Google has so many employees that they need training to limit the damage from random chatter and speculation.
It’s more cumbersome to have to talk about some things via video chat, but it’s not about limiting thought.
Even U.S. government officials have used private email servers to avoid having to serve them up via requests.
And when that failed they destroyed the hard drive that contained the exchange server db. (See IRS scandal)
You already completely answered the perfectly standard and reasonable business reason: "to limit legal exposure. "
In fact, this legitimate business reason is 100% the reason for the policy. Increasing legal exposure for no reason is a bad idea, for companies and for individuals.
Eric Schmidt’s retention policy was 72 hours.
More than likely Schmidt may have said something along the lines of deleting anything more than 3 days old because at the pace of his business, it's 'time out' and not relevant. But that's just a matter of his peculiar communications style. That the label has changed to 'archive' doesn't mean anything really from a corporate perceptive.
So yes, illegal to actually delete, and seemingly impractical to bump from one's inbox, but perhaps at 'Google Speed' there's some reason for it (and maybe there's a big caveat i.e. anything that's 'starred' or whatever doesn't get deleted, or, maybe anything older than 3 days that's opened or unopened gets deleted).
EDIT: For those who are wondering, here is a quick summary [1].
Eric Schmidt's emails are definitely kept around a very long time, for very legal reasons, and whatever he happens to do with his own personal 'inbox' is not relevant to the subject at hand, and amounts to a kind of personal email/habit choice.
To suggest '72 hours' in response to a discussion about legal discovery etc. is basically misleading in that regard.
[1] https://www.spamtitan.com/web-filtering/email-retention-laws...
Eric Schmidt is not deleting his emails after 72 hours for the reason you mentioned and certainly the company is not, which is the salient issue.
One could say 'oh that's just from his inbox' but that's pointless in the context of this conversation because we're talking about 'If the corporation has a copy or not' i.e. 'IT' etc..
Scmidt deleting maybe a local copy after 72 hours doesn't really have anything to do with anything other than his personal email habits.
[1] https://www.spamtitan.com/web-filtering/email-retention-laws...
Yes, it has: GDPR requires that you delete PII in reasonable time. I have a lot of customers contacting me by email for example, but also the JIRA notifications which all end up in emails with extensive PII. It must be deleted in a controlled way according to GDPR.
But you are correct that this excuse goes away with Google, since they don’t do support ;)
Depending on context, these are either shared with the other person (and usually editable by them) or is accessible only by me.
> There is no legitimate business reason for that policy.
So the policy is about limiting the chance of potentially very expensive lawsuits, and has no legitimate business reason? Choose one.
I wish people stopped overusing "Orwellian": the term is so overused that you could use it next to "agile" and I wouldn't notice.
"Orwellian" is an adjective describing a situation, idea, or societal condition that George Orwell identified as being destructive to the welfare of a free and open society. It denotes an attitude and a brutal policy of draconian control by propaganda, surveillance, disinformation, denial of truth (doublethink), and manipulation of the past, including the "unperson"—a person whose past existence is expunged from the public record and memory, practiced by modern repressive governments. Often, this includes the circumstances depicted in his novels, particularly Nineteen Eighty-Four[2] but political doublespeak is criticized throughout his work, such as in Politics and the English Language.
[1] https://en.wikipedia.org/wiki/Orwellian
I defend my use of the term. Disappearing the past absolutely is Orwellian. Down the memory hole!
Many business have auto-delete for the simple business purpose - when someone hacks your email (which will happen somewhere in a large business) - why do you need to keep all that crap around forever? And yes, people email payroll details, passwords, logins and the list goes on - stop with the preaching about how to email securely.
So you auto-delete, which reduces the blast radius. In most cases folks are not looking at emails past 3 years old.
>"Orwellian" is an adjective describing a situation, idea, or societal condition that George Orwell identified as being destructive to the welfare of a free and open society. It denotes an attitude and a brutal policy of draconian control by propaganda, surveillance, disinformation, denial of truth (doublethink), and manipulation of the past, including the "unperson"—a person whose past existence is expunged from the public record and memory, practiced by modern repressive governments.
https://en.wikipedia.org/wiki/Orwellian
Stop overusing terms you don't understand
This explanation was given to me by a corporate lawyer who was trying to figure out whether the same kind of expiration could be put on tickets in bug and project trackers, which would have been even more harmful to institutional memory than an email expiration policy.
The training/policies just codify that.
Have you ever been affiliated with a political party that was highly popular (or a monoparty even) in your country but was held in contempt by the rest of the world because of how totalitarian/inhumane it was?
In both cases, you could read whatever, even critical information about your values. But you would have an explanation ready — enemies envy and slander us, they either know they lie or they are repulsed by the God's light because of how corrupted they are, they are not aware of the whole truth... You would have a whole arsenal to explain things away, because you are committed, and your commitment makes it hurt to realize that the purpose your values serve is not very noble, or that you're a part of something atrocious. It's the human nature.
But there are also a lot of employees who have strongly opposed various Google policies and engaged in various political activity based on that, so the groupthink doesn’t seem to be working very well? Also, the company leaks like a sieve these days.
Even before that, there were a lot of internal debates. (They just didn’t leak as much.) It’s in part because of these debates that you need policies; people sometimes say careless things in heated discussions.
(Former Googler, but it’s been a while.)
I suppose you could make a case for continuing if the policies are/have been/could realistically be changed.
But if that's unlikely?
Similarly, at a large enough company, if your work is good, but some other division of the company is bad, should you leave the company(city/country)? Or just avoid the bad division?
[0] Leaving aside the question of whether it's a good thing for companies to get that large.
And it's limited, by necessity, even outside working hours, lest your tongue/fingers slip and you utter a bad word in your Googler capacity so that a liable deed gets a liable name and there won't be any lawyering around this.
Heck, it's almost, though not entirely, like a brainwashing cult!
I guess in China they also force their Uighur camp operators to not even think about what they do as "torture", but "reeducation". It makes them happier in their workplace.
The overarching concept is "don't make it hard for the company to do business." The point of those trainings is that the words to avoid have legally-defined meanings that may or may not be what the Googler intended, but are likely to be interpreted in an antitrust sense in a court of law. The underlying concept is "don't talk like a lawyer if you're not one of our lawyers."
Watching what you put in email (as described in this article) is in the same training where Googlers are given the overarching advice "always communicate via email as if those emails are going to show up on the front page of the New York Times tomorrow."
The fact alone that there exists such an extensive training specifically about monopoly-related stuff hints that there is extensive monopolistic behavior liability for which only hinges on whether it's acknowledged as such in the internal communication.
Also, it adds a whole new (new?) meaning to any press release or a blog post from Google using any of the terms from the right column if you substitute them with the terms from the left column. They say "dedicated to providing the best services to our users", you see "dedicated to eliminating our competitors".
But they do, and the good ones are really good at it.
Any competent company is going to train its people with some sort of variation on a course named "The Ten Dumbest Things You Can Write In An Email So Don't".
The language, at best, just makes the cognitive dissonance a little easier.
When I started at Google in 2015, in my first week here chatting with some peers, some of them were complaining about some of our policies around Android and that they much preferred Apple (the person didn't work anywhere near Android, but was complaining about it more as a user).
There are many people at Google that have issues with various parts of Google's businesses. Some are more vocal about it than others. One great example was Brad Fitzpatrick complaining about the first-gen Nest smoke alarms (2015): https://twitter.com/bradfitz/status/566072337020112896
I'm not saying Google is the great Satan or anything, I'm just saying it's impossible for most people, especially the typical Googler, to simultaneously work at a place for a nice paycheck and think it's bad for society. Everyone justifies, whether it's Phillip Morris or Google.
I don't work for Google or have much of an opinion on "Googlespeak".
However, that the practice of law is allowed to exist in its current state is an indictment on our society. The legal profession is one that polices itself, has no proper oversight (judges are just lawyers with a more refined superiority complex), raises barriers to entry with a level of zeal only matched by medicine (to which it is not actually comparable), and is also allowed to maliciously and limitlessly wield this power over the people who do real work is a foundational problem with governmental design.
In a different vein, when I was younger I didn't understand why people preferred phoning.
I as a Deaf engineer rely on written communication. This exposes myself and people communicating with me to «showing what has been said». I am sure that in my career I missed some important information just because people weren't willing to create a persistent record of communication.
My friend «solved» this problem, but I am sure his interlocutors would be miffed if they knew that.
Yes, it stands to reason that if you’re engaged in a potentially unlawful conspiracy you need to be careful what you put in writing.
However if this is coming up constantly and prevents you from using common sense words for your regular business operations then it’s a pretty clear red flag that your actions may be subjecting you to legal liability.
The other side of "Be careful what you put in writing because lawyers, lol" that is always ignored is:
"If you think we need to dress up the way we talk about this one particular thing we're doing, then maybe we should reevaluate whether we should be doing this thing. If you think we need to dress up the way we talk about literally everything that this company does, then maybe it's time to step back and reevaluate the ethics of what this company stands for."
A company is a machine that is going to do whatever it can to print money, including brainwashing its employees. You and your colleagues are the only entities capable of ethical reasoning. The company and its executive functionaries are not going to do this for you. In fact, they're more likely going to try and stop you.
It's your responsibility to do it anyway.
At Google, the team responsible for deciding whether a given project is legal is the legal team. Googlers are encouraged to get a member of legal on board as soon as a project gels far enough to have a concrete description that could have legal consequences. At that point, a set of attorney client privileged communications could begin where any of the words listed here can be on the table (because that communication is not in discoverable media).
But in general, Google doesn't encourage its software engineers to think they're experts in law any more that it encourages its lawyers to think their experts in BigTable performance tuning.
I'll grant you that not every corporate policy will agree with me, but I would argue that every human with a brain has a responsibility to think about whether what their boss asks them to do is ethical, and a responsibility to raise hell if they think it isn't.
I don't believe it's ethical to abdicate this human responsibility to a corporate legal team.
Part of what these corporate policies are deliberately designed to do is condition employees into believing that "deferring to the legal team" is where their responsibility ends. They want to convince you that this checks the box for both "legal" and "ethical" so that you feel like you've done your duty, and now you don't need to think about the ethics of your work anymore. This is what I meant by corporations "brainwashing" their employees. But you're always on the hook for the ethics of your work.
That's the key difference and the purpose for constraining what ends up in discoverable media.
There is, perhaps, a meta-ethical question of whether companies should, in general, be factoring into their calculus ways to minimize the government's capacity to hinder their activities. It's a good question. I don't have an answer that's universally true. I suspect if we sit down and consider it, we find lots of circumstances where it's not in the best interests of anyone to just hand the government a company's throat to be slashed. After all, especially if we're talking about the United States, it's not like the government itself has proven a bastion of ethical reasoning either.
Since legality in a corporate context is not typically a binary evaluation, it would be far more accurate to say that their job is to ascertain the relative financial and business costs of potentially illegal behavior so it can be effectively compared to that behavior’s potential profits