The counter argument that I'm sure Google will bring to the table is that (a) that advice was given in the context of helping engineers understand that legal definitions and vernacular definitions are different, and it is the nature of adversarial court systems that words they think they're using with one meeting can be taken out of context by an opponent in a court case, so think about what they are saying, and (b) all of the above is just facts and the government suppressing a corporation's right to state facts would be a clear and obvious violation of the First Amendment.

But these are two legal arguments and it's hard to predict how the court will decide.

“Rather than use words like ‘insider trading’ we should say ‘based on market research’”

Well, I have.

At a large well known tech company I used to work for, we were told not to use terms like "killing our competitors", "smoking them", etc. or any words that implied we had any dominance or monopoly in our industry. We even had a "War Room" undergo a name change. The reason is exactly the same as Google's, where they were worried about reporters using it to twist narratives or in discovery by lawyers who came across emails.

I think my concern is that the important thing is not to do illegal things. If training focuses on just not sounding like you do illegal things instead of just not doing illegal things, that’s the problem.

And that's the point. Lawyers can twist things written in emails to be whatever they want. People who say "if you're not doing anything illegal, then you have nothing to hide" has never been at the hands of a witch hunt. I have a friend who lived in the US on F1 and H1B for over 10 years and was banned from entering the US over jokes with her ex-bf about marrying for a green card.

Don’t worry, soon it’ll get done by AI instead and it’ll match on “acceptable” language too.

That's insane, especially after her being in the US for 10+ years and having established a life there. Sorry to go off-topic, but I have a friend whose SO is on a work visa and they have joked about the marrying for a green card thing before (a very common joke it seems). How did the government find out about the joke? (in-person, through texts, etc.) I ask because I wouldn't want the same to happen to them and to caution them.

What? That's also fine.

No, that isn't the problem.

The problem is that the difference between violating the law and not violating the law is whether you used what the court/jury felt to be inappropriate language. "Not sounding like you do illegal things" is how you avoid violating the law. Your actions are less relevant than your commentary on those actions directed to your friends.

That's why the training focuses on using the right magic words and tabooing the wrong ones. The training is there to help you follow the law, no matter how stupid the law might be.

only if the regulators investigating you are complete morons, which may be the case once in a while, but I doubt it is for any large case like this one.

regulators see straight through doublespeak and careful wording. buzzwords and corporate lingo are completely transparent to anyone who spends a single moment thinking about the message that the words are conveying, and regulators have far more experience cutting through that stuff than just about anyone else.

> Moron

Not so much because they might making some incorrect conclusions but because they don’t know when to keep their mouths shut.

When one actually suspects that something is anticompetitive, subverts the privacy policy, or is against the law, the done thing is to go talk to an attorney. Either product counsel inside the company if you are just curious or concerned in good faith, or your own attorney if you think the company is acting in bad faith.

No they don't, Microsoft didn't do this training and got nailed for bundling internet explorer with windows. Then Microsoft and all other big tech companies started to police words and look, no big anti trust like that has happened in US since. As long as Microsoft says "we always open edge no matter the settings when you click links in our products to ensure the best customer experience", instead of "we always open edge to increase edge market share", they wont get nailed in USA.

you are saying that regulators can't see through this then giving an example of them seeing through it.

They spend a lot of time parsing just what Gates and others on his executive team meant by things like “cut off [Netscape’s] air supply”.

Regardless of who you think “wins” these exchanges, the government thought that putting Gates in a position where he had to explain this kind of language would be embarrassing and discrediting.

It worked.

Describing something in terms of helping the competition while actually doing the opposite, will get you dunked on doubly so. If you think equivocation is they to doing illegal things, my dear poster, snap out of it and feel the Terra firma, because you're so high, the FAA can do an airspeed check.

Actually anti-trust training I have gone through mostly emphasizes that you should never ever help the competition or talk to them, because that sounds like forming an illegal oligopoly.

He's saying banks/funds don't do this. The words were "From a bank/fund perspective".

You're saying large tech companies (not just Google) do.

Hence the impression is, big tech is not being careful to follow rules, they are just trying to avoid getting caught.

I got the same takeaway when I was at Google as when I was in the hedge fund space: be careful what you say, regulators and judges might not be very charitable. In both cases, the concept of something you say being in the New York Times was present for me.

The Googlewide training and messaging when I was there did focus on not using anti-trust-triggering _language_ over the specific acts, but I don't think it's fair to interpret this as "Violate anti-trust regulations" but rather a recognition that the common fuck-up some random employee can make isn't to commit a material anti-trust violation, but to provide nasty evidence by writing a puffy email about "crushing the competition". Telling people to write emails about "making users happy" rather than "owning the market" is not telling people to behave badly.

I don't know what you're doing in finance, but it's my understanding that folks who are actually involved in stuff where their job could look like insider trading probably do get subtler lines drawn than I ever sought (I was in systematic/program trading.) Places where doing my job well could be confused for something bad, e.g., tax optimization, we did discuss how to properly communicate in order to avoid misunderstanding if a regulator, court, or journalist ever read our communication.

For example - all emails to above go through compliance first & include a disclaimer that you are a market participant and do not wish to receive MNPI. Meetings with above have compliance sit in, etc. People are urged to self report if someone does volunteer something stupid to you that they shouldn't have, and then you sit out from trading it for some period, etc.

They make it very clear that the upside/downside risk is asymmetric. Hey maybe you can cheat and make 10% more this year. Or you can go to prison for 10 years, lose your home, have your wife leave you, and never work in the industry again.

    you can go to prison for 10 years

Raj Rajaratnam is one of the most notorious in the current generation. He only did 7.5 years. And Stephen Cohen had so many layers in his onion legal strategy that SEC could never nail him down.

It’s like an investment banker back in 2006 saying that he was doing everything by the book, which he might of well have done, but the whole thing was crooked nevertheless.

that's fair - did they also provide training on actually not violating anti-trust regulations, or was it just how to avoid leaving evidence of violations?

The additional rank-and-file training and process that ties into actual anti-trust-related behavior is how to contact a lawyer as well as the fact that a lawyer and executive needs to review and approve all of the most minor of launches.

Do you have an idea of what the relevant training you would prescribe for a random mid-career engineer or UX researcher or whatever would be?

If they can develop training that says "don't say bundle", they can develop training that says "don't bundle", and in both cases they might want their employees to talk to a company lawyer if they have any questions anyways

also curious if the lawyers themselves had rules like "don't bundle" that they would pass along to employees with questions, or if google was okay with bundling

You may not believe it, but no financial firm I've ever worked for gave me training on "don't use insider training language". All the training was emphatically that they did not want us to do it, it would destroy the company and destroy our lives explicitly. It was drilled into our heads with live talks, videos, handbooks, attestations, etc.

This GOOG training seems to be like "how to pass a sobriety test" rather than "don't drunk drive".

There's a ton of stuff that has been allowed to happen via consolidation, merger, or pure homegrown that simply hasn't been prosecuted because recent administrations had an alternate theory that monopolies are OK if it benefits consumers via lower prices. Tech gives out a lot of stuff free or where its cost is hard to directly measure.

Tech also enjoyed bipartisan support for the last 20 years or so, but that has completely flipped over the last 4 years or so. We will likely see a lot of this stuff is no longer ignored.

That's really not incompatible with violations in other circumstances. For example, Google releases Android for free, resulting in low prices for mobile operating systems. Competing operating systems may be put out of business, but customers benefit from lower prices and this should be fine if anyone can still choose to use a competing OS -- or fork of Android -- should they prefer it.

Whereas, Google uses tying to its dominant services to ensure that Google Play is the only viable app store on Android, then takes a 30% cut. High margins don't result in lower prices and neither does excluding competitors who might take a smaller cut, so they should be slapped down for this.

Google adds remote attestation and encourages third parties to rely on it even though it makes third party apps dependent on Google's version of the system, creating a barrier to entry for forks or competitors. This kind of thing should be prohibited, because otherwise people are not choosing Google's offering because it provides the best value for money, they're choosing it because Google manufactured a way to shut out competitors who might do better.

The theory isn't the problem. Taking a competitor's market share because customers choose your product for having a better price is not bad. The bad is making it harder for customers to choose competing alternatives they might actually prefer.

The problem is the application. The company comes up with some claim that their anti-competitive moat is there to benefit customers by providing security etc. and get away with it even though it doesn't really provide security or there are viable alternatives that provide security without restricting competition.

A good start would be don't do things that create friction for a customer who wants to choose a competitor's product instead of yours.

the article says employees were instructed to avoid the word "bundling", were they instructed to avoid the action as well?

With that said...

> You may not believe it, but no financial firm I've ever worked for gave me training on "don't use insider training language"

These are apples and oranges. You may not believe it either, but I don't think Google tells anyone "don't use insider trading language" either. That's the apples-to-apples comparison.

AFAIK being a monopoly is a corporate civil matter, not a (strictly personal?) criminal one like insider trading. Expecting people to treat them the same way and claiming they're guilty if they don't makes no sense.

Second, Google wants to withhold from the public certain provisions of its standard agreements, including definitions contained in such agreements, many of which were signed more than ten years ago. Id. at 56, 65-69. These agreements and definitions are foundational to the publics understanding of how Google established its dominant market position.

Third, Google wants to clear the courtroom during the overwhelming majority of Googles examination of Apple witnesses, which could span several days. Id. at 69-70. There is substantial public interest in Googles deal with Apple and the termination of the agreement. There does not appear to be any good reason to close the trial completely for this testimony other than to shield Google and Apple from potential embarrassment."

https://ia902501.us.archive.org/21/items/gov.uscourts.dcd.22...

But until we have that, people will keep studying what they can and can't say.

Not to be paranoid, but that's also how a bank that wanted people to insider trade would act. The idea is, you see the people who got caught, learn what they get caught for, and carve the apple just a little bit less than they did.

A principled (naive) person imbibes the meaning "Don't do bad stuff", and an ambitious (sociopathic) person learns how to do it and what gets caught.

“Politicians and regulators are always playing catch up. They are so dumb. We are creating the future” - some tech bro

They don’t take society seriously. They don’t take the law seriously. At best they aren’t thinking about what they are saying. At worst they feel they are justified.

Treat the law seriously. Don’t break the law. If you do those two things you will not need to send memos around to get people to use phrases which regulators will grep your comms and transcripts for.

From a consulted a bank perspective - I noticed that the examples were pretty darn intuitive that hey this is obviously illegal but for some reason they wanted to have people sit through a multiple choice form and 45 minute tutorial on banking ethics to make sure nobody did it, and also that one of the negative outcomes that was mentioned in breaking the code of ethics was that the bank would experience problems in the media that would make the whole thing which seemed like a good idea not worth it.

Imagine a scenario where a PM or engineer writes down in a memo something like "Option #1 is a breach of antitrust regulations" or "This violates EU privacy regulations". These are often well meaning statements but the reality is most (all?) of the time these people have no idea what they are talking about (which EU privacy regulation? what are the elements of the antitrust law and how does the action implicate each one? etc).

However, a statement like this is highly prejudicial to a court as it indicates fore-knowledge of wrongdoing especially if it is made by someone fairly senior in the organization (eg middle management and higher).

The better option is to pose the assertion as a question directed to your lawyer. This has the benefit of (a) attorney client privilege so the document cannot be discovered in court and, more importantly, (b) getting someone who has the relevant expertise to weigh in on whether what you think is in fact true so you can base your decision making on facts and not random persons opinion of the law.

NB: Imagine your legal team weighing in on whether you should design your architecture as a microservice or monolith. That is the equivalent level of hubris for a non-legal person weighing in on the legal implications of particular courses of action.