Earlier quoted context omitted.
> but didn't secure themselves against a defector Still a surprisingly amateur move. Once worked for the Federal Government where literally everything you write is potentially subject to a FOIA request. The message all new hires were told was, very clearly: "Never put anything into writing you wouldn't be happy to be see published on the front page of the Washington Post" Anything that was even vaguely close to faili…
> The message all new hires were told was, very clearly: "Never put anything into writing you wouldn't be happy to be see published on the front page of the Washington Post" > And this was for an org that was doing nothing sneaky or underhanded in the slightest I disagree. That culture is one of routinely doing sneaky and underhanded things. If it wasn’t, the rule would be “don’t do anything you wouldn't be happy to…
Plenty of examples of companies that did nothing wrong, but some written comment, by some employee is enough to convince a jury that "something probably happened".
I remember one pharmaceutical company was on trial for a drug that was suspected to have a bad side effect (later analysis show it didn't). They had done nothing wrong, all data was provided to the FDA, they diligently collected post-marking data, shared with the FDA, etc.
A civil suit was brought and during discovery they found a comment that said "maybe our dose was too high?". The person who made the comment had nothing to do with the trials, didn't have the skills to interpret the data, didn't even have a medical background.
Once that was found? Boom, the company settled because that was enough to convince a jury that "they probably knew it was wrong".
It's very easy to put something in writing now that has no impact, but in a few years time is a smoking gun. It doesn't matter if you testify that around the circumstances when you wrote it ("oh, I didn't mean that"), the jury won't believe you, the prosecution will just flash that one sentence up and say "see, it obvious the company was hiding the problem".