Could this be something that is a bit of a mix of age/gender bias? I can't help but think that someone 40+ years old would be more likely to speak those kind of things.
For men of my age (my perspective/experiences), the examples look more like this:
- "That young person over there? Must be an intern or just came into the field"
- "Back in my day we did it with punch cards, boy"
The article is confusing a prior with a bias. They aren't the same thing. A prior is a belief you hold before any evidence is present. A bias is a failure to properly update your beliefs after obtaining evidence.
Believing a woman at a tech conference is a recruiter (before interacting with her) is merely a prior. In my experience it's also a reasonably accurate one. If one were to fail to change that opinion after a woman claims she is a developer, that would be a bias.
By most definitions, bias includes the negative prior as well as the refusal. For example:
"Bias is an inclination of temperament or outlook to present or hold a partial perspective, often accompanied by a refusal to consider the possible merits of alternative points of view."
This is where I rant about "common sense". Common sense is a first approximation of reality. It's actually right the majority of the time. If "the majority of the time" is sufficient for your purposes, it's fine. If it's not, then you're a fool for relying on common sense when you need accuracy.
So basically, you're arguing that "common sense" tells you that women at tech conferences are recruiters or HR. And from a common sense perspective, it may be right. But you didn't say common sense. You said "How is the prior negative if it is accurate?"
By definition, the prior is not going to be accurate for a significant minority (if not a majority) of the women at the conference. And every time you're wrong, you are negatively affecting individuals. Don't want to talk to recruiters? You avoid them. You don't bring them into conversations, or don't assume they can keep up. Your avoidance harms their networking opportunities. You're hurting them.
I believe this is the heart of the problem of all of the bias flamewars. I think we can agree that it is reasonable to make statements about the group as long as the statements are true (women are more likely to be recruiters). It becomes much more fuzzy when trying to figure out how to apply that to an individual. That is where I draw the line. When you say an individual is a recruiter because they are female, you are doing something harmful.
The thing is, this sort of thing is pretty easy to manage in real life - just don't make assumptions, and ask people about themselves. This also falls right in line with the classic advice from "How to Win Friends and Influence People". People like you more because you're interested in them, and you don't make harmful assumptions about them.
Sadly, too many people think "Well, I'm not sexist/racist/homophobic", and make excuses for continuing their pattern of bias rather than really questioning their own behavior and finding better ways to act.
>The thing is, this sort of thing is pretty easy to manage in real life - just don't make assumptions, and ask people about themselves.
Assumptions are useful, which is why we use them. Yeah you should ask people about who they are and what they do, but do you really want to spend 15 minutes talking with a recruiter that you could have spent talking with a programmer? No, so you have to avoid the recruiters (lets you be stuck with them) and the best you can do is work based on your assumptions.
If you don't like it, try to replace the word assumptions with Bayesian weighted probability.
A prior (e.g., P(recruiter|woman at tech event) = g ) is accurate if the actual portion of women at tech events who are recruiters is g.
Secondly, suppose the prior is accurate. Lets take a very simple model, suppose g_woman = 0.25 and g_man = 0.05. Further, suppose networking with a developer has a value of 1 utilon and networking with a recruiter has 0 utilons of value.
If I network in order to maximize utility, based solely on my prior (i.e. ignoring any posterior info), I've added 95 utilons to the world for every 100 people I network with. If I behave irrationally and network with men and women equally, I've added only 85 utilons to the world. I've harmed 47.5 men in order to benefit 37.5 women - on net I've harmed 10 people.
(If posterior information is available, then you can even increase utility beyond 95/100.)
This is why math matters, and why carefully thinking things through rather than spouting incorrect soundbites (as the author does) is important.
How do you go from a simple abstract model that is rhetorically convenient to actually guiding concrete behavior?
If you walk into a conference and use that model you aren't using anything very meaningful to guide your behavior, you're using a model that probably isn't very true (I would presume that the modal value of networking is ~0, with the occasional valuable introduction bringing the mean up above that).
Going from models to reality is basically a process of expanding the model until it accounts for enough that you are confident it will work.
Also, the particular model I use only requires a mean positive utility - even if we take a model like yours, the conclusion is unchanged. Variance simply goes up, but the best option is still not talking to women.
I mean, if you only talk to men and then measure where you derived utility, I'm not sure you've properly evaluated the model yet (if you interact with a certain percentage of women at conferences and keep track of all this in order to make sure that your model is working out properly, well then, more power to you).
I really can't write an entire textbook on Bayesian decision theory in an HN comment. Honestly, if I could, I'd write the actual textbook - we really need a good one. I'm not claiming to have a rigorously tested model of exactly who to network with at a conference. I'm claiming that ignoring base rates results in worse decisions.
And if I made this point in a non-political context - e.g., "you must account for the base rate for $disease when interpreting a $disease test" - no one would be disputing it.
If you or anyone else here can even present a (utilitarian) model where ignoring base rates leads to a better decision, by all means do it. I'd love to see this, though I suspect the actual outcome of such an effort will merely be the person attempting to do it gaining a much better understanding of Bayes rule.
My issue isn't with the point that the base rate matters, it's with the mathematricality.
And I do think that is fair, if there is not a simple way of actually measuring the utility and such (re your complaint about $diseases, medicine has at least somewhat reliable tests), all the stuff about the modeling is just theatrics.
Am I correct in interpreting this post as saying "I agree base rate matters, I just wish you didn't provide a toy example illustrating that?"
Also, I now strongly recommend you go through the exercise I suggested. Then you'll realize that accurate tests do NOT somehow eliminate the need for models or accounting for base rates.
Also, I now strongly recommend you go through the exercise I suggested. Then you'll realize that accurate tests do NOT somehow eliminate the need for models or accounting for base rates.
I'm reaching for the opposite point. Some sort of meaningful evaluation of actual outcomes is necessary to make utilitarian decisions. If the evaluation of the outcome is arbitrary, then so is the decision.
First, morality matters. And you've forcibly ejected morality from your equation.
Second, network effects matter. You're not just creating utilons for yourself, you're creating a system that creates utilons. Even within the small and inaccurate world of your invented model, you're not following through to conclusions.
Third, you're making some very arbitrary assumptions, and ignoring other reasonable assumptions. For starters, do you give everyone you network with equal time? If you encounter a recruiter, do you give them the same amount of effort you would give to an engineer, or do you extract yourself and move on to the next person? The expense of equal input is not nearly as high as you're presenting here, assuming you don't "behave irrationally" and give everyone equal time whether it's effective or not.
Pretending that bias and excuses are intellectual rigor by inserting arbitrary, invented numbers into an imaginary equation is just an appeal to authority fallacy.
Yummyfajitas seems to be using utilitarian morality in his argument, not ejecting it altogether. That's what the concern with the net number of people harmed would suggest anyway.
Not really. It looks like utilitarian morality on the surface, but it lacks a rigorous analysis of consequences (as I pointed out on at least two fronts). Utilitarian morality's bootstrap definition requires rigor in order to use it - otherwise, there's a real danger of arriving at an immoral conclusion, which means it's not utilitarian.
Putting a pig in a suit doesn't make it a gentleman.
Which gets back to Occam's Razor. Which is more likely... that this was a failure of insufficient rigor, or that it was using utilitarianism and math to appeal to authority? Given that there were multiple violations of rigor, Occam's Razor suggests that this wasn't utilitarian morality at all, but rather mere defensive rhetoric.
Of course, this doesn't imply intent - the author might not realize that his formal-sounding justification was actually rationalization, because of a failure to understand the underlying moral issue. Which is exactly how bias works in most cases.
Someone put it really well recently, in the context of racism and racist police behavior. They said racism isn't waving a Confederate flag around. Racism is looking for excuses every time the police shoot another unarmed black man. People who don't think of themselves as racist or sexist, who actually find those ideas repulsive, are actively racist and sexist all the time! This is because they don't see the bias in their own behavior.
Blfr is right, I'm taking utilitarianism as my morality. Specifically, I believe networking with a developer (regardless of gender) is moral, and networking with a recruiter is useless. What morality do you take?
Secondly, I didn't make any assumption that the utility is all mine. The 1 utilon can be split between both parties in some arbitrary manner, it doesn't change the result.
Third, you are correct that I my constraint may not be #recruiters + #developers = 100. It might be alpha x #recruiters + #developers = 100 for alpha < 1. That doesn't change the optimal course of action - my best bet is always to minimize time I spend with recruiters.
Now if you think my model doesn't work, present a better one. But if you are making a fundamentally moral and non-utilitarian point ("networking with lady developers is intrinsically good no matter how many puppies get killed!!!!!"), make that point and don't waste time on positive claims if the truth of the positive claims is irrelevant anyway.
Also, you seem to wildly misunderstand what an appeal to authority is. An appeal to authority would be "I asked Eliezer Yudkowsky and he said I was right." Writing down a simple mathematical model is not remotely an appeal to authority, that's just careful reasoning.
Applying a pretense of mathematical rigor is an appeal to authority - the timeless purity of mathematical truth. there are countless historic examples of false rigor to justify immoral behavior as moral - it's the heart of pseudoscience.
I am flatly making a non-utilitarian argument for the morality of not making assumptions. That doesn't mean, however, that a rigorous application of utilitarian morality would not come to the same conclusions. I've made good arguments that your utilitarian equation is inadequate, and will arrive at false conclusions. You can think about those shortcomings, or argue that they aren't (as you did with your third point here), or you can write my argument off as mushy do-gooding because it's not "utilitarian".
Ignoring my criticism because it's not intrinsically utilitarian would be utilitarian. It would not, however, be rigorous.
Utilitarianism without rigor breaks down, almost inevitably. See the problem?
Math is not an authority. By this logic, all arguments based on reason are an appeal to authority. An appeal to authority is when you appeal to a human who is highly likely to be correct, but who's reasoning is unavailable for examination. https://en.wikipedia.org/wiki/Argument_from_authority
Since I presented every step of my argument, and you examined it, it is by definition not an argument by authority. It's simply an argument.
If you want to make a rigorous utilitarian case, do it. Simply pointing out some (non-)problems with the model I presented is not the same thing. All you are doing is arguing that there is more uncertainty than I believed, and then making an unjustified assumption that the uncertainty somehow supports your case.
Also, I didn't "ignore" your moral argument. I specifically asked you to make it - "What morality do you take?"
Okay, I'll give you that one. Not an appeal to authority - just a weak argument. Again, I'm saying that utilitarian morality requires rigor in order to be valid, or it risks putting the approving stamps of both morality and reason on false conclusions. There are some serious rigor problems in your original argument.
Beyond that, I do question utilitarian morality, for exactly these reasons. If it were software, it'd be a code smell. It's very easy to turn into justification for all sorts of foul things, and the track record of utilitarian morality is very ugly - like millions of dead ugly. It sure sounds good, especially if you're smart and used to being right on logical issues that don't involve squishy emotions. But it's a dangerous path.
Again, you have yet to identify a single problem that lack of "rigor" caused. I'm aware all models are incomplete. On the other hand, swooping in, declaring a model flawed without actually identifying a single problem, and alluding to some unspecified alternate morality is a little silly.
I also have no idea where you get "millions of dead ugly" applied to utilitarianism. I also don't get why you think "squishy emotions" like tribalism, desire to affiliate with high status people, or envy of others will somehow save us.
Maybe if you actually wrote down a model I'd be able to understand what you are trying to claim.
Utilitarian morality fueled Leninism and Stalinism. Surely, you recognize the damage. It was also popular in the eugenics movement.
I've already alluded to problems with your model - for example, the idea that equal resources are given to good and bad connections. Another problem is the idea that all benefit equally from connections, when it's obvious that the resource-starved benefit more than the resource-rich. Your model is deliberately starving valuable connections (female engineers). It's also reducing the quality of interactions, adding unnecessary noise to the system in the form of distrust. There are many, many shortcomings with it.
Now, you could just try to keep adjusting your theory to conform to reality, making it increasingly elaborate. Or, you could do the engineer's approach and find a solution that works far better, even if it isn't as pretty.
I already showed a simple extension incorporating this that yields the same result. You are still best served by focusing your attentions on folks who are most likely to not be a recruiter, the delta will simply be lower.
If you want to apply some diminishing marginal returns theory to connections, you'll get a concave optimization problem. You'll still need to multiply the derivatives by 1-P(recruiter), skewing your networking efforts towards men (though not 100% towards men anymore).
The model is, indeed, deliberately starving anyone deemed more likely to be a recruiter. That's the whole point.
Or, you could do the engineer's approach and find a solution that works far better...
You have not even begun to show this.
I strongly suggest you try to cook up any toy model where ignoring base rates helps. I'm dead certain you'll fail (there are theorems), but you'll learn a bit about decision theory in the process.
Reasonable proposal here: both of your POVs can trivially be reconciled as follows: developers are marked in some special way, such that they can trivially be separated from non-developers. Perhaps the badge is a different color.
This improves the position of everybody in the network: those who want to talk with recruiters can do so, recruiters don't have to talk with anybody who would only waste their time and there be no reason to assume that women developers are not since gender would be a strictly inferior signal to the (non-discriminary) badge color.
If we were only concerned with making accurate predictions, then all of that would be fine.
But the point of the article is less about the accuracy of our predictions, and more about the effects that our priors may have on others.
If a female technologist attends a tech conference and gets asked 100 times "are you in HR?", it's very likely that she'll feel unwelcome and discouraged. And her reaction to the questions will have absolutely nothing to do with whether the people doing the asking had a "prior" or a "bias".
Creating a welcoming atmosphere for women in technology means that you avoid asking that question despite the fact that the prior may be accurate.
If that's what the article is arguing, it should just say so. I.e., it should say "you should make bad decisions, hold incorrect beliefs, and waste time talking to recruiters in order to make female developers happier".
I'm not arguing with the actual message, I'm merely calling for accurate language.
> Believing a woman at a tech conference is a recruiter (before interacting with her) is merely a prior.
The article didn't say anything about believing a woman might be a recruiter, it said stuff about treating her like one. I mean, guys in suits are often recruiters, but that doesn't mean I walk up and ask if they're in HR..
Fluid intelligence peaks by the early 20s and then declines with age, so "age bias" is rational. When the law mandates non-economic behavior, some people will ignore the law.
I'm 50 and I'm far smarter than I was in my 20s. I find that's true of most people who keep their mind sharp and don't let life beat them down.
edit: A lot of people are beaten down by the time they're in their 40s. They feel like failure and they've lost hope, or they're stuck in a game of Prisoner's Dilemma. Those people aren't as smart as they were back when they were making the mistakes that led to their sorry state.
Fluid intelligence is a valuable characteristic for engineers, but so is experience. It's great to be able to learn quickly and react to a changing environment, but it's also great to be able to draw on past experiences to predict and preempt future challenges.
What isn't great is a lack of variety. Why would you want everybody on your team to be the same type of person? You want everyone's abilities to be complementary, not identical.
Even if that's true (citation needed), "fluid intelligence" isn't the end-all of important intelligence at work. You need people who aren't distracted by the newest shiny reinvention of some old thing. You need mastery, which can only be gained by time in the trenches.
'Fluid intelligence, like reaction time, typically peaks in young adulthood and then steadily declines. This decline may be related to local atrophy of the brain in the right cerebellum. Other researchers have suggested that a lack of practice, along with age-related changes in the brain may contribute to the decline.'
Many of the intelligence studies that this myth were based on have been debunked. They were done in the 60s and compared young college students with a random sample of older people.
More recent studies have shown that, barring disease, there isn't a noticeable cognitive decline until well past 65 because experience more than makes up for declining abilities.
While, for most people there isn't a noticeable decline, some studies do show older people take slightly longer to make decisions, but the decisions are usually better/more accurate. However, this isn't noticeable in all but the most extreme situations (top tier competitive chess for example). Newer research shows that this extra time is probably a combination of conscious choice to trade speed for accuracy, and more available information to sort through.
>Hot meals prepared on site, free drinks, fruits and snacks
Young people are dumb and think of dinner at work as a perk, rather than a red flag
that the company abuses its employees and steals their personal time.
Why have friends and family when you can spend your nights and weekends at the office with your coworkers and manager?
Company I work for gives hot meals too. Saves me a bunch of time and money cooking and cleaning and prevents me from dying of scurvy and/or grease fires.
Fruits and snacks are great, but none of it prevents you from going home when your work is over, as you should.
I've been reading some overviews - I'm already disturbed.
The assumption is that the difference in response times between the association of categories reveals preferences. And they tested by using groups that were expected beforehand to have certain preferences (boys liking insects more than flowers and vice versa for girls correlated with their reaction times).
But when faced with circumstances where these reactions DON'T correlate with self-reports, the conclusion is that the test reveals unconscious biases. If they confirmed their hypothesis about reaction times with groups that correlated - what entitles them to infer this further variable of unconscious bias to defend their hypothesis where the correlation fails?
Uggh... I can see it's going to take an enormous amount of time to find in the enormous literature a decent answer to this question. I can't be bothered.
This characteristically mixes two different political projects that have varying levels of support in the more general community:
One is bias: take, for example, the archetypal example of orchestra musicians sampling their work behind a curtain so directors can be less unfair when hiring: gender distributions have indeed significantly shifted, because we've removed gender out of view for a critical moment. This, of course, doesn't work if said director is a raging misogynist who will scream "no! no women!" when the curtain reveals the looks of whom he had approved.
The other is underrepresentation. This takes as a given that {women, racial minorities, the mentally ill...} should be represented in important positions in rough proportion to their general share of the population. Now, there are reasons to support this in some cases: preventing that injustices from one generation pass on to the next, for example, or ensuring that the unique contribution that epileptic cyclops can give to the world of cinema. But:
-- not everyone that's in opposition to bias is also in opposition to underrepresentation per se. It's just dishonest to mix the two; particularly in tech circles that have had a long, there's-one-right-answer culture that might not be perceived as benefiting from increased diversity at all. There's also the wider issue of whether diversity should be means to an end or an end unto itself.
(I mean, these are positions that exist. What bothers me is the Trojan horse mechanics at play; it predisposes me against one side when I really haven't thought enough about the subject to have a proper opinion.)
One of the points in the article is that the first often leads to the second: if your idea of a programmer is white and male, and that predisposes you to trust the capabilities of male tech candidates, you're going to end up with underrepresentation of people who are already minorities. It'll be subtle but it will be present.
It's a good idea to clearly articulate the two situations and how they interact, but it isn't like you can decouple them in practice so some mixing is natural.
Well, not in the stuff of policy, but yes in its feedback rules.
There's (1) symptom/diagnosis distinction (sure, treating the disease eventually cures the symptoms, but we're evaluating it against associated causality and prognosis concepts) and (2) a matter of ethical structure (i.e. deontology vs. consequentialism).
#1 is why they should be separated in theoretical terms. #2 is why they're distinct political programs.
I don't see anything nefarious here, and don't understand your "Trojan horse" comment. The article is about learning to identify and compensate for one's biases in order to create a more friendly environment for minority demographics in tech.
The concrete reason to do this is to make life easier for those who are the targets of bias. The hope is that reducing bias will reduce underrepresentation naturally.
There are many conversations about diversity that stray into controversial territory. People start talking about quotas, which leads to arguments about reverse discrimination and the true meaning of a meritocracy. This is not one of those conversations!
I see no evidence that the author is getting anywhere close to suggesting that equal representation is more important than evaluating people based on merit alone. She's saying that most of us carry biases around, and that it's important that we do our best to prevent them from making life and work shitty for our peers.
There's a difference between wanting equal representation "by any means" and wanting it "the right way". This article is advocating the latter--and even then only as a secondary goal--the primary goal is to make sure we don't let our biases create hostility toward those who've already proven they have a right to be here.
EDIT: Semantics follow. Also, trigger warning -- someone has an opinion on the Internet. Since it isn't identical to yours, butthurt may ensue.
I don't want to be a better ally, and it's condescending for anyone to demand it.
Definition:
The enemy of your enemy is your ally. (Remix of "the enemy of your enemy is your friend".)
If I and another person are fighting a common enemy, I don't suddenly earn the privilege of telling them how to conduct themselves in their own affairs. Nor does it similarly empower them to make demands of me.
We agree that the enemy we're fighting is a problem. That's all being an ally means. It doesn't imply any sort of deeper relationship, just a common goal.
The concept of "a better ally" is not compatible here.
"Well, we're on the same side, but you're not good enough of an ally for my standards."
Maybe you should try to understand even the most basic concepts before you embarrass yourself massively by mansplaining about common sides, enemies and who is allowed to judge you.
First of all, who is demanding it? This article is telling people who do want to help people who are underrepresented and face discrimination in technology a way in which they can do better.
I, for one, like knowing how to do better. Even if I'm doing an adequate job, I appreciate when my employees tell me that I could be more productive if I changed how I worked. I enjoy acting on that change and becoming better, and seeing how changing my actions makes things run more smoothly. This situation is not very different.
I think the brunt of what you're saying comes down to saying 'I don't want to be a better ally' -- why not?
When you're working on a team or towards a shared agenda, you should try and communicate and best understand how to interact with the people that you're working with -- this is just basic teamwork... When you're a member of the group that's part of the problem, you should probably grant more weight to what the group that is suffering discrimination/bias is saying.
Let me decode what 'you're not a good enough ally for my standards' means -- it means you're still being a dick, and not listening to what people are saying. It means you're still part of the problem. The things that were talked about in this article are things that my fiancee faces in her job as an SDE at Amazon with an unbelievable, near daily frequency, and it really irks me when people decide to put their head in the sand about it.
> I think the brunt of what you're saying comes down to saying 'I don't want to be a better ally' -- why not?
No, I'm arguing against the use of the word "ally".
I'll do whatever I can to help people who are discriminated against, but as soon as the measuring stick of "quality of alliance" comes out, I check right the fuck out and hop on the express train to Nopesville. I refuse to have any part in this stupid torture of the English language.
What would you prefer to call it? If you'd prefer to use feminism as an abstraction you can break it down as follows:
Feminism is a women's movement for women by women. I am not a woman, so I cannot be a feminist, but I can be a feminist ally.
The same thread of logic follows for the other movements.
As we're talking about the movement of feminism, the "quality of alliance" can be clearly shown by the alignment of your attitudes with the movement's and your support for women you see facing issues related to patriarchal systems/biases. I fail to see how this is a torture of the English language. The same is done with international politics -- clearly we have ways that countries can be better allies to the U.S., if we're making the U.S. the center of focus. Whether or not we should do that is a different discussion.
No, the enemy of your enemy is just your enemy's enemy -- they might be still an enemy of yours, might be an ally of yours, and might be entirely neutral toward you.
An ally is more than just the enemy of an enemy, and an ally is exactly the kind of relation that, if you actually were one, you'd be likely to be concerned with how to be a better one.
> No, the enemy of your enemy is just your enemy's enemy -- they might be still an enemy of yours, might be an ally of yours, and might be entirely neutral toward you.
Let's say you and I both hate Nazis.
We might join forces to take down Nazis, but I'm not going to invite you over for dinner. And I wouldn't expect you to accept if I ever did offer. And that doesn't entitle you to tell me how I should behave (especially if we have significant cultural differences).
The common saying is, "The enemy of my enemy is my friend." I don't think this quite captures the nuance between friends (people you actually care for) and allies (people whom are on your side but you couldn't be bothered to spend time with).
That is the distinction I've drawn here. Telling me to be "a better ally" becomes meaningless and needlessly entitled.
TL;DR I'm deeply interested in being a better FRIEND, but I don't care at all about being a better ALLY.
And that would make us allies -- not the fact that we are both enemies of the Nazis, but that we make the decision to work together toward the goal of defeating them. And "better allies" are people who, having made the decision to work together toward a common goal that makes them allies, more effectively work together toward their common goal. If you are going to choose to be allies, its usually for effectiveness in attaining the common goal, so its normal to want to be better allies if that is possible.
OTOH, lots of parties with mutual enemies aren't allies and don't choose to work together against their common enemy (a common, but not the only reason, for this is that the two parties that share a mutual enemy also have a mutual enmity that is as strong, or stronger, than that each has for the common enemy).
> The common saying is, "The enemy of my enemy is my friend."
This is a common saying, and its even more wrong in practice than "the enemy of enemy is my ally". But that "...is my friend" is wrong doesn't make "...is my ally" right.
> I don't think this quite captures the nuance between friends (people you actually care for) and allies (people whom are on your side but you couldn't be bothered to spend time with).
I would say that your description of "friends" is fine, but your description of "allies" misses elements of the usual definition; while alliances may be tactical, they involve more than just sharing a common goal at arms lengths, they involve the decision to work together to achieve the common goal. You seem, in this description, to reduce allies to mere "people who share a goal, perhaps at arm's length". If that's all the commonality you have with someone, you aren't an ally at all, and there message directed at those who would be their allies about how to be a better ally isn't directed at you.
> I would say that your description of "friends" is fine, but your description of "allies" misses elements of the usual definition; while alliances may be tactical, they involve more than just sharing a common goal at arms lengths, they involve the decision to work together to achieve the common goal. You seem, in this description, to reduce allies to mere "people who share a goal, perhaps at arm's length".
If someone is going to go through the trouble to cooperate with someone enough to care what happens to them, then why wouldn't they try to be their friend instead of just an ally?
> If that's all the commonality you have with someone, you aren't an ally at all, and there message directed at those who would be their allies about how to be a better ally isn't directed at you.
Fair enough. I don't want to wear that label anyway.
Author here. The term 'Ally' is often used to describe men supporting diversity (especially gender diversity) in technology. Some examples: the ADA initiative does an 'Ally Skills Workshop' and the Grace Hopper Conference had a "Male Ally Panel".
75 comments
[ 4.6 ms ] story [ 158 ms ] thread- don't speak, let those you want to support speak instead
- remove names from resumes on first few passes
For men of my age (my perspective/experiences), the examples look more like this:
- "That young person over there? Must be an intern or just came into the field"
- "Back in my day we did it with punch cards, boy"
- "So easy even your grandparents could do it."
Believing a woman at a tech conference is a recruiter (before interacting with her) is merely a prior. In my experience it's also a reasonably accurate one. If one were to fail to change that opinion after a woman claims she is a developer, that would be a bias.
"Bias is an inclination of temperament or outlook to present or hold a partial perspective, often accompanied by a refusal to consider the possible merits of alternative points of view."
http://en.wikipedia.org/wiki/Bias.
How is the prior negative if it is accurate?
So basically, you're arguing that "common sense" tells you that women at tech conferences are recruiters or HR. And from a common sense perspective, it may be right. But you didn't say common sense. You said "How is the prior negative if it is accurate?"
By definition, the prior is not going to be accurate for a significant minority (if not a majority) of the women at the conference. And every time you're wrong, you are negatively affecting individuals. Don't want to talk to recruiters? You avoid them. You don't bring them into conversations, or don't assume they can keep up. Your avoidance harms their networking opportunities. You're hurting them.
This, this is why bias matters.
The thing is, this sort of thing is pretty easy to manage in real life - just don't make assumptions, and ask people about themselves. This also falls right in line with the classic advice from "How to Win Friends and Influence People". People like you more because you're interested in them, and you don't make harmful assumptions about them.
Sadly, too many people think "Well, I'm not sexist/racist/homophobic", and make excuses for continuing their pattern of bias rather than really questioning their own behavior and finding better ways to act.
Assumptions are useful, which is why we use them. Yeah you should ask people about who they are and what they do, but do you really want to spend 15 minutes talking with a recruiter that you could have spent talking with a programmer? No, so you have to avoid the recruiters (lets you be stuck with them) and the best you can do is work based on your assumptions.
If you don't like it, try to replace the word assumptions with Bayesian weighted probability.
Secondly, suppose the prior is accurate. Lets take a very simple model, suppose g_woman = 0.25 and g_man = 0.05. Further, suppose networking with a developer has a value of 1 utilon and networking with a recruiter has 0 utilons of value.
If I network in order to maximize utility, based solely on my prior (i.e. ignoring any posterior info), I've added 95 utilons to the world for every 100 people I network with. If I behave irrationally and network with men and women equally, I've added only 85 utilons to the world. I've harmed 47.5 men in order to benefit 37.5 women - on net I've harmed 10 people.
(If posterior information is available, then you can even increase utility beyond 95/100.)
This is why math matters, and why carefully thinking things through rather than spouting incorrect soundbites (as the author does) is important.
If you walk into a conference and use that model you aren't using anything very meaningful to guide your behavior, you're using a model that probably isn't very true (I would presume that the modal value of networking is ~0, with the occasional valuable introduction bringing the mean up above that).
Also, the particular model I use only requires a mean positive utility - even if we take a model like yours, the conclusion is unchanged. Variance simply goes up, but the best option is still not talking to women.
I mean, if you only talk to men and then measure where you derived utility, I'm not sure you've properly evaluated the model yet (if you interact with a certain percentage of women at conferences and keep track of all this in order to make sure that your model is working out properly, well then, more power to you).
And if I made this point in a non-political context - e.g., "you must account for the base rate for $disease when interpreting a $disease test" - no one would be disputing it.
If you or anyone else here can even present a (utilitarian) model where ignoring base rates leads to a better decision, by all means do it. I'd love to see this, though I suspect the actual outcome of such an effort will merely be the person attempting to do it gaining a much better understanding of Bayes rule.
And I do think that is fair, if there is not a simple way of actually measuring the utility and such (re your complaint about $diseases, medicine has at least somewhat reliable tests), all the stuff about the modeling is just theatrics.
Also, I now strongly recommend you go through the exercise I suggested. Then you'll realize that accurate tests do NOT somehow eliminate the need for models or accounting for base rates.
I'm reaching for the opposite point. Some sort of meaningful evaluation of actual outcomes is necessary to make utilitarian decisions. If the evaluation of the outcome is arbitrary, then so is the decision.
Second, network effects matter. You're not just creating utilons for yourself, you're creating a system that creates utilons. Even within the small and inaccurate world of your invented model, you're not following through to conclusions.
Third, you're making some very arbitrary assumptions, and ignoring other reasonable assumptions. For starters, do you give everyone you network with equal time? If you encounter a recruiter, do you give them the same amount of effort you would give to an engineer, or do you extract yourself and move on to the next person? The expense of equal input is not nearly as high as you're presenting here, assuming you don't "behave irrationally" and give everyone equal time whether it's effective or not.
Pretending that bias and excuses are intellectual rigor by inserting arbitrary, invented numbers into an imaginary equation is just an appeal to authority fallacy.
Putting a pig in a suit doesn't make it a gentleman.
Which gets back to Occam's Razor. Which is more likely... that this was a failure of insufficient rigor, or that it was using utilitarianism and math to appeal to authority? Given that there were multiple violations of rigor, Occam's Razor suggests that this wasn't utilitarian morality at all, but rather mere defensive rhetoric.
Of course, this doesn't imply intent - the author might not realize that his formal-sounding justification was actually rationalization, because of a failure to understand the underlying moral issue. Which is exactly how bias works in most cases.
Someone put it really well recently, in the context of racism and racist police behavior. They said racism isn't waving a Confederate flag around. Racism is looking for excuses every time the police shoot another unarmed black man. People who don't think of themselves as racist or sexist, who actually find those ideas repulsive, are actively racist and sexist all the time! This is because they don't see the bias in their own behavior.
Secondly, I didn't make any assumption that the utility is all mine. The 1 utilon can be split between both parties in some arbitrary manner, it doesn't change the result.
Third, you are correct that I my constraint may not be #recruiters + #developers = 100. It might be alpha x #recruiters + #developers = 100 for alpha < 1. That doesn't change the optimal course of action - my best bet is always to minimize time I spend with recruiters.
Now if you think my model doesn't work, present a better one. But if you are making a fundamentally moral and non-utilitarian point ("networking with lady developers is intrinsically good no matter how many puppies get killed!!!!!"), make that point and don't waste time on positive claims if the truth of the positive claims is irrelevant anyway.
Also, you seem to wildly misunderstand what an appeal to authority is. An appeal to authority would be "I asked Eliezer Yudkowsky and he said I was right." Writing down a simple mathematical model is not remotely an appeal to authority, that's just careful reasoning.
I am flatly making a non-utilitarian argument for the morality of not making assumptions. That doesn't mean, however, that a rigorous application of utilitarian morality would not come to the same conclusions. I've made good arguments that your utilitarian equation is inadequate, and will arrive at false conclusions. You can think about those shortcomings, or argue that they aren't (as you did with your third point here), or you can write my argument off as mushy do-gooding because it's not "utilitarian".
Ignoring my criticism because it's not intrinsically utilitarian would be utilitarian. It would not, however, be rigorous.
Utilitarianism without rigor breaks down, almost inevitably. See the problem?
Since I presented every step of my argument, and you examined it, it is by definition not an argument by authority. It's simply an argument.
If you want to make a rigorous utilitarian case, do it. Simply pointing out some (non-)problems with the model I presented is not the same thing. All you are doing is arguing that there is more uncertainty than I believed, and then making an unjustified assumption that the uncertainty somehow supports your case.
Also, I didn't "ignore" your moral argument. I specifically asked you to make it - "What morality do you take?"
Beyond that, I do question utilitarian morality, for exactly these reasons. If it were software, it'd be a code smell. It's very easy to turn into justification for all sorts of foul things, and the track record of utilitarian morality is very ugly - like millions of dead ugly. It sure sounds good, especially if you're smart and used to being right on logical issues that don't involve squishy emotions. But it's a dangerous path.
I also have no idea where you get "millions of dead ugly" applied to utilitarianism. I also don't get why you think "squishy emotions" like tribalism, desire to affiliate with high status people, or envy of others will somehow save us.
Maybe if you actually wrote down a model I'd be able to understand what you are trying to claim.
I've already alluded to problems with your model - for example, the idea that equal resources are given to good and bad connections. Another problem is the idea that all benefit equally from connections, when it's obvious that the resource-starved benefit more than the resource-rich. Your model is deliberately starving valuable connections (female engineers). It's also reducing the quality of interactions, adding unnecessary noise to the system in the form of distrust. There are many, many shortcomings with it.
Now, you could just try to keep adjusting your theory to conform to reality, making it increasingly elaborate. Or, you could do the engineer's approach and find a solution that works far better, even if it isn't as pretty.
I already showed a simple extension incorporating this that yields the same result. You are still best served by focusing your attentions on folks who are most likely to not be a recruiter, the delta will simply be lower.
If you want to apply some diminishing marginal returns theory to connections, you'll get a concave optimization problem. You'll still need to multiply the derivatives by 1-P(recruiter), skewing your networking efforts towards men (though not 100% towards men anymore).
The model is, indeed, deliberately starving anyone deemed more likely to be a recruiter. That's the whole point.
Or, you could do the engineer's approach and find a solution that works far better...
You have not even begun to show this.
I strongly suggest you try to cook up any toy model where ignoring base rates helps. I'm dead certain you'll fail (there are theorems), but you'll learn a bit about decision theory in the process.
This improves the position of everybody in the network: those who want to talk with recruiters can do so, recruiters don't have to talk with anybody who would only waste their time and there be no reason to assume that women developers are not since gender would be a strictly inferior signal to the (non-discriminary) badge color.
But the point of the article is less about the accuracy of our predictions, and more about the effects that our priors may have on others.
If a female technologist attends a tech conference and gets asked 100 times "are you in HR?", it's very likely that she'll feel unwelcome and discouraged. And her reaction to the questions will have absolutely nothing to do with whether the people doing the asking had a "prior" or a "bias".
Creating a welcoming atmosphere for women in technology means that you avoid asking that question despite the fact that the prior may be accurate.
I'm not arguing with the actual message, I'm merely calling for accurate language.
The article didn't say anything about believing a woman might be a recruiter, it said stuff about treating her like one. I mean, guys in suits are often recruiters, but that doesn't mean I walk up and ask if they're in HR..
edit: A lot of people are beaten down by the time they're in their 40s. They feel like failure and they've lost hope, or they're stuck in a game of Prisoner's Dilemma. Those people aren't as smart as they were back when they were making the mistakes that led to their sorry state.
What isn't great is a lack of variety. Why would you want everybody on your team to be the same type of person? You want everyone's abilities to be complementary, not identical.
'Fluid intelligence, like reaction time, typically peaks in young adulthood and then steadily declines. This decline may be related to local atrophy of the brain in the right cerebellum. Other researchers have suggested that a lack of practice, along with age-related changes in the brain may contribute to the decline.'
Many of the intelligence studies that this myth were based on have been debunked. They were done in the 60s and compared young college students with a random sample of older people.
More recent studies have shown that, barring disease, there isn't a noticeable cognitive decline until well past 65 because experience more than makes up for declining abilities.
While, for most people there isn't a noticeable decline, some studies do show older people take slightly longer to make decisions, but the decisions are usually better/more accurate. However, this isn't noticeable in all but the most extreme situations (top tier competitive chess for example). Newer research shows that this extra time is probably a combination of conscious choice to trade speed for accuracy, and more available information to sort through.
>Hot meals prepared on site, free drinks, fruits and snacks
Young people are dumb and think of dinner at work as a perk, rather than a red flag that the company abuses its employees and steals their personal time.
Why have friends and family when you can spend your nights and weekends at the office with your coworkers and manager?
Fruits and snacks are great, but none of it prevents you from going home when your work is over, as you should.
I would have thought hot meals was amazing when I was single. Now I eat with my spouse and thats really important to us.
To be honest - I tend to take any psychology results with a grain of salt... given the crisis of replicability that it is currently facing.
The assumption is that the difference in response times between the association of categories reveals preferences. And they tested by using groups that were expected beforehand to have certain preferences (boys liking insects more than flowers and vice versa for girls correlated with their reaction times).
But when faced with circumstances where these reactions DON'T correlate with self-reports, the conclusion is that the test reveals unconscious biases. If they confirmed their hypothesis about reaction times with groups that correlated - what entitles them to infer this further variable of unconscious bias to defend their hypothesis where the correlation fails?
Uggh... I can see it's going to take an enormous amount of time to find in the enormous literature a decent answer to this question. I can't be bothered.
One is bias: take, for example, the archetypal example of orchestra musicians sampling their work behind a curtain so directors can be less unfair when hiring: gender distributions have indeed significantly shifted, because we've removed gender out of view for a critical moment. This, of course, doesn't work if said director is a raging misogynist who will scream "no! no women!" when the curtain reveals the looks of whom he had approved.
The other is underrepresentation. This takes as a given that {women, racial minorities, the mentally ill...} should be represented in important positions in rough proportion to their general share of the population. Now, there are reasons to support this in some cases: preventing that injustices from one generation pass on to the next, for example, or ensuring that the unique contribution that epileptic cyclops can give to the world of cinema. But:
-- not everyone that's in opposition to bias is also in opposition to underrepresentation per se. It's just dishonest to mix the two; particularly in tech circles that have had a long, there's-one-right-answer culture that might not be perceived as benefiting from increased diversity at all. There's also the wider issue of whether diversity should be means to an end or an end unto itself.
(I mean, these are positions that exist. What bothers me is the Trojan horse mechanics at play; it predisposes me against one side when I really haven't thought enough about the subject to have a proper opinion.)
(ETA: who is being underrepresented.)
There's (1) symptom/diagnosis distinction (sure, treating the disease eventually cures the symptoms, but we're evaluating it against associated causality and prognosis concepts) and (2) a matter of ethical structure (i.e. deontology vs. consequentialism).
#1 is why they should be separated in theoretical terms. #2 is why they're distinct political programs.
The concrete reason to do this is to make life easier for those who are the targets of bias. The hope is that reducing bias will reduce underrepresentation naturally.
There are many conversations about diversity that stray into controversial territory. People start talking about quotas, which leads to arguments about reverse discrimination and the true meaning of a meritocracy. This is not one of those conversations!
I see no evidence that the author is getting anywhere close to suggesting that equal representation is more important than evaluating people based on merit alone. She's saying that most of us carry biases around, and that it's important that we do our best to prevent them from making life and work shitty for our peers.
There's a difference between wanting equal representation "by any means" and wanting it "the right way". This article is advocating the latter--and even then only as a secondary goal--the primary goal is to make sure we don't let our biases create hostility toward those who've already proven they have a right to be here.
I don't want to be a better ally, and it's condescending for anyone to demand it.
Definition:
The enemy of your enemy is your ally. (Remix of "the enemy of your enemy is your friend".)
If I and another person are fighting a common enemy, I don't suddenly earn the privilege of telling them how to conduct themselves in their own affairs. Nor does it similarly empower them to make demands of me.
We agree that the enemy we're fighting is a problem. That's all being an ally means. It doesn't imply any sort of deeper relationship, just a common goal.
The concept of "a better ally" is not compatible here.
"Well, we're on the same side, but you're not good enough of an ally for my standards."
I, for one, like knowing how to do better. Even if I'm doing an adequate job, I appreciate when my employees tell me that I could be more productive if I changed how I worked. I enjoy acting on that change and becoming better, and seeing how changing my actions makes things run more smoothly. This situation is not very different.
I think the brunt of what you're saying comes down to saying 'I don't want to be a better ally' -- why not?
When you're working on a team or towards a shared agenda, you should try and communicate and best understand how to interact with the people that you're working with -- this is just basic teamwork... When you're a member of the group that's part of the problem, you should probably grant more weight to what the group that is suffering discrimination/bias is saying.
Let me decode what 'you're not a good enough ally for my standards' means -- it means you're still being a dick, and not listening to what people are saying. It means you're still part of the problem. The things that were talked about in this article are things that my fiancee faces in her job as an SDE at Amazon with an unbelievable, near daily frequency, and it really irks me when people decide to put their head in the sand about it.
No, I'm arguing against the use of the word "ally".
I'll do whatever I can to help people who are discriminated against, but as soon as the measuring stick of "quality of alliance" comes out, I check right the fuck out and hop on the express train to Nopesville. I refuse to have any part in this stupid torture of the English language.
Feminism is a women's movement for women by women. I am not a woman, so I cannot be a feminist, but I can be a feminist ally.
The same thread of logic follows for the other movements.
As we're talking about the movement of feminism, the "quality of alliance" can be clearly shown by the alignment of your attitudes with the movement's and your support for women you see facing issues related to patriarchal systems/biases. I fail to see how this is a torture of the English language. The same is done with international politics -- clearly we have ways that countries can be better allies to the U.S., if we're making the U.S. the center of focus. Whether or not we should do that is a different discussion.
> What would you prefer to call it?
I'd prefer the use of the compound noun "feminist ally" over the generic and ambiguous noun "ally".
> I fail to see how this is a torture of the English language.
https://www.youtube.com/watch?v=vuEQixrBKCc
Slight exaggeration for the sake of humor.
No, the enemy of your enemy is just your enemy's enemy -- they might be still an enemy of yours, might be an ally of yours, and might be entirely neutral toward you.
An ally is more than just the enemy of an enemy, and an ally is exactly the kind of relation that, if you actually were one, you'd be likely to be concerned with how to be a better one.
Let's say you and I both hate Nazis.
We might join forces to take down Nazis, but I'm not going to invite you over for dinner. And I wouldn't expect you to accept if I ever did offer. And that doesn't entitle you to tell me how I should behave (especially if we have significant cultural differences).
The common saying is, "The enemy of my enemy is my friend." I don't think this quite captures the nuance between friends (people you actually care for) and allies (people whom are on your side but you couldn't be bothered to spend time with).
That is the distinction I've drawn here. Telling me to be "a better ally" becomes meaningless and needlessly entitled.
TL;DR I'm deeply interested in being a better FRIEND, but I don't care at all about being a better ALLY.
And that would make us allies -- not the fact that we are both enemies of the Nazis, but that we make the decision to work together toward the goal of defeating them. And "better allies" are people who, having made the decision to work together toward a common goal that makes them allies, more effectively work together toward their common goal. If you are going to choose to be allies, its usually for effectiveness in attaining the common goal, so its normal to want to be better allies if that is possible.
OTOH, lots of parties with mutual enemies aren't allies and don't choose to work together against their common enemy (a common, but not the only reason, for this is that the two parties that share a mutual enemy also have a mutual enmity that is as strong, or stronger, than that each has for the common enemy).
> The common saying is, "The enemy of my enemy is my friend."
This is a common saying, and its even more wrong in practice than "the enemy of enemy is my ally". But that "...is my friend" is wrong doesn't make "...is my ally" right.
> I don't think this quite captures the nuance between friends (people you actually care for) and allies (people whom are on your side but you couldn't be bothered to spend time with).
I would say that your description of "friends" is fine, but your description of "allies" misses elements of the usual definition; while alliances may be tactical, they involve more than just sharing a common goal at arms lengths, they involve the decision to work together to achieve the common goal. You seem, in this description, to reduce allies to mere "people who share a goal, perhaps at arm's length". If that's all the commonality you have with someone, you aren't an ally at all, and there message directed at those who would be their allies about how to be a better ally isn't directed at you.
If someone is going to go through the trouble to cooperate with someone enough to care what happens to them, then why wouldn't they try to be their friend instead of just an ally?
> If that's all the commonality you have with someone, you aren't an ally at all, and there message directed at those who would be their allies about how to be a better ally isn't directed at you.
Fair enough. I don't want to wear that label anyway.