Google's Answer to Filling Jobs Is an Algorithm

An anonymous reader tipped us to a New York Times article about Google's newest HR tool: an algorithm. Starting soon, the company (which gets roughly 100,000 applications a month) will require all interested applicants to fill out an in-depth survey. They'll be using a sophisticated algorithm to work through the submitted surveys, matching applicants with positions. The company has apparently doubled in size in each of the last three years. Even though it's already 10,000 employees strong Laszlo Bock, Google's vice president for people operations, sees no reason the company won't reach 20,000 by the end of the year. This will mean hiring something like 200 people a week, every week, all year. From the article: "Even as Google tries to hire more people faster, it wants to make sure that its employees will fit into its freewheeling culture. The company boasts that only 4 percent of its work force leaves each year, less than other Silicon Valley companies. And it works hard to retain people, with copious free food, time to work on personal projects and other goodies. Stock options and grants certainly encourage employees to stay long enough to take advantage of the company's surging share price. Google's hiring approach is backed by academic research showing that quantitative information on a person's background -- called 'biodata' among testing experts -- is indeed a valid way to look for good workers."

Comments

Bias

[Threni]

It will be interesting to see if any company using this technique ever get accused of racial,sexual etc bias.

"But the computer chose them! You're not going to sue my computer, are you?"

Re:Bias

[Shmooze]

It depends how good their algorithm is - let's say it looks at what proportion of your life since graduating you've been in work, where more is better. That's a disadvantage to women because they (generally) take time off to have/raise kids and so on, even though the algorithm isn't specifically designed to discriminate against them.

(OK, so it's a trivial case, but you get the general idea)

I suspect there could be plenty of arguments in court about whether some nuance of the algorithm treats some group unfairly or not...

Re:Bias

[AuMatar]

Simple- don't have a question that asks about race, sex, etc. If the algorithm doesn't know it, it can't choose based on it.

Re:Bias

[FoXDie]

if (race.black() || gender.female()) {return 0;}

Re:Bias

[__aaclcg7560]

I'm more interested in how the computer responds to an applicant turning down a job at Google. I been turning down Google recruiters for the past year since they're not offering enough money or a good enough benefit packager for me to move down the street from my current job to work for them. If a recruiters are stunned by my answer, I bit the computer will just flat out seg fault.

Re:Bias

[inviolet]

It will be interesting to see if any company using this technique ever get accused of racial,sexual etc bias.

What if google's statistical data (drawn from its database of performance reviews) shows that some ages, genders, races, and cultures are objectively better at a particular job than others?

Google's test will obviously avoid asking any direct questions about age, gender, and race, because that's illegal (even when objectively justifiable). However, if the test is powered by a statistics engine drawing a database of past performance reviews, then the test could unintentionally evolve to ask about such things indirectly.

An example: perhaps cat-ownership is correllated with femaleness, and femaleness is correllated with superior performance in writing technical documentation. An automated test-generator would unwittingly evolve to ask applicants about cat-ownership, in order to unwittingly select superior female candidates.

It's an amusing possibility. Indeed, it would be the free-market's way of legitimately selecting candidates based on age/gender/race while remaining underneath the legal radar.

Re:Bias

[profplump]

Except "people with less work experience" is not a protected group, so it's not unlawful to discriminate on the basis of previous work experience, unless you do so with the intent of discriminating against an actual protected group. I'm just guessing, but I'd say it would be awfully hard to win a case based on such "discrimination", short of someone admitting that they did it to avoid hiring women.

Re:Bias

[general_re]

Except "people with less work experience" is not a protected group, so it's not unlawful to discriminate on the basis of previous work experience, unless you do so with the intent of discriminating against an actual protected group. I'm just guessing, but I'd say it would be awfully hard to win a case based on such "discrimination", short of someone admitting that they did it to avoid hiring women.

No. See Griggs v. Duke Power Co., 401 U.S. 424, 431-2 (1971). A plaintiff can show that some employment criterion or criteria results in a disparate impact upon a protected group, regardless of whether discrimination is overtly intended or not. The burden of proof then shifts to the employer to show that said criteria are a necessary requirement for the job(s) in question. If they can't, they lose. Even if they can, if the plaintifss can come up with an alternate business practice that satisfies the employer's interests without resulting in a disparate impact, they lose. Good or bad, that's the law.

Re:Bias

[Dunbal]

if(race == black && gender == male){
      assign->job(employee, DRIVER);
      return;
}
if(race == black && gender == female){
      assign->job(employee, CLERK);
      return;
}
if(race == white && gender == male){
      assgin->job(employee, EXECUTIVE)
      return;
}
if(race == white && gender == female){
      if(employee->cute) assign->job(employee,EXECUTIVE_SECRETARY)
      else assign->job(employee,TECHSUPPORT);
      return;
}
if(race == hispanic && gender == female){
      assign->job(employee, JANITOR);
      return;
}
if(race == hispanic && gender == male){
      assign->job(employee, MAINTENANCE);
      return;
}
return 0;

Re:Bias

[Threni]

> Except "people with less work experience" is not a protected group

In the UK, after recent tightening of anti-agist discrimination, you need to make sure that you aren't going to get into trouble for asking for people with over n years experience, or similar.

Re:Bias

[Dunbal]

If the algorithm doesn't know it, it can't choose based on it.

      Unless of course, as someone pointed out earlier, you have access to statistical and demographic data that lets your algorithm figure out religion, race, sex, etc indirectly from the answers with an acceptable margin of error, say +/- 3%?

Re:Bias

[profplump]

As I read it, Griggs v. Duke Power applies more specifically to selection requirements rather than ranking, but I guess I could see it made into an argument about the later if the proper context was presented.

Still, "years of related work experience" is pretty easy to put into the "reasonable measure of job performance" bucket, and given that, the requirement of intent to discriminate against a protected group stands.

Re:Bias

[profplump]

That's true in the US too. You can't ask people when they graduated or other such things (no matter how useful or relevant they are) for fear that you'll be sued for age-based descrimination.

Re:Bias

[jherekc]

Judging by your comment, I'd say that Google don't want to employ your for your linguistic skills...

Re:Bias

[__aaclcg7560]

Linguistic skills? I been telling those Japanese recruiters that while I worked at several Japanese companies and understand the culture, I don't speak or translate Japanese. I can proofread an English translation of Japanese text. A big difference! :P

Re:Bias

[general_re]

Congress expanded the doctrine in 1991 - I forget the title, but you can probably look it up to get a better idea of how it applies to hiring.

Re:Bias

[dagamer34]

You're program does nothing for Asians and Indians. Way to leave out 1/3 of the human population!

Re:Bias

[abigor]

Hint: "I been telling..." --> "I h___ been telling..."

Re:Bias

[Lord_Dweomer]

Or the flip-side...could these statistical preferences be uncovered and gamed? Seems to me there is an army of SEO experts out there willing to try for a buck.

Re:Bias

[gregraven]

Google would be miles ahead of everyone else if it looked only at IQ. Alas, this has been deemed racist.

Re:Bias

[__aaclcg7560]

Try explaining that to the Japanese! ;)

Re:Bias

[yali]

As you might guess, there is a whole tangle of legal and ethical issues surrounding testing in personnel selection.

My understanding (IANAL etc) is that you are supposed to assess only the skills, aptitude, etc. that you can defend as related to the job. If that happens to be correlated with sex, race, age, etc., the correlation is not a problem, but you cannot use those things as a proxy for what you're really interested in. For example, in a job that requires quick responses, you can test people's reaction times, but you cannot automatically exclude people based on age (even though age may be correlated with reaction time).

More direct assessment is better anyway. Suppose you are hiring for a job that requires math skills, which you believe is correlated with gender, which you believe is correlated with cat ownership. Even if those correlations exist, you'd still get more accurate results measuring the math skills directly rather than measuring cat ownership which is correlated with something that is correlated with what you need.

Re:Bias

[GlassHeart]

What if google's statistical data (drawn from its database of performance reviews) shows that some ages, genders, races, and cultures are objectively better at a particular job than others?

The law does not assume that there's no relationship between job performance and age/race/etc. What the law assumes is that the relationship is not causal. That is, just because you're over 50 you can't do the job, even if most 50-year olds really can't. Therefore, we protect the one 50-year old who could from unfair discrimination.

Re:Bias

[rickb928]

If the survey asked for age, I'd expect that to be prima facie in the court case complaining.

Google isn't incompetent in this. I would be surprised of the survey even asked for age or sex. Let the answers do the work.

Or, put another way, if the survey can't select well by measuring a candidate's attitudes, aptitudes, experience, and personality, why bother with the survey?

ps -- IANAL.

pps -- If I answer the experience questions with '>25 years', have I given them my age unwittingly? Probably the legal defense is to limit experience times to something less and below most age discrimination limits, say 8-10 years... After all, if you're not good at it after 8 years, maybe it's time to learn how to bag fries...

Re:Bias

[Dahamma]

An example: perhaps cat-ownership is correllated with femaleness, and femaleness is correllated with superior performance in writing technical documentation. An automated test-generator would unwittingly evolve to ask applicants about cat-ownership, in order to unwittingly select superior female candidates.

Funny you should use that example, as a TV news story I saw on this topic yesterday mentioned that "Do you have a dog?" was actually one of the questions.

Though a bit scarily, FTA...

  "We wanted to cast a very wide net," Mr. Bock said. "It is not unusual to walk the halls here and bump into dogs. Maybe people who own dogs have some personality trait that is useful."

Or maybe Google doesn't have a higher percentage of dog owners than most companies, but a more tolerant "bring you dog to work" policy?? This sounds like with proper study of the correlations between traits and employees it could be illegal, and without study it could be just plain misleading...

Re:Bias

[Dunbal]

You're program does nothing for Asians and Indians. Way to leave out 1/3 of the human population!

      Yes it does. return 0;

      It's a racist program, that's the whole point.

Re:Bias

[snaptography]

Ha - I'd also agree with you there. It's really unfortunate about how little people pay attention to their grammatical abilities and believe that their communication skills really do not impact their ability to perform well in a job, let alone receive a job offer.

Re:Bias

[zCyl]

What the law assumes is that the relationship is not causal. That is, just because you're over 50 you can't do the job, even if most 50-year olds really can't.

I think what you mean to say, is that the law assumes the relationship is not universal. Even if a causal relationship is shown, the law still protects. For example, it's well known that there's a causal relationship which causes women (in general) to be unable to lift weights as heavy as average men can. But we also know this relationship is not universal. Therefore if a woman is rejected from a job requiring lifting because she is female, then this is illegal, but if a person is rejected from such a job because of failing a required strength test, then this may be legal if done in a non-discriminatory fashion.

Re:Bias

[TheoMurpse]

That's a disadvantage to women because they (generally) take time off to have/raise kids and so on, even though the algorithm isn't specifically designed to discriminate against them.

Any good lawyer would counter this argument with, "If we allow women to take more time off between jobs, then we discriminate against men." This is why men are given paternity leave now.

Re:Bias

[Anonymous Coward]

Out of curiosity: why does the law stop employers discriminating based entirely on statistics, when it does very little to stop insurance companies ripping you a new arsehole based entirely on statistics?

Re:Bias

[katsiris]

Racially speaking, I don't see how that could be possible unless they were asked to input their skin colour. Otherwise, it would, by definition, be treating races equally, wouldn't it? Genders, however... more difficult, especially when you're looking at backgrounds.

Re:Bias

[xant]

But this algorithm doesn't know whether you can do the job--the end result. The algorithm only knows that you said you wear a hat on Sundays. The fact that that means you're probably over 50, and therefore less likely to be suitable for the job, makes this strongly tend toward illegality.

In other words, the algorithm won't find that one guy and hire him, it'll just reject him for having the characteristics (over-50ness) that predict a bad fit for the job.

Re:Bias

[sacrilicious]

It depends how good their algorithm is - let's say it looks at what proportion of your life since graduating you've been in work, where more is better. That's a disadvantage to women because they (generally) take time off to have/raise kids and so on, even though the algorithm isn't specifically designed to discriminate against them.

Damn! Even though I keep my gender a secret, they'll be able to tell I'm a woman because of gaps in my resume when I took time to be with the infants... krap, it was so nice thinking I was flying under the radar. Maybe I need to fill those gaps: "Dec97-Mar99, Lactation Dispensation Consultant".

(ps - I *am* joking. And, I'm not really a woman... I'm really a horse. Well... I'm just pretending to be a horse... actually, I'm a broom.)

Re:Bias

[Anonymous Coward]

More direct assessment is better anyway. Suppose you are hiring for a job that requires math skills, which you believe is correlated with gender, which you believe is correlated with cat ownership. Even if those correlations exist, you'd still get more accurate results measuring the math skills directly rather than measuring cat ownership which is correlated with something that is correlated with what you need.

Fuck you guys. So what if I own a cat???

Re:Bias

[Jonnty]

But what about the fact that many US minority groups, especially Blacks, have, on average, a lower standard of education. This leads to they getting lower qualifications. Therefore, by these grounds, if you discriminate on the basis of qualifications aren't you discriminating against these minorities? Can you be accused of sexism if you require some qualification more men have than women?

Re:Bias

[Anonymous Coward]

What if google's statistical data (drawn from its database of performance reviews) shows that some ages, genders, races, and cultures are objectively better at a particular job than others?

Then they'll definitely have a nice visit with the EEOC, followed by some friendly litigation and settlements. Several years ago, a trucking firm in Chicago was sued by the EEOC for refusing to hire truck drivers who had grand larceny convictions. Since the firm specialized in shipping electronics (computers, electronics for Best Buy, etc.), their insurance company required they screen their drivers to make sure they weren't convicted thieves or else they wouldn't afford their insurance.

But the EEOC looked at statistical data showing that blacks committed higher larceny rates than whites in Chicago, and nailed them for discriminating against blacks. Fortunately, a black judge threw it out and said if truck drivers of any race want jobs, they shouldn't steal.

In the case of Google, their hiring formula is amusing and further proof they're headed for the big surprise. In 1998, the finance world's top quants were running a company called Long-Term Capital Management (LTCM) and believed their math could do no wrong. Everything was wrapped up in their brilliant formulas. It all was magic until standard error and unexpected volatility crept into their model, putting over $1 trillion dollars at risk in the financial markets. It took the New York Fed and the CEOs of most of Wall Street to assume the liabilities of LTCM in order to prevent a likely catastrophic meltdown in global markets.

Google's mistaken belief that it can quantify hiring processes reflects a deeper flaw which should be fun to watch - just as long as one isn't a shareholder or employee.

Re:Bias

[Jahz]

I think that really won't be a problem and Google might not even need to include race/sex/etc information. If they're looking for self-motivated, intelligent, and approachable super geeks with good programming skills, why should the bias information matter? In fact, the ideal system could actually (forcebly) remove some of biases that might be imposed by manual recruitment. Just ignore sensitive but useless data points like gender.

On another note, the official Google employment application does not ask for race or gender information (though a name can usually reveal those) and starts with the ubiquitous statement: "Google is an equal opportunity, affirmative action employer. Qualified applicants are considered without regard to race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability or any other status protected by applicable law." So it really comes down to interviewer bias, in which this system has NO bearing at all.

Re:Bias

[ClosedSource]

Why? Has somebody found a correlation between high IQ and the ability to sell Internet Ads?

Re:Bias

[lindseyp]

It's an amusing possibility. Indeed, it would be the free-market's way of legitimately selecting candidates based on age/gender/race while remaining underneath the legal radar.

Or rather, it may be the free market's way of legitimately selecting candidates that are the best for the job, without having to venture into un-PC selection policies

I can't see how this process would *not* discriminate for bias with race, social status, or gender. Just as insurance companies are allowed to discriminate for those things when setting policy premiums, this process may force society to accept discrimination in the workplace by showing that the factors *do* have legitimate consequences

Whilst I'm no fan of PC anti-discrimination, I'm not sure I like the way this is going

Re:Bias

[hdparm]

Oh, for a second I thought outsourcing is covered by another algorithm.

Re:Bias

[bitt3n]

"But the computer chose them! You're not going to sue my computer, are you?"

first the computer hired only other computers, and I was pissed, because I was not a computer.

Re:Bias

[hobo sapiens]

/me falls out of chair

Re:Bias

[Anonymous Coward]

It depends how good their algorithm is

to a point. i think google will be disappointed as the most manipulative people learn to manipulate their algorithm.

Re:Bias

[hauntingthunder]

problem is there using PR reviews to derive the alogorithm and PR revies are adjusted to fit all the time so the data is false.

Re:Bias

[Anonymous Coward]

No.

Re:Bias

[tehcyder]

In the UK, after recent tightening of anti-agist discrimination, you need to make sure that you aren't going to get into trouble for asking for people with over n years experience, or similar.

Er, wouldn't asking for "over n years experience" only discriminate against young inexperienced people?

Re:Bias

[WilliamSChips]

No, they really wouldn't. IQ stops being a reliable indicator of intelligence once you reach the intelligence level Google needs. Past about 120 IQ tells you more about the test than the person.

Re:Bias

[tehcyder]

it would be the free-market's way of legitimately selecting candidates based on age/gender/race while remaining underneath the legal radar.

Oh, well as long as the fucking imaginary "free market" triumphs over legality and morality that's all right then.

Re:Bias

[VJ42]

Yes, ageism isn't only ageism if it's against old people; being anti-young people is also ageist.

Re:Bias

[inviolet]

Oh, well as long as the fucking imaginary "free market" triumphs over legality and morality that's all right then.

Nice package deal there. :P Obviously it's a triumph over legality, but since when is it immoral for employers to select what they regard as optimal employees?

Do you equate morality with legality?

Do you equate morality with transfer payments, such as (in this case) forcing employers to hire less-than-optimal employees?

Do you select optimal friends for yourself? How is that different?

Re:Bias

[inviolet]

I can't see how this process would *not* discriminate for bias with race, social status, or gender. Just as insurance companies are allowed to discriminate for those things when setting policy premiums, this process may force society to accept discrimination in the workplace by showing that the factors *do* have legitimate consequences.

How is that not a Good Thing?

At the very least, if we all learned the real socioeconomic cost of forcing the hire of suboptimal employees, it would completely change the debate over such laws. I say this because anyone will sign on to a law whose advocates sound good while blustering with an air of moral superiority... but it's another thing entirely to see exactly what a law does to final product prices on a store shelf.

Re:Bias

[Stradivarius]

Except that "proportion of life employed" is not the same as "years of related work experience". The latter has direct bearing on job requirements. The former is more indirect and runs a clear "disparate impact" risk.

Re:Bias

[profplump]

I'd say 1 is a indication of particular skills you have, and the other is an indication of your general employability (i.e. do you show up for work and every day and not get caught stealing). While general employability is often assumed for professional positions it is quite possible the single most important factor to consider when hiring someone for an entry-level position.

Re:Bias

[5of0]

But wouldn't google want people who can manipulate an algorithm to work for them? Until, of course, there are step-by-step instructions on "How to Outwit Google's Application for Dummies" floating about the net. Then they change the algorithm.

Re:Bias

[WilliamSChips]

And then when the computers killed everybody, there was nobody to be pissed for me, because all nonrobots were dead.

Re:Bias

[Anonymous Coward]

> Except "people with less work experience" is not a protected group,

it is in the Uk/EU - unless there is an objective reason - ie it's ok only in certain cases aprentices come to mind.

Interesting one of the UK jobs on googles website probaly falls foul of this  no wonder they are recuiting allmost as many Human Resorces types as engineers

Re:Bias

[Anonymous Coward]

I wasn't aware that unproven measures have a useful range. How would one possibly know?

Re:Bias

[GlassHeart]

Causality means that if a condition is true, then another must be true as well. For example, if we assume that no woman can lift heavy weights, then if you are a woman you cannot lift heavy weights. The cause is your womanhood; the effect is your inability to lift heavy weights.

If you have to append a "(in general)" escape clause to your "causality", then it stops being one. This is because you are hiding the real causality (in this case, "(except for women who trained to lift heavy weights)") there. In fact, the causal relationship is between having trained to lift heavy weights with the actual ability to lift heavy weights.

Therefore, if you could actually establish causality, then you could argue for blanket discrimination. For example, in societies that do not allow men to perform full body searches on women, it is pointless to interview male candidates for the job.

Re:Bias

[Anonymous Coward]

Ideally, employers wouldn't have access to that information. We need stronger privacy laws.

Re:Bias

[zCyl]

Causality means that if a condition is true, then another must be true as well.

That's called "implication". See the wikipedia Causality page. Note the sections "Necessary cause", "Sufficient cause", and most importantly, "Probabilistic causation" which is the default when one discusses a causal relationship like above.

Re:Bias

[GlassHeart]

Since you pointed to the Wiki page, you may then have noticed that "probabilistic causation" falls under "theories," not "logic," and probabilistic causation is far from the default when discussing a matter of law. As for logic, a woman not being able to lift heavy weights is neither "necessarily" nor "sufficiently" caused by her being a woman. The fact that other women can lift that weight is sufficient contradiction of the causality hypothesis.

Even by the relaxed standards of probabilistic causality, you may notice that the Wiki article calls it "notoriously difficult, expressed by the widely accepted statement 'correlation does not imply causation'." The observation that most women can't lift heavy weights is precisely only that: a correlation. "In statistics, it is generally accepted that observational studies (like counting cancer cases among smokers and among non-smokers and then comparing the two) can give hints, but can never establish cause and effect."

Any more questions?

slashdot

[jrwr00]

all they have to do is slashdot the survey.

then they will get a flood of some a million people taking the survey.

i would guess anyone here at slashfot would love to work at google, i know i do

Re:slashdot

[Anonymous Coward]

Did anyone else read google (tagging beta) as gagging (beta)?

I must be going into Google overload.

~anon

Today Google Jobs...

[Flimzy]

Tomorrow Google online dating?

Re:Today Google Jobs...

[panaceaa]

I'm feeling lucky.

That should work out well for CowboyNeal

[StressGuy]

I mean, seems like he's always an option....

Psychological profile included ala The Game?

[LParks]

Do you feel guilty when you masturbate?
Do you enjoy harming animals?

Re:Psychological profile included ala The Game?

[gardyloo]

Do you feel guilty when you masturbate?
Do you enjoy harming animals?

      Huh. I must be really efficient. Killing two birds with one stone, no beating around the bush. Er... as it were!

Re:Psychological profile included ala The Game?

[rrhal]

When you see a blue screen you want to:

a) press cntrl-alt-del
b) smash your monitor to tiny bits
c) insert your Knoppix CD

Re:Psychological profile included ala The Game?

[Slippery Pete]

I hope they don't use information obtained from my searches or my GMail account. I haven't read their EULA lately but they could even dynamically create questions from this information.

Re:Psychological profile included ala The Game?

[Infinityis]

Apparently, this sort of multitasking can be taken to a whole 'nother level... http://bash.org/?715644

Re:Psychological profile included ala The Game?

[xdroop]

Do you feel guilty when you masturbate? Do you enjoy harming animals?

Huh. I must be really efficient. Killing two birds with one stone, no beating around the bush. Er... as it were!

Now I'm no expert, but I don't believe it is called "masturbation" if you are harming animals.

Re:Psychological profile included ala The Game?

[GFree]

Needs another option

d) Other

If other, specify:

I would determine the failing driver and rectify the problem. Windows doesn't BSOD for no reason anymore, despite what the Slashdot kids think.

Re:Psychological profile included ala The Game?

[0xdeadbeef]

The Game? I'm sure Google is more interested in determining if their applicants are replicants.

Re:Psychological profile included ala The Game?

[Anonymous Coward]

What if he were "choking the chicken" or "flogging the dolphin"?

Re:Psychological profile included ala The Game?

[Oktober Sunset]

unless you use a vibrator made by trapping a load of angry bumble bees in an empty cigar case.

Re:Psychological profile included ala The Game?

[Anonymous Coward]

e) Wait a few seconds for OS X to finish starting up

Re:Psychological profile included ala The Game?

[Anonymous Coward]

Don't forget choking chickens, and slapping monkeys.

Re:Psychological profile included ala The Game?

[terrymr]

No of course it doesn't, that would be silly.

Even the blue screen message tells you to reboot and try again and to call support if it blue screens again. Doesn't that tell you that microsoft know it blue screens for no reason.

Re:Psychological profile included ala The Game?

[pimpimpim]

thank god for those accurate wikipedia members

Re:Psychological profile included ala The Game?

[Anonymous Coward]

d) Turn off the screensaver

Re:Psychological profile included ala The Game?

[gardyloo]

Sometimes I take the one-eyed trouser-snake for a helmet-polishing expedition. It's always looking forward to a spelunking expedition, but the cave just isn't available at times. So the self-abuse is more of a mental thing.

Re:Psychological profile included ala The Game?

[Anonymous Coward]

uh oh. YOU FAIL IT!

Re:Psychological profile included ala The Game?

[Anonymous Coward]

Do you feel guilty when you use Windows XP?

I bet Google does:

http://graphics8.nytimes.com/images/2007/01/03/bus iness/03google.xlarge1.jpg

Go away or...

[sootman]

Sounds like someone got one of these shirts for Christmas and took it to heart.

Re:Go away or...

[Anonymous Coward]

Unless a position in HR was replaced by these questionnaires which I seriously doubt (TFA didn't mention that) then the shell script joke does not apply and thus the Tshirt is not funny in this case.

Where is this survey?

[Programmer_Errant]

I don't necessarily want to work for Google but it would be interesting to see the actual survey.

Re:Where is this survey?

[Anonymous Coward]

Google for it? :)

Only useful if...

[Lead Butthead]

Applicant is honest in their response to the survey.

Re:Only useful if...

[armada]

I was thinking the same way but that fact is that if the individual is analitical enough to basically figure out the nuance of what google is "looking for" then it basically makes the attractive anyway.

Re:Only useful if...

[Tackhead]

> Applicant is honest in their response to the survey.

Not necessarily. If you get the perfect score, you get hired into the voigtkampf-beta.google.com programme.

What they don't tell you is that you get hired as an interviewer. The light that burns twice as bright, burns only until it asks about your mother.

Re:Only useful if...

[Shmooze]

Although since your manager usually sees your application (at least, they have wherever I've worked at!) they might notice if you put down something that's untrue, even if it takes a while.

I don't know about Google, but I know of quite a few companies that don't like that kind of thing, although it could make them more attractive if they've tweaked the results slightly, true (since I guess that might be quite hard)

I can't see that it allows candidates to be any less truthful than they normally would be to be honest

Re:Only useful if...

[armada]

They could also have pairs of questions that are the same but worded differently peppered all over and then use the algorhythm to "validate" the honesty of the person filling the survey.

Re:Only useful if...

[LostCluster]

It's fairly easy to detect dishonestly in a fairly long survey. Just ask the same question multiple times with slightly different words use. If dissimilar answers are received on these questions, the person clearly can't keep their story straight, and is likely to be giving untruthful answers.

Re:Only useful if...

[HistoricPrizm]

I would think that if a person is smart enough to be able to ascertain what answers are "the best", they would be able to recognize that questions are being repeated and answer them the same way.

Re:Only useful if...

[mattwarden]

Oh, yes, you're quite right. Before this, no one could lie on their resumes or during their interview.

Re:Only useful if...

[zuiraM]

A lot of secondary questions that could be useful in determining the ideal people for you to work with are going to be useful in determining the disparity between your self-rating and the truth.

Include some personality inventory items (to improve gelling of teams), IQ score items (to assay the analytic intelligence), psychiatric inventory items (to correct other scores and improve gelling), a few practically applicable tests (applied intelligence) and some random questions (to make it harder to predict what constitutes enhancing an answer for the person considering cheating), for instance.

Self-criticism corresponds to the inverse of intelligence, and self-rating items (as opposed to the items that are tests) can be appropriately compensated both ways. Embellishment and outright lies can be caught by collecting feedback from how people are doing, and using it to refine the algorithms and assign proper weights. Random questions (submitted by the existing employees) with a database to select them from will allow serendipitous discovery of new factors of importance, etc.

There's really a lot you can do with this stuff, and if you approach it seriously, it will beat any two regular HR people. And with the volumes they're looking to hire, it is no longer feasible to treat people all that individually, so having the tech people quiz skills and so forth is not at all viable.

Re:Only useful if...

[JacksBrokenCode]

Just spitballing...

They already gave surveys to people working at Google for 6+ months. Theoretically those people are secure in their job and stand to gain nothing by lying. If their sample size is large enough there may be some discernable trends by which they can judge applicant surveys. If 90% of current employees who prefer The Office (American) over The Office (British) version also happen to like red Swingline staplers and you get a job applicant who likes The Office (American) but likes Acme staplers, that could be a warning sign. On its own this might not be enough to create a 'this guy is faking' alert, but if enough warning flags were accumulated it could drag down the validity of the final score. Then again, this goes against the principles of hiring a mentally diverse workforce and promotes conformity so it may help show "wrong" answers but defeat the purpose.

More likely there are no "wrong" answers on the survey. The survey generates an complete score & profile so instead of saying "Applicant A isn't good enough for Blanket Job Description" it seems the results would be more "Applicant A's profile fits Job Description A. Applicant B's profile fits Job Description C but not D or B." Then the scores come into play where the algorithm decides that the top X applicants in each category can be used to fill open jobs Y in respective categories.

I'd be curious how much of a role it plays in final selection - if they have an abundance of "good" surveys one month and very few a short period of time later, does the system revert to older survey submissions to find appropriate candidates when a new opportunity opens? Or statistically aberrant responses where a candidate seems too-good-to-be-true how much human oversight comes into play?

Re:Only useful if...

[cp.tar]

It's fairly easy to detect dishonestly in a fairly long survey. Just ask the same question multiple times with slightly different words use. If dissimilar answers are received on these questions, the person clearly can't keep their story straight, and is likely to be giving untruthful answers.

Oh, yes, that's a great way to detect dishonesty.

Not only does it fail to take into account that you can remember answering the same question before, but it also fails to take into account the nuances of meaning if the repeated question asks generally the same thing, but with different wording.

I've noticed in quite a few tests that some questions were almost identical, yet I had to answer them differently.

There is, however, another method, also used quite a lot - a scale of lies (translating from Croatian; I really don't know the proper English term). The test is interspersed with innocent-looking questions to which practically everybody would have to give a "socially unacceptable" answer - e.g. Do you ever lie? or Do you always wash your hands after taking a leak?
If you give "socially acceptable" answers to most of them, you are considered a liar.

Frankly, I don't trust either of these methods.

Re:Only useful if...

[Writlaus]

Only useful if applicant is honest in their response to the survey.

That is not necessarily true. It won't matter to the survey generator whether or not an answer to a question is truthful or not--If there is a statistical correlation where productive people say that they like jogging, then the survey generator would calculate in favor of people who simply said that they like jogging. Whether or not any of those people actually do like jogging doesn't matter.

Re:Only useful if...

[Anonymous Coward]

Then again, this goes against the principles of hiring a mentally diverse workforce

Research has shown that while diversity makes better teams, it only does so up to a point. Too much diversity can actually impair effective teamwork. Though this is more a consideration in team selection than hiring for large companies, since not everyone in an organization is working directly with everyone else.

I think the bigger concern is to make sure people can fit into your corporate culture. An introvert in a company of extroverts is unlikely to do well, for example. Or if your company is very hands-off and unstructured, someone who needs more structure won't do well, even if they would excel in a more defined environment.

Re:Only useful if...

[mackyrae]

I hate that. "You already asked that, now go away!" I hate these HR surveys too though. I have yet to go into a Best Buy and be looking for some piece of tech junk and ask "is this Linux compatible?" ("Linux? What's...um, ask Geek Squad..." not that Geek Squad ever does anything but Google it) or correct the sales guy or ask about when a computer with a newly released processor will hit the shelf and not be asked multiple times if I'd like a job. I've taken their survey thing for a job twice, and I failed both times. The computer hates me, but the alive people want to my hire me. Something's wrong there.

Thankfully they changed the GPA thing

[odano]

I was one of those people who was hired with the under 3.0 GPA, and while getting the interviews was difficult, the people doing the interviewing really didn't care. They asked me to solve problems and show that I could do the job, and that was all they cared about.

Luckily for me I dont have to worry about it anymore, but especially in the technology field why should GPA be more important than actual projects and experience?

Re:Thankfully they changed the GPA thing

[panaceaa]

I've been working at Google for four months, and of all the companies I interviewed at, Google seemed to care the least about my past projects, experience, or my GPA. Google's interviewing process is all about finding very smart computer people. You simply must know the core computer science principles, but it does not matter if you were able to regurgitate them on your college exams. It matters that you can explain them in an interview and use them towards solving a problem. Once I got here, I can understand the reasoning behind the hiring process: Lots of Google infrastructure and technology is unique to Google. Look at the published articles on Bigtable and MapReduce to get a glimpse of the unique systems used every day here. For people to learn these systems and begin being productive quickly, Google doesn't care if you have an MSCE or know the syntax of Apache's httpd.conf. Google just needs you to be smart.

note: These are my opinions and not necessarily those of Google's. And I try not to post on Google articles nowadays, but this doesn't pertain to our business strategy so I'm comfortable sharing it. BTW we had an awesome free lunch today here in Kirkland, Washington. :)

Re:Thankfully they changed the GPA thing

[armada]

That is amazing. I had not heard about that requirement. I, for one, got a 2.8 GPA but a 1400 on my SAT (800) math. Never thought anyone outside of academia gave a damn about GPA.

Re:Thankfully they changed the GPA thing

[panaceaa]

Reading back over my post, I'd like to revisit saying that Google doesn't care about past experience. I meant that it doesn't especially matter what technologies you worked with (e.g. .Net vs. J2EE vs. Lisp). But it's certainly a good thing for you to contribute to open source projects, run your own web sites and web applications, and participate in research projects. It's also great if you know languages outside the norm, like functional languages, Ruby, Python, etc. Just doing your studying or your job is perfectly acceptable, but showing a history of going above and beyond, knowing more than you have to, and doing projects for the fun of it will get you noticed in interviews and when you get here.

Re:Thankfully they changed the GPA thing

[Anonymous Coward]

Smart, but not *too* smart. Too smart people have ideas and go and start their own companies. The Google wouldn't like that, wouldn't like that at all. They want academic smart, as evidenced by their focus on school marks and lack of interest in real world experience. I guess academic prowess not only shows you're clever, but that you're willing to give up some freedom and do what you're told. Funny how when you look at some of the world's most successful people (gates, jobs, branson etc), their academic achievements are modest to poor.

Re:Thankfully they changed the GPA thing

[Johannes]

Having been offered a job at Google with a sub 3.0 GPA (for the short time I was there), I can back up your claim that it's not required.

However, it didn't stop them from holding it over my head during the interviewing process and negotiation.

Re:Thankfully they changed the GPA thing

[panaceaa]

Smart, but not *too* smart. Too smart people have ideas and go and start their own companies.

This statement could not be further from the truth. One of my fellow co-workers is brilliant and he ran his own hosted content management company for years before joining Google. The three people who started a company that eventually became Google Talk are still working here in Kirkland, Washington. One of my friends here at the Kirkland office just moved to San Mateo, California, to work with the engineers at YouTube and learn from their entrepreneurial experiences. And just yesterday we had the founder of JotSpot, which Google acquired a few months back, come to help us with our latest product strategy. The people here are extremely smart, they have run their own companies in the past, and Google's very happy to have them. (And as far as I can tell, they're happy to be here :).)

Re:Thankfully they changed the GPA thing

[Duncan3]

Nobody outside of academia cares about GPA. I've never been asked about mine, never asked anyone about theirs in the hiring I've done either. But Google is almost all academics, so they care.

Lucky for us all, other academics have figured out that GPA is most strongly correlated as a past indicator of family income, not current or future intelligence. (Google for the studies, haha)

Other companies care about what you have done, and what you can do. Experience and skills. Since the only thing Google did is pagerank, and all it does is sell ads with no signs of doing anything else, maybe they should consider the normal methods a little more, and hire some did'ers and do'ers.

Then again, the only thing Americans do anymore is sell Chinese made stuff, or advertise Chinese made stuff so they can sell it, so Google may be right on after all.

Re:Thankfully they changed the GPA thing

[Guido von Guido]

A local employer who makes medical software apparently does. They take a look as far back as your high school transcripts. From what I hear it's also not a great place to work--they like to hire 'em young and work 'em as hard as they can for a few years until they burn out.

Re:Thankfully they changed the GPA thing

[Anonymous Coward]

there is no such thing as a free lunch!

Re:Thankfully they changed the GPA thing

[Anonymous Coward]

Reading back over my post, I'd like to revisit saying that Google doesn't care about past experience.

Perhaps you should just wise up and stop making public statements about your employer's hiring practices. That's a really, really dumb thing to do.

Re:Thankfully they changed the GPA thing

[rossz]

Today I had prime rib here in Mountain View. God I love my job.

Re:Thankfully they changed the GPA thing

[mazor]

>>I've been working at Google for four months, and of all the companies I interviewed at, Google seemed to care the least about my past projects, experience, or my GPA. Google's interviewing process is all about finding very smart computer people.

That's exactly why I quit Google after six months. Their hiring process is/was severely flawed for anyone who has deep specific experience that is not search related. New hires are assigned to the next available vacancy, largely independent of the new employee's skill or interest. "You were a welder, eh? Built ships and stuff. Cool. Now you're a florist. Have fun."

Doesn't it seem strange that Google is only now creating a survey to find out what prospective employees / new hires actually what to do?

Re:Thankfully they changed the GPA thing

[Anonymous Coward]

You're in Madison, right? And if so, I've heard exactly the same thing about Epic Systems. I'd never work there.

Re:Thankfully they changed the GPA thing

[Anonymous Coward]

I'm finishing my master's at Virginia Tech, and I just completed a hat-trick of rejections from Google: UI Designer, Software Engineer, and Web Application Engineer. I graduated summa cum laude with a double major in computer science & math, finishing my master's in CS, have 2 professional internships, and lots of side projects. I came to NY for the SE interview and I thought it went great. I answered their math questions perfectly and I got through a lot of the whiteboard programming material. I also had a phone interview for the Web App position which I also thought I aced. Rejected for all three positions.

I'm not sure what Google is looking for but apparently it's not me!

Only 4% turnover? It's going to rise

[vidarh]

It's easy to maintain a low turnover of staff as long as the vast majority of your staff isn't fully vested, and the stock is moving upwards. As soon as the growth in staff numbers slow down, though, you're going to see the turnover percentage increase significantly as a larger and larger percentage of staff have been there for the full 4 year vesting period of their options, and the company starts seeing pressure for lower refresher grants.

Re:Only 4% turnover? It's going to rise

[Anonymous Coward]

Can I ask what 20,000 people DO at Google?

Size will slow them down, a bad year or quarter will hit them hard and the market will demand heads - a change in CEO will result in a change in culture and the free food will stop. New employees will be given options that aren't as valuable as the old guys and a division between the 'old timers' and the new ones will occur. Employing standards will drop, politics will rise (as inevitable as the sun rising).

The key technical thing google solves is search and the scaling of their other offerings to 'google size' - but what is truely innovative from them?

They're future is held back by the web - their ability to break free from that is hamstrung by browsers, http and html.

They have one product earner - online advertising, and I don't know about you, but all my google ads have cost me money and delivered nothing (for the products we sell obviously aren't something people go looking for on google).

Re:Only 4% turnover? It's going to rise

[cpeterso]

Google needs 20,000 employees to type in all those search results pages.

Re:Only 4% turnover? It's going to rise

[vertinox]

It's easy to maintain a low turnover of staff as long as the vast majority of your staff isn't fully vested, and the stock is moving upwards.

Huh? Most people don't try to get hired or stay at a job just because of stock options. It is a nice perk, but if a company treats you like crap or you feel what you are doing is not appreciated or useful in some way then you are going to quit regardless of how much money they throw at you.

And if you are one of those people who stick around for the money even though you loathe the job, then it will suck to be your coworker, underling, or even manager of you because your performance is going to reflect your true feelings.

Which in turn results in more people leaving the sinking ship...

So rather than throwing wheelbarrows of money at employees to buy their loyalty, you'll succeed more by having a productive and worker friendly workplace.

Heck... If Google asked me I would work for them for minimum wage if they could cover my minimum living expenses otherwise (Well to fair... That would one helluva thing to have on a resume).

Re:Only 4% turnover? It's going to rise

[Anonymous Coward]

Too true.

I worked for a company writing a greenfields package that was going to support another company's business. Without going into details, there was ten thousand dollars (Australian, pre-tax) on the table if I stuck around until the code was delivered.

I quit in September. The code was delivered in December or January. Three months was all that stood between me and ten grand.

I didn't care then. I don't care now.

(Of course, if it had been an order of magnitude higher ...)

Re:Only 4% turnover? It's going to rise

[Anonymous Coward]

10 grand bonus, or 10 grand total payment?

Dunno 'bout AUS$ but in US$, 10K for 3 months isn't a hell of a lot. Pre-tax or not. I certainly wouldn't stick around at a shitty job for that, but then again I'm not hungry or desperate.

Re:Only 4% turnover? It's going to rise

[vidarh]

Most people don't leave a company because the company sucks, they leave because they get a better offer, or because they don't like their manager. High turnover says little about how the company is to work for overall, and high turnover in a large company tends to say more about how people assess the company's economic outlook at hence their feeling of security or how they are rewarded, unless the there's a really bad management culture in the company.

If stock options and vesting periods didn't aid retention, they wouldn't be used, as stock options contribute to tremendous dilution rates for investors in tech companies. If you look at hiring patterns for companies using stock options you will see very clear correlations to vesting periods - a company that sees slowing growth will start seeing increased turnover of staff when more staff get close to the end of their vesting periods. You can try to pretend it isn't so, but in real life it is.

The people who don't care about the money aren't the ones that contribute to high turnover - they quickly find somewhere they are happy and end up staying longer than average, and are unlikely to take jobs in companies that don't look like a good fit for them in the first place.

A lawsuit waiting to happen

[hellfire]

You can't hide bad questions behind an algorythm. The interview process has lots of laws around it now, and it's well established that there are only some questions you can ask. Here's a great example:

The questions range from the age when applicants first got excited about computers...

This question doesn't directly reveal your age, but a clever interviewer can glean much from it. "Oh, got excited in computers at 22, eh? Probably older than I thought. We don't want old employees we want young ones."

It is illegal to ask some questions in an interview. Age related questions are one of them. You are only allowed to ask questions that pertain to your performance of the job at hand. For example, I can ask someone "would you have a problem lifting heavy boxes?" but I can't ask how old you are and make a judgement because you are 40 that you can't lift heavy boxes. The above question you as a logical geek might think is iffy, but to a lawyer, it's shark bait and they'll be all over it, so don't ask it. If you ask a question that falls into this category, you open yourself up to a gender/age/racial discrimination lawsuit. These and many others are protected classes under the law.

And there's a great reason why an interview is a poor indicator of performance... because people lie!!! It's a sales process. They want your job, and you want the best candidate. Last two people I let go both gave great interviews, but when they actually worked, they sucked. They had all the right answers in the interview, but there is no escaping performance reviews.

0% firing rate is impossible, as is 100% retention. 96% retention is a stellar figure, even for silicon valley. I think they should be pretty happy that number.

Re:A lawsuit waiting to happen

[Jerf]

Your post triggered an interesting thought process.

Google knows AI and machine learning; even if they don't use it they'll have people who know about it.

Suppose by asking certain questions, and doing some initial research and calibration, I can determine your age within two years with 97% certainty. Or marital status, or race, or any of the other protected categories. Have I broken the law? What if I don't actually do the computation? What if my computers do the computation but no human ever sees it? What if I do the computation and no human ever directly sees the result but the computer has enough power to say "No" to a hire in practice, thus still incorporating this potentially "forbidden knowledge" into the hiring decision?

(After all, asking someone about their marital status may actually be less reliable in the end; I can easily imagine 1 out of 40 people lying about something like that, or their true age/race/etc. if asked.)

This is extremely likely to be possible, and probably downright easy for Google, so this isn't just a hypothetical. And the problems this raises extends beyond this exact instance into any domain where for legal reasons, we have to cultivate ignorance; exactly what constitutes "ignorance" if you get right down to it?

Re:A lawsuit waiting to happen

[Jerf]

Oh, and for extra double-bonus points, "How will a lawyer representing someone who was turned down for a Google position react to these hypothetical questions?" and "How will a judge and/or jury react to the entire idea?"

Re:A lawsuit waiting to happen

[Incongruity]

But at the same time, pre-employment psychological profiling is allowed in many states -- personality profiling, IQ tests and the like, so I think this system is comparable to using a psychological inventory -- especially if the raw answers are not kept for the people making the hiring decisions, but only the composite/scaled scores... I'm betting a legal team has had this run by them if Google's making it system-wide.

Re:A lawsuit waiting to happen

[JesseMcDonald]

It is illegal to ask some questions in an interview. Age related questions are one of them. . . . If you ask a question that falls into this category, you open yourself up to a gender/age/racial discrimination lawsuit. These and many others are protected classes under the law.

It seems to me that the problem here is the law, not their selection process. What business can it possibly be of anyone but the employers just how they choose to select (or reject) potential employees?

Mind you, I think irrelevant discrimination (as is usually the case for the classes you mention) is usually rather stupid, and most harmful to the one practicing it -- though it may make sense for certain (very rare) sorts of jobs[1] -- but if someone is inclined to practice such discrimination I'd rather have it out in the open where everyone can see it, not hidden in a dark corner to simmer, unexposed and unresolved. If discrimination is prohibited then it will indeed be marginalized, but some employers will continue to believe that they could possibly have done better were they permitted to make their own choices. Permit them to choose freely, however, and the fallacies of this reasoning will soon be exposed by their less discriminatory competitors.

In any event the right to discriminate is implied by freedom of association. No individual has the right to tell another whom they must or cannot hire, or for whom they must or cannot work; neither can any group of individuals (e.g. a government) can hold a right in aggregate which none of its members possesses. Anti-discrimination laws, therefore, can have no ethical or moral basis; no one has the right to impose them on another.

[1] To illustrate, consider age discrimination, which is Constitutionally required for certain political offices. Wouldn't you agree that it's a bit hypocritical for the government to forbit age discrimination to other employers while the practice remains one of its own fundamental employment rules?

Re:A lawsuit waiting to happen

[corbettw]

What if my computers do the computation but no human ever sees it?

A human had to tell the computer that marriage is a factor to consider in the first place. That alone could be enough to open you up to a lawsuit.

Re:A lawsuit waiting to happen

[teal_]

[1] To illustrate, consider age discrimination, which is Constitutionally required for certain political offices. Wouldn't you agree that it's a bit hypocritical for the government to forbit age discrimination to other employers while the practice remains one of its own fundamental employment rules?

Good example. What about when you're hiring a receptionist though? You want to present a certain image for your company when clients come to visit. In other words, you want someone attractive and reasonably young there and, oh yea, female. What if some fat disgusting old dude applies for the position with great office skills, can you turn him down and give the job to a dumb blonde instead?

Hrm... now there's an idea... gain 50 lbs, grow an unruly beard, skip a few showers, and then go interview for a receptionist job... I could retire with the dough I'd make on the lawsuit, cha-ching!! :)

Re:A lawsuit waiting to happen

[Anonymous Coward]

Based on their company, if someone is obviously capable then they are going to recruit and hire them regardless. This algorithm is just going to weed out the undesirable resumes that they probably receive constantly.

Re:A lawsuit waiting to happen

[Bugmaster]

But a human would never tell the computer to explicitly consider marriage. Instead, the computer would be trained (or, rather, train itself) to draw conclusions from all kinds of disparate data, which could amount to inferring whether the applicant is married or not. What happens then ?

Re:A lawsuit waiting to happen

[Jerf]

A human had to tell the computer that marriage is a factor to consider in the first place.

Not in the way that you are most likely thinking of. It's difficult to express what I mean without much math, but basically, that 97% confidence/correlation would come from a statistical profile that would be 97% accurate at guessing, and at no point does that profile ever contain the actual knowledge of the marriage state... yet, an external viewer can only find a 3% variance between this statistical process and simple direct knowledge.

Consider the stereotypical Slashdot geek, who is so very single that they are 25 and haven't had a date yet. If I establish that you are an anime geek, who knows 10 programming languages, plays video games, and so on down the list of stereotypical behaviors, and in addition, that you don't like sports, have a pimply complexion, etc. etc. I may be able to build a model that correctly guesses that you are single 97% of the time.

(I'd like to emphasize that this says nothing about how that is done. There's a lot more to it than this. You'll have to take my word for it that this process is more sophisticated that a set of "if then" clauses, and it's not as doomed to failure as a naive conception would lead you to believe.)

Basically, what is boils down to is: Which is more important? Acting in a way that we currently obtain by forbidding employers to know marital status, or simply not knowing? The actions, or the state of knowledge in human heads? The law basically assumes the two are the same, but they don't have to be.

Re:A lawsuit waiting to happen

[JakartaDean]

Suppose by asking certain questions, and doing some initial research and calibration, I can determine your age within two years with 97% certainty. Or marital status, or race, or any of the other protected categories.

Why bother? I'm an HR consultant (not in US) and I can't imagine why you would want to. If you have a set of legal questions which have positively correlated with on-the-job success in the past, why flirt with illegal questions?

Data like this is like the holy grail for HR departments, but few have the resources or knowledge to do it well. For most companies, hiring people who will do what you want them to do is the most important HR thing they do, but the hardest to get right. Why bother with a manager's unsupported belief that younger, or white, or whatever, people will do better when you already know, statistically, who will do better?

This is just simply excellent HR management, and there's no need to screw around with questions designed to trick you into disclosing information about gender, race, age, etc. As to lying, I imagine in the US lying on your application is a firable offence, so you protect against it as far as possible, but there's not that much to be done.

Re:A lawsuit waiting to happen

[Seraphim_72]

You are arguing semantics...if I can walk into a courtroom and declare that your algorithms find married people and discriminates against them, whether it identifies them as married or not, I have a heck of a good case. If you step on the stand and start to try to explain your algorithms...well good luck with the average jury looking at the guys 3 kids in the courtroom.

Sera

Re:A lawsuit waiting to happen

[terrymr]

The US Federal Government also generally requires US citizenship for employees (except the military). Private employers aren't allowed to discriminate based on citizenship (unless they have a contract with the feds that requires it).

Re:A lawsuit waiting to happen

[DamnStupidElf]

Suppose by asking certain questions, and doing some initial research and calibration, I can determine your age within two years with 97% certainty. Or marital status, or race, or any of the other protected categories. Have I broken the law? What if I don't actually do the computation? What if my computers do the computation but no human ever sees it? What if I do the computation and no human ever directly sees the result but the computer has enough power to say "No" to a hire in practice, thus still incorporating this potentially "forbidden knowledge" into the hiring decision?

The solution is to limit questions to those that reveal only pertinent information. If your questions "leak" information about the age, sex, religion, etc. of the interviewee, then the questions are too broad and not specific to job duties.

Since you mentioned AI, I think a perfect selection function is that interview questions should not form a turing test of the interviewee. After all, if you can't tell whether you're interviewing a smart human who can do the job and an intelligent AI that can do the job, you clearly aren't discriminating based on any protected information. This disqualifies any questions not specifically within the domain of discourse of the job function. Actually formulating such questions requires proving a negative (that no information leaks), so in practice it would be incredibly difficult or intractable to prove. However, simply asking whether or not a given interview question would differentiate between a human and an AI would probably be a useful guide in choosing appropriate questions.

Re:A lawsuit waiting to happen

[Bugmaster]

Couldn't you level the same allegations at any human, though ? I.e., we can look at the entire employment history of CutCo/EdgeCom going back 10 years, see that they hired more single people than married people, then sue them for discrimination. After all, even though they don't explicitly have a policy that considers marital status, clearly their HR managers are using this as part of their judgement... right ?

Re:A lawsuit waiting to happen

[captnitro]

I honestly fail to see to how this reasoning has much logical or historical basis. I mean that sincerely, not just to push back at you.

For 100 years in the United States we had more-or-less open hiring practices after the abolition of slavery -- no anti-discrimination law -- and none of this went very well. Perhaps in a due time, but the thing about discrimination is that it allows bad people to continue doing bad things at the expense of the people they discriminate against. In other words, take a marginalized group, refuse them education on a discrimintaory basis, and you end up with a marginalized class who doesn't qualify for a job. Deny them a job, and you deny them their next job to someone with more experience. It keeps going. You're allowed to have your biases, but if you're not allowed to push them on me as a job applicant.

Permit them to choose freely, however, and the fallacies of this reasoning will soon be exposed by their less discriminatory competitors.

This assumes that everybody had a fair chance to begin with, and it often feels like many Libertarians have to tack this one on to make it clear that discrimination is bad. Discrimination, statistically speaking, creates a loop where the majority tends to reward the majority based on a bias against a minority, and it almost never begins nor ends in a fair market. Again, we had a solid period of time where the fallacies of this reasoning were not exposed, in fact, they merely reinforced existing discrimination. It took a damn-near revolution in the way people thought.

Re:A lawsuit waiting to happen

[evilviper]

It is illegal to ask some questions in an interview. Age related questions are one of them.

It's a shame a company like Google doesn't have some way to scour the internet, to turn up info on a specific person they are interested in finding more info about...

That sure would be nifty.

Re:A lawsuit waiting to happen

[Anonymous Coward]

Your above post proves that privacy in the information age is impossible. Because things can be deduced through the the limited possibilities. I wouldn't be surprised if demographic data is linked or could be linked to the algorithm to actually find out who that person is.

Re:A lawsuit waiting to happen

[gardyloo]

Since you mentioned AI, I think a perfect selection function is that interview questions should not form a turing test of the interviewee. After all, if you can't tell whether you're interviewing a smart human who can do the job and an intelligent AI that can do the job, you clearly aren't discriminating based on any protected information.

      Sweet. I, for one, welcome our silicon-based, monotone Hooters waitresses!

        Oh, wait. . ..

anyone have an example?

[HelloKitty]

does anyone have example questions?
i'd be curious to see how relevent they are...

Re:anyone have an example?

[c0d3r]

I felt that the most difficult question in their face to face interview was:

Given a dictonary word look up function (done on a whiteboard with a marker), write an algorithm to list all of the word possiblities for a phone dial sequence (eg: 2 is a b c, 3 is d e f, etc).

Re:anyone have an example?

[TheRaven64]

Is this really the kind of thing that Google considers a hard problem? It's trivially reducible to to SAT, so should be solvable by anyone with an undergraduate degree in computer science, and about as interesting as writing a sudoku solver.

Out of interest, did you get the job?

Re:anyone have an example?

[gomoX]

They ask you to solve this in front of the interviewer, on a whiteboard. You won't be asked to implement and train an ANN traveling salesman solver on a freaking whiteboard. I think it's a good question. Trivial if you do research in a related field or work in areas related to AI, but for typical Java servlet monkeys I think it could be complicated. It's not as if the admission process consisted solely of that question.

What would be a good interview question, in your opinion?

Re:anyone have an example?

[Prophet of Nixon]

Well, I have no computer science training at all and I managed to solve that in JS in about an hour. I figured I'd attempt since that sounded like a good idea for a useful function. JS follows, I had to really condense the code to get it past the "lameness filter" and then slashdot mangled it anyway by inserting random spaces... it will work if you carefully remove the spaces, I checked. Why does slashdot hate code so much? It took me an additional half hour just to manage to post it...

function numbler(){
var n;
n=prompt("Enter a phone number!"); //get a number
var allowed="23456789"; var i=0; //strip disallowed characters
while(n.length>0 && i<n.length){if(allowed.indexOf(n.substring(i,i+1)) ==-1){
n=n.substring(0,i)+n.substring(i+1,n.lengt h);i=i-1;}i=i+1;}
var words = new Array(); //declare an array of strings
//start the array with the appropriate letters
if(n.length>0){switch(n.substring(0,1)){
case "2":words[0]="a";words[1]="b";words[2]="c";break;
case "3":words[0]="d";words[1]="e";words[2]="f";break;
case "4":words[0]="g";words[1]="h";words[2]="i";break;
case "5":words[0]="j";words[1]="k";words[2]="l";break;
case "6":words[0]="m";words[1]="n";words[2]="o";break;
case "7":words[0]="p";words[1]="q";words[2]="r";words[3 ]="s";break;
case "8":words[0]="t";words[1]="u";words[2]="v";break;
case "9":words[0]="w";words[1]="x";words[2]="y";words[3 ]="z";break;
}n=n.substring(1,n.length);}
//and then start the recursive function
if(n.length>0){words=permutate(words,n); }
//I don't have a dictionary lookup, but here's a quick one:
var dictionary="i|am|batman|and|this|is|a|test";
//lo okup each term in the dictionary, and shout the good ones
var i=0;while(i<words.length){
if(dictionary.indexOf( words[i])!=-1){alert(words[i]);}
i=i+1;}}

func tion permutate(words,n){
var letters;var i=0;var j=0;var k=0;var temp=new Array();
if(n.length>0){switch(n.substring(0,1)){
case "2":letters="abc";break;
case "3":letters="def";break;
case "4":letters="ghi";break;
case "5":letters="jkl";break;
case "6":letters="mno";break;
case "7":letters="pqrs";break;
case "8":letters="tuv";break;
case "9":letters="wxyz";break;
}while(i<words.length*l etters.length){
j=0;while(j<words.length){k=0
wh ile(k<letters.length){temp[i]=words[j]+letters.sub string(k,k+1);
i=i+1;k=k+1;}j=j+1;}}
n=n.substri ng(1,n.length);
words=permutate(temp,n);}
else{r eturn words;}
return words;}

Re:anyone have an example?

[bigbadbuccidaddy]

I have a feeling they weren't looking for the obvious solution of generating all the letter permutations and looking them up in the dictionary. Probably something along the lines of loop over all the words in the dictionary once, calculating the dial sequence, stick it all in a hashtable and have an O(1) lookup on any phone number. Still pretty damn obvious though.

Re:anyone have an example?

[TheRaven64]

They ask you to solve this in front of the interviewer, on a whiteboard. As an undergraduate, we were expected to provide similar algorithms in tutorials in front of a lecturer and half a dozen of our peers. Questions like this are possibly important to check the candidate didn't sleep through their degree.

What would be a good interview question, in your opinion? The best interview questions are those which allow the tester to see how the candidate thinks, not those where the answer is important. As such, they shouldn't be directly related to an area in which the candidate can just call up a solution to a similar problem that they've studied and apply it. One of my favourites is this (of course, they don't work on people who have heard them before):

Imagine a boat in the middle of a pool. In the boat is a brick. If you throw the brick out of the boat and into the water, will the water level go down or up? A candidate who can remember the Archimedes Principle will simply know that the brick is displacing its volume when submerged, but the volume of its mass of water while in the boat, and since it sinks it is more dense than water so the water level will go down. This doesn't tell you much. Most people won't, however, so you can see a lot about their reasoning process. If they get stuck, ask them the same question with a bit of wood instead of the brick, and a helium balloon. If a candidate tells you that the water level goes down, and then once you've explained why they're wrong says 'Oh, do bricks sink?' (yes, it has happened), then don't give them the job...

Don't quit your current job.

[Anonymous Coward]

all they have to do is slashdot the survey.

then they will get a flood of some a million people taking the survey.

i would guess anyone here at slashfot would love to work at google, i know i do

In view of your apparent contempt for proper grammar, spelling, punctuation and syntax, my guess is that they would toss your application after reading the first three words that you wrote.

How do you feel about personality questions?

[sbenj]

I once had an interview for a largish organization in which I only spoke to the HR person (fair enough, he was presumably screening for the tech interview which would've followed). What made this interview notable was that he was largely questioning me on personality.

At the time I thought it was kind of rude, really. What business is it of yours if I "consider myself an outgoing person"? After asing me a few preliminary questions he left the room and had me fill out responses on a computer program. I specifically remember one screen with something like 50-100 checkboxes that asked you to check which ones you felt applied to your personality type. It was then followed by the identical screen, this time to be filled out with "how you thought people saw you". A good half hours worth, many more screens, a personal essay, by the end I was rather ... pissed, actually spent about half of my time deciding whether to be polite or not (I'm sure the test was sensitive enough to detect this and needless to say they didn't call me back). At the time I thought the the HR guy had convinced the company to buy him a new toy and was busy tormenting all the new hires with it.

In any case I'd be curious to hear people's responses to such. Do you think this is fair? As is probably clear from the above, I think it's way out of bounds and personally intrusive.

Lest you think the Google stuff is all technical, here's a quote from the article:
"Some questions were factual: What programming languages are you familiar with? What Internet mailing lists do you subscribe to? Some looked for behavior: Is your work space messy or neat? And some looked at personality: Are you an extrovert or an introvert? And some fell into no traditional category in the human resources world: What magazines do you subscribe to? What pets do you have? "We wanted to cast a very wide net," Mr. Bock said. "It is not unusual to walk the halls here and bump into dogs. Maybe people who own dogs have some personality trait that is useful."

Re:How do you feel about personality questions?

[badboy_tw2002]

It depends on the job of course. If I'm going to lock someone in the basement and have them watch machines all day then I don't really care if you're an outgoing person. Actually, I'd probably want an introvert because they'd be happier there than someone who liked human contact. But if the job involved managing others or dealing with customers, then I want somenoe who's outgoing, and the question is very relevant.

Re:How do you feel about personality questions?

[Control Group]

The company I currently work for does something that sounds similar: you're given a list of a few dozen (maybe 50-ish) adjectives, and asked to check the ones that you believe describe you. Then you're given an identical screen and asked to check the ones that you believe you should exhibit.

When I was called in for the interview, they showed me the results of the survey, and I was astounded at how accurate a profile of my work personality they had come up with. It was almost frightening. Of course, as I assume most people do, I pretty much just breezed through both screens, which might be part of why it worked.

In any event, I got the job, and I'm glad I did; this is actually the best group of people I've ever worked with.

The point, though, is that the test (which I originally blew off as ridiculous HR fluff) was shockingly accurate about me.

Re:How do you feel about personality questions?

[Dunbal]

What made this interview notable was that he was largely questioning me on personality.

      I don't have any sort of degree in HR, but I own a small health care company and do all the hiring myself. And I mostly ask very tough questions to gain insight into the other person's personality. How they view themselves. How they view the world. Why? Because it's all I really care about.

      If the come to the interview dressed like crap, they're automatically out. If they turn up late, they're automatically out. The resume is usually full of a lot of BS anyway - I check on the real important stuff - like - do they actually have the degrees they say they have.

      Letters of recommendation are usually from work buddies, after all, you're not going to ask the supervisor who hates your guts for a recommendation, you'll ask the other one who really likes you. So I'm left with personality - self esteem, self confidence, ability to take the time to LISTEN, and ability to adapt. It's kinda rough on the guys, but hey, an interview is an interview. I have my patients to protect. And I think I've done ok with this technique so far.

Re:How do you feel about personality questions?

[panaceaa]

I personally would not fill out a survey like that. While I'm fortunate to have a nice background of experience, so I can walk out of such an interview without feeling bad about it, there's other reasons for not filling out a questionnaire like that. For example, do you want to work at a company with so much process that hiring requires applicants not to show their ability to communicate clearly about relevant experience, but to fill out 100 question surveys? If they expect you to do them when they're not paying you anything, what kind of bullshit processes will they have once they are? And what kind of co-workers would you work with who put up with the same shit? Are these the types of people who will be revolutionaries and inspire you to be a better worker? Seems to me they're more likely to obey process ... and the company will end up going nowhere.

Re:How do you feel about personality questions?

[mandelbr0t]

I don't mind them as long as they are asked by someone with a psychology degree (or maybe other credentials that show an ability to interpret the results) and the results are kept private. I've seen two things in recent years that disturb me greatly, however.

The first was my experience at a technical college. The first day of class, we all took a personality test, given by a corporate psychologist with a B.A. in psych. We were not warned in any way that the results would not be private. Thus, on the first day of class, the entire class was presented with a psychological profile of their fellow classmates, on the basis of an examination with questionable accuracy and administered by a person with questionable credentials. I believe this information was also provided to the instructors and various business associates of the examiner. I can't believe that's legal!

The second was a company that actually posted the results of the HR personality exam in a public place within the building. I suppose it's possible that you could opt-out from being listed, but even still, it's a disturbing trend.

People put far too much emphasis on the results of these sorts of tests. They also try to simplify personalities to the point where the description is almost meaningless. Certainly, most people who administer the tests are completely unqualified of interpreting anomalies in the results, such as when a person suffers from a multiple personality disorder. What would such a result mean to an employer? Is the employer allowed to discriminate on the basis of a mental illness that may or may not impact job performance?

Personality tests seem to be a new tool for HR these days. But I'd certainly challenge many of their "findings" as being totally subjective guesswork. There definitely needs to be legal guidelines as to how these tests are administered, and what analysis can be done with the data. After all, people's careers can be decided on the basis of such exams.

mandelbr0t

Re:How do you feel about personality questions?

[SixDimensionalArray]

I'm not sure I'd be happy answering these questions at an interview either, but it sounded an awful lot like the Keirsey Temperment sorter (see http://www.advisorteam.org/instruments/KTS-II_orig inal.html) or a Meyers-Briggs (see http://en.wikipedia.org/wiki/Myers-Briggs_Type_Ind icator) personality test. HR departments for large organizations often choose these because they are considered useful for organizational development. I personally am not sure if they really work!

SixD

Re:How do you feel about personality questions?

[AuMatar]

A lot of those are bad ways to hire

1)Dressed like crap- unless they're in a customer facing role, how they dress does not improve performance. My code doesn't get written faster or higher quality if I'm in a suit rather than my Einstein tshirt. If anything, the fact the tshirt is more comfortable improves my mood. I actually dress down for interviews on purpose- to make sure I never take a job from someone who cares about trivialities like that.

2)Resumes full of BS- if there's any BS on the resume this should be a DQ right then and there. If they lie on the resume, how can you trust them later on? As an aside- how do you actually check they have the degree, the last I knew colleges won't just fax you their transcript. Last time I tried to get my own I had a 5 step process of verifying my id.

3)Self confidence- need to be careful here. Confident in their own abilities is good. Cocky is bad. People too confident in their own abilities have trouble taking criticism and ignore input, these make bad employees.

4)Personality- should definitely be checked in an interview, but hiring someone because he seems like he'd be fun to have a beer with is a bad idea. You want the guy you'd have a beer with who also has strong technical skills. Which you don't mention once.

Re:How do you feel about personality questions?

[zxnos]

...by the end I was rather ... pissed...

at the interview? damn you are hardcore, i usually wait until the 3rd or 4th week before i start drinking on the job...

Re:How do you feel about personality questions?

[Anonymous Brave Guy]

Sounds like you'll be very good at hiring salesmen (or con artists) but pretty much trusting to luck as far as hiring technically good people goes.

And I think I've done ok with this technique so far.

You may choose to think that, but the interesting question is: how do you know?

Re:How do you feel about personality questions?

[Anonymous Brave Guy]

Funnily enough, that's pretty much my reaction to that sort of recruitment process as well. I don't care if you're $BIG_NAME_EMPLOYER. I'm good at what I do, and plenty of good places to work will hire me in any moderately employee-friendly market. I have better things to do than jump through apparently pointless hoops for extended periods of time because one particular company is incapable of making a reasonably quick decision about its interviewees.

Re:How do you feel about personality questions?

[Hatta]

The point, though, is that the test (which I originally blew off as ridiculous HR fluff) was shockingly accurate about me.

So are horoscopes.

Re:How do you feel about personality questions?

[Dunbal]

You may choose to think that, but the interesting question is: how do you know?

      Because the coach of a winning sports team knows he has a good team. Profit, growth, and the interpersonal environment at our company makes me sure ;)

Re:How do you feel about personality questions?

[Dunbal]

A lot of it is job specific. You didn't read the part of my post where I said our work is in healthcare. Do you prefer the doctor with the big skull and crossbones tattoo on his forearm, the unkempt green hair, and 3 earrings, or do you prefer the one conservatively dressed? I admit the first one might have had slightly higher test scores in med school (it could happen), but most people will go with number 2.

Everyone's resume is full of BS. No exceptions. Even mine has some BS in it. How do you explain wasting 5 years of your life in some dead end job? You, er, enhance it a bit.

Self confidence - goes with the job. I want the people on the sales side to be self confident to the point of being cocky. A sales rep has to be an aggressive SOB in my book. Courteous, patient, but aggressive. They have to make it happen. I want the people on the medical attention side to be a lot more introspective and cautious. A cocky doctor is a dangerous doctor - we all make mistakes. At least a doc has to be humble enough to admit s/he made a mistake if/when it happens, and try to correct it. I'm not using the same cookie cutter mold for everyone.

As for personality, I don't hire someone who looks fun. I hire someone who I think will get along with the rest of the team, who comes off smart enough to have their own ideas once in a while, and who has the experience that I lack to fill the position I need. But if someone is "umm"ing and "ahh"ing through most of the interview, and has nothing to say - like one poor fellow I spoke to a couple weeks ago - sorry.

Re:How do you feel about personality questions?

[minniger]

1) Even super geeks need to have some people skills to get by in most places. Simply putting on a clean pair of jeans , nice shoes and a shirt with buttons on it say's you at least give a slight crap about getting along with other people. I'll hire someone in a t-shirt for a startup. Never, for a biz like medical care.

2) There's BS and then there's BS. A little polishing is good, too much is as auMater says... a bad sign. You gotta know which is which. You can check the degree by just asking the registrar if the person graduated as they said. No transcript required.

4) Super important in a small biz. I'll take a somewhat less skilled person that's good to have a beer with over super geek who's as ass any day. Kind of the same thing as 3. Good personality, tech skills and the ability to share and play nice with others is ideal.

Re:How do you feel about personality questions?

[wikinerd]

I am a Computer Science graduate now working as a programmer while studying toward an MSc in Management, which includes a good deal of HRM, and I can say this: HR drones that use lengthy questionaries are incompetent. You do not need to complete a full MBTI test to get hired. A good HR manager can understand your personality very quickly by having a look at your face, your CV, and your website or blog if you have one. Also, there is just no logic in preferring outgoing personalities for technical positions: Programmers and sysadmins work with their brains, and outgoing people aren't famous for their intelligence. Outgoing people tend to socialise, and I think a company should prefer its programmers to spend time with computers rather than with that nice new female coworker on the next cubicle. Perhaps I would prefer an outgoing person for customer service, but I think programmers and sysadmins should be allowed to work in a quiet environment without interruptions so that they can achieve "flow mode" or "hack mode" and be more productive.

Re:How do you feel about personality questions?

[leabre]

I hired once someone that was a "fun" person during the interview but not as strong technically but we were willing to take that risk "for the right person". That was the worse decision ever made. I'll never do that again. Actually, I did that twice (the second time I was under pressure to hire someone NOW NOW NOW). Now, when I interview, they must write code to solve whatever problems are proposed during the interview. If they say they did something on their resume they must be able to answer questions relating to it and write code relating to it (example, they say they did socket programming so they should not only be able to answer TCP/IP questions but also were a simple socket server/client). We've weeded out 90% of applicants that way and only the good ones got through. It is a chore getting them to accept an offer, however. The next problem is keeping those good ones at the company because they usually leave for more appealing oppurtunities after 1-3 years. Some have gone to be google employees, architects for major financial instutions, senior people and your favorite social networking startup, etc. These days, I don't care whether they're "fun" or not as long as they aren't a jerk. If they can do the job demonstrably, they're hired, period.

Thanks,
Leabre

Re:How do you feel about personality questions?

[rossz]

In the past, when I had to pick a primary care doctor out of a book for the HMO insurance, I would always pick the doctor who's name sounded most like a disease. I figured he had to be good, and I didn't really have anything else to base my decision on.

Re:How do you feel about personality questions?

[wikinerd]

I am studying for an MSc in Management and I believe that you should not automatically rule out late candidates or people who aren't dressed nicely. Remember that when you are interviewing someone, you ask them to give you their time without any form of payment. Unless they need the job or you are a very reputable employer, people have absolutely no reason to spend time (and perhaps money too) dressing well to appeal to your tastes, or to get an expensive taxi to get on your premises on time. They may have other things to do, they may have been asked to attend 50 other interviews, and they may see no reason why they should spend considerable time to prepare for you. Suppose that you were to hire Einstein in his 20s. You have asked him to come by 1400, but it is 1345 and he is still busy doing his math computations for his new theory. If he stops now he will go out of his "flow mode" so he continues until 1355. At that time he walks out of his home and gets the bus to your office. So busy with his maths, he had forgot to change clothes and he is in his pyjamas, reaching your office at 1415. You rule him out immediately, he finds better employment elsewhere, and your company loses a bright nerd. You see his name in the newspapers 10 years later, but your company, staffed by incompetent monkeys, fights to remain out of bankruptcy. Hire brilliant people, motivate them and convince them to work with you, and you will never lose.

Re:How do you feel about personality questions?

[Vellmont]

Because the coach of a winning sports team knows he has a good team. Profit, growth, and the interpersonal environment at our company makes me sure

Nonsense. In an industry like healthcare where there's enormous growth you'd have to be totally incompetent to not make money and grow. There's plenty of businesses that're horribly managed, have terrible employees, but yet still manage to expand. AOL sucked monkey-dick for years (and still does) and grew at a phenomenal rate until everyone realized they didn't need them (or more to the point, broadband took over).

Re:How do you feel about personality questions?

[Dunbal]

In an industry like healthcare where there's enormous growth you'd have to be totally incompetent to not make money and grow.

      Perhaps that's the case in your country. In my country there is a national health service, and monthly contributions to it are obligatory. The only reason I make money is because people are dissatisfied with the government, and are willing to pay ME a premium in order to get fast, private, quality care. There are many unemployed physicians in this country, who simply can't make it in the private sector. I know that recent medical graduates can spend up to 18 months looking for their first job, and when they get it it's usually a locum tenens, a temporary 2 week position to fill in for someone on vacation. Our situation here is not quite what you expect.

      Anyhow, you seem to be determined to contradict me no matter what I say. Going back to the topic, I think the technical side can be evaluated very easily - either you know it or you don't. Doesn't take all that much in an interview to realize that. Personality is something I pay a lot more attention to.

Re:How do you feel about personality questions?

[minniger]

Exactly my position:

> These days, I don't care whether they're "fun" or not as long as they aren't a jerk. If they can do the job demonstrably, they're hired, period.

"Fun" has nothing to do with it. Being infantile and/or an ass is disqualifying no matter how skilled. We interviewed a guy with straight A's, high sat's, really very smart (even dressed nice). But he spent half the interview bagging on his co-workers and other companies he interviewed with. Everyone gave him a 'no hire' before we even discussed him. Poor guy had worked at a big company for his first 6 years of employment and hadn't learned some basic interview no-nos...

Why should they know some of this stuff?

[Anonymous Coward]

And some fell into no traditional category in the human resources world: What magazines do you subscribe to? What pets do you have?

What business do they have asking that? What difference does it make in your ability to do your job?

Re:Why should they know some of this stuff?

[zuiraM]

That's the nice thing about factor analysis. They don't have a clue why it matters, they just know that it does (because desireable employees have certain traits, and undesireable employees lack them).

You're looking for predictors of good or poor performance, and the only thing that matters is how well those predictions fare in keeping your standards at an acceptable level and making sure people end up in positions where they will be able to deliver, and get to work with people they will gel properly with.

Filling Jobs...

[Anonymous Coward]

And Jobs is so full of himself, how does Google hope an algorithm can fill him up?

Oh... 'Bow-chicka-bow-wow...'

They're not losing money fast enough...

[Prof.Phreak]

Everyone knows that a growing company loses money, no? (ah, Dilbert!)

10,000 employees??? What the heck are they doing? 20,000 employees next year? How the heck do they manage to coordinate anything??? Do they even -have- a corporate culture, or agenda?

Lets see... 10,000 employees, on average, costing the corp ~$200k each... that's... $20 billion a year... in salaries/benefits/office space/etc. Are they even making that much? Are they paying their workers with ``profits'' from stock sales?

Either their salaries are low (and employees work for stock options), or something is fishy.

Re:They're not losing money fast enough...

[Anonymous Coward]

Hey! Quit doing the math! We're not paying you to tell us that we're building another house of cards! We're paying you to tell us your solution to moving the house of cards to Mt. Fuji.

Incidentally, if you could also tell us why manhole covers are round it would help a lot. That's been bugging us!

THX.

Re:They're not losing money fast enough...

[Anonymous Coward]

There's a secret Google bootcamp where new hires are indoctrinated. Those who fail the indoctrination "were never hired in the first place".

Re:They're not losing money fast enough...

[Anonymous Coward]

Well, it's $2B actually, not $20B. And their revenue over the last 12 months was $6B. They're a public company you know. You can look this up.

Re:They're not losing money fast enough...

[Dunbal]

if you could also tell us why manhole covers are round it would help a lot.

      Or how to get out of this damned blender...

Re:They're not losing money fast enough...

[Ludedude]

So they can't fall in the hole of course...

Wait, what's happening?

Re:They're not losing money fast enough...

[Atzanteol]

Why guess? They're a publicly traded company.

Re:They're not losing money fast enough...

[Prof.Phreak]

Well, it's $2B actually, not $20B.

Meh. Confusing zeroes...

Re:They're not losing money fast enough...

[TheRaven64]

Incidentally, if you could also tell us why manhole covers are round it would help a lot. That's been bugging us! Manhole covers are round for two reasons. One is related to user interface; it's a lot easier to remove them if you can twist-and-lift, and it's a lot easier to re-insert them if you don't have to worry about orientation. The other is due to manufacturing processes; it is very easy to make long cylinders of metal, and you can make discs by just slicing a cylinder. This makes round manhole covers cheaper than other shapes.

Re:They're not losing money fast enough...

[Anonymous Coward]

No, moron. Manhole covers were made round back when they were made of wrought iron by hand and they didn't have "user interface" specialists.

As posted above, if the manhole cover was square, it could fall through its own manhole if dropped diagonally into it. With a circle, no matter the orientation, its impossible for the manhole cover to fall in, barring the manhole itself being damaged.

Manhole covers were made round because people were getting killed beforehand.

Re:They're not losing money fast enough...

[corbettw]

10,000 employees, on average, costing the corp ~$200k each

Whoa, no wonder people want to work there!

Re:They're not losing money fast enough...

[dirc]

Lets see... 10,000 employees, on average, costing the corp ~$200k each... that's... $20 billion a year... in salaries/benefits/office space/etc. Are they even making that much? Are they paying their workers with ``profits'' from stock sales?

10,000 employees at $200k each is $2 billion a year, not $20 billion a year. Google is making enough to cover those costs even if they double the number of employees and do not increase revenue at all. You can look at a summary of their revenue, and their expenses as a portion of revenue here: http://biz.yahoo.com/e/061108/goog10-q.html

They are making a handsome profit.

Re:They're not losing money fast enough...

[imsabbel]

Manhole covers are round because square ones could drop down the shaft.
With round ones, this is impossible, thus improving security and easyness of handling.

Re:They're not losing money fast enough...

[Jherek Carnelian]

It's called "overhead" - real-estate, furniture, HVAC, electricity, computer equipment, coffee, security, employer taxes, etc. The rule of thumb is that an employee's annual cost is twice his annual salary.

Re:They're not losing money fast enough...

[mok000]

10,000 employees??? What the heck are they doing?

100 coders, 9900 people scanning books and documents. Can't get enough of those...

Re:They're not losing money fast enough...

[eh2o]

Sliced??? I seriously hope this is a joke.

Re:They're not losing money fast enough...

[wikinerd]

because of this real estate overhead companies, including Google, must embrace telework and telecommuting.

Re:They're not losing money fast enough...

[Rsriram]

1. 200K multiplied by 10,000 employees = $2.0 Billion
2. Not all of the 10,000 staff are in the US at 100K plus salaries. Large number of them are also in India where the average is more likely around 40-50K. That probably brings down the total employee spend to $1.5 Billion.
3. Their current revenue run rate is around $10 Billion with a profit run rate of $4 Billion per year. Even if the employee spend went up to $4 Billion and their revenue is flat, they are still profitable.

Re:They're not losing money fast enough...

[jo42]

It is called the "throw the spaghetti at the wall to see what sticks" method of corporate development. The idea being that with all those people, someone, somewhere, at some point in time, will come up with something useful.

Re:They're not losing money fast enough...

[Anonymous Coward]

Common myth, and incorrect. Sorry.

Perhaps only some misconception?

[butterberg]

100,000 applications a month Wow, are these really all serious applications? I mean, Google just have to put something new on the web, and everyone and his dog starts playing around with it. :)

You mean something like this?

[Anonymous Coward]

http://www.okcupid.com/

The Rise and Fall of the Roman Empire...

[Anonymous Coward]

"Last week we hired six people who had below a 3.0 G.P.A."

This to me is a clear indication that something seriously messed up inside Google HR right now.

I have a 3.65 GPA and couldn't get an interview at Google New York. And they are hiring sub 3.0s now ? LOL

Either there is a rotten apple syndrome, where one dude is helping all his friends get jobs there and hitch onto the Money Train.

This is the start of a trend, that if unchecked, will ultimately result in bankruptcy of this overpriced POS.

I eagerly await the day when Wall Street bends Google over and destroys 30-40 percent of market cap (billions) in a single bloodbath.

In the meantime my browser default search engine is happily set to Yahoo search !... I haven't missed big G for a picosecond !

Re:The Rise and Fall of the Roman Empire...

[TheRaven64]

Google apparently have a problem hiring competent people in the UK. When I heard this, I (as a PhD student who runs an informal seminar series for other PhD students) fired off an email inviting them to send someone to come and talk to us. The response? Nothing; they didn't even reply saying 'sorry, we don't have anyone we can spare.'

If there's someone from Google UK reading this, perhaps you could comment.

Re:The Rise and Fall of the Roman Empire...

[panaceaa]

Mail me your contact information. We have a nice recruiting staff in London that I can try to get you connected with. Thanks!

Re:The Rise and Fall of the Roman Empire...

[AuMatar]

Well, you're obviously just out of college. Unless you want to go to grad school, GPA is pretty meaningless. It gets you your first job. After that, its your work experience and the fact you have a degree that gets you everything else. 2 years out of college, and recruiters barely glance at your GPA anymore.

Which is the way things should be. GPA works as a qualifier among college hires because you have nothing else to discriminate based on. After everyone is out in the real world, your experience on actual projects is a much better indicator of how you'll do on similar projects for another company.

Re:The Rise and Fall of the Roman Empire...

[Aladrin]

Actually, GPA doesn't even manage to get you a 'first job', either. (I assume you mean 'first job in that field' because if it really IS your first job, you had rich parents AND a ton of luck.)

I graduated with a 2-yr computer programming degree and a 4.0 GPA. I went a year and a half without getting a single computer job at all. GPA did not even get me an interview. The job I finally landed, I got by taking a test at Brain Bench and blowing everyone else away. (Company paid for it... Half personality, half technical. They insisted new-hires do well on both.)

Re:The Rise and Fall of the Roman Empire...

[techno-vampire]

In the meantime my browser default search engine is happily set to Yahoo search !... I haven't missed big G for a picosecond !

...and besides, the grapes were sour anyway.

Re:The Rise and Fall of the Roman Empire...

[AuMatar]

You had a 2 year degree. Thats why. Real degrees, from real colleges in CS take 4-5. 3 if you're taking a shitload of classes per semester. A 2.0 GPA with a 4 year degree trumps a 4.0 with a 2 year.

ALthough I will qualify my original post- your GPA gets you interviews for your first job, not your first job. The GPA will get you an interview, from there its personality, technical knowledge, and interviewing skills.

Re:The Rise and Fall of the Roman Empire...

[imageboard]

I believe this link perfectly explains the situation: http://xkcd.com/c192.html

Re:The Rise and Fall of the Roman Empire...

[adrianmonk]

This to me is a clear indication that something seriously messed up inside Google HR right now.

I have a 3.65 GPA and couldn't get an interview at Google New York. And they are hiring sub 3.0s now ? LOL

The fact that they hire people with below a 3.0 GPA doesn't mean their HR department is screwed up. It simply means they aren't being dogmatic.

For what it's worth, my GPA was below 3.0 when I graduated with my BSCS in 2001. Know why? Even though I typically got about a 3.75 every semester my last 4 semesters when I was finishing up school, in my pre-dropout days back in 1990-1992, I got a few semesters of things like 1.5 and 0.5 and 0.0, because I was too stupid to drop classes I'd stopped going to. For what it's worth, I'm 35 years old now. So my low grades are from 15 years ago. Should Google not hire me because of that?

Oh yeah, and this certainly seems appropriate for this Slashdot topic!

Re:The Rise and Fall of the Roman Empire...

[fishbowl]

>This to me is a clear indication that something seriously messed up inside Google HR right now.

A recruiter for Google called me over the winter break, while I happened to be looking for work. I found and accepted another job before I even knew they had called me. I'm trying to decide exactly how to tell them, so as not to burn the bridge.

I noticed you pointed out your GPA. I've found, from being on the hiring side of the table, that people who think their academic stuff is important enough to talk about, don't tend to have a whole lot *else* to talk about. Maybe that's because you're young and straight out of school. But most people who are looking to hire you, want to know what you *can do*, which is not necessarily related to "how" or even "what" you did in school. I'm not belittling your education, don't get me wrong -- I work in academic research now myself and the university is my life these days. I'm just hoping to share a tiny bit of insight.

"This is the start of a trend, that if unchecked, will ultimately result in bankruptcy of this overpriced POS."

If that opinion is in your head, whether you say it out loud or not, you probably telegraph it unconsciously during the interview. I've had an interview where I knew for a fact that the company was running out of money, was losing one of its biggest accounts, and that things weren't looking so good. I tried to be upbeat and positive, but it turns out that I did a poor job of hiding my real feelings (that is, knowing the truth about the dead-endedness of that job.)

I've also interviewed at places where senior developers don't get to program, they are forced into management roles. Once, I actually said what I thought about that. (I would try to use my management authority to create a technical ladder so that senior developers can do career development *as developers* without ever doing management aside from maybe project management.) Wrong thing to say, I guess, but I don't want a programming job where I have to give up programming. What's the point of all these (many) years I've spent learning to program?

Would you hire Yo Yo Ma into your orchestra and then tell him he's been playing the cello too long, he needs to move on??? (That's exactly how I perceieve this "must move to management" theory, and yes, that example was chosen to illustrate the approximate level where I consider my skills to be -- I'm good enough that I am comfortable telling people exactly what I think, even in job interviews.)

Re:The Rise and Fall of the Roman Empire...

[fishbowl]

"Real degrees, from real colleges in CS take 4-5. 3 if you're taking a shitload of classes per semester."

People who haven't been through it, sometimes do not even seem to understand what CS or Engineering majors *do* to get those degrees. For instance, they don't understand the magnitude of university math and physics. They don't realize how much work goes into theory stuff like discrete math and algorithms and automata, or even understand how much they don't know. They go as far as to dismiss all the theory as useless (after all THEY haven't needed it!) And let's don't even talk about capstone courses, and the amount of work and discipline needed to get through one of those.

Not all schools are equally tough. But in some of them, *very* smart people who actually put in a *lot* of effort, still C's. I got a C in Vector Calc, which crushed me, until I realized that nearly half the people in that class literally failed. I've seen people in higher CS and Math courses getting religion after, say, making a 52% on a final exam where 51% would be a "D" and 52% was a "C" -- and I'm NOT talking about slackers or unintelligent people here!

To my mind, this is really the risk of trying a university degree. The chances of failing are *high*. Some people can afford to re-take a class or two, but for others it's do-or-die.

Oh well, I've retreated to the white tower myself after (mumble) years in the industry. I now work in research as a professional staffer in a university. I'm not a professor, but I look like one, have an office that looks like one, and I might take this as a hint that I need to do a Ph.D. (One of the few really good perks of the job, is free tuition, which if you and your spouse both take advantage of it, is like a $30-50,000 bonus... shhhh... don't tell anybody :-)

Oh well, I rant. Back on topic, when I was in industry, I did make hiring decisions. A person with a junior college degree would pass muster for my part, as long as he could actually demonstrate that he could do the work. Experience is everything there. Catch-22, I realize. But even your 20-something person coming straight out of university should have 2 or 3 years of *something* that can be designated as experience. Come on, you were on a college campus and you managed to not work in the lab, or doing web programming for some department, or even some kind of directed research? I don't care about your *grades*, I want to see the game you designed, or whatever.

Re:The Rise and Fall of the Roman Empire...

[cky625]

Q = A . Hint(just in case):

Re:The Rise and Fall of the Roman Empire...

[Aladrin]

I agree that most people who do not want a 4-year degree do not acknowledge the pain involved with getting them. They willfully ignore the work involved so that they do not have to think of themselves as failures. They aren't failures, but admitting to themselves that they can't (or won't) do that much work is a failure in their mind. The easiest way to cope is to put down the accomplishments of others.

And I agree with the GP post that a 4yr degree with the worst grades trumps a 2yr degree with perfect grades in most situations. Entry level is one of the situations that it doesn't mean much more. It can even keep you from getting the job.

What does a 4yr degree mean to a company? It means they'll have to pay you more money for the same job. If the college/uni is worth its salt, you'll be enough better at the job that it's worth the money. But most entry-level computer jobs are very simple. (The job I got was not entry level. I wasn't supposed to be the only programmer on staff, but a series of events caused this to be. I was responsible for all the software systems that had been developed in-house, updating an old one, and overseeing the outsourced programmers. A bit more than I expected, but I managed well enough that I got a huge raise at 3 months, and another at 1 yr. There are more programmers now and the load is shared.)

As for college experience... The college only has so many labs and departments. If every Programming major tried to do that, most would fail to find anything. I do agree that they should have SOMETHING to show for their time in college, though, even if it's a small game or pointless program. They should be able to show they actually care about programming, and aren't just trying to find an easy job with big money. (Which it isn't.) Open source projects are great for this, as they supply motivation. I worked on an open source game engine for a while, and it was quite rewarding. (I actually just tried to contact them again lately, as I want to develop some ideas with it, but their repository is a bit messed up.)

My point is that while degrees matter a LOT in higher-level job hunting, they don't matter so much in entry-level. Degrees are expensive for companies and they don't actually prove you can do the job. (As proved by the guy whose resume looked a LOT better than mine, but I blew him out of the water on the Brainbench testing.)

Re:The Rise and Fall of the Roman Empire...

[tehcyder]

What's a GPA when it's at home?

Re:The Rise and Fall of the Roman Empire...

[Nick_Psyko]

Yeah I'm a little stuck on that one (GPA). American?

Not important enough to Google for it though :-)

Re:The Rise and Fall of the Roman Empire...

[fishbowl]

"If every Programming major tried to do that, most would fail to find anything."

If I'm looking for the cream of the crop, though, it's an indicator. Having been there myself, I know of many, many areas where people try hard to find skilled and motivated undergrads, are willing to pay reasonably well, and yet have great difficulty finding people.

"I do agree that they should have SOMETHING to show for their time in college, though, even if it's a small game or pointless program."

When I've interviewed (mostly interns) from my own school, I *know* what they should have done, and in most cases, what professors they should have had, and I can evaluate certain choices they made, if they are at the junior or senior level.

It's not really a value judgement, but I perceive it differently if someone took the track that included a course in ML and Ruby, a course in high-performance/parallell computing, the automata course and an additional, grad-level algorithms analysis. (This says "actually wanted to be a Computer Scientist", which suggests a passion for the trade.) Other tracks are equally respectable, but some of them say "chose the major in order to become an Income Maximizer").

I enjoy meeting (and working with) the people who took as much math and physics as they possibly could, and who took the CS classes aimed at the hardcore nerd. But then, I'm a hardcore nerd myself, so I'm biased.

In any case, it will be a long time before I'm on the hiring side of the table again. I've gone back into research (Hydrology and Water Conservation. I live in the desert, and this is a pretty interesting field with a surprising amount of computer programming work involved.) This is a whole new gig for me, so wish me luck! I'm pretty happy to be out of the corporate world for once.

Re:The Rise and Fall of the Roman Empire...

[Aladrin]

"I enjoy meeting (and working with) the people who took as much math and physics as they possibly could, and who took the CS classes aimed at the hardcore nerd. But then, I'm a hardcore nerd myself, so I'm biased."

That's me. I took all of the sciences except for 'Earth Science' at a community college for the optional credits involved. I also took all of the math classes. And I also enjoy talking to people who enjoy these things, as they usually have their own viewpoint, instead of being a parrot.

"In any case, it will be a long time before I'm on the hiring side of the table again. I've gone back into research (Hydrology and Water Conservation. I live in the desert, and this is a pretty interesting field with a surprising amount of computer programming work involved.) This is a whole new gig for me, so wish me luck! I'm pretty happy to be out of the corporate world for once."

I can't say I envy you there, but good luck! I used to live out in the Mojave Desert (little town called Barstow) and I loved the town, but I prefer to be within a couple hours of a good mall. (And the tech/game stores associated with that.) I found (was found by?) a nice 'small' company that is now rapidly growing and actually knows what it's doing, with no dot-com capital. It's not corporate, but it's not 'be your own boss' either... A very nice middle ground.

Nitpick

[mandelbr0t]

I get it, but your style leaves something to be desired. I'd write it as:

if (person.getRace() == Race.BLACK || person.getGender() == Gender.FEMALE) { return 0; }

"Black" and "Female" are values for the Race and Gender properties respectively. They don't work well as method names.

mandelbr0t

Re:Nitpick

[infaustus]

I have a better idea! if ( person.getLanguage() == Language.JAVA ) { System.out.println("gtfo"); return 0; }

Re:Nitpick

[abigor]

Google uses Java extensively, so your suggestion is dumb (assuming "gtfo" means "get the fuck out").

Re:Nitpick

[infaustus]

Why would people mod you up for feeding the trolls? lolz. Perhaps google's algorithm should include something along the lines of: public string test (string company) { if(company.getLanguageProminence("JAVA") > .10 ) { System.out.println("kill yourself lol"); return "FAIL"; } else { return "WIN"; }

Re:Nitpick

[Darth Android]

"string" is not a java class, and I see no imports for a custom class, so you get a compile error. Google's algorithm rejects you for your inability to write java.

Don't be over 40

[Anonymous Coward]

I have applied to google several times for positions I was highly qualified for. At some point during the interview process I always get a weird - sorry answer.
Usually for something that makes no sense.

My younger friends get in with no trouble, and advice me to die my hair etc to look younger.

Nice company, nice people but all very young.

Re:Don't be over 40

[Tablizer]

My younger friends get in with no trouble, and advice me to die my hair etc to look younger. Nice company, nice people but all very young.

I wonder if the ratio of hirees to applications is much lower for older people if a class-action lawsuit may not be in order. Lawyer anybody?

Re:Don't be over 40

[tehcyder]

My younger friends get in with no trouble, and advice me to die my hair etc to look younger.

If you had to do a written test, I think I can guess why you weren't hired and it's nothing to do with your age.

ATS

[Cytlid]

Many IT recruiters already use something like this called "ATS". I revamped my resume, highlighting all my skills. One of the tricks is to squeeze as many tech buzzwords (that you know about) on there. I got a ton of hits, I had to take my resume off all the job sites when I landed a job. (The trick is you have to be clever ... I had both CCNP & CCIE on my resume ... literally stating, I have a CCNP certification, partial way to CCIE. This is perfectly honest and correct.) It works well for system admin who can have plenty of words on there, if you're generally a jack-of-all trades. I didn't put say, Java on there because honestly I know practically zero about java.

I'd be willing to bet if I filled out their survey, they'd um, "find" me. For me, it wouldn't be a programming position though, it would have to be networking or general system administration.

I'd love to work for Google, but I don't believe they have an office anywhere around where I've recently moved and settled to (and bought a house).

20,000 vs 200 x 100?

[cei]

At the height of the dotcom bubble, Bill Gross & Idealab! had the philosophy that no company should have more than 100 employees. If your business model got above 100 employees, there was a high likelihood that you were better off dividing and spinning off other business units. (Don't know if they still preach that or not, but that was the thinking "back in the day.")

I don't know that Google would be better served as two hundred smaller companies, but at the same time, it's hard to imagine managing 20,000 employees would be any easier.

Re:20,000 vs 200 x 100?

[EveLibertine]

While there is an argument as to whether or not the 200 x 100 actually provides and increase vs. the 20,000 scenario, it's also important to keep in mind the strength in branding. So, spinning off into smaller companies, however more or less efficient, would be counter-productive if those smaller companies cannot benefit from the google namesake. Contrary to Idealab philosophy, there are benefits to being a single large entity.

Re:20,000 vs 200 x 100?

[cowscows]

So would these guys propose that FedEx start a new company for every 100 of their delivery drivers? How about their warehouse workers? What about the mechanics who help maintain their vehicles? They'd have hundreds of companies, the logistics of all of that would be insane. Coordinating all of them together?

I haven't read anything about their philosophy other than what you just shared, but it's hard to take seriously any sort of one-size-fits-all solution for something as broad as "all companies."

Re:20,000 vs 200 x 100?

[Gumbytwo]

I'd like to see 100 employees maintain the hundreds of thousands of machines in Google's datacenters. That'd be awesome.

But seriously, Google is built on the concept of scalability; even at the management layer.

Re:20,000 vs 200 x 100?

[mattwarden]

A lot of things like this are true. For example, a lot of company owners would be much better off selling all their assets and investing the money in government bonds. Often this alone matches or exceeds the current growth levels, with one very interesting advantage: the owner is now free to do other work (opportunity cost).

I find the fact that companies don't do this one of the most interesting psychological phenomena ever. Just about every other decision in business comes down to which of two numbers (profit) is higher (of course, some people are really, really bad at estimating these numbers and ignore otherwise obvious costs like turnover cost, but I digress).

Re:20,000 vs 200 x 100?

[Nappa48]

You are actually questioning Googles ability to manage 10,000+ "units"?
Pffft, you crazy foo'! Its Google, thats like counting change to them!

Re:20,000 vs 200 x 100?

[phlipped]

The company for which I work pretty much does this internally ...

Every department/group is considered a separate billing unit and must justify a profit in their own right.

The sales departments sell services to external customers. They then arrange for the work to get done by 'hiring' the services of any of the specialty groups inside our company. Presumably, the cost of the internal services is lower than the cost to the external customers, so the sales department makes a profit.

Even the departments which exist purely to facilitate the work of the "real" departments (eg building management, internal IT infrastructure, phones, intranet website) all make a "profit". They do this by charging every other department a small, ongoing service fee, rated against the number of employees in each department.

Ultimately, each manager ends up running a micro-business providing a specific service to other departments inside the main company, and they have to balance their revenue (as collected from other departments), and expenses (out to other departments and employee wages) to maintain a profit.

Re:20,000 vs 200 x 100?

[cei]

Do you think 100 employees could maintain one datacenter? How about one floor of a datacenter?

Re:20,000 vs 200 x 100?

[sacrilicious]

The sales departments sell services to external customers. They then arrange for the work to get done by 'hiring' the services of any of the specialty groups inside our company. Presumably, the cost of the internal services is lower than the cost to the external customers, so the sales department makes a profit.

Very cool idea. In the scenario above, is there "competition" within the company to fulfill the sales department's needs? If not, what leverage is wielded by the sales group when negotiating price? Or is there some other mechanism used to determine the cost?

Re:20,000 vs 200 x 100?

[w9ofa]

Well actually,

Every driver for FedEx is actually a subcontractor working for a seperate company, not an employee.

So yes, in fact, every time they get a new van they are making a new spin-off company.

Corporations

[highwaytohell]

Are we going to see this being sold to corporates? This could potentially be a money spinner for google if it works well enough.

If corporate HR could ask the questions to suit its own corporate profile, and google allows this technique to be tailor made to suit any corporation, then i don't see why this wouldnt be a success.

Why man holes are round

[c0d3r]

Man holes are round because if they were square, they'd fall in the hole, Mr Fuji, and I'd move the mountain with smoke and mirrors or perhaps optical rays into retnas.

Re:Why man holes are round

[dgatwood]

That's actually a common misconception. A square would not necessarily fall into the hole. It would only fall into the hole if the corner-to-corner distance of the interior lip that supports the plate were less than the length of one side of the cover plate. It is only necessary to make the width of the lip at least w * (1/2 - 1/sqrt(2)) where w is the width of one side.

Of course, that comes out to about 14.6% of the width, which means the interior of the hole would be just over two thirds the width of the top part of the hole. That means that the area of the hole would only be about 72% of the area of the plate that covers it. By contrast, with a round cover plate, the area can be nearly 100% of the cover plate's surface area. That is why manholes are round.

Re:Why man holes are round

[dgatwood]

Oops. I forgot to multiply the lip width percentage by two before subtracting from 1. That comes out to 14.6% of the width from each side, or 29.2% of the total width, and thus with a square cover, in order to ensure that the cover cannot fall into the hole, the area of the actual hole (below the support lip) could be no more than approximately 50% of the area of the cover. :-)

Re:Why man holes are round

[Pulse_Instance]

During one of my recent interviews they told me that my answer of the manhole cover being round to keep it from falling in the hole was wrong. I said no I'm pretty sure thats right, having never thought of the question before I stuck to my guns turns out there are a ton of answers to the question, including, one person to move as you can roll it, the tube that it covers is round and many more. I was offered the job but turned it down for other reasons.

Re:Why man holes are round

[vakuona]

So you are agreed that it is not a commom misconception. Theoretically, a round manhole cover only need a lip of less than 1mm, (depending on tolerances), whilst you need a ridiculously large lip to prevent square manhole covers falling into the holes. It's much harder to explain to someone who does not usually have an inclination to that sort of thing. And it usually suffices.

Re:Why man holes are round

[dgatwood]

It's a misconception that anything other than round would fall in by its very nature. Squares would fall in without a large lip, but it is certainly possible to make one that's square without it falling in. It just results in very poor space utilization. Actually, you could do a lot better than 50% utilization by making the inner lip round instead of square. That would increase it by few percent to about 79% of the size of the cover, if my quick-and-dirty math is correct.

Reuleaux triangles are another shape that exhibits the same properties, as are similar variants of any other odd-sided polygon. And odd-sided polygons are an improvement---a triangle would be 56% utilization. It doesn't do as well as the square with a circular hole, though, at only about 60% usable area. For a pentagon, the circular hole gives 86%, and I won't even attempt the math on the pentagonal hole. :-)

That said, if I had to guess the original reason, I'd imagine it went something like this:

Worker 1: Damn, these things are heavy. My back is killing me. I sure wish I didn't have to lift these heavy manhole covers.
Worker 2: You know what would be cool? A manhole cover with wheels.
Worker 1: Wouldn't it be easier to make it round?
Worker 2: Sure, if you want to be lazy about it.

:-D

Re:Why man holes are round

[Anonymous Coward]

So why aren't they Reuleaux triangles?

Re:Why man holes are round

[TexasDex]

I know how to make a manhole cover that is square (helps prevent theft I believe) and yet won't fall down the hole, without wasting as much space. Make the cover square and the hole itself round, with a diameter just a tiny bit smaller than the length of the side of the cover. That way the wasted space is limited to the corners outside of the circle.

There are probably better ways to keep manhole covers from being stolen (locks?) but this would work.

In a related story...

[notnAP]

... Google searches for Myers-Briggs have shot through the roof.

New Algorithm?

[mmxsaro]

Are they now going to implement a new algorithm called PeopleRank(TM)?

Re:New Algorithm?

[Anonymous Coward]

You mean PeonRank :)

No news

[frisket]

What's the big deal? Uncle Sam has been doing this for over a decade, sifting the applications for review using one of the major statistics packages, who developed special text-reporting tools for doing so.

Isn't there already something like this at ....

[Anonymous Coward]

....http://www.aperturescience.com/?

try filling out their application!
the password is 'portal'

Sounds like what the US Military does

[kevinx]

This sounds pretty much like how the military does it.. although the military is obviously less choosy about accepting people and more interested in where to place them. They also have surveys you take once you are in that are paired with your military history. Results are used as indicators for future enlisted; or so I've been told.

Good news

[Anonymous Coward]

As somebody who recently went through their hiring process, all I can say is that compared to what they're doing right now, anything is an improvement.

Re:Good news

[Anonymous Coward]

I suspect the biggest change you'd like to see would be them actually giving you the job.

the code..

[Mogster]

bool recommend = true;

if (surveyResults == Evil())
        forwardResumeToMicrsoft(bool recommend);

hire();

Re:the code..

[Jussi K. Kojootti]

That's still Beta, I hope...

Re:the code..

[Anonymous Coward]

bool recommend = true;

if (surveyResults == Evil())
                forwardResumeToMicrsoft(bool recommend);

hire();

Um... I hope that you don't plan on working for Google cause that's got 1 syntax error and 1 logic error in 4 lines of code. I'll leave the syntax error as an exercise for the reader. The logic error is that you still hire the person with the evil resume - unless of course that was your plan all along and you're just forwarding the resume to Microsoft as a big f*** you to Balmer (after all, they did buy evil from Satan - is Google trying to steal that away too?)

I am not an algorithm...

[Kazoo the Clown]

I'm a HUMAN BEING...


No, really!


-- (alms for the lameness filter)

Re:I am not an algorithm...

[someone1234]

Right, you are not an algorithm, you are just data.

For a second there...

[cliveholloway]

... I read that as (Steve) Jobs, figuring the Apple/Google relationship had gone sour and that Jobs had better not bend over in the shower to pick up the soap.

I really am sick today (*atchoo*)

Or..

[ClosedSource]

They could just flip a coin.

This is the sort of thing that happens when a company has tons of money and can't figure out what to do with it all.

A telling comment:

[chris_mahan]

(last line of TFA)

> "More and more in the time I've been here, we hire people based on experience as a proxy for what they can accomplish," he said. "Last week we hired six people who had below a 3.0 G.P.A."

Arrrgh! It's like saying: "Last week we hired six people who weren't white."

Augurs poorly for GOOG.

Re:A telling comment:

[Anonymous Coward]

88?

Re:A telling comment:

[pflinton]

"More and more in the time I've been here, we hire people based on experience as a proxy for what they can accomplish," he said. "Last week we hired six people who had below a 3.0 G.P.A."

The rest of the sentence was "and someday they might be good janitors."

Re:A telling comment:

[chris_mahan]

I'm not getting that ref, so I'll decline to state.

Re:A telling comment:

[Isthisagametou]

So what you're saying is that only non-whites would get a GPA below 3.0. Augurs poorly for you. The point is: Don't take things out of context.

Timothy Leary

[machine of god]

Reminds me of the story about timothy leary taking his own psychological tests when he was admitted to prison. He got away.

I see it now....

[8127972]

..... After it concludes you won't fit in at Google:

"I'm sorry Dave, I'm afraid I can't hire you."

Why stop at Google?

[starglider29a]

Disclosure: I am an unemployed web developer, and I've never actaully READ PKDick. Only half of this rant is serous. I'll let you decide which half

WOW! My house must be haunted by the ghost of Philip K. Dick! I just got a chill when I read TFA! The first thing that came to mind is "Great... my employability at Google is reduced to the same as a FICO Credit Score."

But then I wondered how long until it grew into "the Google Overall Diagnostic Heurism for Employers to Assess and Determine Suitability" (GODHEAD) where employers would request a "GODHEAD Score" on an applicant. "Well, Mr. Starglider... you're resumé is very impressive, you seem to match our culture, and your references beam about you, but... I'm afraid that your GODHEAD score is too low."

It is difficult enough in the IT world to convince an HR person (who doesn't know the difference between Java and Cocoa) that you are a "good fit" without giving us a one-dimensional point over which to jump. In a world where job listings ask for "UNIX Admin with VB.NET experience and Flash/Photoshop" having one more reason to reject you is NOT what this job market needs.

But maybe I'm being paranoid... Maybe this is a good thing... maybe an entire industy will spring up, coaching people on how to "spoof" the test. GOOGLE Surveys for Dummies. ScoreRunners that "retire" people whose GODHEAD scores drop too low. And what about GODHEAD Score (a la identity) theft? And one could ask of the HR people administering the test "Did you ever take that test yourself?"

And finally, it sounds from TFA that this score is not a very mutable score... it's a test of who you are. It won't change much over time. Maybe after a few tests and the transients level off, you could have it tattooed into your right hand or forehead ;-)

I'm thinking of a number... between 350 and 850... know what it is?

Re:Why stop at Google?

[zuiraM]

There is a difference between paring things down to a score and doing a complex inventory and evaluating it along several axis to find out whether there is a spot for you in the company and, if so, where it is.

Also, what the industry really needs, actually *is* one more reason to reject you. The right reason.

Don't get me wrong. I hope you find a satisfying job. But the most important thing for an employer is, in many cases, to avoid hiring the wrong guy; most of them don't even seem to realize this. Once cruft (in the form of "bad" employees) starts to accumulate, you lose the edge and have to fall back on the usual "10.000 monkeys and a strategy to limit the amount of damage they do" approach of using money as a substitute for the skills of now interchangeable employees.

I've lost two jobs over the years, both times due to incompetent people elsewhere, and both times costing the company money and necessitating a rethink. I also quit two due to being held back by superiors who should never have been hired in the first place; one company went tits up, the other lived on but experienced no further economic growth.

The best way for the "right" people to get the jobs they want, is for the "wrong" people to be rejected, leading to more realistic market pressure (competence, as opposed to sheer numbers) and better results (which translates into further expansion and more jobs).

How do you feel about shootings?

[Anonymous Coward]

"The point, though, is that the test (which I originally blew off as ridiculous HR fluff) was shockingly accurate about me."

Considering the slashdot crowd's attitude towards the soft sciences, as well as anything business, eg. HR, marketing, etc. I'm not surprised by your surprise. Also here's something to keep in mind. Remember all those business shootings? Post office shooting? Even school shootings. Complain about privacy concerns all slashdot wants, but it's cheap insurance to lose a few in the interview compared to the losses in a workplace shooting. Let alone the daily friction an incompatible worker creates.

Quality of hires

[teal_]

I'm shocked that the company hasn't yet started to fade or lose its reputation as a congregation of geniuses, given that with all the reqs they're having to fill, they're bound to be hiring in a less discriminate fashion than they used to. Those new lesser employees in turn conduct interviews, which begets another batch of lesser employees, until eventually you hire just about anybody with a CS degree. Meanwhile, your founding geniuses cash out their millions and go live in Hawaii, leaving their jobs to be filled by lesser talent. Ultimately this leveling of talent begins to show in the quality of your products, which in turn leads to a decline in your company's reputation, and before you know it Google is another bloated bug ridden software company that gets its daily dose of malignment on slashdot.

Re:Quality of hires

[ClosedSource]

There's a big difference between the "reputation as a congregation of geniuses" and the possible reality of it. We've never really known if this was true or not or whether a collection of geniuses increases the probability of having a great company. We know Google is successful, has a lot of money, and talks a lot about how smart they are (explicitly or implicitly), but other companies have done just as well without the genius claims.

The Dot-Bomb Trap

[rudy_wayne]

"Google's vice president for people operations, sees no reason the company won't reach 20,000 by the end of the year. This will mean hiring something like 200 people a week, every week, all year."

Google is falling into the same trap that has hurt so many companies. Right now, profits are high. The cash is rolling by the billions. As a result, nobody (in Google management) is questioning why they need to hire 200 people every week, nonstop for a year. There's plenty of money to pay everyone, so there isn't a problem.

But eventually, profits will level off and then start to decline. Nothing goes up forever. And when the money gets tight, Google will suddenly realize that they've got a whole bunch of people that they don't really need.

Re:The Dot-Bomb Trap

[starglider29a]

No problem. Then, they can devise an algorithm to decide who to lay off.

Re:The Dot-Bomb Trap

[Tablizer]

No problem. Then, they can devise an algorithm to decide who to lay off.

"Ooops, we accidently laid off our termination logic expert."

Re:The Dot-Bomb Trap

[ZeroFactorial]

That kind of power doesn't belong in the hands of mortals....

"We'd like you to write the code that determines who to fire."
or better yet...
"We'd like you to write the code that determines who to promote to what position."

I've always wanted to be a CEO... :)

Hopefully .....

[nblender]

....this will stop the relatively constant stream of unsolicited recruitment letters that I and many others in the OSS world are subject to. Each time I get one I forward it to abuse@google.com (which nets me a canned response) and respond to the sender telling me to remove me from their list. Apparently Google are like every other spammer in that these requests are ignored and the UCE continues unabated.

Re:Hopefully .....

[anticypher]

The solution to this is to interview with them, and somehow screw up.

Google has a strange recruitment process, they never ask what you are currently doing or where you live, they just find a few old web pages in their cache and assume they're current. It was on the 5th interview when the Google interviewer suddenly realised I wasn't a programmer, but I knew enough CompSci to have struggled through 4 interviews. They had the idea I was a major F/L OSS programmer based on all my activity in mailing lists, not a guy just helping test one project. They had also found an old Irish mobile phone number that forwards to my current phone, and assumed I lived in Ireland.

After a few mumbled promises to send my current CV to the right group, within hours I received a "No Job For You" form letter and I seem to have been put on a black list internally. The stream of recruitment emails have trickled off to maybe one every two months.

It's funny, because I run into senior Google people at trade events who try to recruit me because they know my reputation. When I tell them I've already been rejected for a junior level programmer position in an HR blunder a couple years ago, you can see their faces fall. They know that once Google rejects someone, there's little chance of getting them in past HR, but some senior guys are working to reform their broken system.

Getting rejected is a great solution if you never want to work there and limit those spammish requests. Since they are offering you a job, tell them you want to be head of HR ;-)

the AC

Another voodoo interviewer...

[Vellmont]

If the come to the interview dressed like crap, they're automatically out. If they turn up late, they're automatically out.

It's facinating to me the utter-crap voodoo that some people having in making hiring decisions. People like yourself actually believe there's these simple little tests that seperate the good from the bad.

Did you ever consider that all you're doing is just trying to hire people like yourself? You may think that's a great way to seperate the good from the bad... but you may eventually discover that any workplace relies on a variety of people with different personalities, attitudes, and "views of the world". Hire too many people like yourself, and you might just wind up with a bunch of people that can't see outside of the box you've built. If you want a perfect example of this problem, look no further than the Bush administration.

Re:Another voodoo interviewer...

[Dunbal]

People like yourself actually believe there's these simple little tests that seperate the good from the bad.

      Ok, so I should hire the guy who turns up 45 minutes late for the interview - whose excuse is "it could happen to anyone", who has a pierced eyebrow and orange hair, and who has no self esteem? Do you want this person touching your children? I work in health care, remember. Are you implying I should hire the first person who turns up for the job and not make any screening attempt?

      Oh I agree, he might be a really nice guy despite the way he looks. I agree that appearances can be deceptive. I also understand that first impressions are important. I have actually had patients tell me in the emergency room "Doc, I do NOT want that person near my wife" when they see some of our nurses/med students with more radical, expressive attire/jewelry.

Did you ever consider that all you're doing is just trying to hire people like yourself?

      Have you considered that perhaps I am hiring people who I think are best suited for the role I need them to fill? Oh, perhaps I'm a bad judge of character. In that case my team won't work, and I'll end up in bankruptcy sooner or later. It's my perrogative as an employer to hire the people I want to work with and build a team the way I think it will be more efficient.

Re:Another voodoo interviewer...

[Vellmont]

Are you implying I should hire the first person who turns up for the job and not make any screening attempt?

No, I'm saying that making snap judgements based on someone showing up late, having the wrong hair color, or "dressing like crap" isn't a valid way to discriminate. You sound like you're very judgemental based on things that have little to do with anything. Self esteem? You can tell that just by looking at someone?

Have you considered that perhaps I am hiring people who I think are best suited for the role I need them to fill?

Yes, I just think you're making bad decisions based on things that don't matter (and probbably related to your own personal prejudices).

"Doc, I do NOT want that person near my wife" when they see some of our nurses/med students with more radical, expressive attire/jewelry.

So it's your job to validate peoples prejudices by allowing the same predjudices in your hiring decisions? So if people came up to you and said they didn't want that black/chinese/woman/man/jew/mexican/homosexual near their wife you would suddenly stop hiring them?

It's my perrogative as an employer to hire the people I want to work with and build a team the way I think it will be more efficient.

I'm not questioning your right to make dumb decisions. I'm just pointing out they don't appear to be based on any basis of reality.

Re:Another voodoo interviewer...

[mollymoo]

If the come to the interview dressed like crap, they're automatically out. If they turn up late, they're automatically out.

It's facinating to me the utter-crap voodoo that some people having in making hiring decisions. People like yourself actually believe there's these simple little tests that seperate the good from the bad.

In a service industry, turning up late and looking like shit has an economic cost - it loses you customers. Therefore being late and looking like shit makes you bad at the job. That is not voodoo. In any industry being unreliable has an economic cost. What industry is there where saying you'll do somthing by a certain point in time and not delivering not a problem? They even get tired of that in academia eventually.

Have you ever actually been an employee or employer?

Re:Another voodoo interviewer...

[ahuard]

It's my perrogative as an employer to hire the people I want to work with and build a team the way I think it will be more efficient.

I agree absolutely. When a coach has to pick the players on his team, he picks those he thinks will "mesh" with the rest of the group.

What I believe the problem slashdotters have with your post is whether you are hiring personalities based exclusively on outgoingness and (dare I say it) extroverted habits or if you might consider the more typical introverted slashdot personality in a viable candidate.

It scares me to death to think that employers may be prejudiced based on personality. I am an introvert like many of my fellow slashdotters, but I can speak up when need be even though I may still come off as softspoken. In other words, I'm not salesman material at all, which I hope wont be needed in my future career as some sort of technician or engineer.

All other concerns aside, would you automatically disqualify or hold it against the candidate for having an introverted personality like the one I described? If yes, would you consider your line of business exclusive to extroverts only?

Re:Another voodoo interviewer...

[ultraexactzz]

If personal appearance and the ability to show up at a given time are important to the position for which you're interviewing, wouldn't it make sense to disqualify applicants who don't meet those requirements?

Yes, it's a first impression - but isn't that what the interview itself is?

Re:Another voodoo interviewer...

[dcarey]

Your comment: "People like yourself actually believe ..."

Then 2 posts later:
  "You sound like you're very judgemental" ... "and probbably related to your own personal prejudices"

Sigh.

Brains aren't the only thing that are important to being a good employee. A good attitude and being assertive while also being considerate of others wins hands down, every time.

If I get a spoiled brat prima donna who thinks he knows everything about Python in an interview pitted against someone who loves coding, I'll take someone who loves python and who is not a social nuisance any day. Someone who loves to learn and works hard because he or she has to instead of having the world handed to them and who does not look down upon other employees is golden to me. Tell me, how exactly am I supposed to sell a software product using someone who is rude an condescending to my customers?

Likewise, original poster has the full right to make decisions that are completely correct for his business. He needs good personality in his business, which he is defining by being prompt and personable. If you can't make that impression on the very important first meeting, how can he possibly have any confidence in the employee? Just pretend he'll never be late again, or will suddenly start dressing well? Hm, that type of hubris seems familiar ... the cowboy attitude ... hm ... think I'll quote your original reply to the original post: "If you want a perfect example of this problem, look no further than the Bush administration."

From my experience...

[haggie]

The algorithm goes something like this: 1.) Name the Ivy League school where you received your MBA. 2.) Did you do both your undergrad and grad work at Stanford? 3.) If female, is your blonde hair natural? How much over 5'6 tall are you?

More precisely...

[Gorimek]

It is illegal to ask some questions in an interview.

Not exactly. What is illegal is to discriminate in hiring based on certain criterias such as race, gender, age etc.

This usually leads to company lawyers issuing rules forbidding employees from asking about this, on the theory that if the company doesn't know your age it couldn't possibly discriminate based on it.

But in practice, you usually need more to win such a law suit than just claiming you were asked a certain question. And it's never a legal problem to ask people any questions, as long as you offer them employment.

Finally, all this pertains to the USA. Other countries have other rules, and Google are hiring in many of them.

Interesting Boast...

[nick_davison]

The company has apparently doubled in size in each of the last three years.

The company boasts that only 4 percent of its work force leaves each year, less than other Silicon Valley companies.

Let's think about this for a moment:

4% of current numbers leave each year.

Except the overal job pool was only half that size a year ago... So 8% turnover in a year relative to the size of company it was at the start of the year?

Go back two years and it was half the size again... So 16% turnover relative to the number of people that were there only two years earlier?

And they say it's doubled every one of the last three years so we know 4% of current numbers is actually relative to 32% of all those who were actually there three years ago.

Now, to be fair, not everyone necessarily stays a full three years. So maybe half of the last year's 4% (making 2%) is people who were hired within the last three years and the other 2% is people who were hired at least three years ago (when the pool size was 1/8th current levels). So we're looking at, what 16% turnover for people who've been there 3+ years as well as those who simply didn't make three years.

Sure, these numbers are pulled out of my ass. They still illustrate that the stellar 4% turnover claim really isn't worth that much when you're conveniently glossing over 50% of the people here this year haven't even put in a full year's work yet, 75% haven't put in two years and almost 90% haven't yet been there for three years - about the time most people with an eye on their careers and wanting to avoid a reputation as job hoppers would consider looking anyway. All of a sudden, that 4% really isn't much to boast about.

If you think about it, it's a lot like failing to grasp compound interest. Maybe they should ask simpler questions in their intereviews rather than the legendary entertainingly abstract ones that evidently miss out on basics. ;)

Do You Have a Pussy?

[Anonymous Coward]

How is that NOT taking gender into account?

BETA...

[sachu]

I think one should wait till 2010 to apply for job at Google... as this "algorithm" will stay in beta phase for at least 2 years!!

You missed

[xant]

The point the GP is making: All of these questions, taken together in some black box computation, may be able to answer the *other* questions which are illegal to ask individually. To grossly oversimplify: If I ask you the last digit of your age, and then I ask you who was president when you were born, I know your exact age. Nobody would argue that these two questions asked together should be legal.

The GP is probably right that this survey and algorithm could, with the correct calibration, figure out someone's age, hence the debate, here and perhaps in the courts, over legality.

So, what if the computer is *actually* using a neural net that automatically selects young candidates? No human would necessarily know this. Such algorithms aren't really inspected by humans, they're more or less grown in a vat. Is it illegal to use a piece of software that, internally, figures out someone's age and then hires or rejects them with age as the primary factor? Sure. Is it still illegal if neither the people who wrote the software nor the people who use the software know it does this? Well, there's your lawsuit waiting to happen.

Faking

[atlacatl]

This type of profiling can be figured out, i.e., people will start faking answers to fit the "google" profile.

Interviewing is not the best way, but so far it has worked for a lot of companies.

Re:Faking

[apt142]

To be frank, interviewing sucks as a method for acquiring new employees. The only reason that it has been used to so extensively across so many companies is because it's a major Cover Your Ass technique.

Psych studies have shown that you get the best candidates for the job through skill and performance tests. However, you get unintential descrimination through it as well. Via certain demographics having better educational opportunities than others.

Interviewing tends to create a culture of similar people. Interviewees tend to like people most like themselves. Also, they can always change their decisions to keep them out of profiling trouble by periodically taking a candidate who may be sub-optimal skill-wise but keeps the company within a certain range in relation to demographics.

only 4 percent

[Anonymous Coward]

of its work force GO HOME at night. The rest stay at the lovely Google Resort and Spa, formerly Ricky's Hyatt.

Heck, with the expansion, it's the only 200 story high building in the valley.

Regards,

Ricky (I cashed out)

They lower standards because people are leaving

[Anonymous Coward]

Posting anonymous for obvious reasons. After 4 years of vesting I am no longer with the company, but still good friends. Google was an awesome place to be up to last year, but anybody living in the Valley can tell you there are quite a few of xooglers, and for fresh graduates (especially good ones) it's just not a sweet option anymore.

First off, everybody is hiring in the Valley nowadays, so who'd you rather go with - a startup that's offering the same salary but much more attractive options, or a company with $483.26 options, if you started today? I guess you can still cash out and hit it big, if the Google stock hits $900 within the next 4 years.

Second off, every non-engineering area at Google is bordering on disaster. HR, Legal, Finance, Procurement, Travel and other departments are run by people who are Rhodes scholars, but they are freaking incompetent at running the organization day-to-day. The HR VP made a decision to hire only out of top tier universities, and then they found out Stanford Business School does not have HR graduates, but still stuck to the policy, so if you're a Googler and ever tried to resolve a paycheck inconsistency or change managers, the work was probably done by a Stanford history major with no experience in finance. I am not kidding. Or did you ever buy a Prius taking advantage of the 5k grant, and then not get the check? Trying to resolve that is like calling a freaking AOL support line, except the accents are thicker.

Third off, a lot of new hires nowadays will be maintaining and supporting the code, not writing some supercool new stuff as they think they will. There're quite a few groups, where innovation is still relevant, like bigtable, but most likely as a new hire you'll be supporting some code someone else wrote long time ago, and happily retired after the IPO. Or you will be doing some insignificant pet project like Firefox bookmark sync, but getting a new project on the way, from what I know, is near impossible without the founders' support.

Fourth, as of few months ago the indulgent snack rooms with healthy food that kept everybody raving are gone. Welcome to the "entertainment rooms" with large plasma TVs in them. Still good, but the company dropped organic milk, dropped cereal (no more breakfasts), and pretty soon will be dropping most of the expensive things (expect the fresh fruit rack and free smoothies in The Slice Cafe to go away next). Which makes the perks still good, but not great.

Five off, don't let yourself suckered into one of those internal software development groups. The level of engineering hiring for those departments certainly raises eyebrows. One guy I happened to sit at lunch with was surprised that adding indexes to the columns sped things up in mysql. If you ever need something done, no one will take the responsibility, everyone will point to the manager, and endless inane meetings. I understand this is the way most companies operate, but this was something we at google for too long considered "corporate", and unfortunately it's becoming more commonplace.

Re:They lower standards because people are leaving

[b0s0z0ku]

The HR VP made a decision to hire only out of top tier universities, and then they found out Stanford Business School does not have HR graduates

Did they really? How fucking stupid can such a decision be? I know plenty of people who went to Rutgers (state university in NJ) for undergrad. and are *extremely* bright. Their parents just didn't have $100,000+ to blow on a top-tier private or out-of-state education, at least not without selling their home or dipping into retirement. Not did the kids want to leave school $100k in debt.

-b.

Re:They lower standards because people are leaving

[hauntingthunder]

To Be brutaly honest HR doesnt attract the brightest bunny's as direct entrants does it.

Another eye-contact interviewer...

[Anonymous Coward]

"It's facinating to me the utter-crap voodoo that some people having in making hiring decisions. People like yourself actually believe there's these simple little tests that seperate the good from the bad."

Actually there is psychological research demonstrating that people can tell a lot from some simple interaction. You don't need to run people through the wringer for those things that people are good at. You will need to go in more depth for the other aspects. e.g. technical compentency.

They're looking for the Teela Brown gene...

[Anonymous Coward]

They want employees who are unnaturally lucky...so they set up their hiring process as a lottery!

The hope being, some underqualified clod is going to land a plum job at Google, just because the the alog picked him, and s/he will end up making some incredible serendipitous contribution...would make a great screenplay, come to think of it...

Dehumanising

[syousef]

Way to make people feel like their personal contributions don't matter. Right from day one. Amazing. Wait till Google's stock does crash (as it must) and watch the sentiment get repaid as people dessert.

Of course they don't know anything about hiring

[Anonymous Coward]

Google's own video service, created by internal employees, was not doing well, so they acquired an external company called YouTube. Makes you wonder about WHY those internal employees were hired and failed at their job, and then Google had to spend $1.65 billion for YouTube.

Google didn't invent blogging, but acquired Prya Labs.
Google didn't invent "Picasa", but acquired it from IdeaLabs of Pasadena.
Google didn't invent "GoogleEarth", but acquired its inventor, Keyhole Corporation.

See the pattern?

Google's internal employees just can't cut the mustard. They are riding on the coattails of the success of Google Ads (itself a licensed technology from a patent by Overture now owned by Yahoo). If Google needs to make news with technology, it's only because they acquire it, not because it's internally developed. And if it is internally developed, it's almost perpetually in "beta" mode. (Need I remind the GMail fan-bois of the recent problems that GMail has caused for its users?)

On Balance

[NetSettler]

Google's test will obviously avoid asking any direct questions about age, gender, and race, because that's illegal (even when objectively justifiable). However, if the test is powered by a statistics engine drawing a database of past performance reviews, then the test could unintentionally evolve to ask about such things indirectly.

I can't believe they would deliberately make decisions on the basis of anything that was not obviously going to help them. First, they are a global corporation, so institutionalizing a lack of diversity would seem suicidal. And second, leaving someone who could do something cool deliberately on the sidewalk is an invitation to them to start a competing company that does better. So I have to believe they have a genuine desire to grow.

On the other hand, while they might not do something like that deliberately, anyone could do it by accident. People have built random number generators that turned out not to be random. People have built perceptron recognizers for tanks on a battlefield that turned out to be recognizing the time of day the pictures were shot rather than the tanks. People can confuse themselves with their own "intelligence".

The weird thing is that they say they chose to use their own data to seed their algorithm with their own people. If they already have such people, why wouldn't their present hiring practices be fine for finding them? I heard a talk by Amar Bose of the Bose corporation where, among his several messages, was a catch phrase "better implies different". So if Google wants to grow and become better, patterning its growth on "more of same" seems bizarre.

I've also not seen ethical guidelines published by Google that says they're afraid to use their own data. Perhaps they do or perhaps they don't. But absent clear promises not to use data in certain ways, I'm not confident of what they're doing. Surely they receive search strings from people typing to computers at successful companies they admire and would like to emulate. A lot can be learned from examining those strings in the aggregate, I'd bet. (Even if they didn't work back from the IP addresses, they could cross-correlate the searches against "anonymous" information about "all searches from sites that seem business-related" and get similar results that were at least superficially "ethically cleaner"... though it's still second-hand use of data that others who don't own search engines don't have access to). And surely they must have their own internal search data (things their employees have typed) and the results of these profiles they asked for from their employees, too. So they can create a psychological map of the areas their employees inquire about and compare it to what the world is interested in. Surely a cross-match of that will reveal "interests" and "skills" and "areas of inquiry" and other useful stuff that they could beef up on in hiring in order to see and shore up their "weaknesses". Surely something like that would be more likely to reveal what they need to hire for. Not that I think it ethically a good idea, but given that they haven't promised not to, somehow I'd be surprised if they weren't utilizing that vast quantity of knowledge about what people search for in order to know what to hire next, if not what research areas to go into or what products to develop. Search engines already count the number of searches for various things and correlate them to events and products to find out the popularity of all manner of things in today's fashion culture. Sometimes that data is just for coffee station chatter (e.g., "more people searched for thus-and-so sport at this year's olympics than last"), but eventually (or behind the scenes already) it may be more (e.g., "people are asking awfully specific legal questions about thus-and-so kind of genetic research at thus-and-so company")...

I've discounted the hypothesis that, like the "all volunteer" US Army, they're having so much trouble getting v

Re:On Balance

[delinear]

If they already have such people, why wouldn't their present hiring practices be fine for finding them?

The hiring process at Google is very long-winded, involving several interviews (some people claim a dozen or more) and I seem to remember reading that Brin and Page like to be involved in the process personally, maybe vetting the final candidates if not directly speaking to them. That's simply not scalable if they want to maintain the same personality mix they have now while doubling personnel every year - as the number of people with hiring ability grows, there are bound to be branches of new employees hired that have differing personalities, experience, qualifications, etc. Everyone has their own view of what is the most important quality to bring to a role, and this will weight their hiring decisions.

In light of this it would seem to make perfect sense to the Google boys to solve the problem the way they tried to solve search, with an algorithm. If the same algorith is applied universally to vet the vast majority of applicants (and then the ones shortlisted are interviewed by real people maybe), they're likely to hire people with the same general outlook. If they can tweak that the imitate the mix they already have and like, maybe this could work and be scalable as the company expands.

I wonder how long it'll be before we see Job Application Optimisation company's springing up, though...

Oh Great...

[Sir_Lewk]

Now instead of computers taking my job, they're hiring me!

GOOG isn't "surging"

[Anonymous Coward]

Hasn't gone any-damn-where since before the latter part of October

Also, 20% of employees are not happy there

[Anonymous Coward]

In the last internal survey, around 80% of employees reported to be happy working in Google. That indicates that 20% are not. It seems Google might not be so good place to work after all.

The "/"Q Trap

[Anonymous Coward]

"Nothing goes up forever."

Slashdot IQ's

Work experience? they prefer puzzle solvers

[Anonymous Coward]

Work experience seems to be mainly ignored by Google, at least in my experience. It seemed they were just interested in people being able to solve puzzles quickly.

Yes HAL I understand your choice not mine

[Anonymous Coward]

Next thing the computer will tell us is when to eat, sleep and shit. Give me a break. Hiring a potential employee should be a human on human interaction to see how the person actually acts. Surveys are very easy to fake regardless on how sophisticated anybody tries to make them, yet if you are a correctly trained HR employer you will see the subtle signs in people. Such as direction of eye movement when asked question, hand movement, perspiration, etc. Can our wonderful computer algorithm comprehend this "human" data? NO. This is exactly where computing technology should not be involved, if I remember computers are made to make our jobs easier not to take them over.

This is the hiring equivalent to a polygrpah test

[ClosedSource]

You hit the nail on the head. Like nearly all methods for hiring people, it's an open loop system. In order to to evaluate the effectiveness of the approach you'd have to monitor those you hired and those who you rejected to determine their relative performance. This is pretty much impossible to do, but the difficulty of testing an algorithm doesn't eliminate the fact that an untested one may very well be worthless.

This is just the same hit-or-miss approach everyone uses dressed up in psuedo-scientific garb.

Inside jobs?

[tuxicle]

I'm just waiting for the day that someone inserts code into the AI scanner that says

if(!strcasecmp(applicant->name, MY_FRIENDS_NAME)) {
    hire(applicant);
}

Time to start making friends with Google employees...

blind job-interview

[Treates2]

a dating services from google is here to come, within a few months.. swear. i found a lot of girlfriends who's named ended with .jpg and .mpg

Google's Contribution?

[Anonymous Coward]

I think it will be interesting to see if Google can actually add a service (besides a heckuva good search engine) to the economy.

I work in a corporate environment where the users are sometimes working on Intellectual Property, but, mostly wasting time (until the customer needs to be serviced).

I find that most offerings from Google tend to be ones that are either intrusive to desktop security (e.g. desktop search), utter time-wasters (e.g. Google Earth), or applications that actually inhibit the internal workings of my company's intranet. (Yes, probably poorly written web-pages on my company's part.)

Google's inventors are obviously worthy of many bows, and "we are not worthy" chants, however, I truly fail to see how the company really helps the world's infrastructure beyond their search engine. (Yes, they have some really cool toys, but they are just toy's, IMO.)

Lawyer hiring

[Anonymous Coward]

I would really love to work for Google as a lawyer, and I seem to be their kinda dude being a nerdy assembly language programmer who knows the value of pi to a hundred decimal places, but when it comes to hiring lawyers Google won't consider you if you didn't go to one of the so-called elite law schools.

Re:Lawyer hiring

[rigau]

Which schools are those? Where does the cut-off in the rakings happen? Yale, Harvard, Stanford, Columbia, NYU, Chicago, Penn, Michigan, Berkeley, Virginia, etc...

precision

[adrianmonk]

Even though it's already 10,000 employees strong Laszlo Bock, Google's vice president for people operations, sees no reason the company won't reach 20,000 by the end of the year. This will mean hiring something like 200 people a week, every week, all year.

Something like 200? How about 191.7808 people a week, people? How hard is it to compute that? (And if I can figure that out and Google can't, do you think they'll hire me?!)

Biodata

[slasho81]

This selection technique is not new. It's not unique to Google, and they didn't invent this "algorithm". It's called biodata, and it's quite common.

Selling to others = identity aggregation = scary

[scottsk]

Wait til Google offers this as a service to other companies! Why would they develop something this sophisticated just for internal use? Google is becoming an identity aggregator - if you log in, they can track all your searches and group posts. Now they've got google checkout which can track all your purhases online, aggregating all your purchases from multiple stores. They have your gmail that they already index and search through for ads. (Nevermind tracking your embarassing moments on youtube.) Now they want your job application. You could read the terms and services, but what good would that do? Remember amazon.com stabbed its customers in the back - to get a customer base, they promised that they would NEVER share your personal information, and a few years later when they were self-sustaining, they unilaterally changed that. And there was no way you could quit being a customer or have your data expunged. Google's intentions seem pretty clear at this point - become a single point of aggreagation for online identity. So what are they planning to do with all this information?

Google will be recruiting at SCALE 5x in Los Angel

[Anonymous Coward]

Google is a silver sponsor of SCALE 5x. They will be onsite, recruiting on the expo floor. Additionally, they will have 3 speakers on open-source topics.

Just do it.

[Anonymous Coward]

Just post some immature inflammatory post against google. It only takes a second and they'll never hire you, no matter how much they need you. You'll be free and the google freaks will pay for their own arrogance.

It only takes a second.
Just do it!

Re:Just do it.

[anticypher]

Yeah, I admitted during the last interview my /. login was Anonymous Coward. Big mistake.

One thing I forgot to post was that we've noticed getting rejected by Google is now a badge of honour. My 5 interviews don't even leave me in the running compared to some who went through 10 or 15 interviews and never hearing anything back, or getting a reject email 6 months after the last interview.

the AC
How in the hell did my post get modded up to 4? Flamebait, pure and simple, but I can't mod my own posts

Re:the code.. and problems with it

[DigitalDragon]

I am sorry, but I just can't resist.

There are a few problems with this code:

* as pointed out by another replier, when passing parameter to the function (bool recommend), "bool" should be omitted, this won't compile

* why define a recommend variable, if it always is set to true, simply call forwardResumeToMicrosoft(true) instead

* Evil() function does not receive any parameters, so if we have to focus on the surveyResults variable. If this variable is a boolean, than why not simply compare it to true or false, or a global constant? If this variable is a number (indicating the survey score), then comparing it to a function that returns a number does not make sense - either compare it to a global constant of PURE_EVIL = -9 (for example), or pass the result into the Evil() method for valuation. Same logic applies to the variable if it is an object of some sort.

* and finally, what's with mixing notations - either use camel-case notation everywhere, or use capitals, why is surveyResults camel-case, and Evil() starts with a capital letter, notice that hire() method starts with a lower-case.

Well, now back to our every day problems.

Screening for skin-jobs

[ultraexactzz]

You're in a desert walking along in the sand when all of the sudden you look down, and you see a tortoise, it's crawling toward you. You reach down, you flip the tortoise over on its back. The tortoise lays on its back, its belly baking in the hot sun, beating its legs trying to turn itself over, but it can't, not without your help. But you're not helping. Why is that?

Re:Screening for skin-jobs

[Nick_Psyko]

bang bang.

schrodinger

[ohearn]

But that's a prefectly valid question. We know people who are good at math and logic like cats. We just don't know if those cats are alive or dead.

SO WHAT

[ohearn]

A couple points:

1) Having run a business before I would probably not hire you either. Your tone makes you come across as petty, bitter, overly cocky, and rather self absorbed.

2) Yes 3.65 GPA is good, congrats on that. Is that a 2 years degree, a 4 year degree, or from a graduate program? Do you have any work experience to go with it, especially in your field? GPA is not nearly the most important factor in finding a good employee.

3) Guess what, Google didn't want me either when I got out of grad school. This coming from someone who had a 3.48 in undergrad with minors in math and applied physics. My graduate GPA was 3.7 completing both a master's in CS and a Graduate Certificate program (think Master's equivalent to a 2 year degree) in Software Engineering and Project Management. Oh yeah I also worked a full time job all through college (undergrad and graduate), worked a second part time job all through undergrad, taught for the university as a grad student in addition to my full time job at night (in IT) and helped run (as in one of the 5 people who started the company) a small and profitable business during my graduate work.

In the end I had a really hard time finding a good professional level job even with all this. The main reason why was that I was way too cocky. After being turned down for a lot of jobs I knew I was qualified for gave me a little humility during interviews I had multiple job offers to choose from.

From the way you come across in your post, you need to remember that you may have something to offer a company, but so do a few hundred other people that applied for the position.

Re:Selling to others = identity aggregation = scar

[paniq]

simple: advertise stuff to you that you actually want.

Psychological tests vs. reality

[heroine]

Those psychological tests are highly questionable. Give any programmer the test and it says they're better at management. Give any manager the test and it says they're better at artwork. What's happening is these tests are written for an ideal world where everything is 100% efficient and plugged into exactly what it should be for maximum effectiveness. They don't factor in the cultural aspect of knowing the right people, tradition, politics.

For google to use this test means they're not going to get any programmers. If google allowed users to view the results of the tests, which they won't of course, it would be a neat way for people to find out about themselves.

The test is called the MTBI

[heroine]

Google is probably using a derivative of the MTBI. As usual, the fans are going to say it's another Google triumph and Google is going to change the world with this totally original, insanely great idea. The test has been around for 50 years.

I agree

[Anonymous Coward]

Second off, every non-engineering area at Google is bordering on disaster.

You certainly know what you are talking about. I also worked there for a while and left for several reasons. I must admit I have never seen so much incompetence in administration in any other place I have been, including huge universities. I cannot count how many administrative problems I suffered while working there, to the point where it just seemed a joke, where no one wanted to take responsability for their screw-ups. I really believe that it has to be the closest to working in a state institution of a Third World country. Most of the internal technologies at Google are amazing, but there are so many other things that are just below any standard.