search

Planet-Zuda / doge-update

proposed update for doge
4 stars 1 forks source link

Oracle problem #2

Open patricklodder opened 3 years ago

patricklodder commented 3 years ago

Moving discussion from https://github.com/dogecoin/dogecoin/issues/1996

If you're going to import external data into a blockchain, you'll run into the "Oracle problem" (and sybil attacks, and I think there are some issues with censorship resistance, but can touch on those later)

See https://www.datadriveninvestor.com/2019/06/15/what-is-the-oracle-problem-how-does-chainlink-solve-it/ for a start on the subject and an idea about how Chainlink deals with this.

I think this needs to be tackled first. Questions: What is the impact on Dogecoin (or any PoW crypto for that matter.) Will there need to be a coin-at-stake element to penalize bad actors? How does that impact the assets security classification? What does that do towards entry barrier?

Planet-Zuda commented 3 years ago

I didn't want to reply until I could give a well thought out answer, so apologies for the delay. This doesn't fall into an oracle issue, per my understanding of the dilemma because the current design is having the script generate the random word and then the machines would have to match it on Twitter.

If the script was changed to bring in random text from twitter or youtube, then some validation would be useful, that isn't the current design, However that design isn't ruled out, so this is something we have to consider.

What is the impact? There will be a reduction of nodes running by 16 hours per day, which by our early, not finalized calculations is 228 days worth of energy not used on the dogecoin network, so that is saving a ton of energy.

I am unable to speak to the securities classification if coin at stake was implemented beyond saying that with my limited understanding it doesn't seem to violate the Howie test from Howie vs the Supreme Court. There is a write-up on it here https://www.winston.com/en/crypto-law-corner/when-is-a-crypto-asset-a-security-and-why-does-that-matter-part-i.html#_ftnref3 . I can't speak to it, anymore than that and that may be an incorrect assumption.