wannaggie said:
ABATTBQ11 said:
Civil.Savage said:
Isnt this also an indictment of Google that they keep "deleted" items.
Probably not. You typically have a "deleted items" folder that you need to periodically clear to fully delete items. It's like the recycle bin on your desktop.
Laughable. I guarantee you the major cloud services providers have extensive shadow-copies and backups they preserve specifically for the purposes of their own datametrics harvesting as well as complying with TLA/LEO requests.
Everyone needs to remember this one thing:
"In the cloud, there's no such thing as deleting".
They don't need copies for metrics. By the time you delete it, it's already been ingested into their models. They aren't retraining on data they've already trained on. There's also no point in keeping old data. It's stale. Your needs and preferences have likely changed, so they're not training new models on old data. If they want that history incorporated, they'll just take layers from previous models and build I them. It doesn't make sense.
The volume you're talking about is also prohibitive. There's simply no way to economically store it all just to have it. They're not paying to store things just for TLA/LEO. It's not free, so if they're not using it and not getting paid, it's getting dumped.