What is EOS and Why Should I Care?

I have been hearing quite a bit about EOS. Big claims. Big conspiracies. Whatever the case, I generally find the best place to start with any new blockchain project is the White Paper. So I did. Yeah, I sat down and read all fourteen tedious pages. Now, I don't pretend to fully understand everything I read. But I understand enough. I thought I would offer my layman's perspective to anyone interested in learning more about the project, which is in the process of raising a King's Ransom in token sales. As a side note, the white paper begins with a disclaimer that the token sale is not the same as the actual token described for blockchain use with the adoption of the EOS.IO software. Currently, I do not own any EOS tokens nor am I affiliated in any way with EOS. These are simply my observations and thoughts.

Why EOS?

It seems to me that the essential problems that EOS intends to address are scaling, transaction cost, integration with other applications and latency.

These are my key take-aways from the white paper:

• EOS uses Dan Larimer's concept of Delegated Proof of Stake.

• EOS has an account structure that allows for sub-accounts and a domain-oriented structure.

• EOS has a bandwidth management aspect like the Steemit platform.

• EOS utilizes a Constitution with a 17/21 formula for governance that allows for code changes, constitutional changes and other administrative functions without creating forks.

• EOS has a light client for inter blockchain communication.

• EOS supports using one username across multiple applications.

These features are implemented a number of ways. EOS blocks are resolved every three seconds. It only requires 15/21 producers to confirm a block. Communication latency is further addressed within the blocks by further division. Each block is divided into cycles, each cycle is divided into threads and each thread contains a list of transactions. Each transaction contains a message to be delivered. Static analysis will verify that no two threads in any cycle contain modifications to the same account.

Resources are managed. Bandwidth/log storage, computation and state storage are managed by assigning bandwidth based on stake. The stake is calculated on a three-day rolling average. Like the Steemit platform, delegation is permitted.

One of the interesting aspects is the integration of other chains. Blockheaders are used to prove the existence of transactions with a proof less than 1024 bytes in size. 2MB of data can handle a full days worth of proof. Latency is slightly longer on interchain communication. It takes 21 proofs at 100 percent certainty, extending confirmation times closer to one minute. 100 percent validity is intended to maintain validity of the blockchain consensus.

Is this all a big scam???

I used the word all here. It doesn't appear to be a scam. It appears to be a solution to a serious problem that needs to be addressed. Could it be a scam? Of course it could. I have seen a variety of posts on this topic, but it really boils down to one point for me. If not now, when? If not EOS, who? Maybe EOS won't be the final solution, but these are questions we need to ask ourselves going forward. As this technology mainstreams, we need to address issues of scaling, cost, resource allocation, proofing, trust, latency and interaction between blockchains. This is a solid step in that direction. There were a few things in the white paper I had questions about. It may be my own ignorance, as I am barely a lay person trying to understand complex ideas. But let me take a stab at my questions.

The White Paper talks about recovering stolen keys:

The EOS.IO software provides users a way to restore control of their account when their keys are stolen. An account owner can use any owner key that was active in the last 30 days along with approval from their designated account recovery partner to reset the owner key on their account. The account recovery partner cannot reset control of the account without the help of the owner.

There is nothing for the hacker to gain by attempting to go through the recovery process because they already "control" the account. Furthermore, if they did go through the process, the recovery partner would likely demand identification and multi-factor authentication (phone and email). This would likely compromise the hacker or gain the hacker nothing in the process.

This is my concern about the way this stolen key process is set up...it assumes that keys were actually stolen. So the assumption is that a hacker would not go through the recovery process because they already control the account. But what about an account that was compromised, or credentials that were compromised, and the attempt thwarted? The account keys would be changed and the owner comfortable that no losses occurred. In this situation, with prior keys, a bad actor could claim they are the true owner and use the prior keys as proof. By porting the victims phone (as I have seen happen), they could potentially have the ability to pass 2FA to gain control of an account they never actually controlled.

There could be a way or ways to mitigate this. But I was just concerned at reading the assumption that there wouldn't be a vulnerability here. There may not be. But I thought I might play devil's advocate.

One other thing I wondered regarding the white paper was on the ability for nodes to pick which applications they want to run.

Scaling blockchain technology necessitates that components are modular. Everyone should not have to run everything, especially if they only need to use a small subset of the applications.

An exchange application developer runs full nodes for the purpose of displaying the exchange state to its users. This exchange application has no need for the state associated with social media applications. EOS.IO software allows any full node to pick any subset of applications to run. Messages delivered to other applications are safely ignored because an application's state is derived entirely from the messages that are delivered to it.

This is more of a question than an issue. Without fully imagining what applications would eventually run on this type of platform, I wonder how this might affect specific applications and whether this might limit resources for valid applications that either fall out of favor or are otherwise ignored by node operators. Just a thought, and probably overthinking it.

Is any of this really Decentralized

That brings me to my next point. And also goes back to the question "Is this a scam?" No system is truly decentralized. This type of system serves as a representative democracy. There are voting implications to bad actors, but that doesn't stop bad actors from acting, either.

I will give an example of my concerns. I have been using Steemit for about six months and have noticed some anomalies from witnesses. Witnesses have a lot of power. Not just disproportionate. There is a reason why there is a caste system of whales, dolphins and minnows. Most witnesses are in it for the long term benefit of the platform. But I believe (no, I don't have proof), that there has been manipulation of the SBD price and there have been quite a few sock puppet accounts created by the heavy hitters. The synthetic identities are used for a variety of purposes, but all ultimately serve to further some goal of a witness intended to be anonymous. Or at least pseudonymous.

Any system that is a representative democracy has an element of centralization. That is not necessarily a bad thing. But we empower the few to make decisions for the many. The 15/21 formula for improving latency seems okay to me. I don't really have an issue with that. I am a little bit more concerned with correcting blocks with a 17/21 consensus among the delegates. It may be preferable to forking the blockchain, but it seems like a lot of power is being vested in a few people. Based on what I have seen, the temptation for abuse may be too much to overcome.

The way forward

I think EOS is a promising prospect. I know it doesn't sound like it. But I do believe we need to have these discussions. When we can build a foundation upon which to build a broader, interconnected system, we have realized the true potential of blockchain. It is a transformative technology. I like the idea of single user identities having connectivity to numerous applications. I like the idea of interchain communication. I like the idea of low latency, free interaction, bandwidth driven resource allocation and high volume capacity. I don't believe these are challenges that are snake oil as you may read other places. It is viable. There is a balance to be struck. I don't know what that balance is. But it will be interesting to see where all this goes.

I definitely believe that EOS is on the right track. Time will tell.