Hacker News Reader

Comment by habitue | Hacker News Reader

The complaint details the reason it was not allowed in this case, though would be normally: the topics were subject to legal holds.

> Like Mr. Pichai, other key Google employees, including those in leadership roles, routinely opted to move from history-on rooms to history-off Chats to hold sensitive conversations, even though they knew they were subject to legal holds. Indeed, they did so even when discussing topics they knew were covered by the litigation holds in order to avoid leaving a record that could be produced in litigation. As the examples below make clear, Google destroyed innumerable Chats with the intent to deprive Plaintiffs and other litigants of the use of these documents in litigation

Note: that's the argument the plaintiff's lawyers are making, not the final word on the matter

View on HN

> Note: that's the argument the plaintiff's lawyers are making, not the final word on the matter

This is not true anymore. The judge in the case determined that the DOJ was right.

See: https://news.ycombinator.com/item?id=35363095

Thanks, I'm not familiar with what a legal hold is. Would they be obliged to take audio recordings or minutes of meetings due to this?

Not a lawyer. When a legal hold is issued, the company is instructed not to destroy potential evidence related to the case.

They don’t have to perform additional recordings and minutes. But they exist, they can’t destroy it.

Who gets to decide what is "potential evidence"? If we're going to play language games, turning the chat history off doesn't really destroy anything, just prevents the paper trail from being created.

Well, I think the physical mechanism of turning off chat history on most of the modern chat platforms is to have a database delete each message after everyone in the conversation reads the message, which is a lot closer to destroying evidence than not making records.

Also, if you destroy evidence that may be under litigation hold, the judge will instruct the jury to infer that what you deleted was harmful to your case.

> Who gets to decide what is "potential evidence"?

You have an obligation to proactively figure that out yourself. There is then oversight by the court.

> If we're going to play language games

Judges don't play that sort of language games

> turning the chat history off doesn't really destroy anything

What Google called off the record chats were 24 hour history chats. So the history was deleted after 24 hours, when it should have been preserved.

> What Google called off the record chats were 24 hour history chats.

Ah, I missed that part, that's a very compelling argument. I disagree these are not language games though. The way the chat application was kept from preserving history is an implementation detail that doesn't reflect the action the user of the chat performed in the UI.

It sounds like the truth is whatever a human judge can be convinced into deciding is true.

> It sounds like the truth is whatever a human judge can be convinced into deciding is true.

I mean, that’s literally the judges job. It’s why they are called a judge. They make a judgement as to where the truth lies, and where the law lies (except for issues that are delegated to a jury, in which case the jury is responsible for determining the factual record).

But, yes, judges make judgements and a lot of the “well, technically” semantic games that people play will be met with an annoyed scowl and a negative opinion (but, critically not all semantic games; there are some semantic differences that are of critical importance to a judge)

I am getting the feeling you are being disingenuous: a record is a record, a records that gets deleted after 24 hours was still a record. This is a truth based only on the meaning of the word "record". The chat application was not kept from preserving history, it was designed to actively delete history after 24 hours

If there is a court order to not delete history, then the application was designed to do the opposite, and if a technical detail meant the application could not be changed, then the only way to respect the order would have been to stop using the chat application. Instead, there was a decision to use it more ...

I admit I am playing the devil's advocate, but not trying to be disingenuous. The intention behind an ephemeral chat application is to allow private communication without creating a record. From the end user's perspective, the messages are thought to be self-destructing objects that are kept just long enough to be read by their recipient. The way this is achieved in practice is an implementation detail that end users are not necessarily aware of. I don't see how the users of the chat application can be expected to be aware of the implementation or proactively find ways for these private conversations to be recorded.

An analogous real-world situation would be a meeting in a room where someone uses the whiteboard to draw some diagrams and erases it at the end. If we apply the same logic, you would be under obligation to take pictures of the whiteboard and send them to the court? Or even that you should record the whole meeting? Something just doesn't feel right with the idea that you can be under obligation to help make a case against yourself in a legal proceeding.

> Something just doesn't feel right with the idea that you can be under obligation to help make a case against yourself in a legal proceeding.

Well that’s kind of the whole thing with discovery - the court is asking the company to disclose relevant information (some of which will be against the company). Companies here have different disclosure requirements to an ordinary individual.

And if new evidence is being created, that new evidence would be subject to the same data request.

> I don't see how the users of the chat application can be expected to be aware of the implementation or proactively find ways for these private conversations to be recorded.

Most corporate messaging apps have built in functionality to handle litigation holds. This is a requirement for legal compliance, so the company would have to train and brief employees on their legal responsibilities to not destroy evidence during the litigation.

See: https://learn.microsoft.com/en-us/microsoft-365/compliance/e... or https://slack.com/intl/en-gb/help/articles/4401830811795-Cre... or https://support.google.com/vault/answer/7657594?hl=en

Which conversations are subject to legal hold would be discussed in court.

> I admit I am playing the devil's advocate

Are you literally appointed by the Pope? Part of the process of canonisation?

> I don't see how the users of the chat application can be expected to be aware of the implementation or proactively find ways for these private conversations to be recorded.

The company must educate its employees because the company has obligations. A limited-liability corporation is a legal fiction granted extraordinary protections by a society. That corporation had damned well better be a good citizen of that society. If that means erring on the side of caution? Good.

Everything you do on a computer is recorded, at least ephemerally, in the computer memory. Does deleting stuff from the memory count as destroying evidence? If not, how is it different from deleting stuff from somewhere else? I mean, deleting information that is not used is part of how computer operate, if you're going to disallow that, you better have a pretty huge tech infrastructure to store all those memory dumps from every workstation Google has every second or so.

> Does deleting stuff from the memory count as destroying evidence?

Maybe? The line is more blurry here, but there are probably cases where it's more or less easy to make the distinction.

If data is recorded on a permanent medium, it can be arbitrarily retrieved, an explicit delete operation is needed to get rid of it, and this deletion can be scheduled to happen at an arbitrary time - that might easily be understood as "destroying evidence".

If you want asynchronous communication, you will need to produce records of some kind, which can be evidence. That's nothing new.

Maybe if you have a system where the information must be exchanged live and falls off a cliff immediately no matter if it was observed or not, say, like a video conferencing tool - I wonder how the courts see that. It gets much closer to an interesting edge case than something with a 24-hour retention policy.

What if the 24 hour chats are only stored in RAM?

That's not permanent.

RAM is permanent as long as you don't pull the plug. If you've been told to preserve evidence, and still pull the plug, then you did not do what you were told to do.

Silly analogy: person on lit hold writes a message in the sand on a beach with the tide out. Are they obligated to preserve the message before the tide erases it naturally?

If a judge thinks so, yes. So you should err on the side of caution, while being a good and ethical participant in the judicial process, and take a picture of it.

If this would be detrimental to your case, perhaps stop doing shady shit.

You have constructed a hypothetical case where the answer would likely depend on facts and circumstances beyond what's written.

That such cases can exist doesn't mean it's not clear-cut here.

In the vast majority of cases, two engineers will agree on the question "is this data persisted or ephemeral" and the fact that the line is blurry doesn't mean we can't know when it's been crossed.

The judge.

Judges rarely care for people who think they're smart enough to play "language games".

You are expected to act in good faith as an executive or legal officer of a publicly-traded company.

Turning chat history off typically doesn’t stop it from being recorded at all, but rather auto deletes messages after a short period.

If messages are being deleted 1 day after being sent, despite being subject to a litigation hold, that’s… going to often make judges mad.

> Who gets to decide what is "potential evidence"?

Ultimately, the courts do, but if something is possibly related you’re supposed to preserve it.

> turning the chat history off doesn't really destroy anything

IANAL but courts tend to view all written conversions the same way. Turning off chat history is viewed the same as throwing away all the mail you receive, and both are very much frowned upon when the court has specifically asked you not to. These rules are set through longstanding precedent.

What is the chat happens in a software that didn't have chat history feature in the first place? I mean it's Google, surely they could cook up a separate internal chat tool just for this purpose. And if the chat history didn't exist in the first place, can the court order the company to switch to a chat software that has it? If not, how is it that different from toggling chat history on/off in a single software?

IANAL but my reading is that by transmitting chat messages you are generating potential evidence so discarding those is destruction of that potential evidence.

Presumably this does not apply to audio-visual real time communication as you're not transmitting entire messages but fragments in real time that don't convey any information by themselves. Sending back and forth recorded audio messages OTOH would presumably produce potential evidence. Sending each letter as it is typed presumably would not.

That said, usually these things are ruled based on the spirit of the law as much as the letter of the law so trying to game this by working around technicalities would likely be considered ... bad.

> Not a lawyer. When a legal hold is issued, the company is instructed not to destroy potential evidence related to the case.

Not a lawyer, but I have helped implement record management systems, and a requirement is always support for legal holds.

If the reg affairs or legal people were told "oh, yeah, we're deleting all these company records even if they have holds" they'd have strokes. Or quit. Or both.

No, absolutely not. Legal holds prevent you from destroying evidence, they don't require you to produce new evidence and make it available. The legal theory of the plaintiffs in this case is that using 24-hour-expiring chats is tantamount to destroying evidence every 24 hours. I'm not a lawyer and can't comment on how well this matches with previous precedent in this area, but I get the general vague sense that using "off-the-record" to specifically avoid discovery is often something courts take a very dim view of—even when those chats are replacing in-person conversations and meetings the court would have no way of knowing about before the advent of IM.

EDIT: And the court in this case took a dim view of it as well: https://storage.courtlistener.com/recap/gov.uscourts.cand.37...

> At the heart of this dispute is a simple question: did Google do the right thing with respect to preserving Chat communications in this case? There is no doubt that Google was perfectly free to set up an internal IM service with any retention period of its choosing for employees to use for whatever purposes they liked. The overall propriety of Chat is not in issue here. What matters is how Google responded after the lawsuits were filed, and whether it honored the evidence preservation duties it was abundantly familiar with from countless prior cases. The record establishes that Google fell strikingly short on that score. Several aspects of Google’s conduct are troubling. As Rule 37 indicates, the duty to preserve relevant evidence is an unqualified obligation in all cases. The Court’s Standing Order for Civil Cases expressly spells out the expectation that “as soon as any party reasonably anticipates or knows of litigation, it will take the necessary, affirmative steps to preserve evidence related to the issues presented by the action, including, without limitation, interdiction of any document destruction programs and any ongoing erasures of e-mails, voice mails, and other electronically-recorded material.” Google clearly had different intentions with respect to Chat, but it did not reveal those intentions with candor or directness to the Court or counsel for plaintiffs. Instead, Google falsely assured the Court in a case management statement in October 2020 that it had “taken appropriate steps to preserve all evidence relevant to the issues reasonably evident in this action,” without saying a word about Chats or its decision not to pause the 24-hour default deletion.

Normally a document (chat) deletion schedule is completely fine and standard practice at most corporations. The problem is that it continued (without exception for those with data retention holds) once litigation had started.

TBD whether using OTR exclusively (like having in-person real time meetings) for all chats would violate this. I guessing no, it's the choice to use OTR selectively by participants that causes problems.

Interesting, is there any documentation about the legal holds details?

https://support.google.com/vault/answer/7664657?hl=en

mmm i'm looking for the legal documentation/language that requires google to hold the data for this particular topic.

because i'm in agreement with the grandparent comment that why is the topic subject to legal holds vs just having a meeting in person? Are in person meetings rquired to have a voice recorder present? it seems like a weird legal precedent.

You aren't required to make recordings. But if you made a recording, then you can't slate it for deletion. By disabling history, they're setting it up so records are deleted after 24 hours, instead of kept indefinitely.

In the court's eyes, it's analogous to writing notes to each other, only to burn them the next day.

> routinely opted to move from history-on rooms to history-off Chats to hold sensitive conversations

The way this is worded there's no record, so there's no "slated for deletion"?

I guess the question becomes, if it exists inside computer memory, is that legally equivalent to "writing down piece of paper" and thusly is considered destroying evidence?

If I write something in disappearing ink (let's say it perfectly disappears), it's not destroying evidence? is it?

There is a record. It exists for 24 hours. It’s not in just your local computer’s memory. It’s in the cloud, and you can see that history on any of your devices.

> It’s in the cloud, and you can see that history on any of your devices.

For argument's sake, let's say it's not. It's a tunneled connection and all stops in between are only acting as relays.

Is it still wiping it then if it only exists in memory and then gets garbage collected? is that destroying evidence?... maybe...?

I think that would probably be ok, same as if it was a phone call.

You asked for documentation of legal holds. Pardon me for interpreting that as such and providing a link to Google’s legal hold documentation.

It sounds like you are looking for the laws “that requires google to hold the data for this particular topic.”

You can simply have this discussions in a room person-to-person privately.

It gets a bit tricky privacy-wise if you cannot do same online. Though Google is probably the last company deserving any privacy.

Quite ironic...when you consider the lengths Google goes to, to retain everybody else's data.

Is that really irony though? It kind of seems like self-preservation. Based on what I merely suspect Google knows about me, it wouldn't be a surprise at all if I receive a letter of acceptance/rejection for employment there, despite having never applied nor considered doing so. "The algorithm" knows the plans that I haven't even thought about thinking about.

However, these are not people that can merely speculate; they live at the bottom of the rabbit hole.

Imagine you work for a company that has records about you with enough depth and granularity to have already calculated (with fp64 precision) everything you will ever do and contribute in your life (i.e. is your actual future cost/benefit ratio, not an estimated value, favorable enough to allow you to continue exist).

Would you NOT make at least some sort of effort to stop "the algorithm" from consuming you? Would you just keep feeding "the algorithm" pieces of yourself until it eats enough of your being that it becomes you?

It seems they've decided to save themselves by not saving their chats. I'm not a lawyer or anything, yet I'm getting a vibe like this kind of stuff prolly looks "kinda legally bad." Law things like this happen because there simply isn't time to waste worried about looking "kinda legally bad." It's natural to prioritize taking care of higher magnitude existential threats like getting processed by "the algorithm."

No, if you actually read the messages, many of the employees want to turn history on, because it’s useful, but turn it off (or are instructed to turn it off) specifically to avoid being subject to legal discovery.

Google has been getting a lot better.

They let you change collection settings and delete your data.

It’s a lot better than most companies do.

How do you know they actually honor your requests? What about any of your data that's gone into training one of their ML models? You can't just delete the effects of training data from your model weights.

True although in the EU at least I'm pretty sure that retention of the effect of some data after aggregation is ok as long as the originating data is deleted and the aggregation is such that the original data couldn't be reverse-engineered afterwards. So that would cover use of your data in an ML training set. They would need to remove it from the set for future trainings but the weights of existing models wouldn't need to be altered. That's my understanding anyway.

Reply on news.ycombinator.com