Note: this article is now at least two years old and may contain serious errors.

The ICO crapcoin checklist

  1. Does the coin reward users for contributing resources towards the creation of a shared service that will benefit all users?
  2. Can the resources contributed by users be verified by anyone?
  3. Does the service always fall within an expected range of operation?
  4. Is the coin new and no other coin offers the same service?
  5. Is it impossible to substitute the use of the coin for another?
  6. Does the business model for the coin make sense?
  7. Does the team behind the coin have the required experience?

If you answered "no" to any of these questions you may have discovered for yourself a bonafide crapcoin!

A crapcoin is the term for any coin whose main purpose is to make the founders richer. Typically such coins will use misleading technical-sounding jargon to try dupe investors into thinking they are investing in the next Bitcoin when really they are being sold worthless tokens.

Quite often these tokens will be sold as part of an "ICO" - an initial coin offering - which is a new kind of craze where investors give someone they don't know millions in exchange for receiving an empty promise.

To the uninformed blockchain investor a shitcoin may seem fundamentally no different to any other kind of blockchain asset but it can be differentiated from more innovative products because:

  1. Shitcoins can be substituted for any other kind of token.
  2. Shitcoins don't solve a new problem (other than padding the founders wallets - of course! A problem for anyone.)

I'm going to go over the model that Bitcoin introduced by considering it a new kind of economic model unique to the blockchain space. Then I'll expand on how shitcoins violate this model.

Time for a history lesson

In 2009, Satoshi Nakamoto published the Bitcoin software.

The software was unique. It introduced a new design that allowed for the creation of trustless, programmable, value-transfer systems, where previously all such attempts to solve this problem relied on third-party trust.

The way that Bitcoin solved this problem was to introduce a new kind of high-security ledger called "the blockchain" where anyone could contribute their spare CPU clock cycles to make the ledger more secure.

This was genuinely revolutionary at the time and it was more than enough to call it "genius" - but the biggest break-through wasn't that the ledger could be secured between mutually untrusting parties - it was that the ledger introduced a new kind of incentive system to do it.

In the Bitcoin system, new Bitcoins are given out for securing the ledger, in effect inventing a new kind of corporation that:

  1. Values contributions made by anyone towards a common goal.
  2. Requires no trust in providing the quality of the service.
  3. Requires no one to be trusted to pay employees.

The closest word we have for this concept is still bullshit - but in the blockchain space we can say that Bitcoin was the first example of a decentralized autonomous corporation (DACs).

The problem is investors think all shitcoins are DACs

What made Bitcoin so unique was that nobody had to trust anyone. If you think that the idea of a high-security ledger is useful you can always buy your own special hardware to help secure the ledger in exchange for receiving Bitcoins and even if tomorrow everyone decided Bitcoin was worthless you could always keep mining it for rewards.

Economically speaking, the value of a token issued by a DAC is directly tied to the utility that results from doing so. Or in other words - if you were to take away the token the entire thing would fall apart.

Essentially in a DAC, tokens and services are so inextricably linked that taking one away necessarily implies taking away the other. In a far-fetched kind of way, DACs are kind of like a new form of digital organism that feeds off a persons belief that its still worth enough to keep "alive" by providing it with whatever resources it needs to survive (power, hard drive space, processing resources, datasets, networking bandwidth, etc.)

A DAC is more like an idea than a tangible entity

The most important thing is that nobody has to trust it. If you think of a new DAC then you can write the code for it, put it in the wild, and it will survive so long as people think its a cool idea. So you can imagine that if a company were to invent a new DAC then actually raising money to help bootstrap this project might make sense since no trust in the resulting entity is required.

A DAC-based fund-raiser would thus entitle an investor to receive some benefit within the system so that in the future there was no way for the company to stop the service from existing or to reduce the value of rewards. It's kind of like investing in a startup run by robots with a 100% predictable output - that model is extremely unique to the blockchain space.

Shitcoins are the opposite of that. A shitcoin offers no additional value. The token neither helps to create a new service nor does it entitle an investor to any equity in the resulting company. In a worst case scenario the token is speculative and empty (though a well-designed shitcoin can still have a monetary value by locking it to a given platform.)

Coins that aren't DACs

Example 1 - Storj and Golem

Projects like Storj and Golem come very close to being called DACs because purchasing storage space or processing power can be verified without trust. But you have to be careful calling this a DAC because in these systems resources are only bought from the parties who have them so there isn't a single shared service that emerges from collaboration.

The question to ask here is "can the token be replaced with another without the system falling apart?" And the answer with Storj and Golem is definitely "yes" (any other token can be used to purchase resources.)

I am not saying that Storj and Golem are shitcoins but instead I would refer to them as decentralized marketplaces that use smart contracts to pay for services - and that's more impressive than it sounds since 99.999% of current "smart contracts" are based on trusting third-party oracles.

Example 2 - Augur, Gnosis, Oraclize

When Augur introduced their prediction market they obviously needed to create its own token (sarcasm)... Unfortunately, you cannot verify real-world results on a blockchain which leaves anyone who uses the service vulnerable to unforeseen outcomes.

Example 3 - Almost everything else

Anything that doesn't guarantee investors will receive a new product or service is a shitcoin. This applies to virtually every ICO that was ever done.

Now lets look at some things that could conceivably be called a "decentralized autonomous corporations" - both existing and theoretical.

Coins that are DACs

Example 1 - Exchange bootstrapping

A coin could be created to incentivize liquidity on decentralized exchanges. The coin would enforce a system that brought together multiple blockchains (the service) and gave away rewards to help do it (the reward.) This could be done using probabilistic collateral contracts or something like BlockNet.

Shared service: Increased adoption of an exchange for its users.

Example 2 - Search engines

A coin that rewards users for processing information in torrent files to find files that all the participants want in accordance to some formula (they use bandwidth to randomly check files in some hashed set of files and the more bandwidth contributed, the faster the torrents can be "searched.")

Shared service: Finding results for a search query that all users need.

Example 3 - DDoSing people

A coin that rewards users for attacking SSL-enabled websites. The more people who use the coin, the more powerful DDoS attacks become, the more users it gains, the more the rewards are worth.

Shared service: More powerful DDoS attacks.

Example 4 - Ledgers, name registries, permissions, etc

  • A coin that reward users for securing the ledger and the more people mining it, the more secure it becomes.
  • Any new cryptocurrency that improves on the process in some way like scaling, privacy, decentralization, security, smart contract DSLs, etc.

Shared service: High security ledgers.

Example 5 - Math problems

Certain classes of problems may be outsourceable and easily verifiable. If all the parties cared about the solution this would be a good use-case but it differs from something like Golem since the service is shared with everyone.

Shared service: Access to new knowledge

Example 6 - Trustless software bounties

A ledger that allows for arbitrary computations to be run for the purposes of running deterministic software builds against a number of conditions would allow smart contracts to be written that could specify trustless bounties for algorithmic speed improvements or security vulnerabilities.

Shared service: Trustless software bounties

Trust matrices at a glance

Any token that isn't created directly from a DAC by definition cannot guarantee that any utility will be created in relation to owning that token. What's more confusing is that DACs absolutely don't need an ICO for that to happen since tokens are created dynamically by recognising the contributions made towards achieving some common goal.

Thus, all ICOs can be considered legally nonbinding, trusted agreements between an unbounded number of parties in exchange for receiving the promise of some entity to produce a future service. Again, that's not to say that ICO tokens don't have economic value but often that value comes at the expense of usability (and intentionally so.)

Parties Trusted Trustless
Bounded Dumb contracts: Augur, Gnosis (unique distinction of being shitcoins that use dumb contracts) Smart contracts: Storj (low trust), Golem (low trust), Gambling, Lotteries, Lightning Networks, Cross-chain contracts, ENS, etc
Unbounded Crapcoins: BAT, Melon, mostly every ICO token ever... DACs: Ethereum, ZCash, MimbleWimble, Bitcoin, DDoSCoin, BlockNet (or equivalent), Datachains, etc

My conclusion is that most current ICOs aren't compatible with an investors interests because there is no way to guarantee that they will receive anything from the sale; The tokens by themselves aren't legally binding and they're not attached to equity... besides, even if they were you would still have to trust that the company created the final product.

I think in the ideal case there needs to be a legally binding agreement for ICO tokens where investors end up buying some benefit in a future DAC (therefore guaranteeing that the token will have utility) - because only then will you know that you're not just funding a new shitcoin scheme.

Ethereum almost came close to accomplishing this but I don't think I've seen any ICO sale that had a legally binding agreement to create software to back up the value of the tokens sold at ICO.