Vitalik proposed last week that I share my fundamental research and design philosophy in a blog entry, I concurred but expressed that it was still evolving. My companion Jon West mentioned that everyone would truly value it if I disclosed my findings on Casper research, I largely agreed. Then someone on reddit advised me to concentrate on Ethereum.
Thus, here is the Casper technological narrative, presented as a chronological account of the development of the crucial technology, concepts, and terminology associated with “Casper research”. Many of our beloved blockchain figures are part of this narrative. This is my effort to narrate everything in an easily understandable, sequential manner so that you can observe our current position (and where we’re headed) with our research endeavors (so hold your critiques until the conclusion!). I’m aiming to release a chapter daily until it’s finished.
Please bear in mind that this represents my personal perspective, grasping what little I could glean throughout the process of working on proof-of-stake. Vitalik and Greg Meredith’s accounts will differ, for instance, as they each possess their unique perspectives on Casper research.
Preface: My journey into research at Ethereum
March 2013-April 2014
I instantly became fascinated with the Blockchain technology narrative when Bitcoin first (truly) captured my attention in March 2013. This occurred during the price surge related to the “Cyprus crisis”. I discovered cryptographic hashes, digital signatures, and public key encryption. I also learned about Bitcoin mining and the incentives miners have to sustain the network. For the first time in my life, I found myself intrigued by computer science and security. It was exhilarating.
Set against a backdrop of dystopian libertarian economics, it was underground developers (like Amir Taaki) pitted against central bankers in an extraordinary global struggle to liberate the world from the fractional reserve banking system. The blockchain revolution was beyond fiction.
I consumed content on reddit, tuned in to Lets Talk Bitcoin, and engaged with a lot of Peter Todd’s work. I lost funds on BTC-e (once as a result of taking advice from the trollbox). I debated with my friends Ethan Buchman and Zach Ramsay regarding technology. We learned about MasterCoin and the potential for creating systems atop Bitcoin, capitalizing on its Proof-of-Work network effect. When I first encountered proof-of-stake (PoS) in the 2013 alt-coin realm (thanks to PPCoin!), I thought it seemed like heretical sorcery. The idea of replacing miners with coins appeared to be an inherently peculiar proposition. Eventually, I concluded that the long-range attack issue was detrimental, and any solutions would necessitate developer checkpoints in one form or another (an opinion I absorbed from Peter Todd). Being a Bitcoiner in 2013 was one of the most intellectually invigorating experiences of my existence.
In January or February 2014, I encountered Ethereum for the first time. I viewed Vitalik’s YouTube presentations, and I met him face-to-face at the Toronto Decentral Bitcoin Meetups. He clearly had a far deeper understanding of the technology narrative than I did, which drew me in, this time towards Ethereum. Ethereum represented the promise of decentralization made approachable to me, a person with limited background. It was the concept of general-purpose smart contracts capable of accomplishing anything and disrupting any centralized system. It could be and achieve so many things that it was not always apparent what role Ethereum would ultimately occupy in the blockchain ecosystem. The blockchain tech narrative (as I interpret it) took an exhilarating turn with Ethereum, allowing me to be closer to the unfolding events 🙂
Having received an invitation from Russel Verbeeten at one of these meetups, Ethan and I attended the hackathon preceding the 2014 Bitcoin Expo in Toronto. (Vitalik taught me to utilize Merkle trees at this gathering.) I contemplated how to effectively incentivize and decentralize the peer review process for a couple of weeks since I had recently had a paper turned down by an academic journal. At the hackathon, Ethan and I attempted to assemble this type of system. Ethan handled most of the challenging tasks employing pyethereum, while I sluggishly pieced together the first GUI I had ever created. We secured second place at the hackathon (behind Amir’s “Dark Market”, which transformed into Open Bazaar). We got the chance to meet the entire Ethereum team at the Expo, and we secured invitations to the public Skype channels! Charles Hoskinson offered us positions: It was in April 2014 that we began our volunteer work for Ethereum. We even received @ethereum.org email addresses.
Thus, I entered the blockchain domain because I initially became captivated by the Bitcoin technological narrative, and later, the Ethereum technological narrative. I subsequently became engrossed in the proof-of-stake technological narrative, which I now recognize as exceptionally compelling. I intend to share it, maintaining fidelity to the timeline and the manner in which the components have been coalescing, to assist in bringing everyone up to speed on our endeavors. It may require several chapters, but the storytelling isn’t over until it concludes.
Chapter 1: Slasher + Security Deposits: The transition from naive proof-of-stake to modern proof-of-stake.
May 2014 – September 12, 2014
When Vitalik first indicated his interest in PoS to me in May 2014, first over Skype and later at a Bitcoin conference in Vienna, I was doubtful. Then he introduced me to slasher, which I believe he had conceived in January 2014. Slasher was the concept that you might forfeit your block reward if you signed blocks at the same height on two forks.
This provided Vitalik the means to directly address (and arguably resolve) the nothing-at-stake dilemma. (For those unfamiliar, the “nothing-at-stake” dilemma refers to the fact that the PoS miners’ best strategy is to mine on all forks, as signatures are incredibly inexpensive to produce). It also expanded our imagination to new realms of interactive protocols for discouraging unethical behavior.
Nevertheless, I did not feel fully content with proof-of-stake during this period (despite Vitalik asserting a few times that he believes “proof-of-stake is the future”) because I was truly enamored with proof-of-work. Therefore, throughout the summer I predominantly focused on proof-of-work challenges (ASIC-hard PoW, security sharing between PoW chains via “Proofs-of-Proof-of-Work”, neither reaching completion). However, I did propose the application of security deposits to a couple of contract developers on several occasions. This germinated the insight that emerged on the pivotal post-Ethereum-meetup night of September 11th, 2014 (props to Stephan Tual for organizing and inviting me to that event!).
Ethan Buchman and I engaged in late-night discussions about proof-of-stake in the “hacker” section rather than the “party” area of Amir Taaki’s squat in London. I connected various aspects and absorbed the significance of security deposits in the context of proof-of-stake. That night, I became convinced that PoS could succeed, and that bringing it to fruition would be immensely enjoyable. It was also the initialtime I encountered the unexpected magnitude of the PoS design space, through extensive discussions regarding attacks and potential protocol reactions.
Since the early hours of September 12th, 2014, I have ardently promoted (to everyone willing to hear) that blockchains transition to PoS due to its enhanced security. Amir Taaki was not swayed by my passion for proof-of-stake. At least Ethan and I were having a fantastic time.
The implementation of security deposits consistently amplified the slasher’s efficiency. Rather than sacrificing some earnings X, a demonstrably erroneous node would forfeit a security deposit (conceptualized to be around the size of X/r) on which the block reward X was intended to be earned as interest (at rate r).
One places a deposit to participate, and if one plays fairly, a modest return on the deposit is earned; however, if one plays unfairly, the deposit is forfeited. It seems economically optimal, and it is highly programmable.
Incorporating deposits into the slasher indicated that the nothing at stake dilemma was definitively resolved.
At least, I was convinced that it had been resolved to such an extent that we could not fathom why anyone would choose to develop a proof-of-stake system without security deposits, out of concern for nothing-at-stake issues.
Also on September 12th, 2014, I encountered Pink Penguin for the first time, thanks to an introduction from Stephan Tual. I eagerly shared my PoS revelations from the prior night. After I graciously turned down an offer from Eris Industries (now Monax) that week, Pink Penguin commenced sponsoring this research! (Thanks
At this juncture in the narrative, I was unaware of the various independent revelations regarding the application of security deposits in proof-of-stake systems discovered by Jae Kwon, Dominic Williams, and Nick Williamson.
Stay tuned… the upcoming chapter will explore the significant influence that concepts from game theory had in shaping the design objectives that ultimately led to Casper!
NOTE: The opinions expressed herein are strictly my own personal opinions and do not reflect the views of the Ethereum Foundation. I take full responsibility for what I have written and am not acting as a representative for the Foundation.