Edit: Lots of up and down votes. Why is this question so controversial? I'd like to see an argument as to why it's an invalid or flawed question.
To me, this implies that our engineering organizations, and entire company, need to have an appropriate amount of empathy across a broad spectrum of geographies, cultures, experiences, perspectives, and the like. You only get that by having a diversity of talent, and more importantly, inclusion in your engineering practices.
When I talk about diversity, I usually say; "There are two forms of diversity, DNA diversity, the stuff we usually talk about in terms of color, sex, etc, things you can see, and then there's diversity of perspectives and experiences. You need the diversity of experiences and perspectives. In some cases that's conveniently wrapped in some forms of DNA diversity, but is not exclusive to that".
I don't think the question is at all controversial, and we should not be afraid to openly talk about it.
Did they think about what they said there before printing? Because there are plenty of organizations which we definitely do not want to achieve more (e.g. drug cartels, nazi parties).
And are you aware that non-DNA differences also create divisions?
Well, this is a pretty contrived example, but IIRC a major company (HP?) made facial-recognition software that couldn't handle people with dark skin.
If there had been people like that on the dev or testing team, perhaps it would have been noticed earlier.
I would argue there's often correlation between these types of diversity
In the OKR framework, being able to measure your progress towards the goal is important. The real question is, which aspects of people who work for your company do you measure to ensure proper decisions around inclusion to achieve the equity goal?
It's easiest to measure by 'skin deep' factors b/c that is what human beings most easily make poor decisions on (fear, bias, stereotype, self-segregation, NIMBY-ism, etc). It is also required for companies to report to the US government on these factors because of our history of poor decisions (to put it lightly). It is therefore easiest to use that as a metric.
Serious question: If implied in your question that the goal should be equity in 'thought', how do you propose that is measured?
Well, to nitpick even further, those propositions are logically mutually exclusive.
"Does the Chief Diversity Officer include diversity of ideas and experience [as well as surface traits]?" Or is it [only surface traits]"
Abstracted, could be "(p and q) OR (p and !q)" which is literally mutually exclusive.
No, I'm not usually so pedantic, I just thought it was interesting.
Oddly enough, my use of or seems to violates the very rule it is used to describe.
OP mentioned "ideas & experience," not "thought," which seems like an intentional framing as something unbounded and immeasurable.
Diversity of ideas could be measured by the average number of options/solutions that are seriously considered (and investigated/piloted) over the course of multiple projects for a team.
Diversity of experience seems somewhat obvious to me. But if you want clarity on this as well, the idea would be to value various types of experiences in the same way that companies value diverse outward traits like gender, sex, skin color, racial identity, etc. It's a balance. You could hire a person of each gender/skin color combination, but if they all grew up in the New England suburbs and all of them went to either MIT or Harvard, you are generally NOT going to have a diversity of experience, even though everyone _looks_ different. On the other end of the scale, you could hire one person from each type of school, big state school, small technical school, ivy league, "public ivy", liberal arts college, bootcamp graduate, etc. Hire people native to your country/culture, and people who come from a different part of the world. But if they are all white men, you are not going to realize as much benefit.
I think it's important to make an attempt to combine all of these concepts to come up with something that approaches the concept of "diversity of thought."
> gender/skin color combination, but if they all grew up in the New England suburbs and all of them went to either MIT or Harvard, you are generally NOT going to have a diversity of experience, even though everyone _looks_ different
Unfortunately, there are many stories, points of evidence, and history that say people who look different, but come from the same place and education level DO have different experiences. Race and gender are exponentially powerful factors that can change a person's experience and outlook no matter how wealthy they are or what school they graduated from.
It's become apparent to me that diversity is an ambiguous terms and its interpretation can vary a lot over many factors and time. I am not discounting any definition of it. When it comes up, it feels like people are on different pages with it. It could more prudent to state the definition when it is said.
What they seem to be doing is judging people based on surface traits, not the content of character. Their approach to diversity is a regression and not good for progress.
>It doesn't matter what color people are.
I hope you realize the absurdity of this statement in isolation...
Do you believe that a poor black kid from Detroit is only different in "surface traits" as compared to a rich white kid from Orange County?
Or would you be willing to concede that there may be reliable correlations between some "surface traits" and diversity of life experiences and viewpoints? Demographic analysis of things like voting patterns in the US seems to suggest differences of such magnitude that I doubt you'd find another factor more strongly predictive.
I don’t think this is a good goal. Unless you mean equity of opportunity. Equity of outcome is a ridiculously foolish goal in that outcomes will vary substantially and trying to have equity at the end on arbitrary human factors with easy to measure biases (gender, race, etc).
So the goal is not 45/45/10 for gender distribution for all roles. As that is obviously impossible as roles change and then people would need to be redistributed ad infinitum (eg, project managers have “perfect” gender diversity of 45/45/10 today but now the role is changes and split into product owner and product manager. Does this mean that the roles must include the same gender mix?)
I think, especially with improved technology, that 10% will be more common. Cynically thinking, it will be easier if there were some specific quota. Gender is probably the easiest protected class to change after religion, so it’s especially sensitive to outcome quotas.
So, this system is used in the United States, right?
I think we want an equitable outcome and that most people (regardless of any other opinion) would actually agree that equality of outcome is not necessary.
Is it fair that now a sub population has different gender distributions? Is it fair that 90% of programmers are male? Etc etc. I think it is counterproductive and too late to making meaningful changes based on outcomes.
Perhaps if you get to a high enough macro, but even then, I see logical weaknessss in opinions comparing income based on gender because outcome does not, necessarily, mean bias. It’s just easier to measure.
Plus, surface diversity is much easier to measure; and you optimize for your metrics.
[0] Ideally a diversity officer would stand up for Damore. But I have no idea what went on behind the scenes of that decidion.
That claim depends entirely on how you interpret Damore's behavior. Tolerance does not and must not require tolerating intolerance [0].
I'm not stating a claim as to whether or not Damore's behavior should be considered intolerance. I don't have a take on that. But without having some opinion on that, you can't claim a priori that a diversity officer is obligated to enable his views.
The statement I made in my previous comment does not actually require this however. Even if Damore was actually acting in bad faith, fireing him would still have a chilling effect unless it was thought by every Googler that he was clearly acting in bad faith.
Further, I would excpect a Chief Diversity Officer to understand how perception alone can create a hostile environment, so ignorance of this (or disagreemrnent about the facts of the case) is not a good excuse.
You can't help side effects. If someone is acting in a way that is detrimental to the company, you have to fire them, even if you know it will upset some other people in the company.
On the other hand, it's rather convenient how much intolerance can itself be justified by simply quoting that and just a bit of rules lawyering.
I honestly hate how often this is trotted out. Does no one realize that this claim is not based on any evidence whatsoever? Why prefer Popper's claim over Rawls' or Jefferson's?
This is typically used to justify yet more intolerance, and around and around we go.
"In this formulation, I do not imply, for instance, that we should always suppress the utterance of intolerant philosophies; as long as we can counter them by rational argument and keep them in check by public opinion, suppression would certainly be unwise. But we should claim the right to suppress them if necessary even by force;"
Seems Popper would not have much issue with tolerating Damore, but tolerating things like Neo-nazi marches probably crosses his line.
Probably not even then! Neo-nazis are a tiny minority that's suppressed by public opinion. Now if they were to suddenly start gaining political power, then we have a real problem.
And yet it remains an interesting thought experiment, doesn't it? Didn't it work on Reddit, in which they found that driving hateful communities out raised the level of discourse? And a possible way to structure society. Hitler himself admitted that had the Nazis not been tolerated, they wouldn't have gained the foothold they did. Not only do I prefer Popper's claim, but I prefer the Frankfurt School's claim too[0]. Which, in my view, stands firmly in line with the democratic functions of society envisioned by such liberals as J.S. Mill for instance.
But it comes down to ideology; it is not a crime nor an intellectual failing to dismiss liberalism on ideological grounds.
Sure, but that doesn't make it a sound policy for real political discourse.
> Didn't it work on Reddit, in which they found that driving hateful communities out raised the level of discourse?
Or did it simply fuel a growing narrative of censorship and persecution which is driving the counter-PC culture and electing people like Trump? I don't think analyzing this as a closed system is faithful to the point I'm making, because these changes can and do have negative externalities.
> Hitler himself admitted that had the Nazis not been tolerated, they wouldn't have gained the foothold they did.
So basically, had the majority of the country not been sympathetic to Nazi principles, then Nazis wouldn't have gotten elected? That strikes me as absurdly tautological. A minority of people cannot be resist the wishes of a majority via intolerance, and if good people are the majority, then they can safely ignore the minority.
> Not only do I prefer Popper's claim, but I prefer the Frankfurt School's claim too[0].
This claim at that link is pure nonsense: "Tolerance is a democratic principle, since it relies on the idea that nobody has an absolute claim on the truth"
No, simply false. Tolerance is justified by the recognition that all people have intrinsic value ala Kant, that they are thinking feeling beings and that unless you wish to launch your own campaign for genocide against people with whom you disagree, then you should seek to convince them of your version of the truth using non-violent means.
The only time violence is justified is in response to violence. The far left and the far right are both guilty of violating this principle.
I can't see why not.
>Or did it simply fuel a growing narrative of censorship and persecution which is driving the counter-PC culture and electing people like Trump?
No. And besides, on the logic that it might make people angry or convinced of their delusional views therefore we shouldn't do it is very poor reasoning.
>because these changes can and do have negative externalities.
In the same way that naive tolerance does?
>had the majority of the country not been sympathetic to Nazi principles, then Nazis wouldn't have gotten elected?
No.
"Only one thing could have broken our movement – if the adversary had understood its principle and from the first day had smashed with extreme brutality the nucleus of our new movement." (A Hitler at the 1933 Nuremberg Nazi Party rally)
"If our opponents had been clever, considering that political weapons were so unevenly distributed, they could have undoubtedly found ways and opportunities to make our success impossible." (J Goebbels, 1934)
"If the enemy had known how weak we were, it would probably have reduced us to jelly. It would have crushed in blood the very beginning of our work" (J Goebbels, 1934)
>Tolerance is justified by the recognition that all people have intrinsic value ala Kant
Not in democratic society; can you point me to which Kantian thinkers (including Kant himself!) whose thought formed the basis of any modern state? Tolerance as we know it emerged only with the emergence of widespread democracy after the French revolution beginning in Western Europe and the American revolution in the New World. Both republics founded on democracy (and therefore tolerance), not a Kantian maxim. As Marcuse said in 1965,
"Moreover, in endlessly dragging debates over the media, the stupid opinion is treated with the same respect as the intelligent one, the misinformed may talk as long as the informed, and propaganda rides along with education, truth with falsehood. This pure toleration of sense and nonsense is justified by the democratic argument that nobody, neither group nor individual, is in possession of the truth and capable of defining what is right and wrong, good and bad. Therefore, all contesting opinions must be submitted to 'the people' for its deliberation and choice. But I have already suggested that the democratic argument implies a necessary condition, namely, that the people must be capable of deliberating and choosing on the basis of knowledge, that they must have access to authentic information, and that, on this. basis, their evaluation must be the result of autonomous thought."
>The only time violence is justified is in response to violence.
I disagree, and as I said earlier, it's no crime to disagree with liberal dogma like this. But if we must go this route, what does violence say of structural violence, for instance?
Good thing that isn't what I suggested then.
> In the same way that naive tolerance does?
Define "naive" tolerance. Seems like you're either assuming the conclusion, or trying to persuade with loaded language.
> "Only one thing could have broken our movement – if the adversary had understood its principle and from the first day had smashed with extreme brutality the nucleus of our new movement." (A Hitler at the 1933 Nuremberg Nazi Party rally) > > "If our opponents had been clever, considering that political weapons were so unevenly distributed, they could have undoubtedly found ways and opportunities to make our success impossible." (J Goebbels, 1934) > > "If the enemy had known how weak we were, it would probably have reduced us to jelly. It would have crushed in blood the very beginning of our work" (J Goebbels, 1934)
Ah, so if they hadn't been sympathetic or they had been able to read minds so as to discern the intentions of Nazi party leaders. Well that's totally reasonable then.
Honestly, precisely what expressions of intolerance do you think would have prevented the rise of the Nazi party? Everything you've quoted is devoid of anything concrete.
> Not in democratic society; can you point me to which Kantian thinkers (including Kant himself!) whose thought formed the basis of any modern state?
Irrelevant. The question is whether Kantian principles entail the relevant conclusion. We are beholden not only to government law, but moral law as well. Or do you think it's ethical or virtuous to violate the spirit of a law while following the letter of the law?
Furthermore, my claim is not that all opinions are equally valid and deserving of equal attention. My claim is that changing opinions by non-coercive means is right. Doxxing, violence, and online mobs are not right way to change opinion.
> I disagree, and as I said earlier, it's no crime to disagree with liberal dogma like this.
Not a legal crime, but nowadays it's certainly a social crime enforced by mob rule.
> But if we must go this route, what does violence say of structural violence, for instance?
That it can and should be alleviated by non-violent means, taking into consideration social inertia, ie. that it takes time for a message to propagate and persuade enough of the public. As Per MLK, violence begets violence. All the most revered figures of effective social change advocated non-violent means.
There is considerable debate whether the terrorism perpetrated by the suffragettes actually helped their cause. They clearly turned much of the public against the cause, and at best, they gained publicity for the cause, which was clearly possible to do via other means.
I think this is key. Until "diversity" becomes so absolutely fragmented that you get back down to the level of the individual (e.g.you need a non-Hispanic Latino coder, who is Christian but not Catholic, was born poor but still got into an Ivy League, must be over 6 feet tall, and her name must begin with "A"), we will continue to burn resources, mental energy, and time on ensuring that we can't get sued.
There are plenty of Asians in tech (men and women) and yet we're always hearing about a "diversity" problem. This is why tech no longer uses the term minority - they use "underrepresented minority" instead.
Asians are a minority group, but are vastly over-represented in tech, to the point that even white people are under-represented by comparison.
The concern about racial diversity tends to focus on the idea that white hiring managers are subconsciously selective against non-white applicants. This doesn't make sense though when you factor in the minority groups that are actually over-represented.
So what has happened is that the argument has shifted to ignore Asian representation and only focus on black and Latino representation, while still coming to the same conclusion that white hiring managers are subconsciously racist and that the company needs to take corrective action.
The representation of Asians and Indians in tech should not necessarily reduce concerns for other minorities that are under-represented, but should definitely change the way those concerns are framed. The fact that it hasn't suggests that some sort of tomfoolery is afoot.
Of course it makes sense. It means that instead of subconsciously being biased against the non white category, they are instead biased against non white non Asians.
That was an easy logical deduction there.
Maybe it has less to do with bias and more do to with preference and culture and upbringing and other factors. Consider how much asian parents, for example, push their kids in education and toward certain careers. Or the over representation of black people in sports and music. Or the dominance of women in teaching and healthcare. Etc.
If you take popular preconceptions/prejudices around "white" "black" "hispanic" "asian", etc, maybe it doesn't match as well.
The concern is far more nuanced than that. It includes concerns about the the rates of representation all through the pipeline, from hiring back through University CS programs to differences in access to high quality primary and secondary education.
The problems are largely understood as being systemic, arising from unequal opportunity and different social signaling to different groups, not solely from the decisions of hiring managers.
Once all those studies that send out the same exact resume with the names John, Jane, T'yesha, Jamal, Xiaoying, and Shan Shan at the top find you get the same number of callbacks, then we'll actually be in a situation where people are being hired on their qualifications!
Besides, what do you mean by fair recruitment? Because personally in this context I view diversity and fairness as contradicting terms, since it would always be more fair to hire a candidate that has higher qualifications and is a good cultural fit over one that wasn't as qualified, but turned out to be a good match when it comes to meeting present quarter's diversity goals?
I believe that basing hiring decisions on physical appearance, race, religion, gender or political views is simply wrong. I recall there was a company that focused on distorting candidates' voices during phone interviews that effectively prevented the interviewers from distinguishing the interviewees' gender. If companies like Google want to actually be more "fair", perhaps they should move in that direction rather than introducing artificial quotas and justifying them with a vague "need for diversity"?
This almost guarantees the perpetuation of inequality. We need to stop hiring for that kind of fit.
I'm certain that I have heard of research (by IBM maybe?) that determined that more diversity made teams more productive, but the people on the teams were less happy/comfortable.
Maybe because nobody wanted to socialize with each other so they spent more time doing work?
> you're forming up an artificial construct that ultimately results in poorer overall performance (since you had to reject a number of candidates that likely were more qualified for the position in order to reach your equal distribution goals).
That's precisely where I disagree. Again, in a coal mine you might just want to hire for simple traits that can be easily evaluated per individual (like physical strength). But on the types of teams I mentioned, the diversity of experiences and opinions across the team probably matters as much or even more than the sum of every individual's "talents." I would guess that this diversity becomes more important the larger the team gets and the larger the audience for their products/services gets.
To be clear, I also suspect that more diverse teams are more difficult to hire and manage effectively, and that's probably a big reason that some companies either don't care about diversity or actively resist the suggestion to increase their diversity.
Diversity of race or gender is not really important when it comes to working by itself.
Diversity of experience and thought is very important when it comes to working.
What I mean is a very middle-class white family and a very middle-class black family producing offspring in a house next to each other and sending their kids to the same school(s): are not going to be functionally diverse enough for what I'm about to say.
Why?: because people who are thinking differently to you are approaching problems from other sides than you are, if you can be civil then you can build your product in ways you could never do alone.
Now; people who were raised in a different kind of culture or social class than you are basically by default: thinking differently.
It doesn't have to be colour, but in the US (from a brit perspective) your class system looks like black people are sandwiched between middle-class white folk and very low-class white folk. So some people might see that 'hiring black folk would increase our diversity of thought'
or.. more likely, they're being judged on metrics and skin colour/gender is an easy metric to game for.
Giving people an equal shot is noble and we should aspire for it. The business incentives come in because your product will be better with divergent opinions and by increasing the talent pool the supply of coders will meet the demand.. (and then they can pay coders less in the longer term)
There are some advantages. For instance, people with different religions or cultural backgrounds will have non-overlapping holidays, so your office might be open on days that you might otherwise have to close.
Or consider if you're creating a dating app, you would absolutely want a woman's input on how to reduce harassment (see Tinder vs. Bumble).
What you're missing is that many employers don't have the perspective whereby gender, skin color or political views are irrelevant. Biases against gender, race and religion are part of the reason that many professions have been dominated by certain races, genders and religions (although laws have attempted to counteract this effect.)
The point of diversity in the workplace is to force an implicitly prejudiced employment market to be less prejudiced than it otherwise might be, just as the point of labor laws and the minimum wage are to force companies to care more about their employees' welfare than they otherwise would.
Yes, but you can mitigate those biases with gender/race-blind hiring practices. This seems like a much fairer and more reasonable option than enforcing quotas.
- They only work at the screening stages. Most hiring pipelines conclude with an in-person interview, and that can't be blind.
- People are very good at reading between the lines and can often infer race and gender from education and work history alone (or at least, infer deviations from their preferred norms).
- It doesn't address the problem of industry pipelines filtering against diversity before the applications ever get to you.
- A manager who doesn't want to hire women can just treat a blindly-hired woman badly until she leaves.
- It implies to the company that diversity is something that must be hidden rather than something that should be tolerated or celebrated.
- It implies to applicants that you have intolerant managers who need to be tricked.
All I know is gender/race-blind hiring works fantastically for things like music auditions. For technology-based jobs where most of the interview process can be done in writing, over the phone, or over a shared virtual whiteboard or document (for coding or drawing diagrams), there's no reason why everything up to the in-person interview can't be as age, gender, and race blind as possible.
Your assumption here is that the status quo of hiring is an optimum that must be sacrificed for diversity.
The reality is that every hiring program already uses diversity as a criterion, but unless a measurable objective is set, it will invisibly optimize for the comfort of current employees, rather than job performance.
The purpose of measuring diversity in hiring practices is to remove or control a confounding factor, not to add one.
And a lot of that was caused by pervasive discrimination.
Put in the most mercenary way I can think of:
There are and have always also been professions in which hiring decisionmakers screw their own organizations out of talent by ignoring or down-ranking candidates they were bigoted against. There similarly are/have been professions in which talented people with potential to add great value have been driven away by similarly bigoted leaders.
We don't know what might have been, what potential might have been achieved, because of the idiotic/cruel things we've practiced throughout history. Acting like ethical, compassionate, empathetic people seems like a way more pragmatic way than "well, it worked so far, let's keep being bigoted assholes" to see if we could do better in the future.
However, as other have mentioned, there are already biases to hire people "like you" and offsetting those biases is valuable. Often, I'd argue "culture fit" is an umbrella term used to keep people who are different out.
Is that the kind of "diversity of ideas and experience" you're talking about?
Damore didn't do that, by the way.
Your premise is that if the CDO’s role doesn’t include your vision of diversity it is “just skin deep”. Not everyone shares this premise, and focusing on this point also makes it sound like you dismiss all the other aspect his role could have (they’re just skin deep after all, right ?)
To get back to yor question, Google has offices all around the world, I find it hard to fault them for lack of diversity of experiences, nor do I think the thousands of people they employ all have the same ideas. I’d actually think it would be harder to find people all sharing all the same ideas.
Or do you have something a lot more specific in mind ?
Elsewhere in the thread, people are talking about excessive drinking on the job contributing to harassment. In the culture where I currently live (Japan), drinking is virtually mandatory. In fact, I once got an official reprimand for not drinking at a company event. Within the Japanese culture, drinking allows you to relax the way your present yourself. If you are drunk, it is acceptable to clearly say what you think, even if it might be embarrassing for others. This may be the only time to provide feedback up the ladder. Additionally, people higher up in the organisation are allowed to be more familiar with those lower down, which is impossible in normal every day work. This develops an honest camaraderie up and down the organisation and without the social lubricant (or excuse is probably a better word) that is alcohol, the work culture suffers.
Now, perhaps we have several Japanese people who have experienced significant success with this corporate culture. Do we want to grant it a kind of equivalent status within our organisation? Or do we want to have a kind of veto that says, "Despite your previous experience and your cultural background, this is a no-go area for our organisation"?
Even when talking about technical rather than cultural issues, there may be times when we need to limit discussion. I may have hired someone with extensive C++ experience into my Ruby on Rails team. Having that experience is really valuable. The C++ programmer can see things from different perspectives and provide solutions that are different that what the average Rails developer has seen before. However, if the C++ programmer suddenly starts demanding that all string processing should be handled in C++, we might want to limit this discussion. The C++ programmer may have lots of wonderful tools and experience to help them with this task, but the bulk of the developers on the team are not going to be able to cope. Potentially every developer on the team has some niche thing that they would like to introduce. Do we really want to provide a stage for all of these ideas, or do we want to filter them first and work on the ideas that seem most compatible with the team?
It is entirely possible that you disagree with my standpoint. I certainly have met a few people who feel that giving every person in the company an equal opportunity to pursue all of their ideas is a good idea. I have not experienced a successful company that embraced that philosophy, however. Leadership is often about focusing on a few ideas and limiting discussion that appear to be going in incompatible directions. As much as I am frustrated when my ideas get shot down without much air time, I recognise the reality of this necessity.
It's also possible that you have a completely different point that you are hoping to make and it was lost on me due to the brevity of your comment. In that case, perhaps it would be better to try to explain your position in more detail.
at most companies, we all know that the skin-deep diversity is the only thing the HR department cares about because it's all they're terrified about getting sued over. however, given google's unique corporate culture, i think that they might be forward thinking enough to seek out different perspectives, but probably not in a quota-driven fashion.
political ideology can't be a meaningful diversity goal anyway, nor should it be. there may not be a "best" ideology but promoting political ideology alone would be a terrible idea because in the current way that diversity is practiced it would require treating ideologies like nazism as on the same table as others.
My parents have always been right-leaning, working class. I, myself am far more libertarian leaning, with a bit of pragmatism. I'm fairly certain that I wouldn't like the political culture present in Google, and do not feel like my ideas would be respected in that culture at all. It emphatically does not mean that my point of view is less valid, but would more often than not be aggressively dismissed in that culture. Not to mention that I'm a cis-gendered, white male, which seems to be looked down on overall in a few of the large, progressive technical companies by itself.
In the end, it does matter. Culture, for good or bad, influences the makeup of a company. I tend to only have literally a couple drinks a month, or less generally not at work. That said, I don't feel there's shame in having a beer/ale/wine at lunch now and then. It should be about personal responsibility and accountability. It also shouldn't turn into a witch hunt without investigation.
Like most things in life, it depends.
The problem with trying to actually do it is that in recent times, one camp in particular tends to label the other as "Nazis" who must be crushed out of existence the moment they're spotted, although there are essentially no actual Nazis in the world today. In other words the side that preaches tolerance and diversity the most can't actually handle it and immediately tries to get rid of it by using the most overblown pastiches imaginable.