> I don't understand what's going on at Google. If actual data loss occurred, trying to pretend it didn't happen is never the answer. They're not pretending it

> I don't understand what's going on at Google. If actual data loss occurred, trying to pretend it didn't happen is never the answer. They're not pretending it didn't happen though? As per the article they acknowledged it and published a help center article on it. They named the software versions affected (notifying the affected users seems impossible, since the entire problem was that the data had not been synced). Following the links in the help center article, during the incident they posted in a pinned article in the support forum (multiple times) on how to avoid triggering the bug and how to avoid making it worse. That's pretty much what you wanted to see except for a blog post with an RCA, no? > Or if the problem really is solely with files that should have been uploaded but weren't, and nothing from the cloud ever actually got deleted, then explain and justify that as well -- because that's not what people are saying. So the suggestion is that in addition to the bug that they acknowledged, there's a totally different one that appeared at the same time affecting totally different functionality and with different symptoms, and that they're covering up despite not covering up the other bug? That seems like a complicated explanation when there's an obvious and simpler explanation around. That's also the kind of thing that's pretty much impossible to prove categorically, let alone communicate the proof in a way that's understandable to the average user. What are you going to say? "We've checked really hard and can't confirm the reports"? (I mean, I guess it's possible to do it. Collect 100 credible reports of files going missing that can reliably identify the supposedly missing file by name and creation date rather than say that it was probably a .doc file sometime in March. Then do an analysis on e.g. audit logs on what the reality is. How many files were never there at all? How many were explicitly deleted by the user? How many were accidentally uploaded to a user's work account rather than personal account? How many were still in the drive, and the user just couldn't find them? And yes, once you've exhausted all the possibilities, how many disappeared without a trace? Then publish the statistics. But while doing such an investigation privately to make sure whether there is a problem makes sense, publishing the results seems like a stunningly bad PR strategy even if no data was indeed lost.)