DAO ≠ D.O.A.

Before we start talking about the great $79 million DAO robbery, let’s make a quick introduction.

The DAO is a Decentralized Autonomous Organization (“DAO”) – more specifically, it is a new breed of human organization never before attempted. The DAO is borne from immutable, unstoppable, and irrefutable computer code, operated entirely by its members, and fueled using ETH which Creates DAO tokens.

Thus spoke The DAO.
Which is just one possible way to implement Decentralized Autonomous Organizations.

A translation attempt into plain English may sound like this:

In traditional western economies, capital ownership, production and consumption are separated entities:

  1. Uber investors pour billions of USD into a company they own.
  2. The drivers invest into production (CAPEX like cars, OPEX like gas and insurances, their time) and pay their USD tribute to Uber’s shareholders.
  3. Passengers pay USD for the ride.

A decentralized autonomous organization isn’t a shareholder  construct, but a stakeholder model based upon securely transferable crypto tokens.

  • Every token holder is a stakeholder in the DAO’s ecosystem
  • Tokens can be held …
  • … or circulated to pay for services rendered or products received …
  • … or exchanged into another crypto token (e.g. Bitcoin) or any legacy currency


It’s a radically different type of participatory economy and may offer the chance to fix a dangerous flaw of our current monetary system:
– the “real” economy is dwarfed by an unbridled financial system
– the financial sector is pretty much decoupled from the “real”, productive economy
– but both spheres share the same tokens to exchange value: our traditional currencies like the EUR, the GBP or the USD
– those currencies are basically minted and controlled by the aforementioned financial sector.

A DAO token works like a programmable complementary currency. Traditional alternative systems looked like the Wörgl Schilling: a piece of paper used to locally exchange value to keep external problems at bay. Being just locally accepted is the key constraint – and the defining feature. Because the intent behind is purely local.


DAO tokens resemble complimentary currencies in this. They are constrained currencies. Traditional currencies are pretty much universal: highly fungible currencies like the USD or the EUR can be used to pay for any kind of product or service or asset pretty much all over the globe.

The Wörgl Schilling was only valid in Wörgl, the Bavarian  Chiemgauer is only accepted in this beautifully set local economy:

chiemsee_vonkampenwand … but not in Wörgl, located just a one hour car drive further down south in Austria.

Like with traditional complimentary currencies, the DAO token’s constraint is it’s limitation to a specific economy. It may be tied to a locality (like with the Chiemgauer) or a specific private entity (like airline miles, which are a certain form of private currency) – but is much more versatile.

In the DAO, the token is not only used to exchange value.

  1. Every token owner is a stakeholder of the specified economy.
  2. The token itself is programmable. Ideally, it becomes an intrinsic part of the whole process, not just the value exchange.

Token holders are a bit like owners of printed bearer shares: he who owns the physical share is the rightful owner of the asset represented in the paper. The company’s central ledger only lists the shares, but knows nothing about their ownership.

Crypto tokens, be it DAO or Bitcoin, pretty much automate all authentication, validation and transaction processes needed with an amazingly safe technology. Traditionally, all those transactions are safeguarded by a central authority. To buy shares of a company, you need to trust the company as the issuer, the stock exchange as the trade facilitator, the clearing house as the middleman, the settlement process for the exchange of assets (money/shares), and the custodian for administering your held securities.

Crypto transactions are pretty much trustless, meaning: as long as the crypto process is untainted, the whole chain of the transaction, from trade facilitation, clearing, settlement to custody services is inherently secure.

So how come somebody can instigate a rather dubious $79 million transaction?

Let’s go back to the trustless thing. If you read really carefully, you might have noticed I left an important piece out of the trustless specification: the issuer of the share.


And here’s the reason. Meet Victor Lustig. The man who sold the Eiffel Tower – twice. His con was actually pretty hilarious. He convinced a couple of Parisian scrap metal moguls that he represents the French government and they should bribe him for the right to melt down the rusting iron world wonder.

Some misplaced trust in charming Lustig later, the tower was still standing, the government still the owner of the cast iron hulk and Lustig’s target, one of the scrap metal dealers, a bit richer in experience and bit poorer in funds.

A trustless crypto transaction wouldn’t have affected Lustig’s con at all. Like every gifted con man, Lustig leveraged the conditio humana.

Every transaction is a chain of trust. The perceived transaction started with a land register certifying the French government as the rightful owner of the tower and ended with a cash transfer, a trusted means of value exchange.

But in the Eiffel Tower case, the starting point of the trust chain was Lustig and his made-up credentials. Or, to use crypto speak: the Genesis transaction was not building and owning the tower, but Lustig coming up with a fake identity and a masterfully implemented storyline.

Let’s go back to the DAO. In a rather spectacular crowd funding, a quite substantial amount of (crypto) money was raised. The basic premise:

Historically, corporations have only been able to act through people (or through corporate entities that were themselves ultimately controlled by people). This presents two simple and fundamental problems. Whatever a private contract or public law require: (1) people do not always follow the rules and (2) people do not always agree what the rules actually require.

From the DAO Whitepaper.

The offered solution:

The DAO is self-governing and not influenced by outside forces: its software operates autonomously and its by-laws are immutably chiseled into the Ethereum blockchain.


Or, in a nutshell:

  • the problem: people are not always following rules or not always really agreeing what those rules really do mean.
  • the solution: immutable contracts.

Which are a really great solution for many real world problems. But not for the problems they try to solve. Because they missed (3) people cannot foresee all consequences a contract or by-law may have

This is not a new thing, born out of crypto contracts. Matt Levine brings a great example in his Bloomberg piece Blockchain Company’s Smart Contracts Were Dumb.

One more story, one of my all-time favorites. The California electric grid operator built a set of rules for generating, distributing and paying for electricity. Those rules were dumb and bad. If you read them carefully and greedily, you could get paid silly amounts of money for generating electricity, not because the electricity was worth that much but because you found a way to exploit the rules. JPMorgan read the rules carefully and greedily, and exploited the rules. It did this openly and honestly, in ways that were ridiculous but explicitly allowed by the rules. The Federal Energy Regulatory Commission fined it $410 million for doing this, and JPMorgan meekly paid up. What JPMorgan did was explicitly allowed by the rules, but that doesn’t mean that it was allowed. Just because rules are dumb and you are smart, that doesn’t always mean that you get to take advantage of them.

Contracts have always been a complicated affair. Because they have to formalize a stable framework around fuzzy intentions by using language – which as a tool is inherently fuzzy as well.

The proposed solution for this inherent fuzziness created by the mismatches of intent and description and the thereby caused mismatching realities is probably a bit too ambitious: bug free software.



And what do you know: somebody smart quickly outsmarted the contract.

I have carefully examined the code of The DAO and decided to participate after finding the feature where splitting is rewarded with additional ether. I have made use of this feature and have rightfully claimed 3,641,694 ether, and would like to thank the DAO for this reward.

It’s unclear if the text has been written by the hacking trickster, who just wants to add a bit of insult to the injury. But the consequences of his contract are actually rather unclear as well. He may just be entitled to keep the load.

Because The DAO as a non-organisation constructed itself around the premise of its own infallibility. Read this part of self descriptive hubris:

The DAO will be deployed as an exact implementation of the Standard DAO Framework. The Whitepaper therefore describes perfectly how the DAO functions and is a great place to start learning more.

… exact implementation … describes perfectly …
Well. Obviously not that perfectly exact.

In the DAO’s belief system, acts of people are the problem, so let’s move them out of the equation. This created an entity ready to be preyed upon by other people of rather questionable intent. With the attack vector being people not being able to create 100% perfect contracts.

Hard core smart contractors don’t see a problem with this. Win some, loose some: it’s part of the package of immutability. Changing the rules after the fact may be technically possible, but violates the core principle of a Decentralized Autonomous Organization. Rolling back those transactions by an deus ex machina-act would inherently destroy the trust in the perfect engine: mind you, it worked actually without a fault.

Which is probably right. Because in their hubris, The Dao tried to construct themselves as an infallibility engine without any meaningful mechanisms for mediation or arbitration or recourse. And saving The DAO by ex-post changes might really hurt the underlying case for Decentralized Autonomous Organizations.

On the other hand: creating a machine, which enables smart contract-con men to systematically defraud unsuspecting token investors, who wouldn’t have any path of recourse at all … this sounds like a solid way to implement fringe system of very limited reach and effect.

As VC Albert Wenger writes: The Path to  Learning requires Failing: The DAO

Blockchains and smart contracts are amazing new tools in our overall technological toolset. We have to learn how to deploy them to the best uses (many of which have yet to be invented). That will take failures. The DAO is not the first one (e.g., Mt. Gox) and won’t be the last one.

Unfortunately, the first DAO failure might have been somewhat expensive.







Excess Capacity

What’s the economic driver behind the so called sharing economy? Robin Chase points into one direction. Thanks to technology, we can build platforms which enable us to harvest the excess capacities all around us. As a co-founder of Zipcar, the car-sharing trailblazer, she knows what she’s talking about. Owned cars are sitting around most of the time. Rentals you get for 24 hours (which, hopefully, is a bit longer than you’re actually going to drive it). A Zipcar you get by the hour. Daimler’s Cars2Go are even rented by the minute. That’s harvesting the excess capacity of a massive chunk of hardware which usually hangs around at the curb like idling teenage mall rats after school.

Excess Capacity
Robin Chase on Excess Capacity: too much of something can become a good thing.

Her talk was partly a compressed version of her new book Peers, Inc. The gist: linear solutions for exponential problems just don’t scale. Peers, Inc. is about harnessing massive problems with scaling. How was AirBnB able to quickly offer more as much bed-inventory as the largest hotel chains? Because their technical platform leverages the power of the people to pool their excess capacity.

Of course we do know by now, that some of the drivers of the sharing economy are not that benign. Granny letting out her spare bed room once a year sounds nice. But Mr. Greedy creating new inventory by taking 3 bedrooms condos off the rental market to rent them out by the day is a rather excessive approach to excess capacity.

Harvesting excess capacity is the underlying scalable model. Mr. Greedy may be driven more by taking advantage of regulatory arbitrage. But this is a problem which fairly easily can be remedied.
The complicated part starts when sharing economy entrepreneurs understand, that one excess capacity in later stage capitalist societies is man power. Look at Über: the drivers take on the capital expense of buying the car, take on the operating expense of maintaining it, take on all the risks.

And that’s how we come back to Robin’s talk. The occasion was our biweekly Bitcoin Startups Meetup. How does this relate? Currently, most sharing economy platforms are driven by shareholder value. The Bitcoin (or crypto) model works differently. It’s a stakeholder model.

Some small steps are already happening. A couple of Denver cabbies are asking: What If Uber Were a Unionized, Worker-Owned Co-Op? Joel DietzSwarm is pushing forward in many different ways.
Crypto-based decentralized applications don’t need statements like “Don’t be evil”. Their DNA is sequenced and public from the start. This includes the potential switch of the underlying economic model, from rent-seeking shareholders to a revenue-sharing stakeholders.

Addendum: here’s some more material from a talk I gave last year on the topic: “The new decentralized sharing economy and crypto coins”

Don’t Pass Go (go directly to PRISM)?

The genie is out of the bottle, Obama is checking your email, and you can’t push toothpaste back into its blablabla. Welcome to our present defeatist state of surveillance.

The NSA is reading your email? Well, so do Mark Zuckerberg and the Google Bros. And, look at your inbox stats: most likely, they are putting in more effort into this task than you do. OK. Game over, don’t pass go, we all go directly to Prism.

Not so fast. This online world of ours is still in its very early stages. I’m dabbling around there since 1992 or such. As online years count like dog years, I should be about 170 years old (just like Ray Kurzweil, when his supply of dietary enhancements finally runs out). But what are a couple of centuries, if we put things into perspective. Between the invention of democracy (Athens, 500 BCE) and it’s fairly widespread adoption in the late 20th century you can count more than two millenia of feudalism, absolutism, and other -isms. Widespread alphabetism took even longer, took a hit with the invention of tv to finally resurge with the Internet.

Massive societal changes do not happen over night. If not induced by a catastrophy like an asteroid wiping out the dinosaurs (welcome, mammal), a massive war clearing the path for independence day, or 9/11 (good bye nail clippers on air planes, hello total surveillance for safety).

Which leads us back to our current sadly defeatist state of the webs. Let’s put one thing clear: the massive collection of individual data is not a recent bug. It’s a feature of every digital system, where processing power is constantly on the rise and the cost of storing data falling on an hourly base.

The core question is: who owns this data? Who’s allowed to toy around with it? If my baker or corner super market knows about my eating habits, it’s quite OK and helpful. If some secret entity concludes, that because I prefer Halal Döner, my travel patterns should be monitored, we’re entering a very troublesome area.

For quite some time, we mostly did choose to ignore those ramifications of our digital lifes. Or, to put it like that: some proposed a happy hippie hippo lala-land, a united nations of onliners, where the evil forces of meatspace can be safely ignored. Others preferred an Ayn Randian powerplay to achieve the status of robber baron of the virtuality. And, to be sure, military and governement didn’t fall asleep at the wheel. Let’s not forget: the early stages of the Internet was funded by the department of defense. And the global rollout of the Internet steamrolled all national online plays, like France’s Minitel or the German Btx.

In a recent blog post, Emin Gün Sirer, associate professor at Cornell, did something quite overdue: he named the three main stakeholders of our online world.
His three force vectors are Military/Political, Commerce, and the Public. And he tries to calculate a “back of th envelope” vector sum. His rough guesstimate: “the forces are aligned in the ratio 1:1:3, with an alliance of the public and commercial interests that overpowers the M/P establishment in favor of transparency and online privacy guarantees.”

So all’s wrapped up and fine? Most likely, not. Civil rights and liberties are not a product of absolute vecorizable powers, but something you have to work hard for, you have to stand up to (or sit down, as Rosa Parks and Ghandi did).
And, in this case, it’s a very complicated reconcilatory process.
– Government, law enforcement and intelligence services are national entities (except the black UN helicopters, of course, which are after tin foil hat wearing free Americans roaming through Wyoming). There are some very good reasons, why those agencies sometimes should be allowed to wiretap certain individuals or entities. As there are very good reasons why a total surveillance state, a big data GDR on digital stereoids is definitely a thing to avoid. And, let’s not forget: the western democratic idea of government and checks and balances means mostly, that the citizens of a given country are the ones form which all power derives (and that he sometimes has to check, if the balance is still OK).
– Commerce is mostly global. But still bound to national or supranational regulations. If those regulations enforce companies to share their big consumer data with national government agencies (which share their data with some partners in the international intelligence community, because, you know, sharing is caring and NSA and BND do take care for you), they should, in their very own interest, work against this pressure. Because loosing trust means, sooner or later, loosing market share.
– And, finally: we, the people, we are everywhere. We are the one, who have to take care of our governments and our corporations. It took some millenia to gain some liberties. And it’s fairly easy to loose them all. Either against rogue (united) states, or by ceasing to much ground to nicely colored companies.

Community Organizer vs Top Down CEO

How does a business work? In the pre-election days, Romneyites blasted Obama for having no business experience at all. Here’s the former community organizer from Chicago, in the other corner watch the earth shaking CEO. So who’s got business experience? Who can save the economy?

As strong as this narrative sounds, it has a major shortcoming: it’s based upon a perception of business coming straight from the era of the “Mad Men”. Monolithic entities competing in well defined markets. Strong hierachies, wth strong men on top. Who are on top, because they not only are stronger than their underlings, but because they know better (because being on top of the hierarchy, all communications are channeled towards them).

Be it a local grocer or a global behemoth: top down decision making ruled OK. And once upon a time, this might have been a working paradigm. Like any well standing 60 ton cold blooded quasi-reptile had every right to scoff at those pesky mammals, which couldn’t even lay a decent egg.

Top down businesses mimick the dinosaur model: a large body commanded by a tiny head. It’s ruling principle is the vulgar confucianism of the stereotypical kung fu flick: the higher the rank of the guy you have to fight, the better his kung fu. It’s preferred development model is the waterfall process. First define all details, then execute.

The more unstable the environment, the riskier this process model becomes. As daddy already knows best, feedback loops are perceived as a hindrance. Which leads us to the the community organizer model.

Coming from the realm of software development, agile methods are gaining traction in other areas as well. Instead of central planning, those methods are based upon constant iteration based upon the feedback of different stakeholders. It’s somewhat related to community organizing, at least in my understanding.

To be sure, I have not the faintest idea what Obama really did his community organizing days. But let’s first get rid of one misunderstanding. The counter position to Top Down is NOT Bottom Up, at least not in my textbook. Bottom up would be a nicely humming worker’s collective. And just like with top down (AAPL!), there are examples of bottom up entities doing really good (sic). But bottom up and top down share the same problem: both structure are highly inflexible.

Agile methods think in stakeholders. Business owner and customers, marketing and development, production and legal: all those units are made of people, who have to give their constant inputs. It’s all fast iterations, everything’s broken down into easily digestible pieces. It’a a cooperative process with some kind of a community organizer, well, organizing it.

Top down large organization usually do not listen for input. Which sometimes leads to quite fascinating results. In a multinational I once stumbeld into the following situation: as an external consultant I was hired to manage a project. Turns out, there were three different parallel process flows to be aware of. Flow one was the official process as required by headquarter. The second process flow was completely unrelated, not to be mixed up with process flow one, somewhat unofficial, but nevertheless mandatory, as it was the process this formerly separate unit has been following for decades. Process number three was finally the process how the department was handling things, as the other two processes were somewhat decoupled from reality.

Now, back to Obama vs Romney, the community organizer vs the Top Down CEO. Let’s keep in mind: the presidential elections in the US are big business. It’s a $6 billion business, a major stimulus program for local tv stations. And it’s about getting a really large workforce on the ground.

Now, some parts of those makeshift corporations a.k.a. know the campaigns are a fairly well known affair. Buying air time in local tv is a fairly straightforward task, with just two unknowns: will the (outsourced) creative be really compelling? And how to allocate the budget, which is even with the richest of campaigns an important constraint.

But how do you work with the workforce on the ground? Team Obama is a supporter of open source software, the former community organizer defined in 2008 the basics of digital campaigning. So how has the top down CEO been approaching things? Let me give you this purely anecdotal answer: read this example of a worst case, waterfall-like top down process, with no working feedback loops, which might even had some real impact on the outcome of the election.

In a nutshell, the story goes like this: to support volunteers at election day, Team Romney planned a digital tool, which would replace some fairly cumbersome paper processes. @JohnEkdahl, the clearly disgruntled volunteer, gives a good overview on what went wrong. Mostly, it leads back to the problem of not iterating, not listening, of not having working feedback loops. It’s the CEO with the top down business experience at work.

No room to iterate is risk management at its worst: it’s either win or fail. Which looks like an 50:50 bet. But of course not even this is true: you have only one chance to win. But a gazillion odds are against you. It’s betting everything on zero on a roulette wheel with infinite numbers.

Digital Citizenship and Social Network Feudalism

In the real world, we abhor censorship, take many civil rights for granted. But as digital citizens, we happily click ourselves back into the 17th century.

Facebookistan has the 3rd largest population on the globe, just behind China and India. Google+, the new kid on the block, already surpassed Switzerland (big deal), Senegal, and even Australia. Which puts its current rank somewhere between Canada (population of 34.5 million, rank 35) and China (Republic of Taiwan that is, 23.2 millions, rank 50).

Mark Zuckerberg in the 17th century.

Now, those numbers do not make Facebook into a sovereign state, at least not in the traditional sense. Sovereign states are defined by territory. But the Googleplex is not like the Vatican a sovereign enclave in a larger territory. It’s still just a piece of real estate located in the US. And “Business is War” doesn’t mean Google war droids attacking the design soldiers of Jobs.

No land, no armies. The differences between Facebook (more than 10 times the population) and France (real nukes, real food) or Google (credit rating of AA+, just like the US) and Greece (CC, just like me) are obvious.

But so are the similarities. Sovereign nations are defined by their people, otherwise the Antarctica would be a superpower. And it’s we, the digital people, forming those digital Leviathans of the 21st century, which provide us with our digital IDs and currencies. They handle our communications, they might even tax us or control, what’s to be published or not (on their our Kindles and iPads).

350 years ago, Thomas Hobbes’ concluded, that an absolute monarchy be the best way to govern any sovereign. This would be a fringe opinion nowadays, at least in the western part of the real world. Our ancestors fought pretty hard to get us, where we are now: nobody should stay above the law, censorship is bad, sovereignty belongs to the people.

But a look at the digital domain might make Hobbes a happy man. The digital sovereign is not the people, but a corporation.

  • Post your artwork on Facebook, which might offense some bible belters? You’ll get evicted (as it happened to my friend Thomas). Eventually you might be allowed to return (as it happened to my friend Thomas). But no legal recourse here. It’s a little bit like GDR light.
  • Use a pseudonym on Google+? Say good-bye to your Google account.

In a heavily distributed digital universe, this wouldn’t be a big deal at all. Don’t like this bar? There are plenty next door. But Facebook isn’t your neighborhood Hooters, and Larry Page definitely not the soup nazi. There are not even a handful of Digital Sovereigns aspiring to become the operating systems of our digital lives.

The preamble to the United States Constitution starts like this: “We the People of the United States, … secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America.

Our digital selves do not enjoy a constitution or according rights. We the users, have to accept some Terms of Service. And as most of you never really read what you OKed with a single mouse click, we hand now over to Richard Dreyfuss declaiming some parts of the Apple iTunes EULA.