Comment by throwaway7771 | Hacker News Reader

Comment by throwaway7771 | Hacker News Reader

Parent Full thread

I quit Google because I could no longer be a part of the dystopian nightmare they envision as their version of the Nerd Paradise.

You have no idea how strong the reality distortion field is inside of that bubble. Googlers are breathtakingly naive about how the line has been pushed by the senior leadership in order to go from being mere billionaires to mega billionaires.

And it's not that hard to see the mental model and economic conditions that gave rise to this situation.

Assumption 1: Technology is a universal good.

Assumption 2: Centralization is the most efficient mode of operation. (this follows directly from the mission to "Organize the world's information")

Assumption 3: A dumb mantra like "Don't be evil", like intentions matter, and actions don't.

Assumption 4: A centralized data-driven computer system makes the best decisions.

Assumption 5: If the market rewards it, it's moral.

Assumption 6: The more you know about people, the better you can manipulate them (c.f. showing them "more relevant" ads)

Conclusion: Google knows best for everyone, its AI should run the whole world, and since it has a crapton of money, it must be right.

Just pump that 25% year-over-year growth long enough, and you have a sprawling multi-national entity that literally knows everything about everyone. What they search for, what they buy, every website they ever visit, how long they stay there, and how they interact with it, what they watch, what music they like, who they talk to and what about, who they spend their time with, and where, exactly, to the meter, they are every single second of every day, what they look like, what they like to take pictures of, what their voice sounds like, their credit history, criminal backgrounds, and soon, their DNA.

The world doesn't need Google to know this shit. I don't need Google to know this shit about me in order for it to tell me what Katy Perry's second album is named. For fuck sake, we need to stop this.

Your list doesn't really look that damning. All of them are true to a large extent (I suppose they wouldn't be enticing to smart people if they were completely wrong).

The actual cause of Google's bad behavior isn't on the list, though it's hinted by Assumption 6. Namely, Google decided to become an advertising business at its core. When your very essence is poisoned, it's only matter of time when ethics go out of the window, and business practices become increasingly abusive.

Exactly. That's what doomed it. Same with Facebook. There is no way to run an ethical business when your core business model is advertising. It's hard enough when you run a company whose income relies on advertising (like a newspaper), but impossible when you yourself make advertising your primary feature.

I honestly, personally, would not care much if Google had a lot of information about me if their goal was something other than influencing me to buy products and services. I know many others are uncomfortable with any sort of privacy loss, but for me, it depends entirely on the end goal and exact nature of the data usage. And for others, they should have an easy universal opt-out (e.g. disabling all Google Analytics tracking without having to install a Google browser extension: https://support.google.com/analytics/answer/181881).

Agreed. Moreover, if their core business model wasn't related to advertising, their data collection practices could (and probably would) have looked much different. For instance, there's no reason to be opaque about exactly what data you collect, or to make the collection enabled by default, or apply dark patterns coercing people to opt back in when they opt out, etc. - no reason to do all that, if the data isn't used for shady purposes.

> And for others, they should have an easy universal opt-out (e.g. disabling all Google Analytics tracking without having to install a Google browser extension

Problem with that is you'd require installing yet another cookie, which is another data point to fingerprint people.

Or, they could just start voluntarily respecting the DNT header.

Let's not and make that a legal (and legally punishable) mandate.

…which some browsers have already started removing.

Then it should be opt-in. If you don't have the "I opted in" cookie, then you aren't subjected to the surveillance.

> would not care much if Google had a lot of information about me if their goal was something other than influencing me to buy products and services

Ironically, I'm the opposite. I don't care about the advertising targeted at me (I pretty much never bought anything based on targeted advertising). I worry about the knowledge about someone being used to hurt them.

For example, a professor once said that each of us commits 3 felonies a day. With perfect knowledge of people, it brings the specter of selective enforcement. That selective enforcement can be used to blackmail/extort/manipulate anyone you want to.

For a fictional example, check out the TV series "Suits". The "good guys" are always turning the tables on the "bad guys" by digging into their past and finding dirt on them. And vice versa. It's the plot of pretty much every episode. (But of course the good guys do this for good, and the bad guys use it for bad. But in real life, this isn't so simple.)

Even if you think outing people for doing bad things is always a good thing, the data collected on people is often wrong, and innocents get hurt as well. The definition of what is "bad" also shifts over time.

> Namely, Google decided to become an advertising business at its core.

That's a massive bingo right there. And engineers at the company really don't grok this at all.

The content of your response is so divergent than what I have heard from people working at Google that I interact with, both personally and professionally, that I find it difficult to believe you actually spent a significant amount of time (if any) at Google.

1. I've never heard anyone say or imply that technology is universally good. I've heard people talk about how making information readily available and accessible to all is good (though people disagree on how universally to apply it).

2. As far as centralization, from the things I've heard it seems that centralization is considered bad. Everything needs to be distributed, standardized, and redundant. Yes, they seem to like to keep things organized, but apparently they'd rather do it in a heterogeneously distributed manner (like each user request creates many internal requests to different functions and then the result is stitched together).

3. "Don't be evil" seems to be interpreted in the "Don't do evil things" sense instead of "Don't be motivated by evil".

4. I would say a "data-driven system" makes the best decision is the closest I could find to your assumption, and I would tend to agree with them.

5. I've never heard of anything like that. I have heard people mention things that Google could do but doesn't since it is considered not good -- an example of this is that there seems to be high level opposition to using browser fingerprinting for getting around browser cookie blocking for advertising (despite already having developed state of the art techniques that are used in anti-abuse).

6. This might be the closest assumption you have, actually, since it depends on your definition of manipulation. If you ask me if fire is hot and I teach you that fire is dangerously hot (and maybe also provide some example sets of protective items and clothing you could use to work with fire to mitigate the danger), I have likely manipulated you into not getting burned later. So sorry that I'm so manipulative ;)

> Everything needs to be distributed, standardized, and redundant.

Being distributed within a single company's data centres is still centralisation.

For point 2, I think they mean in terms of product and not code, ie. "Google should create a centralized social platform (G+)" vs "Google should create a federated social platform or contribute to existing federated platforms (Mastodon)".

> "Don't be evil" seems to be interpreted in the "Don't do evil things" sense instead of "Don't be motivated by evil".

You are what you do.

>The content of your response is so divergent than what I have heard from people working at Google

Fair enough, but what do you hear from people who don't work at google?

I assume a lot don't care about tracking and the few that do.. though luck? Get used to it?

It's being sold to us as personalized advertising and somehow we're being told that everyone wants it. Is it really that strange for someone to say that they don't want it?

Not gonna get between you two, it could also be that you are a Googler and are thus inside the reality distortion bubble. Or maybe merely a SV resident. It's gonna twist your perception in ways you cannot even see until you're outside of that lens. Google has a tendency to take over people's lives. Googlers so often have only other Googlers as friends, eat 2 or even 3 meals a day there, the company is overly concerned with your well-being and work-life balance, etc. Part of your underlying point seems to be that since Google is made up of seemingly nice and well-meaning people that it couldn't possibly be a horrible thing. I'd argue that it is instead the configuration of people and its incentives that can turn something truly horrible. (something like the infamous prisoner experiment, but milder: Google essentially presents an internal narrative about its own virtuosity and need to scale and swallow the world) And I'd agree with the OP that the senior leadership is mostly concerned with scaling either their own fortunes or the company's bottom line.

1. Sure they don't say that out loud. It's just an assumption baked so deeply into the mindset that no one even considers it a thing worth openly stating or questioning. And others might quibble about making information universally accessible being wonderful. A lot of information is misinformed propaganda. Look at flat earth videos.

2. You are talking about decentralization (and replication) as a means of implementing robustness. You fundamentally misunderstood what this person was saying.

5. There are a massive number of free market libertarian types when it comes to economics, especially among the upper management.

Someone else pointed out that none of these assumptions are all that damning. I agree, they aren't. And I agree that the OP missed that Google's fundamental sin was its decision to be, at its core, an ad company. Googlers definitely don't understand that, either.

> 2. As far as centralization, from the things I've heard it seems that centralization is considered bad. Everything needs to be distributed, standardized, and redundant.

I feel like this has to be a willful misunderstanding of the parent comment. Why would the comment above you, in talking about Google's moral vision and action, be referring to the actual architecture of servers and code? Clearly they are talking about "centralizing" as in "if we can gather all the information together, and all our services can share information, then...."

The wiring of the data centers is irrelevant.

> If the market rewards it, it's moral.

This is a very common stance. It's also, in my opinion, the core rot that is setting the tech industry up for a huge (and entirely justified) comeuppance.

This won't be isolated to the tech industry. This is rot at the core of capitalism itself. As long as people worship money as the only real God, it's fine. But that sentiment is in rapid decline with the young.

A large number of people have left Google after their understanding of the larger context changed.

When I was there in the late 2000's I recognized that their internal and external narrative was very carefully constructed. For me, that is always a red flag as it suggests there is an alternative narrative that might not be quite so flattering, if not downright bad.

That isn't to say that they "lied" or they mislead, just that they consistently worked to emphasize information that fed the narrative they liked, and downplayed information that suggested other narratives might be more accurate.

I related some of this to a coworker at Google using an example that a friend of mine on the police force of Nashville gave me. He said, "I can tell you that this woman just picked up a gun and shot this guy in cold blood. Three people saw her, and she has admitted to doing it. She's a murderer right? That is a narrative. If I add in the fact that just prior to shooting the victim the woman had freed herself from a rope restraint system, and that the victim had kidnapped her and was holding her as a sex slave, now she is a hero right? The narrative changes. The fact that she killed him are the same, but the narrative is much different."

I was having this conversation with my friend in the context of how difficult policing could be. There are people who construct narratives around the facts they choose to present in order to construct an emotion in the listener.

Countering narratives require that more facts be made available. My Grandfather was a US Attorney and a big believer in honesty as the best policy, he would say "The truth of a matter doesn't change when you add additional facts." It took me a long time to appreciate the depth of that statement.

The issue of narratives is further compounded if one side of the situation is constrained in supplying additional facts, like the police are about on-going investigations and Google employees are by their non-disclosure agreements. It gives an advantage to the side not so constrained to construct their narrative the way they want to construct it without fear of contradiction.

I was told by a member of one of Google's many internal committees that I was considered something of a "trouble maker" for asking questions that "suggested the company was hiding something, even though I had no evidence to back up those suggestions." I appreciated that they were trying to help me be more successful at Google, but to be honest that was a reasonable characterization of my actions. Although the questions I asked tended to relate to a particular piece of data or fact that was missing and might change the current narrative.

My assumption going in was that if there was a data point missing from the discussion, and I could see how that data point would change the narrative, then the entity constructing the narrative was likely hiding something. I generally want to know that data point because I prefer to make my own judgement based on all of that facts that are knowable rather than accept someone else's judgement using only facts they provide.

Pro tip, this always bothers people who carefully construct narratives to serve their agenda :-) So don't do it unless you are okay with the person seeing you as a threat.

The fact that the woman did something wrong does not change, but the law make distinctions and things like mitigating or even eliminating the punishment around some situations like you state.

That is there are some new facts that could mitigate or even justify another fact or actions.

For example, if you enter a private property in the US with a gun(in order to steal, or kidnap or kill) the property owner can kill you. Which is a bad thing, but is justified by the fact that someone entered your property armed without your permission.

In some counties in Europe, if you use the force against someone that carries a weapon in your own property, and you kill this person, you go to jail!! And have to indemnify the robber!!

Yes, you do something that is very bad, but you were forced to do it by someone who wanted to profit doing something bad to you. It would be worse doing nothing.

The same happens in wars. Americans invaded Europe and killed (and raped) thousands of thousands of Europeans. That was terrible and wrong. But it was done because someone (Hitler) did it first to other countries in Europe, like Poland and France.(It would have been worse not doing something)

No soldier went to jail for killing enemies(or raping their daughters and wifes).

Another example is police men. They carry weapons(that kill) and they have to use them if necessary, even if it is wrong.

The narrative that you imply is actually creating stories(and facts) that justify your behavior that are actually false.

You know Hitler used german soldiers disguised as Polish in order to justify invading Poland?

He used the fire in the Reichstag fire as Narrative for taking absolute power.

That was stated already by Napoleon, who believed before invading a country you needed a pretext(A narrative) something he learnned from his favorite book, The Prince of Macciavelo.

Today of course those narratives continue existing. Most Americans agreed on Invading Iraq because 9-11. The invasion was planned in words of the secretary of State Defense years before the 9-11 attack. The fact that the country is 2000 miles away from where the terrorist came from (an US allied country) and that they are populated mostly by another faction of Islam didn't matter the least.

But narratives are not bad, if they are true. It is only when they are fabricated when they are terrible wrong.

I mean, this is the dream of pretty much all technologists.

Humans cannot achieve homo economicus status on our own, we need a machine oracle to give us directions.

Reply on news.ycombinator.com