> [...] they cannot be trusted with content It's sad to see that this has not become a standard assumption. I knew people over a decade ago who lost awesome we

> [...] they cannot be trusted with content It's sad to see that this has not become a standard assumption. I knew people over a decade ago who lost awesome websites they created/curated because their hosting provider had a data loss event and no relevant backups. It makes no more sense to trust a hosting service as the sole repository of critical data than a single hard drive. Consider the rule of thumb that important data should have three physically separate copies to significantly reduce the risk of data loss. (E.g. RAID does not count.) As Blip illustrates, organizations have many non-technical SPOFs[1] even if we assume (chuckle) they have taken all possible measures against normal physical data loss. This means that individuals and even online communities need to practice methods to preserve important data under the assumption that all storage devices and/or providers will ultimately fail. [1] https://en.wikipedia.org/wiki/Single_point_of_failure https://en.wikipedia.org/wiki/Single_point_of_failure