just as important: By the time that capacity increases — lets say from 3 TPS (transactions per second) to 30 TPS — the network will be so large that it likely won’t be any less decentralized, even if it costs more to run a node.
At 3000 TPS, Bitcoin would be highly dominant globally, and making use of the millions of datacenters and servers available worldwide. This was always the plan.
The Alternative Vision of Bitcoin Holds Decentralization Risks That Are Worse
Many users are not aware of the decentralization risks that come with the small-node/small-block vision of Bitcoin. Core’s vision for Bitcoin is to transform the peer-to-peer cash system into some kind of settlement network.
While this would be a way to keep node costs minimal, most users would be economically forced off the main chain because they cannot compete with institutions for fees. They would then need to get permission from trusted third parties to transact.
In my opinion, this represents a much more dangerous form of centralization than bigger blocks and expensive nodes.
The Fee-Market Failure Myth
The third primary argument of the small-block philosophy is that eventually, block rewards will run out, and mining fees will be the sole source of funding security. They then claim that without limiting the supply of transaction space, miners will be hopelessly caught in a tragedy-of-the-commons price war, with the users paying rock bottom fees, leading to a collapse of commercial mining.
There’s a few problems with this argument.
First of all, there is a natural market for every good and service in the world. There have been many price wars, but nothing with high demand ever stops being produced.
The concern that the network hashrate will become too low is based on several assumptions and variables, including the number of daily transactions, the willingness of the users to wait for confirmations, the willingness of the users to pay small amounts, the behavior of the miners, the fee policies set by various wallets, the emergent consensus on acceptable fees by the mining community, and other factors, including what actually is “too low” of a network hashrate in the first place.
The hypothetical failure of the natural fee market depends on all these assumptions combining into an unfavorable outcome, as well as the inability of the system to adjust itself favorably using any of these factors.
But, by far the biggest reason that this argument is bunk, is that it will be decades before the majority of the subsidies actually disappear.
Pure Foolishness: Overplanning the Future While Ignoring Urgent Issues Today
Why implement a plan that might help Bitcoin in 20–30 years, if it requires you to damage the user experience and erode the adoption and network effect of Bitcoin, today?
In the case of Bitcoin, it’s completely unnecessary to plan ahead that far, and the destructive consequences are already being seen.
This is the biggest reason why Core’s position should be considered indefensible. Even if their arguments have merit, it is more important to keep Bitcoin healthy right now, stay competitive, and keep the user base growing than to prevent the problems that may or may not happen later.
Even worse, those prevention plans work in direct opposition to the short term goals!
It is no less insane than demanding a bedridden hospital patient, badly in need of rest, to immediately go outside and start running laps because “exercise will help you live longer”.
What About Segwit?
It is my understanding that at “the Hong Kong meeting”, the miners agreed to Segwit PLUS a hardfork blocksize increase because they didn’t trust the Core team enough to offer satisfactory scaling in a timely manner.
I think their decision was smart. Core cannot be trusted. However, if Core changed their mind today, and agreed to the 2MB+Segwit, I would support that as a compromise to break the impasse.
They seem to be unwilling to do this.
Since miners are unwilling to accept segwit on its own, and since Core will not compromise, the only logical alternative is bigger blocks, which is the best option regardless.
What Core Wants
You may be wondering: How is it possible for people as intelligent as the Bitcoin Core developers to fail to see the obvious mistakes in their thinking?
American author Upton Sinclair’s famous quote comes to mind:
It is difficult to get a man to understand something, when his salary depends upon his not understanding it!
The Core team and their supporters want to change Bitcoin into a settlement network. They will deny this, but in my opinion, all of their actions point to this logical conclusion.
This is why they are against on chain scaling, and why segwit offers as little of it as possible while supporting their “HF are bad” narrative.
Additionally, I believe they also want to control public opinion by employing key individuals, by their associates and moderation policies on various platforms, and with an army of trolls.
They also intimidate and punish businesses that don’t fall in line. For example, coinbase.com was delisted from bitcoin.org for supporting Bitcoin XT instead of the Core client.
Despite these shenanigans, companies do support bigger blocks and on-chain scaling.
Most importantly, they want to scare you, the miner, into believing that the community doesn’t really want big blocks and if that if you mine big blocks, you’ll be forked off to a worthless coin and left with worthless ASICs.
Do not let them intimidate you.
What the Users Want
Most users just want a Bitcoin that works. They do not want slow confirmation and high fees. Most Bitcoiners that use bitcoin frequently understand the issues and support bigger blocks.
Despite all the trolling and propaganda, users controlling actual coins vote overwhelmingly in favor of Satoshi’s scaling plan.
The “Healthy Fee Market” is Already Unhealthy
Even IF a centrally planned fee market was a good idea right now, it is being managed poorly. A “healthy” fee market should strive to provide adequate fee revenue while at the same time provide a good user experience and promote growth of the network and user base.
While miner revenue is certainly adequate, the user experience is severely degraded because of slow confirmations and high fees, and this is definitely not attractive or conducive to growing the user base.
If keeping the blocksize at 1mb was an experiment to see how the fee market would develop, it has already played out its usefulness. To keep fees at a level competitive with other coins, supply must catch up with demand (we must raise the blocksize) . But these developers seem to have no interest in doing so. They would rather carry on with their agenda than serve the users.
What About Bitcoin As a Store-of-Value or as “Digital Gold”?
The great thing about Bitcoin is that it can be both a cash-like payment system and a gold-like store of value. These two aspects enhance each other.
Exposed to the propaganda that Bitcoin can’t scale as electronic cash, some users have said “that’s ok. I’m fine with Bitcoin being digital gold only”. The problem with this thinking is that Bitcoin has competition.
If another coin is useful to store value AND to transact cheaply with, it severely undermines Bitcon’s appeal to investors. At the same time, it greatly dampens demand for actual usage.
Sure, its possible that Bitcoin could survive in some form as digital gold, but it would be at a huge disadvantage.
Small Blocks Destroy Miner Revenue
At first glance, the idea that smaller blocks are bad for mining revenue may appear incorrect, since fee rates have recently exploded based on the demand of Bitcoin transactions outpacing the supply of space in the blocks.
However, this trend cannot continue for long, since users will only pay so much. At the same time, new users and new demand are being shut out from the ecosystem.
To use an analogy: Who makes more money — the farmer in town “A” selling milk from one cow? Or the farmer in town “B” selling milk from 8 cows? Townspeople in “A” might pay more per bottle, but they’ll only pay so much for it. They will start drinking something else, drink milk less often, or import their milk from another town.
Bitcoin miners simply cannot meet the demands of users at fees they are willing to reasonably pay if blocks are restricted to 1mb… and users will find satisfactory alternatives which are quickly becoming abundant.
The situation will become even worse in the long run if Core is allowed to create “second layer solutions”, because those solutions will probably not be free, and they will further absorb the money that users are willing to spend in order to transact.
This will be bad for miners, and bad for network security. It will make bitcoin even less competitive, and money will leave the ecosystem.
Price Always Lags Behind Fundamentals
It is easy to look at a high Bitcoin price and think that everything is fine. If things were going so badly, why isn’t the price dropping?
But, price doesn’t always reflect the underlying fundamentals of a market in the short term.
In the long run, fundamentals always dictate the direction of the market. Daytraders are flat at the end of the day. Speculators come and go. In the end, it’s only the long term investors and the non-speculative demand that determines the price.
The fundamental value of Bitcoin primarily comes from its usefulness as a payment system. If that system ceases to be useful, Bitcoin will cease to be valuable.
Time To Act. Let’s Help Bitcoin Grow Again.
It’s always better to fix a problem BEFORE it gets too big. As they say, “an ounce of prevention is worth a pound of cure”.
If we wait until the Bitcoin price crashes because Bitcoin is unusable as a currency, it will be too late. We would have already lost serious momentum, marketshare, users, reputation, and merchants.
This is already happening, but there is still time to act.
I urge you: don’t be complacent.
You are the miner. You have the power. Start signaling for bigger blocks today, and let’s make sure Bitcoin stays #1.
Help Spread the Word
If you’re not a miner, but a concerned investor like myself, then please spread this message far and wide, and ask the miners and pools that you know for bigger blocks.
Addendum
This article is available in several foreign language formats:
Chinese Spanish Japanese German Russian
https://keepingstock.net/an-open-letter-to-bitcoin-miners-c260467e1f0
submitted by german_bitcoiner to bitcoin_offical [link] [comments]

How to choose a Bitcoin mining pool - YouTube What is a Bitcoin Mining Pool? How to mine bitcoins? What is Cryptomining? solo mining -pool Mining EXPLAINED! #AXT BITCOIN: SOLO MINING VS MINING POOL! Bitcoin price analysis!- bitcoin may 29 Bitcoin & Cryptocurrency Mining Pools Explained  Best ...

Pooled mining pools the resources to find blocks faster and split the income between the miners. It would average to about 0.13BTC a day at the moment, but it's dropping about 2% a day at the moment and doesn't look like stopping any time soon. You won't make any income mining, nobody has except for the very first ASIC owners in January 2013. Bitcoin ABC started a fundraiser and three public mining pools have been signaling support for the new full node project BCHN. Bitcoin Mining Pools. There are many good Bitcoin mining pools to choose from. Although it's tempting to pick the most popular one, it's better for the health of the network to mine with smaller pools so as to avoid potentially harmful concentration of hashing power. The hash rate distribution is best when split among more Bitcoin mining pools. Bitcoin Mining Pool Hash Rate Distribution ... Mining Pools are close to SoloMining under the hood. Most computers and hardware would take years to generate a block because they come in groups of 25 Bitcoins. However with a mining pool the bitcoin share goes to the server its self and then it calculates the ammount of work that your hardware personally did. They will then send you that ... DASH Mining Pools: 7. Ethereum Classic: 95950892411491.00 3.104 ETC $5.70 USD 12 sec ETC Mining Pools: 8. Bytecoin: 59202147837.0 1,506.89 BCN $0.00 USD 2 min BCN Mining Pools: 9. Zcash: N/A 6.25 ZEC $62.98 USD 2 min 30 sec ZEC Mining Pools: 10. Bitcoin Gold

[index] [43408] [32922] [30744] [46451] [2011] [33743] [47251] [2262] [4118] [48260]

How to choose a Bitcoin mining pool - YouTube

Mine Bitcoin: https://secure.iqmining.com/tiny/zBYaH What is Bitcoin Mining? Bitcoin Mining is a peer-to-peer computer process used to secure and verify bitc... Bitcoin & Cryptocurrency Mining Pools Explained Best Mining Pools PPS vs PPLNS - Duration: 18:17. VoskCoin 9,965 ... How to start Bitcoin mining for beginners (SUPER EASY) - ULTIMATE GUIDE ... What is a Bitcoin Mining Pool... as Bitcoin mining has become increasingly difficult, individuals join mining pools. You would be required to invest in your own mining equipment and then join a ... Agenda: Livestream for how mining pools work. What is a mining pool, how's it work, what is pool luck? What are the various payout types and how do they work? How do we know the pool isn’t cheating? Bitcoin & Cryptocurrency Mining Pools Explained Best Mining Pools PPS vs PPLNS - Duration: 18:17. VoskCoin 6,118 views. 18:17 . Mix Play all Mix - Crypto Rick YouTube; LIVE market coverage ...

#