3 ms·

Probably not a great idea because: 1. Replacing a DB/server side code with contracts that you pay to run per opcode introduces needless cost (this is general p

Probably not a great idea because: 1. Replacing a DB/server side code with contracts that you pay to run per opcode introduces needless cost (this is general problem with startups that use blockchain too quickly). Sure AWS compute costs money, but smart contracts are way more expensive. 2. If you're building an app and distributing it via Apple/Google, there's no need for a blockchain. Plus the app store itself is already centralized, so if your distribution channel is centralized there's really no guarantee of privacy. 3. Ethereum/<pick your favorite blockchain> txs are slow compared to HTTP requests. We're talking minutes to confirm sometimes, versus less than 100s of milliseconds. 4. If you increase privacy, you must be increasing anonymity, and that's not useful for contact tracing. You need to know who was in contact with who else. 5. Smart contract code is a pain to upgrade. It's costly, complicated, and puts all your "server-side" code right on the blockchain for anyone to mess with it and try and break it.

skitout 6y ago
Thank you for your insight. My (non specialist) answers: 1. It is possibly be a (little) bit more expensive to run, but the cost must be compared with the potential gain: more privacy, and more adoption thanks to privacy (=less cost for treating COV-19). I think cost can be decreased by using existing blockchain infrastructure, and using a gas-free design. 2. If we can achieve much higher privacy without blockchain, I am happy with it, I just never seen any proposal! Crypto-wallet are available on app stores, but still guarantee decentralization and anonymity/pseudonymity 3. Speed is not key here. Knowing now or in 2 minutes that you have been in contact with someone with COV-19 does not seem to be game changing 4. My understanding is you need to know if you have been in contact with someone having the COV-19. But you don't need anything else, and health system does not need to know your name or other info 5. If it is a pain in the neck but allow better privacy and more tracing app usage seems a reasonable trade off to me. Furthermore I think that it possible to do a design without smart contract.
- m11a 6y ago
  1. You don't see this from a layman's perspective: a person who hardly understands Bitcoin never mind blockchain. You may realise the privacy improvements in such a solution. The average laymen will not. How can you expect trust, as far as privacy goes, when people can't even understand how that privacy is guaranteed? The media, who could simplify and spread, have very low trust ratings according to recent surveys on the coronavirus crisis. Trust in almost all organisations, including central governments, has went up - the largest exception was the media. On a sidenote: Governments haven't played around with blockchain enough to be able to rapidly develop and deploy such a feature using blockchain technology. Then they'd probably run into lots of bugs and hurdles on release. Not good when you need to increase public trust, adoption, and deploy effective solutions rapidly.

skitout 6y ago

Thank you for your insight. My (non specialist) answers: 1. It is possibly be a (little) bit more expensive to run, but the cost must be compared with the potential gain: more privacy, and more adoption thanks to privacy (=less cost for treating COV-19). I think cost can be decreased by using existing blockchain infrastructure, and using a gas-free design. 2. If we can achieve much higher privacy without blockchain, I am happy with it, I just never seen any proposal! Crypto-wallet are available on app stores, but still guarantee decentralization and anonymity/pseudonymity 3. Speed is not key here. Knowing now or in 2 minutes that you have been in contact with someone with COV-19 does not seem to be game changing 4. My understanding is you need to know if you have been in contact with someone having the COV-19. But you don't need anything else, and health system does not need to know your name or other info 5. If it is a pain in the neck but allow better privacy and more tracing app usage seems a reasonable trade off to me. Furthermore I think that it possible to do a design without smart contract.

m11a 6y ago
1. You don't see this from a layman's perspective: a person who hardly understands Bitcoin never mind blockchain. You may realise the privacy improvements in such a solution. The average laymen will not. How can you expect trust, as far as privacy goes, when people can't even understand how that privacy is guaranteed? The media, who could simplify and spread, have very low trust ratings according to recent surveys on the coronavirus crisis. Trust in almost all organisations, including central governments, has went up - the largest exception was the media. On a sidenote: Governments haven't played around with blockchain enough to be able to rapidly develop and deploy such a feature using blockchain technology. Then they'd probably run into lots of bugs and hurdles on release. Not good when you need to increase public trust, adoption, and deploy effective solutions rapidly.