Ben Jessel, NJ - Notary Pending


#1

Very excited to be part of the PoA community, having recently learned about the technology and its potential. I am consulting industry veteran, having been in management and technology consulting for Big-4 organizations for over 18 years.

Over the last four years, I have been involved in the pilots, business case development, business and technology solution development of numerous blockchain proof of concepts including: nine bank equity swaps projects with tier 1 wall street banks, syndicated lending blockchain platform with a tier 1 global bank, blockchain strategy for IATA (large international payments organization), and a Fortune 100 healthcare provider.

Today I head up the growth agenda for an emerging blockchain provider.

There are a number of interesting projects that are addressng public blockchain scalability challenges such as Proof of Stake/Casper, Level 2 scaling solutions such as lightening. PoA takes a really novel approach and I’d be excited to learn about how this technology works, it’s overall vision and goals and help shape its development.


POA Validators & Candidates Guide
#2

Welcome Ben - a very impressive background! Given your interest in blockchain scalability, what do you think are the pro/cons with some of the scaling solutions you mentioned and the feasibility in their implementation?


#3

So ultimately with scalability the questions are:

  • Can we make pow more efficient at an L1 level

  • Can we enhance existing solutions through level 2 solutions such as shading, side channels

  • Can we use more effective alternatives to proof of work that are less computationally intensive, avoid the challenge of mining concentration (51% issue) but still ensure distributed consensus?

  • Can we use alternative network models eg DAGs

  • Level 1 scaling: On the first point, there is some good work on level 1 scaling solutions such as chainweb which paralkelize PoW based solutions

  • Layer 2 solutions such as lightening and sharding do help but any time you are doing something outside the PoW network you have some element of compromise of security and fidelity; it opens up attack vectors. The ethereum proposed implementation of sharding has raised eyebrows - it may be sharding in name only.

  • Proof of Stake is interesting as it’s about incentivizing known validators who are selected based on their stake they have in their network where they are awarded incentives for behaving.

  • Challenge here is:

  • The skewing the network towards those with a lot of money, which is bad if not worse than skewing it to those that have a lot of mining power as PoW does - substitution of one bias for another; it’s just a different type of network politics where the powerful have outsized leverage

  • Determistic consensus and legal recourse. If you are a block creator or validating a transaction in PoW - it’s non deterministic and anonymous for the most part.

  • In the real world, if you are involved in the support of proceeds of crime you rightly go to jail. Yet in bitcoin that must happen all the time. However, when transactions are validated, that’s just 50% of the network validating - you cannot point to an individual and state that they were solely involved in that supporting act. Also with key rotation and the anonymous nature of PoW, identification of who block creators and validators are in the real world is hard.

  • Proof of stake is deterministic - if you have the largest stake, you’ll be playing this validators role; far easier to be identified as an individual node supporting crime. Moreover, key rotation is harder in PoS.

  • Now this is all debatable, and is hotly debated in the community but PoW has so far proved to be somewhat immune to money laundering accusations, so there are risks with trying something that is unproven in the regulatory environment.

  • Ethereum was seeking to adopt Casper - a major switch to PoS over PoW on an economy that is $30bn. That’s quite an overhaul. But not only is a fundamental change of approach concerning to some, but because Casper would have taken too long, a less complete version will go live instead; this project is not being led by the same thought leaders as Casper, which has led folks to be concerned.

  • As I understand with a DAG, you are effectively using the properties of the network around how data is distributed meaning you can have different communities working on building out different parts of the graph therefore leading to parallelism (blockchain is sequential referencing, DAG is referencing adjacent nodes).

  • That’s good as long as you have an even distribution of data in the graph. As soon as you start having the community referencing the same data, the graph bottlenecks…all good if everyone is trading all stocks equally on the fortune 100, bad if everyone starts trading Apple to use an analogy. Key here is the difference between maxmimum and minimum throughput - a network with a maximum of 1m TPS is useless if the minimum is 1 TPS - you need to have a narrow range of scalability and throughput bounds.

  • Low energy PoW systems like IOTA have promise but as they have a lower hashing difficulty, the threat of network takeover is high. Fine if you have devices distributed all over the world, so hard to compromise, but it needs adoption for that to occurs, and until then we seem to have a partial solution that contains some centralization. We also need hardware to evolve to adopt trinary operations for the full benefit to be realized, which is nascent.

What do you think?


#4

These are some fantastic points Ben! Very thoughtful comments on the benefits and risks of consensus protocols. It’s interesting because there are obviously trade offs between every protocol in terms of cost, scalability, centralization considerations etc. I think what we have for POA consensus protocol is interesting given that the incentivization model is for us to stake our identity in order to act in the best interest of the network and aims to resolve some of the issues that you may see in PoS.

Again, happy to have you with us and as a consultant myself who focuses mostly on financial services, it’s always great to get thought leadership from people who have had exposure to a variety of industries such as yourself. I encourage you to share some of your experiences on implementing blockchain solutions and some of the benefits/challenges you have witnessed.

Feel free to reach out with any other questions!


#5

Indeed, these are complex issues and ones that will need to be unpacked. I view POA Network as filling the gap around a ready to use Enterprise Ready dApp Platform (Hyperledger tried, but wasn’t able to sustain this push). POA Network is a project that is both accessible (wrt to dev talent pools) and having familiar properties.

The use of PoA is interesting, and when this key component is communicated to Enterprise clients - I think makes them a bit more at ease to explore - perhaps roll out that PoC. PoW’s issues, as you already stated, it being non-deterministic and anonymous makes for a ‘hard’ sell to this group. Also, I think that everyone understands the huge question marks surrounding PoS (whichever variation you’d like to consider). PoA seems to be a proper middle ground.

As for those other algo’s… I will admit this: I am interested (as all scientists should be) in the weird and crazy. So, I do have a soft spot for IOTA. When I saw you mention this, it brought a big smile to my face. I’m probably the only person in the valley that will admit to liking this project. I fully realize that are so many challenges (issues/whatever) - that it is not even funny… but, it is so crazy of an effort, and if just a small fraction is realized then the potential applications are just awesome.