I am becoming seriously concerned at the general level of ignorance surrounding blockchain, its benefits, capabilities, and shortcomings. This article clearly states what is required of its solution:
"That requires relevant agencies and health care entities to share data as quickly and efficiently as possible. During any actual outbreak of infectious disease, there is little time to waste. Unfortunately, the current data-sharing systems do not allow for the secure and speedy input and sharing of information needed to respond in the most effective and time-efficient manner possible."I think it is well overdue for someone to create a comparison table between blockchain and other ledger and database solutions. This would illustrate that blockchain is not a good fit for handling a disease outbreak.
- Blockchain is very very slow. Writing a record to the blockchain can take more than ten minutes, and much longer if suddenly thousands of records need to be written at once.
- Blockchain requires specialised software, which in an emergency situation might take between hours and days to setup to the point of running a full node.
- Blockchain does not allow records to be deleted. It does not allow passwords to be changed, or access to be revoked. Someone from the CDC could be fired for misconduct, but because they have the private keys, all CDC health records are permanently available to this person.
- Blockchain does not handle multiple tables or manage referential relationships. It does not enable data to be procedurally updated. So if it turns out that all body temperatures entered by some region were in Fahrenheit, not Celsius, those records must remain wrong.
But this is only scratching the surface of why blockchain is a dreadful fit for the CDC. And it is bizarre that they are looking in that direction when there are so many off-the-shelf proven solutions to provide a fast, flexible, powerful, full functioning database with tiered levels of access. A MySQL database, with both desktop and mobile webpage access would be instantly usable by any authorised person in the world with no software installation or training required.
On this matter it should be noted that they are still in the conceptual stage, as is nearly everyone else who works with blockchain. In the CDC's 3 goals with blockchain (data sharing, tracking, and privacy law compliance), I think they will eventually be successful in all. But this isn't to say that there aren't better solutions in the interim until all engineering challanges are met, making blockchain more effective in database management.
The largest flaw I see with todays DBMS' is that any which have remote connections enabled are prone to attack. There has actually been ransomware floating around recently which specifically targets and locks MySQL dbs. I would take some additional latency any day of the week over the risk of a system wide lockout.
That being said, the latency in writing to a blockchain DB could easily be solved by making it a back end network (instead of imagining it as a standalone), where the data access layer periodically sends requests to collect activity logs from all front end users, and proceeds to permanently record it on the blockchain. Vaguely similar to how the bitcoin lightning network worked to alleviate some congestion on their blockchain.