Hacker News Reader

> You get a range for your calibration score. Whether your manager tells you the specific number is up to him, and you have no way of knowing whether he's telling the truth.

No, the calibration score can be displayed in the internal performance tool based on a manager's preference. And if it's not and you believe your manager is lying to you about your specific score, I'm sure he'd be happy to get you to STFU (hard, I know) by showing it to you directly. Regardless, as long as one is "Meets Expectations" or higher, they will have no issue transferring teams after their 1.5 years of service to their current team. Your problem was that you were both new to the team and were performing at the level below "Meets Exceptations".

> If you go to HR, you get fired ...

More bullshit about Google from someone who has no clue. I (and plenty of people I've worked with, including those that have successfully gone through a PIP) have had very pleasant experiences with HR and their boss^2.

> At least you admit to that problem. Yes, that is correct.

I don't believe requiring someone to stay on a team for 1.5 years before transferring is a problem. The honest truth is that no team wanted you, because if they did, there would have been an exception made.

We get it, you had a horrible time at Google, performed incredibly poorly, and continue to blame others/the system for your issues. Even worse, your attempt to garner sympathy and support from other Googlers didn't work out as your illusions of grandeur would have suggested. Now you continue to jump at any opportunity to smear Google as if getting agreement from people who've never worked there will somehow make it all better.

Dear Cowardly Anonymous FAL-Dude,

Name yourself or stop with the ad hominem attacks. Thanks.

You know nothing about my situation, nor why I left Google. Please stop speculating. Even I never found out what calibration score I got. My manager promised a 3.4, gave something lower (his boss confirmed) and it was being appealed at the time I left. Call it -6.2 for all I care.

I was in the process of internal transfer and it was going alright; however, I got an offer from another company and left. I figured that a clean start would be better than an HR file that, while it wouldn't affect me in the short term, would have long-term implications come promo time and that it'd take months of grueling, bureaucratic effort (some of dubious ethical character) for me to fix.

It was like this: Option A is that I transfer, stay at Google, but may have an early-term smear of my Perf history and might have to cut bribes to people with HR database access in a few years when I rise to a level where even early, cosmetic, irrelevant stuff counts. Option B is that I get another job. Option A is unethical and unlikely to work. Option B is much easier. So I chose B.

Some time after I left, Google did something about a few horrible managers on the issues that I'd raised, but I don't know how much progress they've made and it makes no damn difference to me at this point.

> It was like this: Option A is that I transfer, stay at Google, but may have an early-term smear of my Perf history and might have to cut bribes to people with HR database access in a few years when I rise to a level where even early, cosmetic, irrelevant stuff counts

Sorry but that's complete and utter crap.

If you had transferred to another group and consistently performed well no promo committee would have cared about what happened in your first few months. If anything they would have looked at it and thought something along the lines of "well he had a rough first few months, but since then he's been kicking butt. Either his first team was a bad fit, or he learned and grew since then."

Promo committees (and hiring committees for that matter) have to sort through conflicting feedback all of the time. When you're on a promo/hiring committee you are constantly "reading through" the feedback to look for patterns and trends, and trying to build a broader context to help interpret any outliers. A bad patch early on wouldn't have mattered, as long as your trajectory afterwards was positive.

Many (very) senior developers offered to sit down and talk with you privately. I'm sure if you had taken them up on their offers they would have said the same thing. There is nothing that happened in your months here that couldn't have been fixed by transferring to a group that was a better fit and focusing more on producing great software than on posting to the internal mailing lists.

BTW, piaw has several blog articles about problems with performance review committees: http://piaw.blogspot.ca/2010/04/promotion-systems.html

"stop with the ad hominem attacks"

Michael, you're complaining about rude attacks the day after you wrote "Choke on a fucking taint, Google. Choke. On. A. Taint" at https://news.ycombinator.com/item?id=5508058 ?

Rude != ad hominem. Different categories altogether.

Ad hominem attacks usually involve asymmetry of public identity and often place a target between two undesirable options: (a) letting the attacks stand, or (b) further exposing information that's irrelevant to the discussion and possibly compromising, especially with the persistence of the Internet.

I am fine with rudeness. In fact, in many circumstances, I encourage it.

You understand that pretty much any Googler who wishes to defend their employer's practices against you is placed in the same position? I don't like the personal attacks that are made against you either, but you have to realize that how you see your experience at Google is different from how many people who were there at the time see your experience at Google, and that every time you make some assertion about how things work that, well, isn't exactly how we would've analyzed the situation, we're forced to choose between (a) letting the attacks stand or (b) airing dirty laundry.

Maybe it isn't so clear, but I thought it's been clear that I acknowledge the wide variety of Google experiences. Some people have good ones, and that's great.

If I said, "Google is a hellhole and you'll hate your life", then I'd be a liar because some people get great managers and the best projects and Google is really awesome for them.

What I do intend to say is that Google's HR infrastructure and upper management (at least on the people-management front) are irresponsibly bad at their jobs. That usually won't affect you if you have a string of good managers. If you land in the wrong spot, though, look out because there won't be any competent to fix the problem.

Google has a lot of ways it could become a great company again (open allocation) but for now, I'm going to come out and continue to say that their upper management (again, at least on people management) is disastrously bad and they are singularly responsible for ruining what used to be and still could be a great company.

I'm sorry, but I'm not going to let these assholes destroy billions of dollars in value and get away with it.

> I am fine with rudeness. In fact, in many circumstances, I encourage it.

It'd be really nice if HN was not one of those circumstances.

I don't support rudeness for its own sake, but there are some circumstances that call for impoliteness and abrasive questioning.

Seriously, you make good counterpoints, but the adhoms are not called for. Not cool.

Dear Googler123. Telling someone STFU is not nice, in any universe, not even HN. Also, it is quite inappropriate, and may be illegal as well as unethical, to discuss a person's performance while at a company.

The real culprit is Google, no matter what. See https://news.ycombinator.com/item?id=5296146 for my take on management's responsibility for what happens in a company.

Does Michael have a beef? Yes he does. Does Michael have a right to express himself, give his opinion publicly? Yes he does. If Google is the subject of his smears, Google only has Google to blame.

He is not saying Michael to STFU, he is telling him that if he had accused his manager of lying in the face, he would have been shown what the scores were just so that he would 'STFU'.

He's not? It sounded a lot like he was telling him exactly that.

> And if it's not and you believe your manager is lying to you about your specific score, I'm sure he'd be happy to get you to STFU (hard, I know) by showing it to you directly.

It seems entirely unambiguous that he's saying "You can get your manager to show you your scores by making enough of a fuss that your manager will do it just to make you STFU".

Read the whole post as a unit, and realize he's telling him to STFU. He's also hiding behind this man's manager. He assumes the manager would also tell him to STFU, when in fact the manager would probably not have done that. In fact, the manager didn't do that.

The manager may have wanted to do that, although that is purely conjectures, but the manager didn't, perhaps because it's against company policy, but also perhaps because his manager didn't want to go all out on Michael, perhaps realizing that the faults in the matter might not have all been on Michael's side.

So, then, I think Googler123 is telling Michael to STFU not by telling him directly, but by putting words in his manager's mouth, which is I think even worse, because not only is this denigrating to Michael, but also to his former manager, who may still be a Google employee.

Makes sense?

Yes, that's literally correct. But I read it as him also implicitly telling Michael to STFU. Interpretations can differ, but that's part of the point: one can be unambiguous by not writing 'STFU' anywhere in the message and since not being unambiguous is a choice....

Telling Google to "Choke on a fucking taint, Google. Choke. On. A. Taint." [1] is also not nice, in any universe, not even HN. Not that two wrongs make a right, but I don't agree with your read on the "STFU" line. I think Googler123 meant it conversationally and casually, not like a not-nice direct challenge to Michael like you read it.

Your argument about management is flawed: a company is made up of people, who ALL have to act the way they want employees to act, and expect the same out of each other. Michael absolutely has a right to express himself, and since he was a Google employee, he had the duty at that time to try to make it the best place in the world to work. That he didn't even confront his manager about his real number doesn't speak well to his willingness to try to improve the place. Especially since he had a safety net job offer. What was his risk?

[1] https://news.ycombinator.com/item?id=5508058

  > Telling Google to "Choke on a fucking taint, Google. Choke. On. A. Taint." [1] is also not nice, in any universe, not even HN.

Maybe this is just me, but I think hyperbolically telling a for profit corporation to stuff it (e.g. screw AT&T), is far more socially (normatively) acceptable than making a personal attack against an individual.

Neither are _nice_, clearly.

It's not just a for profit corporation, it's his former employer.

So? Why does that detail matter?

In my first digs at Google I specifically avoided exposing any relationship (it was exposed by an HN commenter, to deflect). All I said was that there was mismanagement at a level approaching the criminal (which is true) and I gave a couple details (accidentally leaking calibration scores-- I thought that had been common knowledge-- and the death of 20% time).

Neither of these are confidential. How a company treats its employees is (with lots of legal precedent) covered under whistleblower protections-- not a trade secret. Therefore, the existence of any employment relationship is irrelevant.

If I leaked a product, I'd be unethical and deserve to be sued. Exposing the existence of calibration scores is fair game.

You're completely ignoring the context of our conversation:

"Choke on a fucking taint, Google. Choke. On. A. Taint."

It is not okay to say that about your former employer. In any context.

If you think it is okay to say that about your former employer, there's something wrong with you. I certainly would never want to work with you.

Google never treated me with any loyalty. What do I owe them? Nothing, as far as I can see it, as long as I don't do anything unethical. (Exposing bad practices and using rude language aren't unethical.)

Still, you're right in a way. Google is 55,000 people and I have no problem with 54.9+ thousand of them. In fact, that company often makes great products because, in spite of the executive malignancy, they have some top-notch engineering talent under their roof.

I shouldn't have said, "Choke on a taint, Google". I should have said, "Choke on a taint; parasitic, psychopathic, incompetent and entitled executive caste that implemented closed allocation at a once-great company, developed a mean-spirited, Kafkaesque 'calibration score' system worse than Enron's, and has destroyed billions of dollars of value."

Sorry, but the people who played a direct role in the managerial destruction (e.g. calibration scores) of Google are human garbage. They took what used to be a great company and destroyed it, and for no good reason. They deserve much more than I have thrown at them.

It's not because of LOYALTY that you're supposed to avoid telling your former employer to not "choke on a taint."

It's because it makes YOU look bad.

Can I ask a question? You keep complaining about closed allocation. What EXACTLY would you want them to have, instead? And I don't mean just say "open," say specifically what you mean, how you think it would work in practice. At a company with 55,000 employees.

Also, you're putting way too much emphasis on calibration scores. Companies record what managers think of their employees. How ELSE would you want a company to do it?

It's because it makes YOU look bad.

After Google, I worked for a company with the most unethical management I've ever encountered. True evil. (Calibration scores are just criminal stupidity; not the same class.) Also, that company is notorious in NYC for having unethical management. Furthermore, I was there for 3 months... after a 6-month period at Google. Now that is a blotch on my CV.

I'm done worrying about cosmetic shit. I already fail on that front. Too much substance, and it's too visible by now.

I also don't buy into the "never make an employer look bad". If you're there for 2+ years, then I agree and would say that the rule applies, even if you were laid off or had a crappy manager at the end, because there's nothing embarrassing about a 2-year stint. If you are at a company for less than 12 months, either you or they like horrible. Make it them by airing enough to prove them the bad guys. (Then you still look bad, but not horrible, because you show it to be their fault.)

Can I ask a question? You keep complaining about closed allocation. What EXACTLY would you want them to have, instead? And I don't mean just say "open," say specifically what you mean, how you think it would work in practice. At a company with 55,000 employees.

Actually, being large makes it much easier to afford open allocation. If you're small, you often have real technical deadlines that put the business at risk, so full-on open allocation isn't always possible.

If you're at Google's size, your moral responsibility to your engineers is open allocation, because you can afford it. It's the 20-person companies that have deadlines and clients that have good reasons not to do full open allocation.

So, that's easy, at least for the engineering organization. That's where you start, and you branch out from there as you build a knowledge base.

Companies record what managers think of their employees. How ELSE would you want a company to do it?

Well... for starters, take that garbage out of the transfer packet.

Does this seem radical? Actually, the law is on my side, here. Until about 1995, most companies didn't have performance reviews in the transfer packet. At all. Those were separate and only came out when someone actually needed to be fired. Enron changed that. It was Enron that made stylish the "innovation" of making performance reviews part of the transfer packet, and Google is carrying the torch.

A manager who interferes with internal mobility is guilty of harassment (plenty of legal precedent here) because internal mobility is considered part of a worker's job performance, and interference with job performance is...? Well, it's one of the most common subcategories of workplace harassment. Ergo, a manager who communicates anything negative about an employee (except a breach of law or professional ethics) is guilty of harassment and can be charged.

That's why the calibration scores and manager-level-only feedback are secret. It's to allow Google (or, more specifically, managers) to break the fucking law.

Yes, companies need to fire severe underperformers, unethical people, and law-breakers. That's true. Performance review for mutual benefit also has some value. Creating a system where performance reviews are part of the transfer process is, while not technically outlawed, essentially giving managers the right to break the law.

When I say that closed-allocation management in the Google/Enron-style company is often extortion, I'm not exaggerating. It's technically civil rather than criminal extortion, but now we're getting into details.

"Make it them by airing enough to prove them the bad guys."

I don't believe I know anyone who would see it your way. This is like complaining about your ex-girlfriends on match.com, it's self-defeating in a way that just makes me feel bad for you that you think it will be effective and help you achieve your own goals. It won't. And I honestly think that if you found someone who it DID convince, that they're not a good person to work for, and that will become apparent in time.

Back to my question - You talk a lot for not actually answering my question:

How do you think open allocation would work in practice. Lay it out for me like I'm 5 years old. It would also be to your benefit, if you could point to companies, of roughly equal size, that do it the way you want Google to.

Everyone in a company evaluates actual worker performance. Yes, it's possible to game that system maliciously, and that would probably be harassment. But it's perfectly reasonable to communicate performance evaluations when considering any promotion, increased responsibility, or transfer to another team.

That he didn't even confront his manager about his real number doesn't speak well to his willingness to try to improve the place.

I don't know where people get this idea. There's a lot of story and I haven't told most of it.

My boss promised a 3.4, then gave me something lower (probably 3.0-3.1). When I confronted him, he said it was because Google+ out in California wanted me to suffer for criticizing a product (that later failed, as I predicted). So I asked him to put a note on the HR file noting that the ding was political and not related to performance. I didn't care what the score was (the difference in bonus between 3.0 and 3.4 is minor) but I wanted the record to show what had happened, and an agreement on 3.4+ for the next 4 quarters to compensate for throwing me under the bus. That's all I wanted, and it's a small, reasonable, request. Everyone knows that "performance" reviews are about politics anyway, so I have no qualms about using one as a negotiation token. It's a game, so play it.

In response to my request, he went and revised my calibration score downward. He also made speculations about health problems (I have very-high-functioning hypergraphia, which is often a symptom of something else but in my case it's just standalone hypergraphia) that were irrelevant to work performance-- just to label me. I was in the process of deciding whether (a) to get attorneys on the case-- most attorneys don't want to appeal performance reviews because there's no money in it-- (b) to offer a side bribe to someone with performance DB access, or (c) to just get another job. The winner was (c); (a) and (b) meant $10k+ for something that might not work and would only pay itself off if I were there for 5+ years.

Also, HR investigated and found out there was no ding from California. Completely made up.

So, I dug around and found out that a mysterious departure of someone (a serious CS heavyweight) who had this same manager was a case of outright bullying that had gone on between approximately Oct. 2010 and July 2011. More digging found a pattern; this particular manager had a pattern of lying. In one case, he was actually reported to HR for repeatedly (and probably intentionally, though he denied it) using last-minute 1:1 reschedulings to conflict with someone's therapy. That was before I came to Google. Predictably, HR did nothing that time, too.

This was a manager who had a years-long history of bad behavior, especially toward people with (otherwise manageable) health problems. It wasn't even hard for me to piece together the story, because so many people knew that it was going on. If he didn't see you as vulnerable, he was very affable and supportive, but if he smelled blood, he attacked. He actually admitted, in one proceeding, that he enjoyed "testing" an employee with panic disorder to find triggers. In my case, I'm fairly normal but I do have hypergraphia and can tell a decade-long story (with increasingly high levels of function; from 2002-09 I had a serious trolling problem, now I focus on writing coherent and useful stuff) about that.

I don't blow whistles over small shit. This was a big fucking rat.

When Google stops allowing Evil, I will stop attacking it.

"I don't know where people get this idea."

They get the idea that you didn't confront your manager about your real number because:

1) You've repeatedly told us that you don't know what your real number is.

2) Googler123 has made the declarative statement that your manager could have easily shown you the real number in the internal perf tool.

3) You haven't contradicted Googler123 that it's not true.

4) You haven't told us, "I confronted my manager about showing me my real number in the internal perf tool."

From that, we're quite correct in concluding you didn't confront your manager about showing you your real number in the internal perf tool.

I'm sorry you had a lousy manager.

If there's something wrong with your manager, you're supposed to go to HR and you're supposed to go to your boss^2. In any company. It actually sounds like that's what you did, so why are you telling us all that the outcome is guaranteed to be "fired" or "similarly fucked", "In any company." ???

No, in the company I want to work at, if I have a lousy manager, I'm going to talk to HR and my boss^2. That's what I expect from myself, and that's what I expect from you. It even sounds like you did it - so why are you telling us not to?

Google is a company made up of people. Some of them have even bothered to respond to you in this thread. And yet, you're telling them NOT to do what people SHOULD do.

How exactly is ANY company supposed to get better over time, if the employees aren't brave and do what's right, even if it might have consequences for them?

Or should they just quit, and then take every opportunity to smear the company on websites?

It sounds more like you had a bad manager rather than Google is a bad company.

Any one with work experience and common sense can tell that talking to boss^2 and HR is a recipe for disaster. You are also unwilling to challenge michael's claim that he was meeting expectations and in the midst of a transfer. You also think restricting transfers for 1.5 years is not a problem. Is there any problem within Google that you are willing to admit?

You sound like an HR drone tasked with attacking michael's record with false data. You haven't responded to his claim that his record met expectations. Unless you use a real identity, your claims will have no credibility.

I have work experience and common sense, and I've had success talking to boss^2 and HR.

You get out of life what you put into it. Also, at some point, you have to be willing to take the risk that things won't go well. What's the worst thing they can do? Fire you? Michael was already thinking about leaving anyway.

At some level, you just have to ACT like the company you work for is the company you WANT to work for - and if they don't see it the same way, no hard feelings.

If Google was the company he wanted to work for, and he and his manager didn't get along, what SHOULD he do? Stand up for yourself. Don't quit for the sake of your "permanent record." If there's more to it, then that's fine - but if that's your reason for quitting...?

It's a company, and you have a job. If your manager sucks, say something about it. If that doesn't work, vote with your feet.

Talk to boss^2 and HR.

Is anyone saying that "Talk to boss^2 and HR" will never work, or just that it's very unlikely to work?

I too have work experience, but perhaps not common sense when it comes to dealing with people, and I never got anything accomplished by talking to boss^2, who as Church noted was more invested in their choice of hiring your boss than in you the underling, and indeed then voted with my feet.

Yes, I think Michael is saying it will never work.

Michael: "If you go to HR, you get fired." "If you go to your boss^2, you are similarly fucked..."

I think that's terrible advice.

I'm sorry you had to vote with your feet, but I'm glad that you did. The only way boss^2 or HR have a chance to learn is if employees do what they're supposed to, and Michael is actively telling people to not do what they're supposed to.

    *it's almost impossible to transfer teams before you've spent at least 1.5 years on that team.*

    At least you admit to that problem. Yes, that is correct.

Actually, I've seen two people transfer teams in their first couple of months because they were unhappy with their job, and it was not a big deal at all (find potential new job, talk to new team, talk to manager, notify HR, done).

I think the 1.5 years is the time you're expected to stay on a project because below that you might end up being more effort than gain to the team, but as long as you don't make a habit out of job hopping across the company, it's not an issue.

I've seen plenty of instances of the same thing - new hires where it was clear he team wasn't a good fit. They had an adult conversation with the current manager and did their best to be productive until they found a team that was a better fit and transferred.

No drama and no hard feelings.

Google depends a lot on your manager. If you have a good manager, then it's still the Old Google and the company is still open to you. With a good manager, you can transfer as you wish, and as long as you don't have a complete don't-give-a-shit attitude toward your assigned project, you're OK no matter how things play.

I'll be the first to say that there are good managers at Google. However, the major reason why people work for large companies is to be protected from the event of a bad manager. Startups fire same-day. Large companies offer transfer for the no-fault "fit" cases. Google does not extend such protection, which means it lacks the one upside of a large company.

That's why I always tell people that if they get a chance to work for Google, they should give it a shot. Nonetheless, their HR infrastructure and executives have failed to do their jobs; that much is clear. The matter of whether that will affect a specific individual is unpredictable.

That's what they told me it would be like when I interviewed at Google, but that's not what actually happened. I believe that it does work that way for some people, but the system is not transparent; it did not work for me and I don't believe that my manager or Google HR were ever being completely straight with me about what was going on.

Perhaps there is some narrow, undocumented window during which you can do this kind of transfer, and I unknowingly passed it? It did take me about three months to understand what was going on well enough to realize just how much I hated the project I'd been assigned, another month to figure out where I wanted to go and to talk things over with the team I wanted to join, and another month for them to deal with some headcount freeze and some apparently-brand-new interdepartmental transfer process (!?). By that time the manager who'd hired me had moved on to something else, and the team was run by some guy in another city who was clearly overextended. When I told him I wanted out, there was no "sorry the starter project didn't work out, enjoy your new team"; instead it was "what the hell do you think you're doing" and a PEP, blocking the transfer.

If there was a PIP involved, you were likely not meeting expectations, and either the new team didn't want the liability, or the transfer was killed by HR to prevent under performers from avoiding a PEP by continually transferring to a new team. It's OK to hate your project but it's not OK to underperform.

The problem is the measure of "performance". Some companies including Google do it via a central committee, but that has it's own problems.

If there was a PIP involved, you were likely not meeting expectations

Or, his boss was resentful of his desire to transfer and, because each manager has a limited number of "review points" (that's how Enron/Google-style performance review systems work; the forced curve means that a manager's generosity in reviews is limited by his credibility) he was thrown under the bus. That's much more likely.

This is called the Welch Effect. "Underperformers" tend to be average-performing, junior members of underperforming teams (who had the least to do with the team's underperformance). The reason is that that the manager (who is also desperate, but in slow motion) doesn't have many "calibration" points to give out and tends to focus everything he has on senior people who are likely to stay because, without superior review scores, they'll flee.

Credible, less desperate, managers look out for all their reports. The ones who are on the bubble tend to scapegoat one every year, and spend all their review points on the senior people who hold the project together in spite of mismanagement.

the transfer was killed by HR to prevent under performers from avoiding a PEP by continually transferring to a new team.

Then the people in HR who blocked the transfer should be fired in disgrace because they're mean-spirited pieces of garbage and it's irresponsible to let them make decisions that affect other people.

It's OK to hate your project but it's not OK to underperform.

Do you know how disgustingly self-righteous you sound? You know nothing about the guy and you're telling him that he's "not OK".

If you actually think more than 10% of people labelled as "low performers" in any company are real problem employees, then you're either (a) under 23, or (b) fucking stupid. Management uses "low performer" labels to brush away its own failures and messes. Most problem employees and true low performers never get caught, and that's why companies tend toward mediocrity and failure over time.

You spin a good yarn, and maybe your theories do apply to someone, somewhere (and maybe to yourself, possibly only in your own head, as your ego attempts to reconcile congnitive dissonance), but in my own observations at Google, I have not found this alternative reality to match reality, ever, so I must conclude your explanation highly improbable. I think Occam's Razor would agree with me.

As an aside, I have found a large number of underperformers in every organization past a certain size. The reason is that hiring is inherently hard, and once employees have exhausted their immediate social circles, it is extremely error prone.

I've found that effectively weeding out non performers, who are inevitably hired, is one hallmark of a healthy organization. As mostly hire As but B players mostly hire Cs, and Cs put you out of business. This goes double for managers (who, incidentally, are continuously rated by their reports at Google).

I agree with you insofar as companies ultimately have to get rid of people who've ceased to contribute (or never were).

The problem is that "low performer" witch hunts rarely do anything about the actual problem employees. If you don't know that, you either have worked for less than 2 months in total, or you're completely fucking clueless. The true problem employees, who have a lifetime worth of experience at hiding their toxicity, stay. It's people who get unlucky (who don't have a whole career of underperformance to inform their strategies) that are swept out.

I'd rather companies just have an honest layoff, the way Wall Street does it. Layoffs have the Welch Effect, but that's not as bad as a system that actively selects for toxicity.

See, here's the impact distribution, where 1.0 is the average employee contribution:

    'A' players :  1.5 to inf
    'B+' players:  0.7 to 1.49
    'B' players :  0.3 to 0.69
    'C' players : -0.2 to 0.29
    'D' players : -2.5 to -0.2
    'F' players : -inf to -2.5

Okay, so the As and B+'s are usually fine (unless they are pre-emptively attacked by competing co-workers). B's are vulnerable and beholden to their bosses when a witch-hunt starts, but there's no good reason to fire them. C's need to shape up; those are the ones you need to improve or get rid of. D's and F's should probably be fired yesterday. It's too costly to keep them around.

It tends to be the B and C players (who could turn into A's, with better projects) who get smacked in a low-performer witch hunt. D's usually survive, and F's become managers in tough-culture systems because they tend to be the ones with severely toxic personalities who love the power.

Interesting--can you elaborate as to how a Google engineer could subvert peer review and the revision control history?

Official policy (when I was there) is 18 month minimum before you can transfer. There can be exceptions, but they are exceptions.

About 3 months in, I talked to my manager about transferring off a project I disliked. His response a month later, which I believe came from the site director, was literally "We don't care."

Which is fine, I just think it's important to know that Google is not generally flexible about initial allocation, and you will generally be expected to stay on your first project for 1.5 years--regardless of how you feel about it. And you may not know what your first project will be until after you're hired, or even after orientation.

I'm sympathetic, although your concern about transferring teams in a short timeframe might be misconstrued. I was hired to lead a certain project. If I tried to transfer to a different project shortly after being hired, that would be considered rude.

When I hire someone, I hire them to work on my project. If they immediately transfer, or get sniped by another project, that's a little rude. As a result, the general tone is that you "pay your dues" on the project you are hired on before you start thinking about transferring. I think a year and a half to two years is a good amount of time.

We're a cooperative, non-competitive development environment, so I can imagine that what is a "little rude" in our environment could be interpreted as highly aggressive in a more competitive environment like Google.

When I hire someone, I hire them to work on my project.

Google is different, because employees are hired on a whole-company basis, which I think is the right policy. It means there's a uniform hiring bar, instead of the typical big-company dynamic where hiring standards are relaxed to fill less desirable projects.

I agree with you that, in the typical company where one's immediate manager had to fight to make a slot/get budget, leaving at < 12 months is (with some exceptions) a bad behavior. I don't like the concept of "paying dues" but I think that returning the favor is appropriate.

I think you have to be careful with the whole-company hire business.

It's important to note that not everyone will be able to do every job well, and some people will not be able to do some jobs at all. So while it is important that their ethics, personal drive to learn, and cooperation skills be top-notch, one has to be careful not to try to hire expert generalists, because one ends up with people who are neither true experts nor true generalists.

I think in Google's case, and I speak from an outsider's point of view, that Google tends to try to hire the experts at the expense of the generalists. Generalists, while less technically adept at any one task as experts (I was thinking of using the term specialist there), many times have more experience with communication, managing changes, and, hum, diplomacy.

Again, from the outside, Google looks like a regiment of special forces: each member individually highly competent, and together, when deployed appropriately, quite capable. But one would not expect them to be able to run an orphanage, or, to wrap up the analogy, have very good customer service.

Although, thinking of it, a regiment of special forces would be very effective at running orphanages, since the great majority of soldiers would feel very compassionate toward the children, and not many things would get in the way of them getting the resources needed to care for the kids properly. Not sure Googlers would do as well.

Generally the hiring pipeline at Google does not involve a hiring manager until after the individual has been accepted. There are exceptions for special roles and people, but for the software engineer case, we require that someone be generally competent across the board. Once they're in, then their specialties and interests factor into what team/project they start on.

By generally competent across the board, you mean adept in a variety of disciplines, but only within the general discipline of Software Engineering. Of course, this covers a lot of ground, from networking to kernel development, to web frameworks, database system (RDBMS and NoSQL), and even old-school assembly...

Or no? Perhaps you look at the fundamentals of algorithms, such as types of sorting, binary data operations, that sort of thing...

Either way, many people will not be able to know enough across the board to make the cut, who in one or two particularly focused areas may be quite talented, and there may be people who, having spent all their waking hours in the deep study of bits and bytes, pass all such tests with flying colors and who, once in the company, poison the well of goodness because of their inability or unwillingness to adapt to rapidly changing surroundings, because of their inability to endure tenaciously under, shall we say, difficult managerial arrangements, and because of their inability to genuinely care about the welfare of their peers, Google customers and clients, and Google investors, both in word and deed?

Perhaps this is not what happens, but, from the outside, as far as I can see (and I remember the pre-google days of Lycos, HotBot and metacrawlers) Google the Company behaves as one would expect a talented yet socially awkward and uncouth twenty-something; brash yet caring, uneven in his affections, vacillating between opposites, such as standing up for freedom and privacy one year, then chucking freedom and privacy to the wind the next for a sizable windfall. This person can charm yet irritate, seem brilliant in one area but somewhat novice in another.

This is how I see Google, and I see very little, standing on the outside as I am, that leads me to believe that there sill be change for the better anytime soon.

A year and a half is an eternity in software development. It makes sense if the developer likes the project, feels good about his peers and his manager and feels technically challenged. If those two or more of those criteria aren't met, it just breeds resentment in everyone. As a manager you shouldn't even want someone on you team that is becoming resentful. It's toxic. If a manager were willing to keep someone on their team that is becoming more resentful and tried to change the situation by changing teams and that manager prevented it, then I'd fire that manager. That's not how you lead a team or company.

Thank you for cutting through the ad hominem crap and to a real point.

I'll do unpleasant work on occasion for a manager-- and more often to keep my own promises and see through projects I care about-- but managers who expect reports to subordinate their careers (rather than just doing a bit of work) and keep their heads down for 18 frickin' months are either entitled jerks or extortionists (depending on whether their attitude is subconscious or conscious).

Does Google have mentorship as an official responsibility? As an engineering led organization, I imagine it should. Ideally a mentor should be someone politically separated from your people manager and technical lead. They would be an advocate for your development as a professional and would be like a good college counselor (Between two majors and switching majors and schools once I had five college counselors and two of them were indispensable.). If you're good enough to make it through the Google hiring process for engineers, then you are good enough to deserve an advocate/counselor that can coach you and represent your position as an outsider to a conflict.

Ideally a new engineer would get to choose their Google Counselor and they wouldn't be allowed to transfer to work for their counselor without finding a new counselor. If they didn't find a counselor within 2-3 months they could choose to attend "counselor speed dating" events where people who enjoy coaching and advising can show up and meet promising new googlers and take them under their wing. People who are really good, knowledgable and wise are often not only happy to mentor/coach others, but actually enjoy going out of their way to do so.

In the case of a conflict between an employee and manager (or tech lead) like you experienced, the counselor would act as an advocate for the employee and suss out the truth like a lawyer helps a defendant in prosecutorial proceedings where there is a clear asymmetry of power between the prosecution and defense (like federal cases); The counselor provides a system of checks and balances against the types of managers you've railed against in the past. Having been victim to such a manager myself several years ago, I know that such a mechanism is sorely needed as a company grows and certain poor managers become entrenched in a defensible position despite their toxicity to the organization.

That's a great idea but, no, Google doesn't have anything like that.

You can find mentors, in the sense of people able to teach you things, and there are a lot of great people at Google, but there's nothing like what you described, and managers have unilateral power and the system seems to be described that way (project expediency, I assume).

I would reckon it's more likely due to lack of trying than project expediency. A C-suite sponsored mentorship directive would be a managerial innovation that few is any companies have tried.

A bad manager like the one you've described previously doesn't get a product done faster, cheaper and better. Quite the opposite actually.

In a previous discussion in a subthread started by you there was a concise reply with the most crystal clear example I can imagine about the stupidity of inflexible closed allocation from "russell" (https://news.ycombinator.com/item?id=5177884):

This describes my son's experience at Google exactly. He is a computer engineer with strong software and chip level design experience. He sought and was offered a hardware design job at Google. When he got there he was given a job writing python scripts to manage hardware, not the same thing at all. When he tried to change jobs, he was told he had to wait 18 months, so he quit.

Wow, that sounds like... Microsoft. What happened to Google?

It didn't happen only to Google.

Most companies start-off half-decent and sell off their culture to hire executives, which usually involves zero-sum autonomy transfers and globally undesirable cultural changes.

Usually, if this is going to happen, it happens early in the startup phase (~50 employees) but Google managed to hold it off for several years-- which is admirable-- but eventually hired some evil execs who did the culture in.

I feel like the hiring of executives is where most companies lose their culture. Bringing in a semi-retired burnout with a sense of entitlement, and giving him all the keys, turns out to be a bad move.

Nothing happened to Google.

Michael had a bad time, for reasons that most people believe were due to his own impatience, and now he likes to talk about Google as if his situation is the norm. It's not.

No one minds if Michael talks about his own bad experiences at Google. People just get upset when he pretends his experience is representative of Google as a whole.

My experiences may not be representative. My observations are. I've met countless people who've been screwed over by its anachronistic closed-allocation regime.

Some people have good experiences, some don't, and for 95+ percent of the "don't" category, it's because Google screws the pooch when it comes to people management.

I've never met a Googler who (a) got smacked by its antiquated HR system, and (b) actually deserved it. There probably is one or two out there-- it's a Poisson distribution with parameter around 2.5-- but I've never met one. Most of Google's problem employees (I've been personally attacked by a few) seem to do just fine.