EOS is a culminating effort to generalize smart contract infrastructure using a delegated governance model, featuring parallel processing and inter-blockchain communication (IBC),
scaling to handle high-transaction rate commercial applications having large sets of concurrent users.
As of June, 2018, over 130 people have contributed to the open source-project hosted on GitHub, with over 7,500 commits to the repository.
Dawn 4.0 was released on May 4th, 2018.
Dawn 4.1 was released on May 18th, 2018.
Dawn 4.2 was released on May 25th, 2018, a single-threaded implementation with sustained benchmarks over 500 TPS (Transactions Per Second) on a heavly loaded global test network, 33-times faster than Ethereum and 71-times faster than Bitcoin.
EOSIO 1.0 was released on June 1st, 2018, featuring bnet plugin which provides a
"beast" P2P protocol based on websockets using multi-threaded networking to achieve over 3,000 TPS, 200-times faster than Ethereum.
The latest version is v1.1.0.
Eventually, a multi-threaded version will be released with side-chains, scaling even higher across clusters using an inter-blockchain communication protocol with theoretical million+ TPS throughput.
Distrupting The Status Quo
EOS.IO is designed to disrupt existing centralized, large-scale enterprise technology.
It provides a scalable blockchain platform for a new generation of decentralized applications designed to disrupt current technology incumbents, including Facebook.
Besides $4 billion in ICO revenue,
Block.one has announced several venture capital partnerships to help finance their ambitious plans.
July, 16th, 2018: Blockchain platform Block.one is close to announcing a new fundraising round, bringing in Peter Thiel and Chinese billionaire Jihan Wu’s Bitmain Technologies Ltd. as new investors.
June, 1st, 2018: Block.one, publisher of the ground-breaking EOSIO blockchain protocol, and SVK Crypto, an industry-leading, City of London-based investor in blockchain technologies, announced today that they have partnered to launch a new US$50 million fund to accelerate the growth and development of the EOSIO blockchain ecosystem.
April, 6th, 2018: Blockchain veterans Michael Cao and Winnie Liu formed a new US$200 million joint venture fund, EOS Global. EOS Global will make strategic investments in Asia-focused projects utilizing EOSIO, and is the fourth injection of capital through Block.one’s EOS VC initiative which now totals 600m USD.
March, 21st, 2018: FinLab AG, one of the first and largest company builders and investors focused on the FinTech sector in Europe, today announced the signing of a letter of intent regarding the formation and capitalization of a new $100 million (€81 million) fund managed by FinLab, that will make strategic US$100m investment throughout Europe in projects that utilize the EOSIO open-sourced blockchain software.
January, 23rd, 2018: Block.one, the developer behind the leading blockchain software EOSIO, and Galaxy Digital LP (“Galaxy Digital”), a full service, digital assets merchant bank, today announced the formation and capitalization of a joint venture that is focused on developing the EOSIO ecosystem and making strategic investments in projects that utilize EOSIO blockchain software. As part of the new relationship, Block.one and Galaxy Digital will deploy capital for future investments through the capitalization of a new US$325 million EOSIO Ecosystem Fund (the “Fund”).
January, 16th, 2018: Block.one today announced with TOMORROW BLOCKCHAIN OPPORTUNITIES (“TomorrowBC”) the creation of a US$50 million fund to exclusively invest in opportunities leveraging EOSIO software. This is the first announcement for Block.one’s EOS VC partnership program to stimulate EOSIO innovation. TomorrowBC aims to invest in trailblazing entrepreneurs and companies looking to shape our future utilizing new technologies, like the EOSIO blockchain open source software.
Block.one has partnered with leading technology hedge fund managers and venture capitalists:
EOS has chosen a delegated proof-of-stake architecture (DPOS).
A concensus of EOS stakholders with governance rights endorse a blockchain constitution that establishes community standards, including an arbitration process when disputes arise.
The blockchain is supported by 21 primary block producers distributed throughout the world, chosen by ongoing community voting and rewarded daily.
Primary block producers are determined through a recurring vote tally every 126 seconds.
Votes decay over time and need to be re-cast weekly to remain fully counted.
Block producers found to be in violation of the constitution can be voted out of the active producer set.
There can be any number of stand-by block producers, currently numbering over 350 registered producers in June, 2018.
Stand-by block producers that garner more than 0.5% of total stakeholder vote are also rewarded using a graduated scale.
Larimer's experience with prior DPOS blockchain projects has identified a "sweet spot" of 21 active block producers.
There are three fundamental design levers available, but blockchain architects can only pick two: Security, Performance or Decentalization.
Security is mandatory, but all blockchains next face a trade-off between transaction performance and decentralization.
Greater decentralization provides higher trust but leads to propagation delays and longer transactions that result in unaceptable user experiences in many commercial applications.
The EOS infrastructure includes:
Delegated block production (no mining or mining pool cartels)
The EOS infrastructure provides user accounts, authentication, memory-mapped file system databases (MongoDB), IPFS storage, and the asynchronous scheduling of DApps across many CPU cores and clusters.
The use of memory-mapped persistence via MongoDB is a key design decision that allows for very fast EOS transactions.
Sub-second performance is attainable because the data/structures allocated in EOS (including the contracts/apps running therein) are actually held resident in RAM.
In effect, data is cached in memory for super-fast access-- not in slower SSD storage or even slower hard drive/array storage.
If physical RAM is depleted, MongoDB will page memory blocks to slower SSD/disk storage, but transaction performance drops.
Ledger Capital has written a good overview.
EOS Germany has also created an objective SWOT (Strengths, Weknesses, Opportunities, Threats) analysis.
This Steemit article by eosinsider presents an economical way to create a new EOS account.
Tang Hongbo posted an article on Medium that describes how to use his
signupeoseos Dapp to create a new EOS account.
Use the latest version of Greymass when generating public+private keys,
since Greymass provides safe offline (cold storage mode) key generation.
NEVER reveal private keys, especially to account-creation services/websites
Creating a new account isn't free, since each account requires enough EOS to purchase approximately 4KB of RAM
Each account has owner and active public/private key pairs; if you haven't changed them you can assume both pairs are identical
Block.one initiated their ICO on June 26, 2017.
The ICO ended nearly a year later on June 2, 2018 at 22:59:59 UTC, when tokens were frozen on the Ethereum blockchain during preparation for the EOS mainnet launch.
The ICO utilized an ERC20 token contract running in the Ethereum blockchain.
The ERC20 distribution totaled 900 million tokens over a 341-day offering, 2 million tokens every 23 hours, with Block.one reserving 10% of the distribution.
Besides directly participating in the ICO, tokens could also be purchased indirectly on several exchanges, including Kraken and Binance.
All Registered ERC20 tokens were converted to EOS tokens in a genesis snapshot taken during the launch.
During the ICO, Block.one recommended either of two Ethereum wallets to store EOS ERC20 tokens: MetaMask or MyEtherWallet (MEW).
Wallets were required to be registered with
eos.io prior to launch.
It's important to emphasize the wallet itself must be registered, not the individual ERC20 tokens stored therein.
The registration process using MetaMask is relatively straightforward.
However, people in the United States and China were restricted from participating in the ICO due to protective legal measures taken by Block.one to avoid EOS being considered a security by regulators,
complicating the registration process. Workarounds included using a virtual private network (VPN) product like CyberGhost during the eos.io registration process, selecting a VPN server located in Canada or another supported country to mask the browser location.
Alternatively, some exchanges (Binance,
and Kraken) announced registration support for customers who purchase and hold EOS ERC20 tokens.
After completing EOS registration of the Ethereum wallet, token holders could visit eoscountdown or EOS Authority to confirm whether the wallet is registered properly for the EOS genesis snapshot.
There is a recovery option after launch for those who do not register.
An EOS contract named eosio.unregd tracked unregistered ERC20 balances identified during the genesis snapshot at launch,
allowing owners of unregistered ERC20 tokens to prove ownership and claim their tokens.
Ethereum public addresses can be converted into EOS public addresses without registering.
Scatter announced support for converting ERC20 wallet private keys to EOS private keys within its wallet.
NEVER divulge private keys unless the source is absolutely trustworthy!
Encrypt private keys and keep them offline with their corresponding public keys, preferably stored at two seperate physical locations, like a home safe and bank deposit box.
There are several good encryption tools, some of which are freely distributed as open-source, like VeraCrypt.
This ecosystem diagram focuses on the token lifecycle.
Requirements around account names and namespacing are presented in GitHub issue #3189.
Community benefit projects are migrating to a worker proposal system, tracked by issue #1005.
Acknowledgement to RiverKingfisher for providing many refinements:
EOS DApps require CPU cycles, network bandwidth and physical RAM to run.
CPU cycles and bandwidth are in relatively abundant supply at reasonable prices.
RAM is in relatively short supply and rather pricey.
Staking EOS tokens automatically reserves a proportional amount of CPU and bandwidth, since these resources are plentiful.
However, RAM is treated separately due its scarceness.
Automatically reserving RAM would unnecessarily lock up a limited resource, since most users don't need to reserve RAM.
Developers who want to run applications need RAM and must use it efficiently in their code.
To avoid hording and excessive speculation, RAM supply will be initially limited and gradually increased as more applications are released in the community.
RAM must be explicitly purchased with a 0.5% transaction fee, with the same fee applied whenever RAM is eventually sold/released.
Just prior to launch on June 1st, there were 189 registered candidates vying to be voted into the active set of 21 block producers.
On May 30th, EOS GO published the final EOS Block Producer Candidate Report.
Visit EOSWire to evaluate the latest block producer compliance report, scoring BP disclosure, code of conduct signature, ownership and bp.json compliance.
Whales have become entrenched in the top-21 active block producer set.
The token distribution is concentrated at the top, with 90% of all tokens being held by 1.6% of the accounts.
Dutch EOS has published guidelines for voting in high-quality, underrated block producers.
EOS Authority maintains very useful voting analytics page.
EOS relies on a DPOS governance model, so stakeholder diversity and voter participation is critical to avoid block production collusion and cartels that threaten community stability.
EOS is a double-edged sword: Using 21 active block producers has many advantages, including low-latency DApps that scale;
however, having only 21 active block producers is a temptingly small number of entities to corrupt, susceptible to collusion.
For the ecosystem to thrive long-term, stakeholders must vote carefully and frequently to establish a widely distributed, independent set of block producers that are technically solid, ethically sound and independently funded.
Suggested block producer (BP) selection criteria:
Avoid voting for mining pool cartels having ownership in more than one block producer
Avoid voting for exchanges that frequently vote for themselves using customer tokens
Avoid voting for block producers controlled by investment cartels having ownership in more than one block producer
Avoid voting for whales or block producers supported by just a few whale accounts
Vote for 30 block producers distributed across a wide geographic area
Vote for self-funded BPs free of whale control and collusion
Vote for founding BPs involved in standing up the mainnet during the arduous launch window
Vote for historically active BPs that participated in prior Dawn releases
Vote for BPs operating in stable geopolitical locations that protect individual liberty and support free markets
Vote for BPs with dedicated staffing, preferably with 7x24 support
Vote for BPs knowledgable of governance and arbitration systems, experience with Bitshares and Steemit
Vote for a mixture of BPs using bare metal servers, cloud-based and hybrid architectures (bare metal + cloud)
Vote for BPs with load balancers, redundant networks and servers for reliable failover
Vote for BPs that make significant efforts at outreach, including meet-ups and social media content
Vote for BPs that make significant technical contributions, including applications, adminstrative tools and GitHub commits
Avoid voting for whales.
Careful consideration is required when voting; be especially wary of block producers associated with whales.
Avoid organizations that fund a cabal of block producers, exchanges, mining pools and venture funding organizations.
Avoid candidates that are exchanges or are associated with exchanges, including Eossey Hanbitco, EOS JRR (shared funding source with Binance) and Bitfinex.
Avoid candidates that are backed by mining pool cartels, including Bitmain Antpool, Huobi Mining Pool, eos.fish (backed by F2Pool) and viaEOS.
INBlockchain is a whale that actively funds a cartel of BP candidates, including EOS Mao Lao, OracleChain, EOS Gravity and eosONO.
Flexibility is another important criterion to consider.
Favor BPs that integrated/tested on both the Ghostbusters/CORE and EOS.BIOS testnets, as opposed to being fixated on just one chain/solution.
Block producers like EOS Blocksmith and EOS SoCal supported both approaches and exhibited the kind of flexibility and competence needed in the active producer set.
It is important to vote for 30 block producers, not just a few.
Based on the criteria noted, consider voting for the following recommended block producers, in no particular order:
Community and technical contributor, experienced team, self-funded and transparent, AWS cloud, future bare metal
EOS is live and block producer voting is an ongoing, critical stakeholder responsibility.
Re-evaluate and cast your votes on a regular basis, since voting strength begins to decay after one week.
There are three avenues available for voting: CLI (Command Line Interface), desktop/network-disconnected and web/browser portals.
CLI tools are the most secure.
Desktop GUI voting tools are nearly as secure and much easier to use.
Web (browser-based) voting portals are the least secure means of voting.
Until the new iOS wallet is ready, CLEOS is the officially approved voting tool by Block.one:
For technically skilled stakeholders, there are two frequently used command line tools available:
and EOS Canada has provided eosc, a more accessible cross-platform (Windows, Mac and Linux) CLI tool that supports block producer voting and other common tasks:
Amongst the available community desktop voting applications (supporting network-disconnected signing),
Greymass is a popular desktop voting platform (and light wallet),
endorsed by Dan Larimer and at least 10 block producers:
Voting tools provided by Tokenika and Cypherglass
perform the critical transaction signature process offline:
Other block producer candidates are also developing web voting portals, including
Everpedia is also working on a voting portal.
Bitfinex is also creating a voting tool for their exchange customers.
Voting by proxy is convenient. Instead of indiviually researching up to 30 block producers, stakeholders can allow a proxy to vote their stake.
Research is needed to select an appropriate proxy that matches each stakeholder's voting preferences.
For more information about proxy voting, refer to this Medium article by Blockgenic.
For an ongoing discussion about proxies, join the EOS Voter Proxies channel on Telegram.
EOS Titan provides a very helpful voting analysis tool that ranks BPs and reveals major stakeholder voting:
EOSeoul is monitoring the vote and the mainnet with EOStat, which includes a web voting portal:
The mainnet is now fully operational, launched on Sunday, June 10th, 2018 at 1300 UTC (9AM EST).
ERC20 tokens became non-transferable on Friday, June 1st, 2018 at 22:59:59 UTC when the ICO ended.
Exchanges began locking down EOS ERC20 token deposits and withdrawals two days before launch.
The window for wallet registration expired on June 2, 2018 22:59:59 UTC when tokens were frozen on the Ethereum blockchain.
The mainnet launch sequence began with a genesis snapshot of all registered ERC20 tokens less than an hour after the freeze on Saturday, June 2nd, 2018 at 10:00:00 AM UTC.
First, several block producers will generate and validate their seed-token genesis files and arrive at a concensus file to boot from.
This JSON file maps EOS wallets created during registration to associated Ethereum wallets storing ERC20 token balances.
Two block producer launch groups have formed using differing bootstrap approaches, which is to be expected when standing up such a large distributed platform.
One appointed block producer (chosen from 11 candidates) will seed the mainchain, with other block producers validating and subsequently meshing in.
Finally, after the mainnet is validated it will be opened up to stakeholders to begin the block producer voting process.
Before any blockchain can actually become fully operational, EOS core requires stakeholder approval representing at least 15% of the issued and outstanding EOS tokens.
More than one mainnet may arise during launch.
More than one chain may attain 15% voter endorsement, enabling the token snapshot and most functionality.
Eventual consolidation down to a single mainnet is the prevalent expectation, possibly during launch day or soon thereafter.
At some poiont, block.one will vote their 10% stake which should result in consensus on a single chain.
Thomas Cox has written about his launch expectations on Medium,
asserting it will be difficult for even two blockchains to garner 15% of the stakeholder vote.
It's reasonable to surmise there will be at most two competing chains on launch day.
Chances for one mainnet have improved given recent improvement in communication between the block producer bootstrapping camps.
As the flowchart above depicts (courtesey of EOS Tribe), Dan Larimer's third crypto-child is going through a launch gauntlet.
June 4th: Block producers have been working extremely hard for two days straight, many are pushing themselves beyond exhaustion.
The snapshot is good. Many block producer teams seem fully capable of standing up testnets rather quickly.
Efforts are centered on testing the blockchain itself and its numerous life cycles (un-staking tokens, voting, auctions, etc.), and also evaluating network security with attack/penetration testing.
Half a dozen or more tools/scripts are being utilized for this testing.
The work is time-consuming to perform and difficult to coordinate across a multitude of BP teams, with participants distributed around the world in numerous time zones, speaking various languages, etc.
There is still noticeable dissension with the usual political maneuvering when money/power is up for grabs, which doesn't help the velocity.
Essentially, Dan's software, the network infrastructure and each BP team are all being set through their paces.
June 6th: Block producer candidates have been testing the security infrastructure and blockchain functionality for 4 days.
A large group of BPs in China announced the start of an independent audit to review the entire project (network, server configuration and EOS.IO source code).
Two outside auditing firms were hired in China to perform the work:
SlowMist and Joinsec.
Several BPs in China originally requested 7 days for the audit, but soon lowered the request to 2 days with a status provided each day.
China obviously has considerable influence (with 500 million stakeholder tokens, according to EOS Silicon Valley) and several BPs there do not want to rush into a launch.
The earliest launch date appears at this point to be Friday, June 6th, unless testing/auditing identifies critical issues (identified as P0 or P1) that need to be fixed and then revalidated.
June 7th: Block producer candidates have been testing the security infrastructure and blockchain functionality for 5 days.
The 0100 UTC go/no-go meeting was extremely long, disorganized and therefore chaotic.
Lingering disagreements were the rule, including how to fund RAM for accounts at inception and the severity of a few other issues.
The ongoing independent audit in China had not produced any P0/P1 issue.
The first vote met the two-thirds + 1 requirement for a launch.
However, disorganization pervaded the meeting, especially over voting procedures, which resulted in a second round of go/no-go voting.
Prior to the second vote, Dan Larimer joined the meeting to discuss the issues, an upcoming 1.0.2 patch (bi-weekly recurring patch cadence) and to take questions; he clearly indicated there were no show-stopper issues, no P0/P1 issues.
According to Larimer, the highest unrecovarable risk is private key loss at inception, with most of that risk existing in external browser-based wallets and voting portals, not in the core chain itself.
Thereafter, a second round of voting resulted in less support for going live, not enough votes to meet the two-thirds + 1 threshold.
The aftermath was continued frustration for the majority of block producers who voted in favor of launching, along with increasing stakeholder dissaproval, especially given the zero-show stopper feedback provided by Dan Larimer, which a significant number of block producers in the minority chose to disregard by voting no.
June 8th: Block producer candidates have been testing the security infrastructure and blockchain functionality for 6 days.
Unlike prior meetings, this 0100 UTC go/no-go meeting was streamed live on YouTube.
Unlike the prior meeting on June 7th, this meeting was brief and appeared to be a formality with no discussion over recurring disagreements.
The single round of voting resulted in unanimous approval to launch.
An appointed block producer (ABP) is preparing the genesis snapshot and EOSIO 1.0.2 for launch.
On Saturday, June 9th at 1300 UTC, 10 additional seed nodes will begin to mesh with the ABP.
On Sunday, June 10th at 1300 UTC, the live chain will be announced and BP voting commences.
Thereafter, EOS becomes fully operational whenever the mainchain gathers 15% voter approval, a threshold that will likely be crossed rather quickly.
June 9th: The independent audit report was released in public form on June 9th, 2018:
The EOS mainnet went live on Sunday, June 10th, 2018 at 1300 UTC (9AM EST).
Voting commenced immediately with EOSphere,
EOS New York and EOS Authority tracking the vote tally.
EOS stakeholders voted past the 15% minimum threshold (150 million tokens, 15% of the total distribution) and unlocked the mainnet on Thursday, June 14th.
EOS Authority processed the genesis snapshot.
For those who were not able to register before the ERC20 freeze (3,301,220.3641 tokens), EOS LaoMao created a script and validated all of the unregistered EOS holders and their balances.
Block.one and several exchanges represent the top-10 accounts, holding nearly half the total distribution.
Besides the 10% stake held by EOS.IO, other top accounts are exchanges with token counts that represent an aggregation of many individual customer accounts.
The overall distribution is concentrated at the top: 90% of all tokens are held by 1.6% of the accounts.
Top 10 Holders: 20,675,047
Top 100 Holders: 646,595
Top 1,000 Holders: 42,941
Top 10,000 Holders: 3,312
Top 100,000 Holders (of 169,930 accounts): 55
Top 10 holds 496,735,539 Tokens or 49.67% of the total 1 billion ICO distribution
Top 100 holds 748,176,831 Tokens or 74.82% of the total tokens
Top 1,000 holds 858,120,383 Tokens or 85.81% of the total tokens
Remaining accounts (1,001 through 163,930) hold a total of 138,570,296 Tokens, 13.86% of the total distribution
Ignoring 10% Block.One Stake
Top 10 (excluding B1) holds 396,735,539 Tokens or 39.67% of the total tokens
Top 100 (excluding B1) holds 648,176,831 Tokens or 64.82% of the total tokens
Top 1000 (excluding B1) holds 758,120,383 Tokens or 75.81% of the total tokens
Remaining accounts (1,001 through 163,930) hold 138,570,296 or 13.86% of the total distribution
Block.One stake: 100,000,000 tokens (10%)
Top 2 through 1,000 holders have 758,120,383 tokens (75.81%)
Remaining 1,001 though 163,930 holders have 138,570,296 tokens (13.86%)
3,309,321 tokens remain unregistered (0.33%)
Grand Total: 1,000,000,000 EOS Tokens (100%)
Block.one, DApp providers and block producers have already provided or are working on wallets.
GreyMass offers a popular light wallet and voting portal that's endorsed by Dan Larimer:
Scatter is a Chrome/FireFox browser extension wallet, and the desktop version is now in beta:
SimpleEOS has been released by EOS Rio.
EOS Wallet Pro is available, having annunced their beta a week prior to launch.
EOSYS has announced (timepoint 1:09:13) an Android wallet that will be released in June, followed by IOS and Chrome extension wallets in Q3, 2018.
Pomelo for Apple iOS devices is a joint effort by several block producers.
Dan Larimer recently announced a very secure hardware wallet for select Apple devices is being developed at Block.one:
FeeXplorer also tracks RAM prices with a histogram:
RAM Allocation Proposals
During the July 2nd EOS EMLG Meeting,
Dan Larimer discussed various RAM allocation proposals to address increasing RAM costs driven by current RAM speculation.
One key take-away to help govern RAM pricing was the need to slowly introduce new RAM capacity over time instead of dumping large amounts suddenly.
Eyal Herzog published a subsequent analysis that
examined another proposal for a continuous 10% burn rate from the RAM contract to drive prices down.
Larimer has also published his thoughts on the present RAM market on Medium.
Greymass recently proposed the separation of RAM into two separate components: live storage (current) with read-write permission, and persistent, backed-up storage with only activation permission.
Proposal for Partitioned RAM with Tiered Quality-of-Service Costs
Similar to Greymass, this proposal further stratifies storage into four tiers instead of two, with side-chains having a declared storage quality-of-service level and associated relative cost.
Not all applications require sub-second transactions.
Some can work asynchronously and tolerate 2 or 3 second end-user response times.
Some applications are batch-oriented entirely and have long execution windows (minutes, not seconds).
Many applications fall somewhere in between.
EOS uses MongoDB, a memory-mapped file system document database that performs best when everything remains resident (cached) in memory.
MongoDB is flexible enough to support all of these applications.
Whenever RAM runs short as the number of applications and/or user activity grows, MongoDB can swap part of the "working set" of infrequently referenced documents out of memory into SSD/disk storage, which slows performance down noticeably whenever the data in these swapped-out blocks are referenced/needed again.
EOS itself and its RAM marketplace should be equally flexible and support various performance tiers with relative pricing.
RAM could be partitioned in terms of performance, with the highest prices determined by the Bancor marketplace in the top RAM partition that guarantees the contract/app will always remain resident in RAM, assuring fast transactions.
Another leased RAM partition would ensure 66% residency and 33% SSD storage for contracts/apps that don't require sub-second transactions, but response times in a 1-to-3 second window.
A third leased RAM partition would ensure 33% residency and 66% disk storage for contracts/apps that support asynchronous transactions lasting between 3 and 10 seconds.
A fourth partition would not ensure any RAM residency and use cheaper disk/array storage for contracts/apps that process information in the background in "batch" mode.
The lower three RAM partitions would use leasing to allow application developers more choices and lower costs for their particular application.
Note the thresholds/metrics in these tiers are hypothetical; actuals would be based on observed application performance testing/benchmarks.
Once side-chains are supported in EOS,
each side-chain could belong to a tiered service offering that supports a targeted RAM partition with a designated quality-of-service level.
Applications that require high-performance will stake a lot of tokens chains belonging to the top Bancor marketplace tier with 100% RAM residency.
Background/batch applications can be deployed on side-chains configured for the bottom performance tier using significantly fewer tokens.
Many applications will fall somewhere in the middle two tiers.
Side-chains offer the opportunity to stratify service offerings in terms of application performance, which is directly tied to RAM or the lack of residency therein.
Many developers may not be able to afford 30 or 50-MB in the open Bancor marketplace used by the first RAM tier.
Many of these developers won't require sub-second response times, some may need 100-MB or more, and others are dealing with a totally batched design with minutes or hours-long execution windows.
These application domains shouldn't have to pay a premium price for fast transactions when they aren't needed.
Developers will want several storage options priced according to performance characteristics.
EOS applications would be comprised of three main components: Ricardian Contract, Arbitration Specifications and Designated Side-Chain RAM Partitioning:
Partitioning like this seems almost inevitable if true mainstream adoption takes place, as the number and variety of applications and users increase.
EOS Titans has released a block producer ranking explorer:
EOS Beijing has created a block producer voting analysis tool:
EOS Titans provide a page that summarizes block producer voting patterns:
EOS Nation hosts Block Producer bp.json Validator with endpoint, resource, API version error reports:
Note how many of the top ranked producers have a handful of whales voting them into the payout zone.
Some BPs have just 3 or 4 whale accounts voting them into the active producer set.
However, whale support does not equate to block production competency.
Dawn 4.0 established 21 active block producers and any number of standby producers.
In order to receive rewards, block producers must receive at least 0.5% of the total stakeholder vote.
Producers with vote tallies below this threshold receive nothing.
Each block producer claims their share of the per-vote reward once per day.
For more, EOS Canada has provided a block production revenue analysis.
An important GitHub proposal was made recently to add support for
TypeScript software development, using
Binaryen for compilation to
Expanding access into the TypeScript marketplace will significantly increase the rate of new applications for the EOS community.
Dispute resolution and the initial
EOS Core Arbitration Forum (ECAF) are in their infancy.
As EOS New York stated, arbitration and governance processes/tools are work in-progress.
Initially, ECAF is the only hamstrung option available, but it will evolve over time and more arbitration options are coming.
Just as an appointed block producer was chosen to bootstrap the blockchain, arbitrators involved in ECAF have been similarly appointed.
EOS did not launch with a full suite of community tools; it will take time to stand everything up.
ECAF is just the initial/default arbitration service.
Additional arbitration organizations will appear, some specializing in various sectors, and stakeholders will be able to choose from a menu of arbitration options.
Arbitration services/choices are established within contracts.
Arbitration will be independent and self-funded with fees collected from the claims process.
Future community arbitration enhancements will allow stakeholders to discuss arbitration cases, nominate and vote for arbitrators.
Watch this BP meeting where arbitration is discussed with Dan Larimer.
For a variety of reasons, on June 26th Dan Larimer announced his desire to replace the present constitution with a more narrowly focused referendum contract at the application level, including a reduction in arbitration scope.
Just 1.6% of Eos holders own 90% of the tokens. The token ownership, governance, and block producer selection are all compromised by the existence of massive
"whale" addresses that can push Eos in any direction they want without regard to the majority of token holders, developers, or even the overall health of the network.
Telos Whitepaper, July, 12th, 2018
In naval terms, this is Dan Larimer's third and latest crypto-vessel: H.M.S. EOS.
Larimer is putting the ship through its paces, a shakedown cruise to evaluate the mainnet and resolve issues before major applications begin rolling out.
Besides evaluating performance in the field, shakedowns also familiarize the crew (block producers, application developers and stakeholders) with ship operations.
We are witnessing a harrowing, yet effective trial that relies on the abilities of Block.one and block producers to patch and enhance EOS:
Expect bugs and ongoing patches, security enhancements and hard wallets to appear near-term.
Thereafter, expect tools for improved stakeholder voting/participation in constitutional ammendments and worker proposals.
Expect a major constitutional overhaul, and major arbitration changes that affect/limit the current ECAF arbitration process.
Performance enhancements for higher scale will likely appear afterwards, towards the end of the shakedown cruise.
This maiden voyage is a high-risk/high-reward strategy in preparation for upcoming applications, higher loads and community tools.
Timing is critical to foster early developer adoption and emphasize first-mover advantage.
Put your life vest on (secure your private keys) and prepare for heavy seas.
Fully participate and do your best to bolster the EOS ecosystem.