Comment by EL_Loco | Hacker News Reader

Comment by EL_Loco | Hacker News Reader

Parent Full thread

I absolutely hate, hate this NFT scam. I even hate all the metaverse talk. I've been watching every iteration of VR since the early 90s, and it still looks like a gimmick. But how is vaccine tracking, with temperature information, lot numbers, location, etc, being logged to a secure, fast, carbon-neutral DLT not a good use case? Maybe not any better than a regular database owned by the vaccine buyer, but I can see quite a few benefits (many different buyers, each with their own quirky, sometimes less robust database, for an example). A lot of people are losing money with this moon propert...NFT fad, but I'm seeing quite a few use cases of real value popping up here and there. The example above is not my area of expertise, so I might be wrong, and would really like to hear a good rebuttal, but it's difficult. People seem to eithe love crypto/DLT too much or hate it too much to even bother discussing beyond "The whole entire planet will use crypto for everything" or "There's not a single use for this crap".

Append-only, cryptographically verified, distributed databases are a good idea with some useful applications, and I'm sure they'll spread as tech infrastructure just like e.g. stream processing has in the last few years. But that seems barely related to the whole crypto phenomenon which inherently relies on destructive proof-of-work (maybe proof-of-stake will turn out to work at scale but I'm dubious) which in turn requires everything to have financial value and be pumped and presumably eventually dumped.

Oh, I agree with you, I can't even read discussion on crypto online, it's 99.9% about which or coin will value the fastest. However, I will still keep an open mind. For example, Silvio Micali works on proof-of-stake and he says his stuff already works at scale. The guy won a Turing and a Gödel award. I'll keep an open mind.

Proof of Stake is already working at scale. Ethereum has been running Proof of Stake in parallel to Proof of Work since December 2020, with plans to do away with Proof of Work in the upcoming eth 2.0 merge this year. Not to mention all the other chains running live on Proof of Stake (eg. Polygon, Harmony, Avalanche, Fantom, etc) with billions in TVL.

You can do it without blockchains. This is roughly what Certificate Transparency achieved. The only difference is that you cannot access a consistent state at any given moment.

> Append-only, cryptographically verified, distributed databases are a good idea

More often than not they are a bad idea.

Because often you are required e.g. by the courts or legislation to truly delete something.

> But how is vaccine tracking, with temperature information, lot numbers, location, etc, being logged to a secure, fast, carbon-neutral DLT not a good use case?

Trust isn't an issue; a conventional database will be perfectly fine. Like, if you wanted to sell a product for tracking vaccines (assuming, for a moment, that health services and other large vaccine buyers don't have decades-old systems for this which everyone already knows how to use, which, if course, they do) why would you bring The Almighty Blockchain into it?

Around eight years ago, there was a power failure in my town (a small town in a third world country). The freezer that stored the local health facility's vaccines stopped working for two days because the backup generator was broken. After some investigation, it was found that when power was restored, the health officials altered the temperatures on the database to show that everything was fine and the vaccines kept their temperature in proper range for those two days. If it had been logged to a DLT with an immutable ledger, this wouldn't have happened. That's the reason I'm particularly interested in this use case (something that was done in the UK recently), because kids ended up taking these expired vaccines.

If it had been logged in any sort of correctly designed system it wouldn’t have happened; no blockchain required.

I agree, but what I'm asking is if it isn't easier and cheaper to have a robust logging system with a public DLT. I'll give you a real example of how bad some of the database systems are in my country: a study on police gun and ammunition control found that many police stations keep track of ammunition by logging the numbers on a.... ready... a chalkboard. So I don't trust the cops who sell bullets to drug gangs if tomorrow they tell me they are not doing it anymore because they are using a computerized database to keep track of their equipment.

We don't need a blockchain to solve this problem. It would be enough to make the database public and allow third parties to make backups and subscribe/log all updates.

Agree with you, there are many systems like this. What I'd like to know is if a system using public and decentralized distributed ledger wouldn't be cheaper and/or easier to implement effectively (on a poor, corrupt country). Because you're adding more steps, like having third parties. How many? What if said third-party decide to quit checking after one year. Which database supplier do I trust to make the database public and final? To me, it seems more complicated and less transparent.

It’s not a technology limitation, it’s a “get people to agree to do it in a transparent way” limitation which crypto doesn’t address.

> there was a power failure in my town

> If it had been logged to a DLT with an immutable ledger

.. while the power was off?

The vaccination rate is about a million an hour: https://ourworldindata.org/covid-vaccinations (!!), no blockchain is going to keep up with that.

Yes, the first temperature times were typed onto a laptop, while it was still running. They were the correct temperature (unsuitable for the vaccines). I think they did temperature logs twice a day, don't remember well. Then it was found that afterwards they altered those first times and added the ones that were missing.

That sounds like a database access problem.

Sure, the temperature should just be logged automatically to the database, and the value shouldn't be allowed to change. However, it isn't. And I can see clearly how a database supplied by some vendor would yield to pressure from a government to 'tweak' some value, just this once, so that they'd be sure to get a nod on the next contract. I would trust a public ledger on its immutability more than a government run database. Or a maintainer of a database that relies on government money to exist.

Reply on news.ycombinator.com