How would the network handle corrupted data?

Let’s say we have an important but unpopular file. As it’s not highly in demand, there might be the minimum number of host computers storing it. Let’s say it gets stored in the year 2050, and then isn’t accessed until 2060.

In that time, it’s possible that all copies of that file will have corrupted over time. And I assume it’s inevitable that all copies of at least one file will have corrupted over time.

At the moment, it’s easy to brush this aside, because corruption isn’t that common. But the longer the interval between accesses, the harder it is to dismiss, and that’s only partly because of the time-span.

If the network becomes popular, then storage-producing companies will make Safe-specific storage devices to farm tokens for profit. If it becomes very popular, then laptops and smartphones might include storage-for-farming. These storage-for-farming devices won’t need anywhere near the reliability of modern hard drives to be profitable. If they lose data, then the company only suffers a tiny loss, which could be offset by cheaper manufacturing processes. It could be that bit corruption on these devices becomes like dead pixels on screens: increasingly less noticeable if kept below a threshold.

The only way to avoid this problem would seem to be regular corruption checks.

Anyway, could someone familiar with the low-level stuff give a timeline for what would happen to an unpopular file? Like, if I upload a 1 GB video in 2050, then only try to access it again in 2060, what happens to the chunks on the network in that time-span? Do they ever get checked for corruption?

6 Likes

Oh okay, this puts my mind at ease. Thanks!

Also content hash == name so data is checked on every movement.

6 Likes