Why so complicated?

When we introduced our auction mechanism or our dual token one of the frequent responses that we have received has been along the lines of: “Why is it so complicated — why didn’t you stick to the traditional model?” Let’s answer those questions. The main reason is that innovation is in our blood. The Gnosis team is constantly questioning everything that exists; finding ways to improve efficiencies and creating systems with more sound incentives is what made us build a decentralized prediction market in the first place. Striving to achieve this will always drive our decisions. However, we believe that innovation is not a goal in and of itself, rather, it is the answer to existing problems and inefficiencies. Here is a short overview of that problems we address:

Gnosis Token Auction vs. Traditional Token Sales (ICOs)

Fixed Token Supply — No Cap (e.g. Augur)
From the moment that a buyer makes a purchase, the number of tokens they receive in return can only go down and there is no lower limit.

Our model:
From the moment a buyer makes a purchase, the number of tokens the buyer receives in return can only go up and so purchasers are guaranteed to pay a specific maximum price per token. At the point of purchase, a buyer has already made a decision regarding the amount that they want to spend and the price at which they want to buy the token (i.e. what is their fair cost to purchase GNO relative to the utility GNO offers them). Since the price of the tokens can only decrease from their initial point of purchase, the buyer benefits by getting more tokens than the amount they decided would be an acceptable deal should the auction continue to subsequent blocks.

Unlimited Token Supply — No Cap (e.g. Ethereum)
A similar problem is presented by unlimited token supply sales. A purchaser might know the amount of tokens they will receive but the utility of a single token is usually meaningless if the buyer is unaware of the total number of existing tokens.

Our model:
At the moment the buyer makes a purchase, they know they will own at least exactly X% of the total tokens.

Fixed Price — Cap (e.g. Golem)
In this model, the seller is playing a guessing game upfront to set the right price. This can lead to token sales that sell out in minutes and to tokens selling at a very different price days after the sale, encouraging short term speculation. We’ve designed a platform and token that is based on utility, so this outcome seemed both inefficient and at odds with our core mission.

Our model:
Gnosis aims to be a global markets platform! We believe that markets are best at price finding — instead of playing a guessing game we empower users to find the price (i.e. what is fair value for the utility a purchaser will receive from GNO) with a proper market driven mechanism.

Early Buyer Bonus (e.g. TheDAO, but almost all others as well)
For the buyer, paying up to 50% more for the same token just because they looked to purchase a few days later almost feels like a scam. We don’t blame anyone for doing it because it should counter the dilution problem mentioned earlier but we still don’t like those bonuses.

Our model:
No pre-sale, no bonus. We haven’t sold a single token yet and we will not do so before the auction starts. Every token buyer will be able to purchase tokens at the same price as everyone else.

A Short List of Main Advantages of our Token Model:

Pay Once For Access Instead of Usage: Our model
You can gain access to the GNO platform simply by holding GNO tokens (and activating their WIZ generation) — you don’t need to spend GNO. This is appealing for heavy platform users, or for 3rd parties that decide to build atop of Gnosis. It allows them to use Gnosis with a one time cost vs. paying a usage fee in perpetuity.

In Comparison to Ethereum
Ether has a similar structure as ETH and GAS. However, GAS can not be stored and is only purchased at the moment it is needed. Imagine holding ETH to give you enough GAS so that you can interact with a contract every day. For Ethereum this model was not possible because the marginal costs of a transaction are not zero but in Gnosis they are.

Platform Users Should Have Predictable Costs
Let’s say you want to build a business on a platform like Golem or Ethereum and you can roughly forecast the usage of the platform. However, it is not really possible to buy tokens upfront that will cover your demand. In the case of Ethereum we have seen bigger changes in the gas price and we will see those more often (as in Bitcoin) as soon as market mechanisms develop here. With GNT it might even be worse — since as long as people offering computation have to pay their utility bills with Fiat money they will also price it in Fiat (e.g. $). So holding GNT will likely not be helpful for predictable costs(*). The logical consequence for consumers of platforms that require payments for its services in a volatile token is that they only buy the token at the time needed. If no one has an interest in holding a token, however, its value would go towards zero. Read more about this topic at this post from Coinfund.

Our model:
In contrast we created a WIZ token with an absolute predictable value within the Gnosis platform. WIZ is exactly worth $1 in fees on Gnosis. If you know you are going to use Gnosis you can hold WIZ. If you know you are going to use Gnosis regularly you can hold GNO that will regularly generate WIZ.

A very interesting side effect is that because of its stable value, WIZ can optionally be used in markets on Gnosis as collateral. Read more in our token post (bottom).

In summary our auction model makes the process of buying tokens much more predictable. Tokens models are still a topic with a lot of ongoing discussion and research. Many reasons to have a token haven been identified. We believe that not all of functionalities of tokens can be covered by one token. Our dual token model puts us in a position to cover them all.

(*) Don’t get us wrong, we are for sure excited about a decentralised super computer but we nevertheless think that the token model can be improved.

A few people that we want to acknowledge for the good discussions on those topics: Amsel, Alex Van de Sande, Jake Brukhman and Aleksandr Bulkin from CoinFund, Jack du Rose, Vitalik Buterin