Hacker News new | past | comments | ask | show | jobs | submit login
Google Sued by 3 Female Ex-Employees Who Say It Pays Women Less Than Men (nytimes.com)
167 points by lmcnish14 on Sept 15, 2017 | hide | past | favorite | 272 comments

> (women) hold only 20 percent of the company’s higher-paying engineering jobs.

I believe women make up less than 20% of comsci graduates? So wouldn't this be about right then.

Also I'm assuming women in comsci have lower workforce participation than men do (as in broader economy) so Google must be hiring women at a higher ratio than men from the available pool.

I'm 100% in support of open workplaces for gender/race/politics/sexuality/dress and whatever 'work ability' irrelevant preferences. And cases like Susan Fowler are absolutely disgusting. I also feel a bunch of people increasing complain about their lack of success due to gender or race and are not willing to see they have a fair crack at their career but dont have the talent/drive etc and cant see this lack of ability in themselves.

This. Men make tons of excuses for their lack of success as well, it's just they can't get in on the discrimination complaint wagon. The Lake Wobegon effect and all that. I think a bit of the anti-"workplace diversity" movement is driven by this as well though. "I'd totally be division chief if the higher ups weren't working so hard to make their diversity numbers work out".

In software engineering, the figure is much lower than 20%. It is at most 15%. The 20% figure includes anyone within a Tech department in the org chart. This means PMs, UX/UI designers, etc.

I don't mean to minimize the level of technical competence in these roles. It is great that there are SWE-adjacent roles with 30%+ women. It irks me deeply that Google puts out these misleading statistics which then get parroted by the media and then everyone else.

> dont have the talent/drive etc

Don't have the talent/drive/connections/hustle/luck …. There are more possibilities than just sexism vs. meritocracy.

Edit: hey downvoters, I'm talking about why a random given person of any gender may not be progressing in their career. Read the context.

What percentage of programmers at Google have compsci degrees? What's the split like within that subset?

What percentage of HIGH PAY ENGINEERING jobs in google have comp sci degrees, that's the true comparison. And then there is bigger issue here. It's really the obsession about percentages of women in X position. Here's another percentage what percentage of women and men didn't waste a thought on how they looked in their teenage years because they where so obsessed on how computers worked? I believe it is both very small and men are over represented there, and they are over represented in high paying computer jobs.

>>Here's another percentage what percentage of women and men didn't waste a thought on how they looked in their teenage years because they where so obsessed on how computers worked?

Most people are not in the job because they like create things or build things. They are there because they want a job and nothing more. And they want a job where (unit of money)/(unit of work) has a higher value.

That means they have to either figure out a way to make more money or do less work, or both.

Therefore the question of early interest or passion is meaningless to most people. So as far as they are concerned, it rarely matters how set of people of identity A got to it. They think regardless of that if set A got it, other sets should get it too.

This also creates other problems. Set A is likely to do side projects, write programs and hacks out of personal interest. Other sets looking at job as a return/effort metric will likely see why they are expected to do anything all all apart from working 5 hrs a day between 9 - 5.

The problem is merit is heavily at the side of A and other sets want the reward to not go with merit, but rather with participation itself.

So true.

People didn't give two shits about the nerds and the tiny useless portion of the world known as computing they occupied 20 years ago - back then it was less glamorous and less money involved. They scoffed at the morons who not only did computers at work but continued to think, write and do "work" after work!

Now theres money involved everyone wants in! Now those same leeches demand to be given the same roles/money as those who worked much harder for it , because its so unfair that people who put have more experience and passion are rewarded while they are not!

So true.

People didn't give two shits about the ladies and doing the tiny useless portion of the world known as software they occupied 50 years ago - back then it was less glamorous and less money involved.

Yes, women used to dominate software development. Until the money and prestige started coming in, then men started dominating the profession.


The article you cite is fluff and doesn't support the statement that "women used to dominate software development." It mentions the "Computer Girls" article from the 1967 Cosmopolitan, but here is what's in it: "At one point the author speculates, seemingly without irony, about the “ the chances of meeting men in computer work. ”(The conclusion she comes to is that these are “ very good, ” as the field was currently “overrun” with men.)"


Apparently, the best guess is that between 11% to 50% of programmers were women back then.

"Mandel suggests that one out of every nine working programmers was female. This is probably overly conservative. The exact percentage of female programmers is difficult to pin down with any accuracy—even figuring out the total number of programmers in this period is difficult—but other reliable contemporary observers suggest that it was closer to 30, or even 50, percent.3 The first government statistics on the programming profession do not appear until 1970, when it was calculated that 22.5 percent of all programmers were women—an estimate more than twice Mandel’s.4"

"Of course, computing itself is a very broad term covering a multitude of occupational categories, including high - status jobs like computer programming and systems analysis as well as low - status jobs such as keypunch operator. Women tended to congregate in the lower end of the occupational pool in computing"

Thanks for providing this alternate perspective and invalidating my source. I will reconsider my position.

Can you account for the significant amount of women computer scientists making ground-breaking discoveries? Grace Hopper, et. al. Were the outliers?

I'm asking because I genuinely am unsure.

Not justifying anything. But wasn't early programming two distinct tasks?

1. Writing the Program(On Paper).

2. Feeding the program to the computer.

People in 2. were called 'Programmers' because that is what they did literally- 'Program the machine'. Also early programming had all sorts strange situations where a large part of programming was actually getting results by plugging values in largish math formulas. So there were a range of people who did just that. Generating graphs, being the human equivalent of source control etc.

The 2. part wasn't exactly a very glorifying role and was more like borderline stenographer.

By some definition that is still true. Notice how many programmers there are who probably write MVP web apps for a living copying code from stackoverflow/internet, but there are also people doing all the real work thinking about stock markets, security, medical devices, writing compiler patches, building tools.

I've always thought as coding as a mere ritual the real work always happens on the paper.

This was my understanding too. Women were given the the enormous and tedious task of transcribing programs onto punch cards and such. It required a large "typing pool", much like secretarial pools preparing letters dictated by others.

I don't know what's more funny (or sad): that you think engineers are no longer as looked down on or compensated at a lower rate than their "work" should otherwise merit or that you seem to have completely missed the part about "equal experience" in your haste to poohpooh them that desire equal treatment for it as inferior.

Thanks for the condescension sensei.

I think your position is equal work experience should merit equal pay.

You seem to have missed my point also - experience outside work also counts as experience. And millions of things besides experience determine your salary. So we should not be so quick to assume that 2 people with the same years of experience "should" merit equal pay - its obviously false.

If all people with equal work experience are simply equal in the value they bring to the company then why do we conduct interviews or ask for resumes or anything? Lets just set everyone's salaries to "$100K + 10K * (years worked)" no?

Are you saying that the plaintiffs in this case didn't do side projects?

Well, majority of software developers don't do side projects, especially not those in high paid difficult positions. Those people put all strength they have into work itself. Employers don't even care all that much about side projects. So, it is safe bet.

This whole thread is stupid red herring. It has nothing to do with reality of working in corporation. It has zero to do with skills required in those high positions - just about only reusable from teenage years is linux internals.

It is basically assumption that a.) women are surely lazier b.) since they are girls they did not played with boy toys c.) if you discovered tech as part education instead of in playroom you can't be good.

>>Well, majority of software developers don't do side projects

Most do. The fact that bulk of the software infrastructure from tools to top notch production software is open source says something.

>>especially not those in high paid difficult positions.

If its really about money, then there are better ways to get paid in a software company.

>>Those people put all strength they have into work itself.

Which is often a placeholder for a side project.

>>Employers don't even care all that much about side projects. So, it is safe bet.

But doing these projects does make you a fairly good programmer, and you tend to gravitate towards valuable work and hence good pay.

>>just about only reusable from teenage years is linux internals.

Bad news. Starting early matters. Want to build a good retirement fund? Want to play big leagues sports? Want to be a concert musician? Want to be a doctor? In fact want to be big in anything? Most certainly you have to start early.

>>It is basically assumption that...

Nobody made such an assumption. We only said a particular group A(mostly nerds) starts early and works hard. What the other group is for them to decide.

Note, Nerds is a group of people irrespective of gender, race, or other identity(even nationality or religion).

It just comes down to one thing. You can't progress beyond a point without work. Reservations only take you that far.

Most open source is paid work, according to statistics. That not a bad thing, why would you want that work unrewarded. Then there is a bulk done by people on universities who have time - not employed as software developers.

Moreover, even if open source would be written as side projects mostly, it is still just a tiny part of all software out there.

You assume they are lazy, for no reason. No, not just nerds work hard. No, not all nerds work hard - many spend majority of time playing with something easy they like. In particular, many nerds are unwilling to learn what they don't like.

As for your bad news starting early matters, it don't. I am old enough to have seen where people ends. Many men stared later or changed career and it was no disadvantage after initial year or so.

I personally have all the things you talk about as necessary advantage. They are not and that is not even bad thing. The only difference is that I am not as bitter as rest of thread who want reward for being associated with them (I bet they were not young geniuses they want themselves to be) while real world does not work like that. The other annoying thing is that you would assume I don't have them, cause I am women whole dudes who changed careers yesterday are assumed to be interested since childhood no matter what facts are.

Most full time developers have side projects? I find that hard to believe. If you said "most single, childless, under 28 year old developers living in San Francisco have side projects" then that's more believable. Most people don't have time to go to work all day then come home to do unpaid work.

Open source isn't always someone's side project. Lots (most?) open source is professional paid work.

Thank you for getting to the crux of the issue. It is simple gate keeping.

Indeed, startup founders and great engineers get absorbed in their work and spend time on it.

>>that desire equal treatment for it as inferior.

The kind of equality that you are aiming for is meaningless, it doesn't exist and attempts to create this have lead to far bigger problems than they have tried to solve.

This is the equivalent of asking why we don't handover gold, silver and bronze medals to the people who came last in the race instead of the top 3. This isn't discrimination. The sheer concept of effort vs reward in human psychology is designed such that "Humans are unequal by merit of our actions"

Also the fact that some people are better than others is here to stay and not just restricted to software. Most of us are not going to be Neil Armstrong or Richard Feynman. That's not discrimination.

There is only equality of access and opportunity. Outcomes are not going to be equal.

I find it interesting that HN commenters frequently take it for granted that someone who tinkered with programming at an early age and do side projects is a better performer at work.

I have never seen any multiple-data-point evidence presented to support it. Sure, various prominent entrepreneurs used computers as hobbyists (Gates, Zuckerberg), but does that apply to the typical programmer?

We all encountered students in classes who didn't study yet completely grokked their mathematics and algorithms coursework. Why is it so hard to believe that those quick learners use their time effectively at work?

Additionally, a multiplier to good engineering is strong communication and organizational skills; those can be enhanced through social recreational activity, and diminished by spending leisure time in solitude.

I'm not claiming either method is better, and intuitively the passionate engineer should win, but we shouldn't take it for granted until we get some actual information.

Edit: there seems to be some belief that I don't value experience. It is extremely important. We just shouldn't take it for granted that programming at home is as valuable as work experience.

In every single industry and craft, those with more work experience are typically paid more than whose with less.

There is little difference between "experience programming during a job" and "experience programming for fun". It is the same activity.

So of course those with more experience should be expected (on average) to better than those with less. Of course those who seek out more experience due to passion will (on average) be better than those who don't.

It's not an unusual thing to expect at all. People who care more do better.. in every human endeavor, and this is widely accepted by society. It only seems unusual when newcomers cry "unfair" when they see others enjoying the fruits of their labor.

You can ask for data - great, we don't know the answer. But if I had to guess one way or the other, based on all human experience, yes I would lean heavily towards experience. Its why professors know more than students, why Edison invented a lightbulb after a thousand other failed inventions, it is the basis for the very concept of an expert - its why we appoint a doctor instead of a physicist to run a hospital. Its pretty goddamn fundamental - people get better at things with time, so those with more time tend to be better.

There is little difference between "experience programming during a job" and "experience programming for fun". It is the same activity.

Over of my more recent hobby projects was writing an RSS->IMAP bridge in maybe a couple hundred lines of Python. It has one user (me). This is fairly typical.

Until this year, my primary work project was a toolset for streamlining one-off ETL projects. It has parts written in I think five different languages, it has parts that run on Windows user machines and other parts that run on unix servers, it was built to reduce the annoying parts of a business process, it evolved over the course of almost a decade, etc. It has a couple dozen users (the team I used to be on).

The two are about as different as adding a new attached garage and workshop vs building a dollhouse.

Yeah, exactly. With the work frequently being the dollhouse.

At my various $dayjobs, I spent most of the time building various flavours of the same CRUD (web CRUD, desktop CRUD) in dumb mainstream languages. At home, I do everything from video games to data processing to AI, in actually productive languages.

The difference I see is that with hobby projects, you're at least free to make something that's actually useful. At work, you might get that if you're lucky. If you're not, you'll be implementing in code some dumb inter-management politicking.

This reminds of a situation I faced early in my career. We worked on a network alarm management tool, and some of us used to rewrite the tool or bulk of its features. To a point we finally arrived to invent our own tools to fill up the gaps in the feature set offered by the current tool.

It used to be one of the biggest reasons why we learned so many thing in comparison to the rest of the team(which was fairly big), to a point we could make a lot of very critical design decisions or write an application that could save hours of time for our users.

Its one of those axioms of software development I've learned, in order to do good work you have to do mountains worth of waste work.

> Its one of those axioms of software development I've learned, in order to do good work you have to do mountains worth of waste work.

It suggests an obvious question though: can we do better? Can we avoid having to do "mountains worth of waste work" before getting to do the "good work", or is the former a necessary prerequisite for the latter?

I believe we can reduce the amount of waste work by doing deliberate practice.

This means really analysing what you did and how it could be better.

Also doing a small amount of work each day is much better than a large amount once a week.

Take a large task and break it down into tiny manageable chunks that you can analyse.

There's tons of material out there that has best practices for mastering anything.

It's not just about skills, but also about knowing what problems to solve and being in the position to solve them. I wonder if we can side-step having to do waste work in the latter cases.

> In every single industry and craft, those with more work experience are typically paid more than whose with less.

This isn't true. Those who do more with their work experience are paid more.

We all know engineers who have been at the same company for over a decade, but have practically nothing to show for it. There is an enormous disparity between engineers.

There's a reason why Google and Facebook allow engineers to stay Senior Engineers for an entire career: most don't actually progress beyond a certain point.

Jeff Dean and Rob Pike aren't good because they've been programming for a long time. They're good because they've developed their organizational skills, architectural skills, communication skills, and excellent public speaking (this is vital as a tech lead and higher). None of those can be developed by programming on something cool on the weekend.

None of what you said actually disputes my statement; we both value experience. I'm just trying to communicate to you that extra hours programming at home might not be as intrinsically valuable as you might think.

Well I agree that hours don't always equate to value, in many cases they don't for the reasons you pointed out. Ultimately it matters what kind of experience you have and how valuable that is in a given context.

My point was if I had to pick one broad generalization - "value increases as hours increase", "value is independent of hours", "value decreases as hours increase" - it certainly makes sense to pick the first. Extra hours are more likely to make you more valuable than to make you equally or less valuable.

This was what I was replying to

> I find it interesting that HN commenters frequently take it for granted that someone who tinkered with programming at an early age and do side projects is a better performer at work.

Thats why I do take it for granted (on average) that people who do side projects do better than those who don't. With 2 equal resumes I would confidently pick the one who started programming early. Nothing is certain when generalizing like that obviously, but then again no signal ever in any interview is a guarantee of future work performance.

>>We all know engineers who have been at the same company for over a decade, but have practically nothing to show for it. There is an enormous disparity between engineers.

Please note that this has nothing to do with gender.

>>There's a reason why Google and Facebook allow engineers to stay Senior Engineers for an entire career: most don't actually progress beyond a certain point.

Or they have golden handcuffs, and its very hard to get paid that much else where. And throwing away money for some measurement of ethics which no one cares about is foolish.

>>There's a reason why Google and Facebook allow engineers to stay Senior Engineers for an entire career: most don't actually progress beyond a certain point.

Oh they have been. They don't build furniture, they build the machine that makes furniture. So they are valued more.

>>'m just trying to communicate to you that extra hours programming at home might not be as intrinsically valuable as you might think.

Ever heard of the proverb: "Opportunities multiply as they are seized".

To go some where, you might have to travel a lot of distance going nowhere. The destination isn't always visible from where you stand.

Tinkering in teenage years does not give you more professional experience 10 years later. It gives you a bit advantage around early 20ties. It is essentially irrelevant at 26. As technologies change and develop, that get lost and become useless.

Also, imo, important predictor of how good you will be later on is more your willingness to learn tech you don't like in the beginning. If you don't have that, changes will leave you behind.

Agreed with the latter, don't agree with the former.

Tinkering gives you out-of-domain experience which subtly improves the decisions you take - both architecturally and organisationally. It gives you at least a bit of an insight of what the right tool for the job might be even if it is outside your domain.

If that tinkering is in hacking IT security, and you do CRUD app development (or project management) for a living, your app might just coincidentally avoid SQL injections or obvious buffer overflow errors. If that tinkering is in image recognition and machine learning and you are a business manager, you just might know the difference between "feasible" and "impossible" AI projects (avoiding this XKCD situation: https://xkcd.com/1425/).

Obviously you will also avoid these if you have actual professional experience in IT security or AI - but very few actually can get involved in a dozen fields professionally at once.

The thing is, if it is your job to do crud applications, then having a look at security development at least once a year is a part of job. And at that point, you will learn about sql injection if it is a new thing. If sql injection was something known for 10 years, then you would learn it in the process of learning crud.

All that assuming you take work seriously.

I think that when young people spend their time doing something, it says somethi nd good about them and their environment. However, you can learn the things they learned later on if you have aptitude and study the topic seriously.

Its not necessarily starting early in programming. Its being in general being good at 'scoring marks' vs 'doing projects'. The former and latter take very different skill sets, and their returns too are often very different.

If you are good at doing projects. Along the way you learn a lot of other very important life skills. Things like resourcefulness, persisting at things, immunity to failure, trying many times etc. And these come handy and are usable to to many things that actually matter in the real world. That is building things.

These things are harder to gain at a later stage in life because expectations from one's life at that time are different and you have to worry more about monthly payments and putting food on the table. You don't have 10 - 15 years lying around to do what other programmers have done in their early years where it was cheap to that in terms of time.

You are also discounting the accumulative effects of these things. After a while due to years of practice, early starters are likely to get very good at things in a far more disproportionate way than those who come later.

Would participation in athletics or math competitions fulfill your requirements for early hard work and adversity? I do agree with your premise. I just don't see where you actually disagree with mine.

I'm not discounting anything. I'm just saying that it shouldn't be a given that engineers who got into the field late are automatically less able.

> early starters are likely to get very good at things in a far more disproportionate way than those who come later.

We think that but it's not actually demonstrated. Entering a field at the age of 18 or 20 isn't actually late. Until we actually have evidence that people who programmed before college have better outcomes, we need to stop talking about it as if it's a known truth.

>>Would participation in athletics or math competitions fulfill your requirements for early hard work and adversity?

Yes, but in School and a little higher- math is about learning heuristics and ready made procedures to solve problems in algebra and plugging numbers in formulas to get answers. And that is one of the biggest reasons why Nerds go onto to do things like projects to bail from the what is largely a useless academic exercise that gives you nothing real at the end.

Sports has a binary distribution, you either make the big league money or you don't.

>>I'm just saying that it shouldn't be a given that engineers who got into the field late are automatically less able.

They won't be treated as such. They will be evaluated on the same grounds as the long slog people. And that is precisely the problem. Somethings come only with time.

>>Entering a field at the age of 18 or 20 isn't actually late.

Its isn't late in absolute terms. But everything in the world is comparative.

The world is essentially a stack ranking system. You can do anything at any age you want. But there are costs associated with starting late, and they are in comparison with people who are long there.

It would be a hard study to do, many of the non-tinkerers will never enter the industry in the first place. For a fair evaluation I think you'd have to look at first year comp-sci students and then see where they are in 10 years.

Personally I think the correlation I've seen personally is strong enough that I'd be shocked if a study proved it otherwise.

>>It would be a hard study to do, many of the non-tinkerers will never enter the industry in the first place.

They will if there is enough incentive. They won't other wise. For example: CS definitely sees more diversity than mechanical engineering, simply because its easier to get paid and higher in CS.

Even in CS a lot of people eventually hop to MBA as that is even more better in terms of making low-effort big-money.


If we did manage produce a study with multiple-data-point evidence supporting it, then should we legislate to compel all businesses to adopt the conclusions and recruit the same way?

The free market already allows founders and backers to back their theories with money and compete in the marketplace.

The teenage years are completely utterly irrelevant to this. They are super irrelevant to high tech positions and definitely anything related big data software. Video games are irrelevant too.

Who play with dolls and who plays with action figures or cars is even moremember irrelevant.

High paying computer jobs are not filled with nerds nor high funstional autistic while we ate at it too.

> High paying computer jobs are not filled with nerds

I beg to differ

I'm just going to go out on a limb here based on personal experience and say like 95%+ have a bachelor's degree or higher.

Sure, lack of advancement indicates a lack of ability. It's an old rationalization, but it's a post hoc fallacy.

That is: lack of ability to do what? Can you clarify?

Or just a lack of drive/desire. In my office almost all of the management positions are filed by women while almost all the technical people are men. I asked about it when I was first hired and most of the people in the office had been offered management positions at some point during their career and they tuned them down because they'd rather do engineering work than human work.

Looking at the demographics you might suppose that there's a lot of bias and discrimination but it couldn't be further the truth. We just have lot of mobility and many options for salary increases without 'climbing the ladder' so everyone individually wound up where they wanted to be.

>>In my office almost all of the management positions are filed by women while almost all the technical people are men.

This again creates a new problem of 'Smart engineers' Vs 'Dumb managers'. To a point eventually you get to reducing management to routine supervision work.

In the past I have seen a situation where a program manager was routinely pissed because he was barely able to understand what engineers were talking about. After routinely under estimating time estimates he came to a point where the entire argument on him could be reduced to a rude statement: 'Why don't you stick to your spreadsheet cell filling work, and let us engineers do real work. Work that matters isn't your cup of tea'

I can see where this would go in case of women managers. In only some time, men would be accused of things like 'mansplaining'.

It sounds like your workplace doesn't suffer from the same issues addressed in the article. The women managers at Google were not being offered the same compensation increases, despite performing at or above the level of their male counterparts (all allegedly).

You shouldn't take it on faith that the women were being underpaid. The case will reveal what happened.

You also shouldn't derive from your experiences that just because you have only seen good outcomes, another workplace or team is similarly good. You only know what you actually know.

This is a broad and complex question with many variables, especially the human element and needs deep context to answer. From this I am going to avoid the question as as I get the feeling this becomes 'internet nitpicking' at specifics on a broad issue. But if you'd be more specific in the question I'd be happy to discuss. Apologies if I have misunderstood.

The Fundamental Attribution Error comes to mind, but the last time I mentioned it someone jumped on me a little about it.


It seems more like a faulty null hypothesis. Instead of assuming zero correlation between ability and pay, some people start from the assumption of perfect correlation.

With regards to ability, we seem to intuit the existence of an unmeasurable "ability factor" that underlies real, measurable metrics of performance. I have no idea how Google measure this kind of thing or any statistics qualifications, but a very dumb first attempt might be to get all the engineers to recommend five top co-workers, break the recommendations up into male/female groups and discard cross-group recommendations, then look at the attributes of the most recommended co-workers in each group.

> lack of ability to do what

To be recognized as more valuable.

It seem there are two issues at hand:

1) Women who are in tech get paid less then men.

2) There are less women in tech to begin with.

They are related, in multiple ways perhaps and it's possible tease some links and accompanying explanations:

a) Maybe women know they'll be paid less so they don't go into tech to start with. It's kind of a chicken and egg problem.

b) As you've identified, because of so few women in tech, individuals who end up in well paid positions by chance are less likely to be women as well. So it's a statistical explanation in way.

c) There is sexism. I've seen it, together with xenophobia and other prejudices. And because of that women are rejected, as in not getting offers to start with, or even if hired not taken seriously.

d) In order to fix 2) women were hired but they didn't have an equivalent skill set or experience as men, so they got the job, perhaps to fill some diversity metric, but they did worse in performance reviews later on so they get a smaller salary, less bonuses etc.

Regarding d), I remember going out of the way to bring more women for onsite interviews trying to "fix" 2), that is I wanted to provide more opportunities for success and wanted to increase the diversity in the office. Most of the women failed the interview. I am sure c) was a contributing factor in some cases why the higher ups rejected them. But many simply didn't have their skill set at the level we were looking for. Had they've been offered and taken the job, they might found themselves in the position of getting a smaller salary and poor performance reviews down the road.

Now this doesn't present any good solutions, and doesn't address the lawsuit at hand because I don't know enough details about it. So it is mostly a breakdown of the issue as I see and from personal experience, as I was in charge of recruiting and interviewing for a good number of years.

One of the women suing complained that a man was hired a level above her with the same years of experience. They both had four years experience.

I joined one of the big 4 eight years into my career. Similar to her I was placed at an entry level position. When I started I saw that I had the same ability as people higher than me. I made sure that I displayed that ability to my peers and manager. And what do you know, I got promoted into a more suitable position.

Hiring is broken, we all know this. Sometimes we might not get what we deserve. If we want more then you need to work for it. Plain and simple.

Google can't win with its current culture. This is what happens when a company or group is too left leaning. You always get attacked by the very people you're trying to help.

Everyone who worked at Google or a company like Google knows that 90% of people do a job well below their skills. That's how it is in those places.

To sue them because you have back-end skills, but you were doing front end is laughable. There are so many PhDs in ML there writing SQL queries, or PhDs in CS changing the color of some text for an A/B test. That's how it works in the best companies in the world. You get the money and the prestige, but the job is bad.

My guess is that this is true of a lot of finance and law jobs as well. Actually, I once talked to a dermatologist who said he wished he got to use his mind more often (he said this on hearing that I was a programmer! I guess he thought it was a lot of deep thinking and logic or something? Seriously made me realize that my impressions of other fields is probably just as flawed).

Of course, there are jobs in all these fields that are fascinating, and people with boring jobs often have had a project here and there that was interesting, so that's what you hear about and how your impressions are formed when you aren't part of another field. After all, people at parties don't like to talk about how boring they are.

I think many people complain about this too. I remember a very famous HR person of a famous IT services firm in India once told in a public event that software was one of those areas where every one expects to be treated special and expects to be given a job where they feel they are generating tremendous value, by the virtue of which deserve a very high salary.

While only a handful such jobs exist in any company.

The standard response for these things used to be 'Go find a different job which you think works for you'. But these days its either 'do as I say' or 'prepared to get sued'.

there is probably some kind of sweet spot where you get to do whatever really cool stuff you can think of and your company is just pretty good

Probably in R&D departments of tech companies.

This is actually very helpful to know, thank you.

this is very true.

> Sometimes we might not get what we deserve.

Sometimes we deserve less than others. Gender/etc aside, some people are less talented/effective. Society seems to be afraid to say this, but not everyone is created equal upstairs.

> They both had four years experience.

Obviously, on the job years is almost entirely unrelated to pre-work experience, so this is just silly. I hope there's more to it.

I'm curious how their resumes/accomplishments compare for those four years.

Thanks for making these points explicit.

Years experience is not an indicator of seniority. I have two friends that graduated at different years.

One has more than 10 years experience. The other has three years.

10 year guy has been working with jquery and HTML the whole time. He avoids any other problems or domains.

Three year guy wants to work in every area possible. He actively pursues internships and work that provides a wide range of experiences and technical variety.

Now who do you think is more valuable as an employee?

> Now who do you think is more valuable as an employee?

For which job? For some breadth of experience would be better but for a job working with html+jquery all day every day the depth of experience in that area would be preferred. Sadly the industry seems to be too focused on "full-stack" interchangeable components lately.

> And what do you know, I got promoted into a more suitable position.

This is a bit of a ridiculous statement. You seem to be claiming that everybody who shows off their skills rises to their correct level. I don't see it that way. I see that there are way more people capable of doing a senior position, but only a few positions are available.

As far as I understand the larger companies have a clear trajectory for promotion in software engineering. There's no limited places. Prove you have skills a, b and c and you get promoted.

Where I work, the level is determined by how well the did on the interview, regardless of years of experience.

It might be that the plaintiff didn't do as well on the interview as her colleague. Just because both have have 4 years experience doesn't mean that both did equally as well on the interview.

> One of the women suing complained that a man was hired a level above her with the same years of experience. They both had four years experience.

It's impossible to say without more information but it could just be a case of the old 4 * 1 year experience for her and 1 * 4 years experience for him. Comparing people on years of experience is the sort of irrelevant laziness I'd expect from a recruiter, not a developer.

> Google can't win with its current culture. This is what happens when a company or group is too left leaning. You always get attacked by the very people you're trying to help.

Yah, this sums it all up, Google makes this assumption that everyone they hire has the same mindset, and this is not true.

> Hiring is broken, we all know this.

Then what is your problem with them suing ? They go a different way than yours, but if it's broken in the first place why not ?

Well, the suit only has merit if hiring is broken in a way that unfairly disadvantages a protected class. (In this case, women.)

I think the poster's point was that hiring is broken equally for everyone, which is inefficient but not illegal.

> complained that a man was hired a level above her with the same years of experience.

> You always get attacked by the very people you're trying to help.

helping how?

Google puts lots of effort into hiring and promoting women, and attracting them into tech.

Companies pay some employees more for pretty good reasons. I don't think they're thinking "Jim's a dude so we should pay him $50K more than Sally."

No, they're paying somebody more because they don't want them to leave.

And if they are paying you less than somebody else, it's probably partially because they don't think you are as much of a flight risk.

And if Google is wrong about these decisions, I'd guess we'd see a lot more female engineers leaving Google to start their own companies. But is that what we see?

> Companies pay some employees more for pretty good reasons.

I think you give companies too much credit. Those pretty good reasons are typically that that's how much they asked, it wasn't unreasonable, and they passed the interviews.

People's salaries are all over the map for the same sorts of positions, and it's not just IT and development.

But, even though it might not be a clearcut answer to whether there is discrimination or bias in salaries, it should be explored. Even data with a wide, overlapping range doesn't mean that the mean and median can't be compared and valid conclusions drawn about probable bias.

Google seems to take bias seriously, and I think if it could be proven with adequate certainty that salaries were affected by the sex of the employee alone, they'd do something about it.

I tend to think that successful companies tend to know what they are doing better than observers.

But the problem is "know what they are doing" can still be unfair. Consider: you're staffing up a company and need folks now. You've found two candidates who seem equally qualified and want to hire them. Both are asking for only a modest raise over their current salaries, but one happens to make 50% more than the other due to arbitrary reasons[1] having nothing to do with qualifications.

What do you do? Realistically you give them both what they ask and pay them unfairly. Congratulations, you're now part of the problem.

[1] Yeah, one is a dude.

And if we change [1] to "Yeah, one is a gal", then I'd expect a lot of pro-diversity people suddenly stop complaining that there's a problem.

Ultimately, the only alternative is to pay people equally. Which won't fly generally, as companies always want to pay the minimum amount that can keep the employee from leaving.

> I'd expect a lot of pro-diversity people suddenly stop complaining that there's a problem.

The fun thing about making a straw-man is you make yourself immune from criticism. I can't attack this point without having to completely define someone who is "pro-diversity" (what's that even mean?), and give evidence that no one in that group has the traits you claim of ("look at this feminist, see I was right!").

More importantly, what's even your point if this straw-man were true? I could just as easily say this about anything.

"Oh your life is shitty? Well I bet you wouldn't complain if things were working out well." It's an interesting case of vacuous truth!

The argument is to make things better for the people who have it worse at the moment, correcting for previous inertia. In this case, the inertia is that men were previously paid more or gained an initial advantage that allowed them to be paid more, and then that advantage has percolated across their various positions until the difference becomes more stark all because as you say "companies always want to pay the minimum amount that can keep the employee from leaving". If that's the case, then if companies could pay women less for the same work, why wouldn't they? Similar arguments apply to affirmative action and similar programs. That's the argument at least, whether I agree with it is an entirely different question.

What irks me is you dismissing the argument with your little straw-man and moving on to "the only alternative". You completely disregarded the points of everyone opposite the aisle of you in the most condescending way.

> The fun thing about making a straw-man is you make yourself immune from criticism.

I don't. You just criticized me at length :). Some of it I could even agree with :).

My point is, the current narration coming from "opposite the aisle of me" is, on the one hand, "gender discrimination is bad", and on the other hand, "we need more discrimination to correct for existing one". Even if we leave out that I don't think there's that much discrimination happening as feminists would want you to believe (in fact, I think they're just reinterpreting everything in terms of gender, whether it is actually related or not) - still, you can't have it both ways.

I'm just against hypocrisy, especially this extreme. When someone admits that they are advocating literal sexism as a corrective action (hopefully temporary), I'll accept that (even though I don't agree it's needed). What irks me is condemning sexism when it goes in one direction, while at the same time advocating sexism in the other direction.

> we need more discrimination to correct for existing one

I like the following analogy. Your body has cancer. The cancer is bad, and you don't want it. Obviously killing your own cells is also a bad thing, but you choose chemotherapy because it's the best way to get rid of your cancer.

How do you quickly get rid of the cancer that is discrimination?


I know you're itching to say, "Make everything equal!". But hold on, how do you do that? Can you ever do that? In my opinion, the answer is no.

First let's suppose we do it through regulation. The logistics involved in such a thing would be astronomical and more importantly would involve massive governmental regulation of basically all employment.

The other option is that we deregulate everything and just let the free market handle it all - wait it out till all the racists and sexists are dead! It's so easy. But that doesn't work either - humans tend to cargo cult. We haven't had slavery in this country for a long time, yet the effects of it are widespread and evident everywhere you go here. It's in the way people talk, the movies they watch, and the music they listen to (or don't listen to). You can't just shut off the lights, pretend none of that bad stuff ever happened and move on. People still walk and talk with that knowledge in their head.

More or less, we can't easily eliminate discrimination without drastic actions, and it definitely can't be done immediately. One of these drastic actions is to counteract the discrimination. In other words our chemo.

In my view, this is fine morally and not really hypocritical. It's also one of the things that's less difficult to actually implement in practice. Let me spin just one more analogy to illustrate:

Say a dude beats your face in at a bar. Completely wrecks you. Puts you in the hospital. The whole thing. Three years later you've finally regained your ability to walk. It still hurts every once in awhile but here you are. So you go back to that bar, and that guy is still there. He's gotten a lot weaker, doesn't have as many friends, but he's still there. Then he tries to mess with you again, so you punch him in the face and move on.

Are you a hypocrite if you think he shouldn't have beaten you to a pulp even though you punched him later on?

The point is that the thing we're counteracting with policies like this is so astronomically bad ("Black people are more suited to physical labor" "Women can only handle childbirth" "Women can't vote") that the policies that correct them ("favoring some black kids with equivalent records as the white ones because in theory they might have had a harder life" "favoring a woman for a position because she might have had a harder time getting there and you want to encourage more women") aren't nearly as bad in comparison.

Sure if you take the "any decision based on [race/sex/religion] is bad and always bad" then yeah, I guess it would be hypocrisy. But I'm not sure I believe that myself. Until we actually hit that mythical day of equality, we're always gonna have imbalances like that. In my eyes, it's better to try and correct them (while not actively hurting anyone of course - there's a big difference between Johnny getting into Yale but not Harvard and sending Johnny to a work camp for the summer).

These are of course debatable points. Any one could be argued about through an entire undergraduate course, but they're still arguments. I highly recommend you become more familiar with the points you are debating and the merits / downfalls of those arguments. It's much more productive than creating straw-men to argue against.

bell hooks' books on feminism are especially good for someone new to the field and interested in such topics.

How does class fit into your view of inequality? Is it a bigger predictor of a person's future than say race or gender? I'm curious.

> "...aren't nearly as bad in comparison."

Certainly from the perspective of rich whites, poor whites, on the other hand, are probably the least advantaged group we have today, but I could be wrong.

> What irks me is you dismissing the argument with your little straw-man and moving on to "the only alternative". You completely disregarded the points of everyone opposite the aisle of you in the most condescending way.

"little" straw man is also condescending.

Plus, a straw man is "an intentionally misrepresented proposition" - not only can we not conclude what OPs intention was, but also it's not clear to me that it is a misrepresentation of the original comment.

> And if we change [1] to "Yeah, one is a gal", then I'd expect a lot of pro-diversity people suddenly stop complaining that there's a problem.

Find me someone (anyone, anywhere) who defends (even in the abstract!) a situation where women are paid more than men for the same work. That's a horrific strawman, and it tells me that you're looking at this as a war (with feminists as your "enemies" I guess) instead of a problem.

And no, you don't have to pay people equally. You have to pay them fairly. If there differences between individuals, that's fine. If there are systemic differences between easily classified groups of individuals, that's discrimination.

>>I'd expect a lot of pro-diversity people suddenly stop complaining that there's a problem.

Isn't that their very argument?

Their point is a few need to be paid more even if they don't work for it, to make up for what is lost.

They are asking for inequality by very argument. Equality by their definition would be injustice to them.

This assumes venture capitalists don’t exhibit a bias towards male founders. Depending on the extent of bias (if any), female engineers would lose a bit of bargaining power with Google.

Also, empirically, women tend to be more risk-averse than men on average, so Google (and other companies) could be paying less knowing the chance they’d leave is lower than a man of similar skill.

It doesn't even have to be for the startup life. They could just leave to work at another established company. Though the rest of your point about being risk averse and so sticking with a job (and therefore being less of a flight risk) is more interesting and would be worth exploring.

This is an interesting thought!

If an employer pays people who are a flight-risk more (a-la "the squeaky wheel gets the grease"), and women are on average a lower flight risk because they are more risk-averse - is the resulting pay disparity a case of illegal discrimination?

My intuition says no, but I'm not sure everyone (including the legal system) necessarily agrees.

Is venture capital a requirement to start a company?

The biggest one I've seen is "we pay them Y because they already made X."

So if it's biased externally it's gonna look that way internally if you're not proactive.

I feel you missed something important in the article. The plaintiffs are asserting that there is a pervasive irrational cultural bias by managers that assumes people like them are superior engineers.

The entire premise of the lawsuit is that the managers are working contrary to the best interest of Google. Companies are made of people, and no matter how much CEOs might want employees to behave a certain way, employees are still beholden to their cognitive biases.

The optimum situation for Google would be to pay all employees the market rate for the level of quality they desire. This would basically yield a meritocratic pay structure. But this is not what happened. The plaintiffs were awarded different pay grades for measurably comparable ability. This is suboptimal for Google and for society; that is why Google is being sued.

> But this is not what happened. The plaintiffs were awarded different pay grades for measurably comparable ability.

The article doesn't say that. It says they were awarded different pay grades given an equal time spent in the worforce, not that their abilities were comparable.

What you're talking about is referred to as a 'market opportunity', which in this case doesn't hold water since I don't know of any successful all-female (or majority-female) tech companies.

In order to believe that you have to believe that literally 100% of tech hiring companies/managers are sexist. If even ONE was not, they'd wind up with all the women because they'd be willing to pay those women more money than every other company. I'm very open to the idea that there's sexism in tech, but the idea that literally 100% of everyone is sexist doesn't seem plausible.


It's not that the companies are sexist

Or the managers are sexist

Or the engineers are exist

Or you're sexist.

That's not what anyone has been saying this whole time - ever. You have completely misunderstood the point.

The system is sexist. Not the people in the system, but the system itself (similar arguments exist for police departments being racist - it's not the cops that are, but the system). In this case, the general consensus is that women are paid less than men in tech due to a variety of reasons.

A culture can also be sexist. While the people within it can harbor no ill will whatsoever, their behaviors and group dynamics can actively prevent a woman from being welcome. In this case, people are calling the tech "culture" sexist.

Here's a couple examples of a sexist culture I've seen in my own workplace. Note that none of the people here meant any harm, nor were there actions actively "sexist" in the common parlance, but they still did things that made people want to GTFO if they're a chick:

* I had a coworker who used to grill the woman on the team much more fiercely than the men. I never asked why this was - didn't have a great way to broach the topic. But it was always extremely obvious, and I have noticed it on HackerNews, Reddit, and pretty much every video game I've ever played. Men are very critical of women in jobs that men are already the predominant workers in. Female doctors and lawyers have had this issue for decades as well. It's not a sexist action. In my eyes it happens totally accidentally (My pet theory is that because we aren't used to seeing women in the field, we tend to be more involved when giving them criticism)

* My coworker had this anime poster in his cube with some half-naked bikini clad girl. Do I even have to mention why a girl might feel a little awkward here?

See how these things aren't actually bad in isolation? But over time, and in great numbers, they add up and make women feel like shit working at your company because they can never truly "fit in" with the guys. That's sexism in work culture and whether it's ever truly fixable is a great question.

I highly recommend you read some literature from across the aisle to become more familiar with the arguments you're facing, rather than the arguments you think you're facing. I highly recommend bell hooks' books on the subject. She's extremely clear and lucid and helped me grok a lot of the logic my first time around.

First of all, the two things you listed are clearly sexist behaviors, even in isolation. Treating women differently than men, whether it's more harshly or with "kid gloves" is sexist behavior. Putting a pinup on the wall of your office is pretty inappropriate for a professional work environment and the reason it was allowed to stand is most likely sexism.

But again, it only takes one workplace, one manager that doesn't treat women more harshly and doesn't allow pinups or have these other sexist behaviors to foil the entire sexist system, because that one employer would get all of these great women who are just as productive for a lower price and would take over their market. It only takes two such employers in a market who are competing for those women to get women up to income parity with men.

I'm not making any arguments whatsoever about sexism in the workplace. I'm only talking about the wage gap. "The system" having these latent sexist rules and behaviors in place almost certainly drives women away. But that's a different problem than whether women engineers make the same amount of money as men for the same work.

You seem to have missed my argument so I'll try and rephrase it.

> that one employer would get all of these great women who are just as productive for a lower price and would take over their market. It only takes two such employers in a market who are competing for those women to get women up to income parity with men.

Yet companies will always want to hire someone for the least possible amount they can - regardless of the manager's personal views.

Two potential employees walk in the door at your hypothetical perfect company. They are perfectly equivalent. One was originally paid $40,000 while the other was paid $55,000 at their previous position. They each want a 10% raise to come to your company. One is a woman, the other is male.

Now you end up paying the woman less than the male, purely because her previous position paid her less. And the cycle continues until eventually you get to some sexist manager back at her first company that thought she wasn't as skilled for some reason or another (apparently).

That's the point I'm trying to make. Your hypothetical situation would not be the panacea to these problems because companies don't offer salaries like that. You're also making the astronomically huge assumption that every person wants to change positions in the first place. Some people like the job they have and stay there for much longer than would be competitively optimal for them.

Systemic sexism is the sexism that manifests in these cycles.

I get your argument entirely. Your argument doesn't hold water at all. Because it demands that every woman be more willing to work in a toxic, sexist work environment for lower money than a non-sexist work environment for more money. What are the odds of that exactly? If even a small percentage of women engineers are willing to move, then the one exception company will be loaded up with massively disproportionate amounts of women.

> and I have noticed it on HackerNews, Reddit, and pretty much every video game I've ever played

what do you mean?

> My coworker had this anime poster

And that's "culture"?

> because they can never truly "fit in" with the guys.

Your examples don't really back up this point.

A) Women are overly criticized when they're doing anything that's a male-dominant activity. Videogames, Reddit, and HackerNews all have examples of this. Announce you're a woman, and you'll suddenly face more scrutiny for unknown reasons.

B) Yes. As it turns out, the behaviors we normalize in the workplace (like putting up scantily clad women) are part of a workplace culture.

They feel unwelcome, different, and unfairly questioned on the basis of being a woman and a woman alone. How do my examples not back up this point?

> How do my examples not back up this point?

You give an example of one guy with a posted, and one manager, and then conclude "culture" in general, and that they can "never truly fit in with the guys"; But why is this the case with the anime poster guy? How does the managers treatment of the employee relate to their relationship with the other guys on the team?

> the behaviors we normalize

This is along the lines of the "role model" argument; all must conform to prescribed behavior lest a slippery-slope epidemic of deviance should arise. Do all the cubicles have scantily clad women?

> Videogames, Reddit, and HackerNews all have examples of this

The first two are verymuchnot the workplace. Reddit has many different cultures, depending on which subreddit you are in.

You list HN too (which is maybe a little closer to the workplace), as in "Women are overly criticized on HN when they're doing anything that's a male-dominant activity" - I'm not sure I agree, can you give an example?

It is definitely not that straightforward. You are probably right when referring to the top talent, but for the majority of the workforce it comes down to negotiations and raises. There is plenty of guilt to be spread around for everybody involved (like, "why don't women negotiate more"), but there are also known well-measured biases against female engineers (SWE has a neat "research" page listing reproduced studies on the topic, same with Project Implicit).

I think to complete the picture, we need to consider "what percentage of ex-google engineers leave.. and fail". The difference might make up the pay gap.

Unfortunately for Google, paying women less because "They can get away with it" (ie, women are less of a flight risk), is still illegal.

I don't think that's the situation I'm describing. If a company could replace an engineer that is more likely to leave with an equally productive one who is likely to stay, they wouldn't hesitate to wave good bye.

What amazes me is the confidence with which people equate the work of engineers when they actually have no stake in making that judgement correctly.

You see, if you are paying the money and living with the results, you have a much greater incentive to make this call correctly.

And those who are unhappy with these judgements have a much greater incentive to equate engineers who aren't actually comparable. It's called politics.

The business reason as to why a company would discriminate against certain groups is perfectly rational.

I mean, if a company can get away with paying women 70 cents on the dollar, because women are worse at negotiating or have less negotiation power (ie everyone ELSE is also discriminating against them, therefore they have less counter offers), then of course they would do it.

It has nothing at all to do with "making the correct call". It has to do with making the profit maximizing decision to pay a group of people less because you can get away with it.

That doesn't change the fact that this is still illegal.

Your argument is irrational. If a company could get away with paying someone 70%, they would only hire the 70 percenters, or at least disproportionality. Google has roughly 20% female engineers, in line with the 20% female CS grad population.

Seriously... if there's some nefarious bias, it would mean a company hiring more women than expected, bc they can "under" pay them. Silly irrational people.

Well no, Google wants to hire all the best engineers they can find, even if they have to pay more. Obviously they don't want to pay more than they have to, but their primary concern is capturing as much of the top talent as they can.

Which is how you end up with worker demographics reflecting the grad demographics, as it is expected when there's no bias involved...

Agreed, I don't think there is overt bias, however it doesn't rule out industry-wide systemic bias which I think is real.

Bit in the same breath if you can get away with getting an equally talented employee for 70¢ on the dollar then why would you ever hire a man?

The way that it works is that you hire anyone at all who can pass the hiring bar, and then negotiate salary afterwords.

The problem is quantity. At the end of the day, if only X people that you interview pass your hiring bar, it is impossible to hire more than X. You can't just say "Lets double X!".

Now, would companies do things like try to reach out to women groups, so as to increase the amount that interview with them, and therefore increase the proportion that are hired?

Yes, absolutely they would do that. As has been demonstrated by all the diversity reachout efforts that companies are doing to women and minority groups. Diversity efforts are PROFITABLE.

Wait, what? How in the world does anything you mentioned relate to that ending protestation written in capital letters?

All diversity efforts that I've ever seen are the exactly opposite of profitable and result in sexist/racist/otherist selection of candidates. The unfortunate end result of worse selection is that worse candidates were hired.. every. time.

Because it is not worse candidate selection. It is better candidate selection.

Why wouldn't a company want to target groups that other people are missing? It means that you can get good candidates for cheap.

Obviously you don't lower the bar. You kept the hiring bar exactly the same, but get more people from certain groups to apply.

IE, if you can find someone who passes your hiring bar, but no other company has noticed, that means a cheap candidate.

If you target the group that everyone else is going after, they are going to be much more expensive for a given quality level.

I knew a person who when looking for work they wanted, also sent out application to other places that they was less interested in but then demanding close to unreasonable pay (around what someone who had worked 10 years would get).

And they did land one of those less interesting jobs, earning similar to those coworkers who had been there for 10 years, and there were nothing illegal in this. Have a bunch of employable credentials, send out enough applications demanding above average pay, and you are going to likely end up with above average pay.

The media should practice more circumspection when reporting about lawyers seeking plaintiffs or suing for class action status or at least mention how it’s a money making endeavour for that lawyer/law firm.

Many of these lawyers are vultures trying to capitalize on topical matters and they rely on this sort of coverage for free marketing, it’s as likely for this to be thrown out as it is for any other outcome, and the media should consider that before obliging with the reputational damage.

Edit: not sure why I'm being downvoted, I don't think I've made any non factual or offensive statements.

Of course class-action lawsuits are a money-making endeavor for law firms. We live in a capitalist society, where the way we make things happen is that we make them profitable. In particular, class-action lawsuits exist so that law firms can find it profitable to take on cases where there is a wrong to be righted, but any single individual isn't going to find the cost of going to court and paying a lawyer to be worthwhile.

If you think that lawyers being motivated by money corrupts the legal system, well, I would agree and apparently so would Judge Richard Posner, who retired this week because he suddenly realized how unjust this all was. But, well, this is the system we have now.

Posner's an interesting case. He claims that he is sick of how the under-lawyered get mistreated in the courts, but he's been a consistent advocate of many legal theories to harm marginalized people and animals: he's anti-privacy for citizens, but pro-privacy for police, anti-souveillance but pro-surveillence-by-State, anti-animal-rights, anti-antitrust, in favor of a free market for selling children, opposed to equality of educational opportunity even though he doesn't believe in genetic racial intelligence and does believe that Black children are systematically oppressed, and even though he thinks copyright/patents go to far he believes hyperlinking and paraphrasing news should be illegal!


My more pressing (no pun intended) concern is with the press' coverage of these suits, when they don't lay bare the underlying machinations they are basically choosing sides.

>The media should practice more circumspection

They won't for exactly the reason the lawyers do this. This is a trendy, hot topic (look how many comments it has on HN so far). This is "news" fodder for media outlets. They don't mind that they're being taken advantage of as long as they get a non-zero amount of ad views out of the deal. Win/win for the lawyers and the media, although definitely a loss for those reading.

The fun part of these articles is reading all the comments that literally post proximate causes for the gender wage gap and somehow conclude from that that the gap (which they just validated) does not exist.

This whole process infuriates me. The gender wage gap is an empirical question. It exists. If we believe it's a problem (I do; you may not), it has the easiest solution in the world: give women more money. Seriously, I just solved the gender pay gap, right there: give women more money.

There's all kind of ways you could do that (wage mandates, tax credits, etc.) and they all have pros and cons, but searching for proximate and ultimate causes here is kind of stupid. It exists; if you think it's a problem, the solution is breathtakingly obvious.

The confusion here is between "women get paid less on average" and "women get paid less on average because of sexist treatment/assumptions". If the latter, more specific statement is true, then we have a problem. However, if women are paid less because of less career-oriented personal life choices, maybe making an issue out of it is an overreaction.

Left-handed people earn less on average than righties. Tall people earn more. With a sufficiently large sample you'll probably find a correlation between salary and hair color, hand size, skin pigmentation, freckle density and a million other arbitrary factors that we've long decided are not worth fighting about.

Why are the motivations of the people paying women important to you? The gap is an outcome of a complex system, and the outcome is what is problematic.

Mandating equality in outcome is probably a stance on fairness that the vast majority of Western society disagrees with. It cuts to the core of what we see as fairness. It's entirely unsurprising that this gets people riled up, though most of it is as the GP said about whether the cause is discrimination or women's own choices.

Personally I think the left in general need to grapple with the fact that we're not all equal, rather than saying that idea is tabboo because it has lead to horrible places before. The fact that the left has largely focussed on suppressing this idea that many people believe (and on an individual level is self evident) rather than tackling it head on is what leads to the backlash against political correctness because people feel like the emperor has no clothes, but they can't say so.

Not to discriminate isn't to treat everyone equally, but based on merit. If I hire/fire an employee because they're black/male/straight/christian, I'm discriminating. If I hire/fire a person because of their work performance, I'm being fair.

I don't care about the motivations of the people paying women. I care about their employee utility function, in which gender shouldn't be a factor.

Because discrimination is bad. Fighting discrimination is fighting for equality of opportunity. This makes sense and we all basically agree it's the right thing to do.

Unequal outcomes are okay. Fighting unequal outcomes is fighting for equality of outcome and it's unfair and makes no goddamn sense in a society where there is free will or any variation at all between the members.

There are plenty of data-driven studies that demonstrate empirically that there is no gender gap in pay. For example, in Google's case, they take many precautions to ensure that compensation decisions are made without knowledge of an individual's gender: https://www.blog.google/topics/diversity/our-focus-pay-equit...

They even remove names from resumes for hiring decisions to avoid unconscious bias from the decision makers.

This is about as close as you can get to a completely fair system and is far above and beyond what can be reasonably expected from a business.

It seems that your solution is to simply increase bias in the labor market rather than try to make it more efficient. Over time, marketplaces tend to abhor inefficiencies and seek equilibrium, so not only would "give women more money" inevitably have unforeseen adverse side effects, but it also runs counter to our fundamental social and economic principles.

Seriously, the fact that otherwise intelligent people claim this with a straight face is kind of crazy. The "data" here are looking at numbers and seeing which one is smaller and which one is larger, and the wages of women are smaller.

Welp, that's the whole field of statistics down the drain! Who needs complex analysis and accounting for confounding variables when the answer is simply "Big number good, small number not good!"

What does a confounding variable have to do with anything here?

The question is "do women make lower wages than men"?

You are adding confounding variables I assume because you are uncomfortable with the fact that the answer is "yes", but are less uncomfortable if there's some reason that it's yes other than "people deliberately pay women less".

Of course there are reasons women make lower wages than men. That doesn't really matter, though: if the gap is itself a problem as an outcome, the fix is very simple, and the reasons don't particularly matter.

If I asked which of the two of us is taller, all that takes is a measuring tape. The fact that one of us might have gotten better nutrition as a child, while possibly interesting, doesn't actually do much about the fact that one of us is really taller than the other.

Suppose women get paid less because they work less hours (this is actually part of the reason); Do you suggest women should get paid more per hour simply on account of being born with a vagina?

What if women get paid less because they tend to pick professions that happen to be lower paying? (This is also true.) Do you suggest we just give them raises? What if the reason the profession is lower paying is because businesses in the field have lower margins? Where are they supposed to get the money for these raises, take it out of the men's paychecks? Speaking of, do you give raises to everyone working in the lower-paying field, or just the women?

See, I think the bone of contention here is that you assume the problem is women being paid less. Most people have no problem with women being paid less; Their problem is with women being paid less SIMPLY BECAUSE THEY'RE WOMEN. This is why very few people are comfortable with the solution of "just pay all women more." Paying someone more because of their gender is literally the definition of sexism.

Why do you think the gap itself is a problem as an outcome?

Because in every economy we've looked at, gender income parity is strongly correlated with overall economic growth. In addition, research on microlending suggests that this link is statistically co-causal (improving either one improves the other).

If you prefer a more intuitive take, most of the attempts to dismiss the gender wage gap say things like "women take more time off for families" and "women choose lower-paying occupational fields". But there's no a priori reason those activities and fields should not be more remunerative than they are, and there is evidence that making them more remunerative is beneficial to the economy as a whole.

>evidence that making them more remunerative is beneficial to the economy as a whole.

How so? What difference economically does it make whether a male or female fills a position

Could you please cite the studies you're referencing?

I should have tried this when I took Regression Analysis.

I don't understand how just giving a group of people more money will solve the problem. That's is an artificial way of addressing it, it's just a band aid.

Let's look at the causes , and address those.

I think that's how you solve it. It is pretty sexist to say , "give someone more money because they are a women"

Ummm... you realize band-aids solve the problem of bleeding, right?

Similarly, giving women more money solves the problem of women getting paid less money.

The same logic would mandate that everyone gets paid the same amount of money, which I personally would be sort of OK with, depending on the details, but it is generally considered such a radical idea that it suggests there's something missing from your argument.

why isn't everyone paid exactly the same? because that would suck for all of us here and be great for everyone that doesn't work as hard. I was messing with strange *nix distros for years and teaching myself programming before I got a tech job. I would be seriously pissed if someone that didn't know as much got my same job and pay just because they were a _____

I could understand getting mad if your pay was lowered, but how are you harmed if the other person's pay is increased?

We make all sorts of incredibly bizarre (from an outside perspective) decisions about who gets paid what; the fact that we've fooled ourselves into thinking that it's because of supply and demand curves may be comforting but is completely unfounded in any data.

Where does all this "equalizing" money come from? Does it just pop into existence because you want it to?

Yes, in fact, for some value of "you"


You can't create the amount of money you are talking about without severe economic consequences. That should be obvious.

Giving preferential treatment to anyone based on their skin color, gender or sexual orientation instead of the value they provide or the skills they posses is exactly how not to address diversity issues. This line of thinking only fuels more outrage from other minority groups.

A band aid is a temporary solution to a problem that fixes itself. Do you expect this problem to fix itself?

Actually I do imagine that more gender parity in management and boardrooms would probably organically fix the problem, eventually. This is, ultimately, an outcome of culture(s), and cultural problems are situations where you can fake it until you make it (though it can take a generation).

But isn't the context here "women in tech"? As in, SWE/STEM positions, as opposed to upper management?

> it has the easiest solution in the world: give women more money

I don't understand why this is so difficult to understand or controversial - especially in an industry with such a skills shortage.

If you want more and better employees start hiring more women and paying them.

That is a market solution - consider it an investment. Pay them now, pay them more, have a larger hiring pool later.

Jesus christ, no. Scroll up.

I wonder if their skills are on the same level as the men who were paid more. And I'm curious, did they ask every single man in Google how much they get paid? Probably, there are some men who make even less than they do (or did).

Please don't consider me a sexist, but this kind of stuff is everywhere now and it's hard to tell whether it's a truth or not.

This is ideological warfare, it has nothing to do with logic.

Has anything happened with the Department of Labor case recently? Last I heard, the specific organization that was pursuing the case was being defunded / split up by the current administration. It sounded like they had some interesting analysis of the data, and I don't think that's ever made it to a courtroom.

I'm pretty unsurprised that leveling is a good way for bias to sneak in. My experience as a man applying for a Google position and also talking to women applying for Google positions is that leveling is extremely opaque, more so than the salary offer, and the same candidate could easily move between L3/L4 or L4/L5 essentially at the whims of the recruiter and the interviewers, and the same role can be filled by multiple levels (e.g. there isn't headcount that's open at L4 but not L5). And this would be consistent with both Google's claims that people of the same role and level are paid consistently, and employees' claims of pay discrepancy.

Also, here's the original complaint: https://www.documentcloud.org/documents/4044053-Kelly-Ellis-...

It's rare you'd have headcount open at L4 but not L5. There's a significant jump between levels, but not to the extent you're saying.

Leveling is a function of interviewer recommendations and hiring committee approving a candidate for a certain level. We usually interview people with a target range in mind (e.g., L4-L5) that's based on your experience and current role.

Base salaries don't have a huge range until you get very senior (e.g. Director). There's a LOT more variance in stock grants. I'm sure there are people two levels below me that have a higher overall comp due to huge stock grants.

Source: I work for Google, interview a lot, but don't serve on hiring committees. This isn't advice, and may not apply to all areas of the company.

> Has anything happened with the Department of Labor case recently?

Last I heard was the judge denying a request for more information, saying DoL was on a "fishing expedition" and didn't have anything to back their case and justify said fishing expedition.

Got a link?

WaPo article (with provisional ruling linked as well):


Well that's odd. Why would DoL need names and email addresses and phone numbers of employees? Didn't they already claim to have a statistical analysis of salaries based on previously-gathered data?

Is this related to the thing where Google says getting the data would be too hard (i.e., do they want to do their own analysis of competitive salaries based on scraping LinkedIn or something)?

Flat salaries may fix this? You basically join at the same bracket publicly. No performance bonuses and all salaries increases are directly based on years put in. Everyone joins with the same options across the board. For each male hired there must be a female hired. No roles. Everyone has the same title: "Engineer". Off sites are strictly prohibited.

Is this what the lawyers want?

Haha that'd be ridiculous! Imagine working all night and getting paid the same as someone who works half as much.

That being said, salaries are not determined by how hard you work but rather how much the company needs you to stay.

>>Imagine working all night and getting paid the same as someone who works half as much.

You don't. And that's where it will all begin to break down. Things as a whole will be mediocre.

> Imagine working all night and getting paid the same as someone who works half as much.

Isn't that the ideal?


> Imagine working all night and getting paid the same as someone who works half as much.

It's the socialist dream!

> For each male hired there must be a female hired.

General grad CS populations are split 4:1 in favor of males. It won't end well when you cut out a significant amount of your candidate pool because you want to promote "diversity". Why 50:50? Why not majority women? Any cutoff would be arbitrary.

Incidentally, this is not too different from how top-tier law firms work.

For example: http://abovethelaw.com/2016/06/breaking-ny-to-180k-cravath-r...

> No performance bonuses and all salaries increases are directly based on years put in. Everyone joins with the same options across the board.

This is how Japan worked for a long while; now it's moving toward a more meritocratic approach with performance bonuses and raises because the fact of the matter is that it sucks for people who work hard to be paid the same as the guy who comes in 9-5 and does average work.

Who would want to work at such a place? Last I checked this is still a free country and you can work wherever you want.

You've described the typical "up or out" professional services firm.

> Ms. Ellis, who left the company in 2014, says that almost all of the female software engineers at Google worked in front-end jobs while men worked in back-end roles.

That seems both highly dubious and easy to verify.

Anecdotally, I spent a couple of months this summer going through team selection at Google NYC, with a specific request for back-end, low-level, preferably C++ roles (app security, build systems, etc.). I didn't meet or talk to a single woman through the entire process, with the exception of reception, another candidate, and someone who was covering for my recruiter when he was on vacation. I asked every manager I talked to about diversity on their team, and they all gave wishy-washy answers (at best; one told me that women just weren't interested in his team's work). If I'm remembering right, of the teams that I could have accepted an offer from, exactly one had exactly one woman.

This wasn't the only reason that I ultimately turned down the Google offer (and spent so long in team selection after passing the interview), but it was certainly one of them.

There are some back-end teams at Google that have lots of women; Chrome security comes to mind (I don't know if that's "back-end" in the common sense of the term, but it was the sort of team I was interested in). But I don't find it particularly dubious that these are the exception and not the rule.

Keep in mind that because of their scale, jobs that would be considered full stack or back end at other companies are probably considered front end at Google.

You're half right. Google doesn't hire purely frontend engineers. They have a job listing for them, but once inside everyone gets the same title: Software Engineer.

At Google, the "frontend" work also usually includes the server which serves the frontend code - this means engineers need to be not only capable at UI development, but also be familiar with the Java systems that exist at Google.

From this perspective, frontend at Google is indeed similar to a full stack role at most startups running on AWS or GCP. Backend at Google is more like working on AWS itself.

Edit: Nvm. Alex3917 defines it better than I did.

At a two-person startup, the front end developer is the one doing Angular or React, and the back end developer is the one doing Django or Rails.

At a mid-sized startup, the front end developers are probably also responsible for Django / Rails, and the back end developers are responsible for things like managing the Hadoop cluster and the machine learning components.

And at a large company, anything built with existing libraries might be considered front end, whereas back end might be creating a new database or whatever. I don't claim to know how Google works, but as companies get bigger the back end tends to go further back, and so the 'front end' encompasses more also.

I haven't worked at Google, but I have worked at a very large tech company, and this isn't accurate.

They still hire frontend devs. Those frontend devs still do front-end JS with libraries. Some of them work on in-house libraries, but their focus is still narrow. Same for backend - they use the industry standard term to mean the industry standard.

However, the differentiation of teams/departments is much higher. To take on your argument directly, a project to create a new database would have software engineers but be on an infrastructure team. They don't call themselves backend engineers or recruit backend engineers, they call themselves software engineers and recruit people with skills writing system software.

Well, they sort of do, because those terms don't map so cleanly when you begin spreading into a multi-tiered architecture.

You could easily end up with 'server side' work, which exists solely to augment or support a UI, and is not part of the 'core' business applications/services.

You could, but at Google "frontend" work means things happening in the browser.

I worked in infrastructure and to us 'frontend' meant user-facing services while 'backend' are services that only other services are talking to.

What about Android?

I am not sure about US trends in Eastern Europe a fairly large % of FE devs are female same for QA roles.

Ladies, please, never take the first offer. My wife did that (she's also a programmer) and she's getting paid shit salary by a shit company to get shit on all day.

From Sheryl Sandberg's Lean In

"But what's interesting," she says, "was that when my brother-in-law and my husband were saying 'negotiate, negotiate, negotiate' – when I finally said OK I'll do it, because no man would take the first offer, I then thought to myself, I felt like I needed a justification for doing it. And it turns out that's what the data says: men can negotiate without apology or justification. It's expected. If women negotiate, they need to justify it. It can't be that you want more for you. Because that's what men get to do." As she writes in the book, "success and likability are positively correlated for men and negatively correlated for women."


On a related note, my company raises salaries yearly as a percentage of the current salary. Even though those with lower salaries get a slightly larger percentage, everyone gets somewhere between 8% to 15%. Having a higher initial offer and asking for a higher raise (and arguing that you earned it) results in better raises. Nobody expects everyone to get the same salary.

A purely capitalist solution would be to fire all the men and keep only the women because they are doing comparable work for less pay. In a capitalist market why hasn't this happened already ? This is exactly what happened when much of manufacturing moved away from USA to china, but for some reason this has not happened in the male female salary disparity. A company can earn diversiy points and save money by not hiring expensive male engineers.

Someone tried this already, didn't go as well as planned.


At best a story about someone running a company with no idea of how to hire or manage for culture fit, at worst a fiction made to bring in some cash from the daily mail, a publication generally known for their lack of journalistic depth.

This might be true and I hope they win.

But unfortunately everything that they say applies to men too. I wonder if this due to lack of negotiation skills. I have seen this problem being under compensation for the same skill and peer levels for men too.

At work, In one case I discovered a colleague being paid almost 50% times higher than I was. In another case in only a casual lunch conversation I discovered a colleague at the same level having RSU's almost triple my entire compensation. I also discovered while I moved to US from India(I moved back for Visa expiry reasons) that some colleagues had even negotiated green cards through really acrobatic legal work. Promotions, foreign travel, bonuses etc.

Over 10 years in this industry I have seen ability to program well, or even do bigger software work like build scalable and stable systems isn't worth two shoes in this industry.

One must have the ability to be politically skillful, negotiate well, know how to be well connected up management and use that leverage to further your career in both money and positions. I've tried to learn this, and failed. Unfortunately this turns out to be not something you can RTFM and learn.

We are all salesman - and a better salesman gets better results. As nerds we tend to discount, dismiss, overlook, whatever, salesmanship skills. But they still count big time.

We sell ourselves all the time, like it or not. Might as well try to learn how to get better at it.

All aspects of ourselves can be tweaked to better our salesmanship - grooming, dress, posture, getting our teeth fixed, tone of voice, words used, email protocol, etc., the list is endless.

This all started with the book "Dress For Success" by Molloy who noted that businessmen wearing tan overcoats did better than those wearing black overcoats.

What you have written is 100% true. Especially about the dress part. A while back I started wearing a shirt and pant to work. People, especially juniors do tend to take you a lot more seriously if you look formal and well dressed, compared to say showing up in denims and t-shirt.

But the negotiation and sales part. That's the issue. Its not easy to get good at that.

I noticed early in my career that people who dressed better got noticed more - i.e. promotions. Habitually wearing flipflops and track shorts to work won't work.

> But the negotiation and sales part. That's the issue. Its not easy to get good at that.

You're right. I work at it all the time, and I have a very long way to go. It will always be a "work in progress" for me.

But at least one can try to not be simply terrible at it. I still shudder at the stupidity of some of the things I did that clearly damaged my career.

> This might be true and I hope they win.

Can you elaborate on this? Namely, the burden of proof is higher than 'might be true.'

I'm guessing you meant something more like, 'If this is true, I hope they win.' I can never tell, anymore.

If this is a merely negotiation problem than the problem gets stated in a different tone.

I have been told many times that I need to work on my negotiation skills. One manager has even asked me whom do I go lunch with. And why don't I go to eat with the big shots. Of course you need to act all pally and buttery, saying yes to anything they say all the while.

What they mean to say is stop expecting the system to be fair and do what it takes to win.

A friend of mine worked for years at a firm, and did well as a salesman. But he would get overlooked for promotions. He finally figured out why.

He was frugal, and would bag his lunch and eat it at his desk. The others would get together and go out for lunch. He failed to realize how much networking and business was conducted over lunch.

A corollary is that working remotely is probably not a good idea if you're ambitious.

>>A corollary is that working remotely is probably not a good idea if you're ambitious.


And also communicating and blowing your trumpet at all opportunities available. You don't want to be overdoing it. But definitely enough of it make every one hear about you and your work.

That makes sense. And, yes, negotiating for yourself is a very good skill to learn. The other person may not be an adversary, but they probably don't have your best interests in mind. So, you'll want to represent your best interests.

You say everything that they say applies to men too — but that seems like a reason that they will lose.

To win, they have to prove that they could reasonably have been paid more but weren't because they were female. If they show that they could reasonably have been paid more but weren't because they are bad at negotiating, they get nothing.

So, yeah, don't mistake discrimination law for something which would make sure people are paid what they are worth.

>>To win, they have to prove that they could reasonably have been paid more but weren't because they were female

I'm only saying if you replace 'female' by:

1. 'not good at negotiations'.

2. 'not good at making friends up management'.

3. 'not having good political skills'.


You still arrive at the same result, regardless of the gender. Which happens with us men too.

Somebody at Google could write a memo proposing that sexism might not be the only possible explanation of why a majority of their top engineers are male.

These ladies are behind the times. Gender politics at Google are now so advanced that feminism isn't even a thing. The "non-binary" and "genderqueer" have thrown the feminists under the bus. Frankly I'm shocked that the informal voluntary salary survey had the bad taste to try to classify gender.

What is the end goal here? Companies being forced to publish and formalize pay tiers?

Can Google publish a report on non-pay differences by sex at Google? Sick (or personal) leave taken, overtime worked, vacation time availed of, etc. by sex?

Obviously with everyone being equal and doing equal jobs, the above shouldn't really be an issue?

> What is the end goal here?

Presumably these women would like more money.

So what? I'd like more money too. Should I simply get more money because I want it?

I'm not sure why you are so grumpy and taking the questioning tone directed at me, given that my response simply proposed a possible explanation for GP's question.

The courts are a lever to improve compensation just like negotiation is, if that makes you uncomfortable that's unfortunate, but it is how the US economic system is structured.

Court cases have significant costs, they don't simply "want it" but hired (presumably) decently skilled attorneys to argue their case, have real social and personal risks associated with being publicly associated with this case, etc.

The end goal is a meritocracy where women arent discriminated against, and are paid what they are worth.

The current situation is not a meritocracy and it needs to be fixed.

In capitalism you are already payed what you are worth. If you agree for a shit salary that is your problem.

Actually, if a company is illegally discriminating against people based on gender, then it is THEIR problem.

If they do that then they can enjoy their millions of dollars in fines for doing something clearly illegal.

Person A asks for 90 dollars per year; Person B asks for 100 dollars per year.

Person A does not ask for a rise after one year; Person B asks for 10 dollar rise after one year.

TBH I don't see how it is a company fault.

They are at fault because they are systematically underpaying women of equal qualifications.

It does not matter what the "reason" is. It does not matter if the bias is because equally skilled women merely ask for raises less often.

If for any reason at all, you are systematically underpaying certain groups that are equally qualified and equally productive, you are legally liable and the onus is on you to fix it.

I think that if the person is not asking for more money company has a right to not give him more. Isn't that how capitalism works?

Paying 2 different groups of people that are equally qualified is called illegal discrimination. It doesn't matter what the reason is for the discriminatory pay difference.

It doesn't matter if it is capitalistic to illegal discriminate against certain groups.

So no, they do not have that right, if the result is an illegal pay disparity. The company would be required to correct the pay disparity, even if the reason why the pay disparity exists is solely because 1 group didn't ask for more money.

There is an empirically proven wage gap, yet almost every comment to this article is people conjuring up social constructs to explain how that is perfectly fair.

It's not fair though, that's the entire point of the inequality in the wage gap.

What I do t understand is why people so fiercely defend this wage gap. If women were paid equal to men, we would have lost nothing.

Disagreeable personalities tend to get paid more. When a man is disagreeable, people usually get wide eyed, but then accept him as kind of abrasive, but lovable. Women are either less likely to be disagreeable personalities, or less liked when doing so.

If women get paid less, why doesn't google hire only women to save money?

Disagreeable people also tend to get fired more, in my experience. Perhaps it balances out, which would explain why there are still disagreeable people.

I gather that Google has pay grades. But are they reviewed in periodic performance reviews?

And to what extent is salary synced to that negotiated at hire? I mean, if women (as I've read) generally don't negotiate on salary as hard as men do, are they indefinitely paid proportionately less?

I don't know much about civil suits... Is the complaint a matter of public record already? Would be interesting reading.



If you wouldn't mind, please send me $20k per year. It's nothing to complain about, after all.

Ok, give me a job that pays 80K and I will send you the 20K without complaining.

How does this point invalidate their issue? All you've done is brought up another problem. Now there are 2 problems.

is not about invalidation, its about priorities; have you seen any articles in the front page about those millions of Americans? Me neither; but dozens about how someone somewhere is winning 10% more than other someone and that the gender was clearly the reason.


We detached this subthread from https://news.ycombinator.com/item?id=15254245 and marked it off-topic.

I don't know if you are trying to be sarcastic, but this is a very hot topic in the nursing field. https://www.nurse.com/blog/2017/03/17/making-the-case-for-mo...

From the article,

"a University of San Francisco study of data from 1988 to 2008, extended to 2013, demonstrated male nurses made an adjusted $5,148 per year more than female nurses. This inequality has persisted over 20 years with no discernible trend toward resolution in the future."

"The percentage of men in nursing has slowly risen from about 2% in 1975 when I entered nursing to almost 12% (or 330,300 nurses) now."

And I'm not trying to imply anything. I just thought it was noteworthy.

Hahahaha it's funny because "nursing" is a gendered term for a class of front-line medical occupation that is almost certain to be vital to the provisioning of cost-effective health care in the 21st century where it is vitally important that we both raise the status of the profession to get more people into the labor force to do it and also to raise its status among consumers so less people have to wait for ludicrously expensive general practitioner doctors when a qualified nurse-practitioner will do and that's totally what you meant, right?

The best joke!

Because one of the several definitions of "nursing" is gendered does not mean it is a "gendered term", especially when used in a non-gendered context.

Unless I'm immune to the irony, this looks like a ridiculous out of place outburst.

You don't think that people are accustomed to thinking of "nurse" as a female role?

Sure but I don't think a mother breastfeeding their child being called "nursing" is the reason

I looked up the etymology of the word and this is in fact where it came from: https://nursemanifest.com/2012/04/24/some-history-on-the-ori...

> The first instance in English of nurse occurred in the early thirteenth century as the Anglo-Norman nurice, derived from the fifth-century post-Classical Latin nutrice, a wet-nurse (hired to provide an infant with breast milk when the infant’s mother would not or could not do so), although by the time it entered the Middle English lexicon, it had already absorbed the figurative sense of any female caretaker of children (Oxford English Dictionary 2010). Etymologically it is related to our modern word nourish, to feed.

Unlike other professions like stewardess, nursing has never broken away from the gendered roots of the term.

Language does evolve, though, for better or for worse. I'm sure the term has contributed to the "nurses must be female" perception, but I doubt renaming the profession is going to do anything

Yes language evolves, but it hasn't in this case and doesn't look to be anytime soon. Compared to much of the language it's remarkably stable.

I'm surprised it didn't get changed back when we were changing actress to female actor, air hostess to flight attendant and all those others.

But a nurse is a nurse whether male or female, it's not directly comparable. And since when is the term "actress" not a thing?

I think you agreed nursing is gendered in favor of females.. no one complains about that.

Construction jobs are gendered in favor of males.. no one complains about that.

So if 20% CS majors are women, and 20% programming jobs are held by women why is that a travesty?

Why do we call diversity initiatives to increase the hiring proportion of one group ABOVE what is naturally expected fair rather than unfair? If they are paid less, sure we need to address THAT. But seeing they are represented as expected in the workforce, biasing your hiring process further in favor of women necessitates biasing it against every other demographic. And why does no one give a shit about diversity in other less desirable occupations?

Thats what OP is pointing out. Its not a joke. There are problems for women in tech. But there are also simply people who want to grab as much special treatment as they can get for their own group over others, when it is in their best interest. And thats why its important that people clamor so much more about diversity in <CS/high-paid profession X> than any other - many of them will support any measure that improves their odds regardless of what it means for society. At that point it becomes pure self-interest not social change for the good.

42 minutes before you wrote your comment, a comment was posted which includes a link to an article that does complain about the lopsided gender distribution in nursing: https://news.ycombinator.com/item?id=15254670

When I read these comments just now, it was literally 2 comments above yours.

Because feminists don't consider nursing and construction to be jobs worth fighting over. They have a target on lucrative tech jobs and corporate leadership positions due to being inside air conditioned buildings with high status, and high pay. Nobody is going to be fighting to be a coal miner.

Nursing is also a bad defense for discrimination in software development because male nurses typically excel in the field. There are no lawsuits by aggrieved male nurses because they get promoted and are approved for special positions at the same rate as their peers.

Not to mention, lifting a 300 pound patient, perfecting an IV stick, or shouting down a raging alcoholic is pretty fucking manly.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact