Truth, Consensus & Skin in the Game: July 16, 2017 Snippets
Hello again! Snippets is back for the second half of the year. We hope you’ll all get a chance to find some time to enjoy the summer, the great outdoors, and your friends and family.
Want Snippets delivered to your inbox, a whole day earlier? Subscribe here.
Cryptocurrency protocols have the potential to do a lot of different trust-related jobs — storing money, executing transactions, coordinating resources, and much more. In the last few months we’ve seen a frenzy of new protocols proposed, where entrepreneurs across the world are proposing radical new ideas for how to use distributed computing to solve problems and perform jobs in the real world. (Or, as others would suggest, radical new ideas for how to separate fools and their money.) And although the technical details of how they work may not necessarily be of interest to the average citizen, and the potential gains or losses may be too dizzying to contemplate in either direction, there’s an underlying philosophical idea about truth that underpins these cryptocurrencies — and it’s an idea we need to discuss.
The idea of the public blockchain, beginning with Bitcoin, addresses one of the challenges of a world in which everyone has a megaphone and no one can necessarily be trusted, and for which the “fake news” phenomenon has become an eye-opening alarm: How do we agree on what is the truth? How do we “vote” on it? And how do we prevent the truth from being hijacked or scrambled by mischief-doers? This is a problem that extends beyond crypto-currency, and will soon concern everything under the sun. So what should we know about it?
Trust and security are inherently expensive, and there’s no avoiding this expense being expressed as some sort of cost. If voting on what is true is free, or otherwise too easy, we can expect mischief, fraud, and a loss of confidence in the system by all. (Consider what has happened with the news media- a different kind of “distributed protocol for determining what is true” whose weaknesses have been crucially exploited. As a second point of comparison, imagine how elections might proceed if the verified final vote count was also validated by the distributed voting public, rather than by an independent electoral commission, and without any kind of cryptographic or otherwise rigorous corroboration of what has taken place. How quickly would people maintain confidence in this system?) It needs to be somehow expensive to vote on the record, as otherwise there isn’t anything stopping people from voting for many different versions of the facts, or seeding misinformation and voting in one’s dishonest interest. In other words, we are really asking the following: what kind of skin in the game should be required in order to participate in establishing what is true?
Quite a few different consensus protocols have been proposed in the cryptocurrency community for this skin-in-the-game problem, but the two that have been applied most seriously in current and in anticipated implementations are proof-of-work and proof-of-stake. The proof-of-work consensus protocol used in Bitcoin expresses it this way: “In order to earn the chance to contribute, you must do a certain amount of objectively difficult work.” This work in practice is solving a certain kind of math problem over and over again until you find a correct solution that gives you the right to transcribe the most recent “status update” to the truth, which other participants will then verify and then on top of which transcribe future status updates. The skin in the game here is that you must consume an objectively scarce resource (energy, in the form of hashing power) in order to vote, and unless you control 51% of the total voting power there’s little use in needlessly burning resources for nothing. Your past expenditure of resources is at risk.
In the proof-of-stake consensus protocol that many anticipate will soon be adopted by Ethereum, it’s a different story: “In order to earn the chance to contribute, you must put up collateral that is at risk if there’s any misbehaviour.” Rather than the enforced consumption of resources to reach the starting line, it’s the threat of losing your resources that serves as skin-in-the-game for anyone who wants to participate in establishing what is true. Your future ownership of resources is at risk. The tricky thing here is, how do you police mischief? Can this be trusted to a set of rules, or must there be something inherently subjective to the idea of evaluating and punishing misbehaviour?
The difference largely comes down to the following: skin-in-the-game in the form of already spent resources at risk for being spent uselessly, or skin-in-the-game in the form of not-yet-spent resources at risk of being lost. Next week we’ll talk about the fundamental differences between these two approaches, why it matters for determining the truth, and what it means outside the cryptocurrency world.
Life and death:
Late in life, Thoreau became a serious Darwinist | Randall Fuller, Longreads
Pesticides could hike risk of catching schistosomiasis, a parasitic worm | Erik Stokstad, Science
Cloud notes:
What happens when AWS goes down: an interview with CTO Werner Vogels | Rob Price, Business Insider
Windows Server is running on ARM CPUs, Azure is next | Chris Pietschmann, Build Azure
Where we live and what it means:
America’s future is Texas | Lawrence Wright, The New Yorker
The invisible life of urban spaces | Rohit Aggarwala, Sidewalk Labs
Is California anti-family? | Joel Kotkin, Orange County Register
The Telecom world:
Both sides are wrong in the net neutrality debate | Dean Bubley, Disruptive Wireless
China tells carriers to block access to personal VPNs by February | Bloomberg News
Microsoft’s rural broadband solution: TV white space | Jay Greene, WSJ
Other reading from around the Internet:
Two kinds of caution | Scott Alexander, Slate Star Codex
A brief history of existential terror | Taylor Pearson, Ribbonfarm
Can Venture Capital be saved? | Mitch & Freada Kapor
Paying professors: inside Google’s academic influence campaign | Brody Mullins & Jack Nicas, WSJ
Mets, Patriots owners said to pay $20M for eSports teams | Bloomberg Technology
Amazon: the company with 100 CEOs cannot be stopped | Anshu Sharma
Elon Musk: there will not be a steering wheel in 20 years | Ina Fried, Axios
How poverty affects the brain: findings from a new study in Bangladesh | Carina Storrs, Nature
In this week’s news and notes from Social Capital, we’re delighted to share news of a new company joining the family: Cover.
Introducing Cover, a consumer-centric insurance company | Karn Saroya
Cover raises $8M to insure your belongings | Katie Roof, Techcrunch
Insurance has traditionally been an industry where you could get away with a lot: since customers are often mandated to purchase insurance and the business model is really good, insurance providers and brokers have been able to get away with being exceptionally consumer un-friendly. But that’s finally changing, and Cover is leading the way with a new product that almost feels like magic. CEO Karn Saroya writes:
“Insurance in the future will be about services + insurance. Our company is geared towards figuring out the right mix of services that resonate with customers such that they’re getting the most out of their premium dollars.
What does this utility look like? Today, we’re launching the first ever implementation of computer vision in insurance, available now to all of our users in the Cover app. Our customers can literally walk around their homes, and Cover will auto-detect, identify, and catalogue their property against an existing policy, or a new one if they need it. We’ll also set up price drop alerts on any policy, and port the insurance you have on your credit cards to Cover. These tools uncover insurance our customers already have, automate behaviors they already perform (shopping rates), and make insurance as intuitive as getting your property on camera. Take our services + insurance view to the extreme, and insurance premiums start looking like the next ad dollars — spinning out free products and services that bring joy to insurance customers, the basis for a consumer-centric insurance company.”
Among the unexpected perks of bringing 21st century tech to a very old-school line of business? A bit of friendly culture clash. Karn notes: “Until a few weeks ago, Cover was being run entirely out of the living room of the house where our co-founders live, myself included. Post Y Combinator, we’ve entertained the C-suites of insurance companies from around the world at our kitchen table. It was pretty funny to ask armies of seasoned insurance executives to take off their shoes when they showed up at our place. They were much cooler about this than I expected.”
With their Series A in the books, Cover is hiring for both engineering and sales roles — you can find more about the company and their values here, and see job listings on Angellist here. Ready to make the jump, or know someone who does? Email Karn directly at karn@usecover.com. And finally, are you a human who has a house or an apartment, and there is stuff in it? If yes, then you should use Cover. You can download it right now, right here, for iOS.
Other notes from this week: Slack has a bright new batch of startups it’s backed through the Slack Fund, as a part of a broader bet on the developer ecosystem that has grown on their platform with every passing week:
Introducing 7 new Slack Fund companies | Slack Blog (a.k.a. Several people are typing…)
And finally, congratulations to Mathilde Collin of Front and Roddy Lindsay of Hustle on the inclusion in Forbes’ Rising Stars of Cloud Computing series. Well earned!
Have a great week,
Alex & the team at Social Capital
