Hashgraph mainnet up, HBAR minted, centralisation fears lingering
Check, check and check. Everything's proceeding smoothly.
The moment has come. Or to be more accurate, a moment has come. These kinds of things take many steps and a lot of tinkering en route, as one of Hashgraph's chief architects has previously said.
This particular moment is the Hedera Hashgraph mainnet going live, the minting of all 50 billion HBAR tokens and ongoing criticisms about some of Hashgraph's centralised governance elements.
What just happened?
The network is now live but only accessible to a select range of testers and dapp developers. The field of testers will grow over the coming months, leading up to the full public launch about six months from now. The public launch will be when anyone can create an account and avail themselves of network services.
In its current private state, the network has three services available for testing: cryptocurrency (transfers, transactions), support for Solidity (Ethereum language) smart contracts and file storage (file storage).
The mainnet launch also saw the minting of HBAR, Hashgraph's main cryptocurrency. There's a total of 50 billion HBAR, and barring a drastic change of plans or catastrophic technical error, no more will ever be minted.
No HBAR has been distributed yet though. Other than the Hashgraph developers and other dapp testers, no one will be touching the tokens until the public launch.
When that happens, the SAFT buyers will be getting their share of pre-purchased HBAR.
A simple agreement for future tokens (SAFT) isn't a token. Rather, it's a kind of legal structure for cryptocurrency fundraising, where a company can sell the promise of future tokens to accredited buyers rather than throwing a legally dubious ICO.
And though it may not be live, the beginning state of the ledger has been cryptographically hashed, with the hash:
This hash might hold some sentimental value for the developers, but it's also an important digital signature for future reference to provide proof of the state of ledger as of 24 August 2018.
What to expect from the public launch
The goal is to launch a public, fully functional, high-performance distributed ledger network with real live applications about six months from now. The next half year is going to be a busy time for Hashgraph dapp makers.
There's been more than enough interest so far, Hashgraph developers Leemon Baird and Mance Harmon have said, to expect a well-functioning network with real usage at launch time.
"DApp developers interested in building on Hedera include IoT applications, music and IP management, supply chain management, medical credential management, casual games, security tokens, smart contracts for escrow, and more."
One of the most exciting killer apps for Hashgraph, they say, is probably micro payments that haven't previously been possible. This was one of the big promises of Ethereum smart contracts, but ongoing scaling issues have so far put a damper on those possibilities.
Based on its starting support for Solidity smart contracts, it's safe to say that Hashgraph is scooping up some Ethereum dapp makers that have been turned away by scaling troubles.
How Hashgraph works
Hashgraph is an intriguing project as an exceptionally high-performance, high-scalability all-in-one platform. But everything's a balancing act, and Hashgraph's performance comes at the cost of "pure" decentralisation.
Hashgraph in a small nutshell
Hashgraph is a proof-of-stake DAG. Anyone can stake HBAR tokens to help power the network and maintain security, and they can in turn get paid with a portion of the tiny HBAR transaction fees. So far, so decentralised.
However, the governance elements are semi-centralised and consist of a total of 39 council members. These councillors will get to vote on future developments and changes in the Hashgraph network and will have the power to take actions in the name of network security, such as if one entity looks like it's accruing a big enough HBAR stake to threaten network security.
Each council member serves a two-year term, and future council members will be elected by the current council. The goal is to deliberately choose councillors from a diverse range of institutions (governments, non-profits, businesses, etc) who have a diverse range of motivations. The point is to prevent cabals from forming and to create and maintain a council that can't agree on anything except moves that will benefit the network as a whole.
Centralisation, decentralisation and Hashgraph
Some might see this council element as an unacceptable technical vulnerability in its own right, which will inevitably lead to a network powered by politicking, corruption and backroom deal making. Others might see it as a necessary guiding hand in the early stages of the network and a far better alternative than the "decentralise-washing" which plagues most other projects.
As it tends to be, the truth is probably somewhere in the middle.
On one hand, Hashgraph aims to become a new world computer used around the world and beyond for almost anything. In that context, a council of 39 might seem quite unacceptable. Plus, even the most democratic civil power structures tend to consolidate into little balls of corruption over time as the incentives of governors start diverging from the needs of the governed. From that angle, any kind of centralised governance system might be seen as a kind of insidious technical vulnerability.
On the other hand, that's still a long way off and getting to that stage in the first place is a lot easier with some kind of guiding hand. Like any other part of the network, the council and governance mechanisms themselves are up for debate and likely to change in the future based on user feedback and new insights.
Plus, literally every single cryptocurrency, to say nothing of literally every other facet of human existence, is centralised to a certain extent.
Distributed consensus does not magically equal decentralisation
One of the most common misconceptions in cryptocurrency is that distributed hashing power magically equals total decentralisation. In fact, this is just one facet which needs to be decentralised.
Consider bitcoin, the "one true decentralised" coin. It has its own rigid governance structure, where all users are subject to the whims of developers, who are in turn subject to the whims of miners because no fork can succeed without adequate miner support. The only freedom available to bitcoin users is the ability to choose whose whims they want to be subject to. But at the end of the day, it's all pretty much the same.
In the case of Hashgraph relative to bitcoin, governance centralisation will be transparently formalised under a large and changing council of diverse motivations, instead of small and largely static mining cabals who all have the exact same profit motivations. The low entry barrier of Hashgraph's proof-of-stake consensus mechanism will similarly be distributed among a wider range of node operators than bitcoin's high-entry-barrier mining scene.
By any intellectually honest interpretation, Hashgraph is more decentralised than bitcoin in both governance and network consensus. If Hashgraph is too centralised for one's tastes, then bitcoin should definitely be right out of bounds.
Ethereum recognised this and went out of its way to solve the problem, but this just led to different elements of centralised governance. Governance-wise, anyone can put forward Ethereum Improvement Proposals, but the community's veneration for Vitalik Buterin and the other lead developers means they're unlikely to succeed without support from key players. The somewhat hilarious end result is a group of developers using their centralised authority to mandate decentralisation against many people's wishes.
EOS is also worth a mention as a clear example of how not to do anything. It made the bizarre choice of consolidating both its network consensus and governance power into the same group of 21 nodes, then just lazily tied it all to voting power given by EOS tokens and expected it to work. It didn't.
Hashgraph uses proof-of-stake nodes to decentralise its consensus, while transparently acknowledging and working on its centralised governance elements. By contrast, the vast majority of other projects will simply call themselves decentralised even while having completely opaque and centralised governance. Simply by openly acknowledging this unfortunate fact of crypto and transparently laying out its governance systems, along with a plan for hopefully keeping its governance at least somewhat decentralised, Hashgraph is head and shoulders above most other projects and much more decentralised than most.
Whether that's enough for Hashgraph's ambitious plans or whether anyone actually cares about decentralisation more than they do about making a buck is anyone's guess. That almost all cryptocurrency projects are highly centralised and that true decentralisation might be an impossible pipe dream is a hard fact to face.
If Hashgraph's central council tastes like a bitter pill, that's just because it hasn't been artificially sweetened.
Disclosure: At the time of writing, the author holds ETH, IOTA, ICX, VET, XLM, BTC and ADA.