[tsumnia]

> The fact that almost no one on this [thread] knows these

Its not that they aren't known, but rather we just came off a long trend of thin-clients and cloud storage. Some companies merely stay in that ethereal space, while others had concerns about their data. Criticizing people for doing what experts were pushing for the past 20 years doesn't need to devolve into calling their expertise into question.

The downvotes are for that, not because "you're wrong".

[lazyasciiart]

I don't think I understand what you're saying here.

[tsumnia]

Around 15ish years ago, there was a heavy push for things like parallel computing, hosting things on 'the cloud', and managing "big data". So the overarching recommendation was for devices and data to be accessible through a server. It was cheaper to use a third-party for high end compute and large storage rather than storing locally. Remember this was a time when Dropbox was still quite popular.

My original comment is mostly saying that it is too critical of staff saying "how did they not know" when we're now starting to return to in-house solutions. The prior solution was "Go Cloud", now its "Stay Home". In a decade, once enough people learn the struggles of having everything in-house, the next solution will be "Go Cloud" again, or whatever the future equivalent is.

The overall purpose of my comment was more akin to "calm down, we're just in a new tech cycle, no one's an idiot for following the last cycle's solution".

[alephnerd]

I disagree with your statement simply because I myself started my tech career in the midst of the Cloud First hype cycle, and even then principles around data management and limiting access (eg. via RBAC) was already well understood.

Maybe a significant portion of the HN base simply never worked with companies that either sold to or were a part of regulated industries, but I do not buy that.

Furthermore, all of the design patterns I am describing can and have be implemented within cloud environments as well.