15

The Parasite Contract

Behold this outline for a Smart Contract:

  1. Offer all the functionality of a pre-determined “Host” Oracle Contract.
  2. Wait for the Oracle to report. In other words:
    1. Access the database of the Host’s blockchain (by tracking things like block number, block date, SPV proofs, etc).
    2. Scan the database for anything perfectly-correlated to the External Data fetched by the Host Oracle (contract state, payouts to certain addresses).
  3. Use that Host-Data to alter the Parasite’s state (without paying the Host anything).

The description above was taken from:
http://www.truthcoin.info/blog/contracts-oracles-sidechains/

14

The parasite problem is as of yet unsolved. Internally, the Augur team has referred to it as the problem of derivative financial markets. The short version of the problem is that in order for REP to remain secure, the system needs to know the total value at risk so it can accurately assess how valuable REP must be in order to effectively secure that much value at risk. If enough people have assets attached to Augur indirectly (in a way that Augur doesn't know how to account for) such that the cost of launching an attack against REP is less than the profit to be made from such an attack then the security model has broken down.

As the link you provided suggests, this is an as-of-yet unsolved problem. The tentative plan is to move forward with recognition of this problem and closely monitor the system at launch. If Augur is wildly successful, then it is possible that its internal volume will outpace derived volume. In that case, we (the community) merely need to watch from the outside and pay attention to how much value is at risk in derivative markets. If derivative markets start to surpass Augur's internal volume, then the first step is to approach the authors of the derivatives markets (off-chain, human interaction) and see if they would be willing to hook in to Augur's Oracle system more closely (and pay fees, which are presumed to likely be very low). The reason one may choose to do this is because if they are dependent on Augur as an Oracle and the Oracle fails, they fail as well. 0.01% trading fee (or whatever it is) may be well worth it to someone who is big enough to be a risk to Augur's security model if it helps ensure their system remains secure.

For smaller fish, who may be less interested in "contributing their fair share" to system security, as long as Augur and big fish derivative markets are paying their fair share then the system is designed to handle a bit off lossiness in the value at risk calculations.

The real problem arrives if you have an actor that not only is wildly successful and dependent on Augur's Oracle, but is also unwilling to pay for the security costs that lead to their success. In general, such an actor needs to either be uninformed as to the risk they are taking or not financially motivated. At the moment, we aren't terribly worried about such actors in the ecosystem, and initially the Ethereum ecosystem is small enough that we can likely run a successful smear campaign against such an actor.

Generally speaking, Augur does want derivative markets (and other interesting things) built on top of the Augur Oracle. As long as people doing this recognize that security isn't free, it is possible (likely?) that they will be willing to contribute the very small cost associated with maintaining enough security to cover their own security costs.

9

I also have some ideas if this ever becomes an actual problem for creating contracts to provide defensibility against derivatives on top of augur markets. These new contracts would take away security/incentive to do this by converting the schelling point to resolve a market on Augur "incorrectly" if enough money can be stolen from the derivatives markets (via buying the other side as a group) to trustlessly pay out the participants in the real augur market correctly & honestly.

0

I solved the parasite problem like 3 years ago. You can see the solution built into Amoveo.

The solution is 2 fold: 1) some initial money needs to be given when the oracle question is asked. 2) we need a good escalation mechanism, so if there is a disagreement, the amount of money at stake will grow exponentially.

By combining these ingredients, we remove any need for trading fees to pay the oracle reporters, and this makes Amoveo secure against parasite contracts.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.