A New Era of Publishing

I always fancied myself an academic. A real go-getter in school. Hardly a teacher's pet - I certainly had a knack for shit-disturbing - but I'd also beat myself up if I didn't get a 95% on everything. For some reason I thought that mattered. Whoops.

By university I'm working in biology labs. Mostly interested in cell biology and using microscopes - watching the basic processes of life unfold before my eye in brilliant red and green fluorescent colours while the grey scale contours of little organic machines dance a biomechanical jazz.

Somewhere along the line I remembered, frighteningly, that I hadn't programmed since high school. So I started coding up various bits of math I knew and running funky simulations, the wholly inferior (yet fascinating) digital equivalent of what I watched through the microscope.

Before I knew it I was building neural networks to pull data out of cellular microscopy videos. But that wasn't really what I wanted to do. What I wanted to do was understand the principles by which complex, highly functional, anti-fragile systems emerged from their much less formidable components.

Then, somewhere in early 2013, I heard about this Bitcoin thing. There was a moment when I suddenly realized that the principles I was after were at play here in a remarkable way, supporting the early stage of a new form of organism with a biochemistry built from internet protocols rather than chemical physics. Well this was interesting.

It was also about the time I started to pay attention to economics, too. To the structure of the global financial system, the nature of money, the blatant corruption and incentive-incompatibility at the intersection of the public and private sectors of society. Fitting. Western civilization is clearly not a healthy organism. And neither, it became increasingly clear, is academia, specifically, the notorious Publishing Industry.

Yuk. Here we had a multi-billion dollar industry with complete control over the publication of research and development; work done for the betterment of our and other species; funded with public tax-payer money; the great accumulation of man-millenia of investigation and discovery; and it was behind a pay-wall. And not just any paywall. It was a "you could feed a family for a day for the price you just paid for single journal article" kind of paywall. Try building a bibliography out of that.

But it wasn't just the paywall. It was the process. They extorted libraries with obscene subscription models. They built a walled garden and an Old-Boys club and an elitist arrogant regime of autocrats. They dogmatized anonymous peer-review and subsequently broke it completely. Most published science today is known to be false. It can't be replicated. The statistics and conclusions are piles of bullshit cobbled together by students struggling to stay a float in the piranha-infested waters of publish-or-perish academic institutions dominated and hamstrung by bureaucrats that would be more productive if they were monkeys flinging shit around ...

Er. Lost my composure for a second there. Anyways, it's clearly a broken system. Backwards incentives. Utterly failed. Aaron Swartz had recently figured all of this out too; in his martyrdom, and even long before, he became an inspiration to many. I couldn't stand to be apart of it anymore. So I defected to the Bitcoin world. Started running Bitcoin ATMs. Ethereum was also up and coming, and I was pretty much all in on that. I was still in-school, a Masters degree in Engineering, but only because I had promised a few people I'd see it through - I learned early in life that a man has two things in this world: his word and his balls.

The degree was supposed to be in machine learning. But it was becoming increasingly clear it would be based in the emerging blockchain technology instead. Nonetheless, machine learning was a ripe environment. Less for the content, more for the context. The medium is the message.

Machine learning researchers, like other computer scientists, had become increasingly aware of the value of open source. They were starting to go big time with it. Not only that, they had never cared much for journals either - these guys held conferences, and published all the work in conference proceedings! Pretty much everything got pre-published to arxiv.org as well. Of course, their algorithms were also starting to become mindbogglingly successful - neural networks are better than Humans at all kinds of non-trivial tasks now; they're even beating experts at our hardest board games.

Yann Lecun, one of the principle figures in the field, decided to try a new approach to publishing for a conference he ran. Put simply, he wanted to create a free market for peer review, where all content and reviews are published open-source, and reviewers compete to provide the most accurate reviews. Most especially, reviewers were rewarded most for positively reviewing relatively unknown content that would later become highly influential.

Now here, finally, was an incentive-compatible publishing model! It was something I wanted to see deployed at a large scale. And obviously, I wanted it to be decentralized - no central point of control, no gate keepers, a fully peer-to-peer review system upon which a new era of scientific publishing might begin.

So I set about trying to design it. With my close friend and collaborator, Vlad Zamfir, we entered the "In Crypto We Trust" hackathon in Toronto in 2014 to prototype the idea using Ethereum smart contracts. We called the thing CryptoSwartz, in honor of Aaron, and we came 2nd place. Notably, 1st place in that hackathon was Amir Taaki's DarkMarket, which became OpenBazaar. 3rd place was a lock controlled by a smart contract on Ethereum - an idea which would eventually spawn Slock.it, The DAO, and the latest drama in the Ethereum community.

Vlad and I have spent the last two years working on various pieces of the infrastructure we feel will be necessary for building out the CryptoSwartz vision. Most recently the focus has been scaling - the system better scale if its going to support the publication, review, and reputation of the world's abundance of research. For me, it means a public and heterogeneous cryptoeconomic system built out of fast consensus engines with clean separation between consensus and application state and secure light-clients - we're calling it Cosmos, but it's really just Proof-of-Stake sidechains with a few nice twists.

In the meantime, it looks like Dan Larimer up and built a production-ready iteration of something like CryptoSwartz, only he called it Steemit. Awesome! I'm thrilled this exists and I hope to see it succeed. Now, to scale the sucker to take on the world's research publishing needs. Ahoy!