Hacker News Reader

Comment by djyaz1200 | Hacker News Reader

To be generous to those proposing blockchain for health records I think their intent to suggest that it be used for the access/authentication layer of a more conventional secure database, but that's just a guess.

View on HN

I've found this diagram to be extremely helpful for thinking about why/why not to use blockchain: https://i.redd.it/uu0qg8t28tq11.png (p.42 https://nvlpubs.nist.gov/nistpubs/ir/2018/NIST.IR.8202.pdf)

One problem is blockchain is good for identity but not so much access/authentication.

When I look at the first flowchart - there’s no way a PMI-certified Project Manager would say No to any question on the flowchart - so Blockchain it is !

How would it actually solve a problem that couldn't be solved more conveniently with current standards?

It would provide an auditable, tamper-proof database with an immutable history, accessible by all health companies and not controlled by any one health company.

Ok... but a database of what?

If it's the access/authentication layer activity, I don't see what blockchain really brings to the table.

Cool.. what about the attached records, are they also accessible by all health companies? And if not, how do you know the ledger is really tied to the records, and if so, isn't this a breach waiting to happen?

Not an advocate for this tech, but the obvious answer is a hash so that someone could verify the data was added to the chain

I believe there were technologies before blockchain that enabled that.

Since the actual data in this scenario would still be hosted by the individual companies, what guarantee is there that they’d correctly log all access into this shared ledger?

Reply on news.ycombinator.com