Regarding ‘Burned on Blaze’
On November 21st, 2019, Mr. Joseph Morton, on his blog called ‘The Digital Decrypter’ posted a critique of Fandom Sports Media’s move towards Blaze blockchain. While I appreciate Mr. Morton’s inquisitiveness that prompted him to share his detailed views, I think it might be only appropriate to answer some of the queries Mr. Morton has raised.
To give you a brief overview, I am the lead developer at Blaze Protocol and was closely involved in creating its underlying consensus algorithm, WPoS. Hence, my response to Mr. Morton on this can be considered as one from the original creators of the protocol.
I will start off with the first observation Mr. Morton had, as he puts it:
“The first point that jumps out at you is “nine-dimensional space.”
“My pet peeve with all of these different companies is the different words used to describe basic blockchain functionality.”
One of Blaze’s primary purpose has been to make blockchain simple. We want the technology to be adopted by millions, and this is only possible if we keep its jargon as words that are used by everyone, and have a precise meaning in the context of blockchain. The term ‘dimension’ is used as it explains the structure of Blaze much more clearly — Blaze does not have a ‘layer’. When we use the term ‘dimension’ — it indicates a form of extension, which is what Blaze’s dimensions are. We initially decided to call it a term popular in the community called ‘side-chain’ — but the definition of sidechains is not clear and its meaning varies from one project to another. Hence, we decided to simply term them ‘dimensions’.
When it comes to ‘Nine Degrees’ — Blaze is designed so that it is limited to 9 degrees of dimensions. Slightly relevant, Blaze has scaled up to 50,000 TPS with only 2 degrees dimensional structures under lab conditions.
Mr. Morton continues with:
“Through a peculiar consensus protocol mix of distributed proof of stake and proof of authority, closes the block, and moves the chain along with the information imprinted.”
I would say that this abridged version of the definition trivializes the intricacy of Blaze’s consensus mechanism. To paraphrase, WPoS is a chain based proof of stake protocol where executives are elected based on voting and prestige score and are allotted different time slots randomly. WPoS is a substantial amount of differences from proof of stake and proof of authority consensus mechanisms (refer whitepaper section 4.3.2).
Further, he says:
“In standard English, it’s a consensus protocol that eliminates needless data to speed the data processing along, while requiring a certain percentage of nodes to agree on the contents of the block to close. This involves communication between the blocks and we have both direct and indirect communication methods to accomplish it.”
There is a clear semantic error here. Needless computations (not data) are eliminated or executed in parallel. Crosswise communication is a slightly more elaborate process than how Mr. Morton puts it. It involves verifiers and endorsers for communication between multiple dimensions. I would encourage Mr. Morton to read Section 4.6: “Horizontal and Crosswise- Transaction”.
In continuation,
“The problem with this entire arrangement lies with the consensus protocol. In order to close the block, 67% of the nodes must agree on its contents. Nothing wrong with that. But the problem is the same as Distributed Proof of Stake consensus in that there’s nothing to stop individual nodes from communicating, forming a bloc, and agreeing to take over the functionality of the blockchain — effectively choosing and re-choosing the same closers every time — regardless of which version is correct.”
Consensus protocol of blaze needs more than 2/3rd of the nodes, it’s a requirement for the system to be byzantine fault-tolerant just like other distributed consensus algorithms. The very foundation of blockchain is all about the honesty of the majority of network and it is the first assumption before any distributed consensus algorithm is designed. Every voted executive is allotted a time slot randomly to forge a block and it’s not possible for a single executive to forge block in more than one slot in a certain period of time. Even if malicious nodes collaborate they will be voted out. Bitcoin and Ethereum are considered secure networks and even there few mining pools control majority of hashing power and can make a 51% attack, small users have no chance of contributing to the network. In the case of blaze, everyone can vote and have a say.
“All that’s needed to be a node operator, and an executive, is the technical specifics and the votes.”
I would say that Mr. Morton is low-key praising Blaze here. This is what makes blaze more distributed than Ethereum and Bitcoin where nodes need mining rigs to mine a block. Everyone who holds Blaze has the power to vote and will vote out all the executives who try to act maliciously.
“The governance model has top down pyramid structure: Those with power reside at the top of the pyramid. They make the decisions, including dispute resolution, after which a 67% agreement is required to make it binding. This centralizes an awful lot of power in the hands of a few people on the network — people who can be persuaded, manipulated, bought-off, or coerced — into doing whatever it is you wanted them to do. Those folks needn’t even be executives in the company — that’s the kicker.”
All the executives are selected based on continuous voting and the votes are counted within hours. It’s not as if once an executive is selected it will always keep on forging blocks for years. The real power lies in the hands of voters, blaze stakeholders, executives are just their representatives. If more than 67% of executives are manipulated than yes they can harm the network just like any other blockchain. But the difference here will be that they will soon be voted out by stakeholders.
And lastly,
“Too many moving parts. If a blockchain is going to scale-up sufficiently to beat Visa, it’s going to have to do more than clear out useless data from the blockchain.”
Mr. Morton, I have read some of your articles. You know more than I do that the scalability issue with blockchain is a complex problem and a solution is bound to have more moving parts. So, we are not just clearing useless data, we are eliminating useless computations, making parallel computations wherever possible, communicating between multiple dimensions via crosswise transactions. Our aim has been simple and clear, we’ve been moving forward focusing only on scalability to support a massive influx of users, which platforms like Fandom Sports need.
Finally, I would like to thank you for writing a detailed critique of technology. It’s always great to discuss tech with enthusiastic bloggers. I hope that this article will help the community to have a better understanding of Blaze.
