Hacker News new | past | comments | ask | show | jobs | submit | more TrueGeek's comments login

Yes, easily. It's the perfect machine for this and you'll love the experience.


Better article with more information:

https://www.bizjournals.com/nashville/news/2024/07/08/girls-...

> In an email Monday, founder and CEO Adriana Gascoigne said “Girls in Tech will be closing its doors due to a lack of funding in 2023 and 2024.”


Thanks. Following a link from there:

https://www.bizjournals.com/nashville/news/2024/06/06/girls-...

> Nashville-based Girls in Tech Inc. may be forced to shut down by the end of summer. [...] needs to raise $100,000 or it faces imminent closure. [...] Girls in Tech has a membership of 130,000 "women and allies" across 50 cities and 38 countries.

Was the membership base already tapped out, or the org didn't reach out to the membership on this, or the org had larger near-term funding needs than the immediate $100K?

Also, is it possible that funding isn't the only consideration? For example, even if the org could be saved with heroics, there's opportunity cost to leadership (personal, professional)?


Like many orgs, probably, their funding model was working, and then got hammered by COVID-19, and were stuck holding on, hoping for return to "normalcy".

It's very hard to pivot. Fund raising costs money. Some one needs an idea, a plan, a strategy. Everyone needs to agree to it. Meanwhile, an org's (remaining) execs and board members are doing triage. To execute a new plan means even more work.

And so on.

I've met and worked with terrific fund raisers. For me, personally, fund raising is just the worst. I've done enough to know a) it's very hard and b) I suck at it.


Well, they seem to offer a "premium membership" for $9.99/month [1] and presumably that hasn't raised enough. If the aim of the charity is to get career resources in front of as many women as possible, they probably don't want to put their most impactful resources behind a paywall - that would be contrary to their goal.

I suppose they could try an appeal to generosity instead? Depends if they've got a network of grateful people they helped 17 years ago who are now making six-figure salaries.

[1] https://girlsintech.org/membership/premium/


I know almost nothing about non-profit fundraising, but this benefits tier membership model looks very familiar as a tech for-profit service, rather than a charitable non-profit for the benefit of all.

(Specifically, in a tech, like a SaaS, the free tier are sales leads and inflated "market share" numbers, and the premium tier are the real customers of the service value you're providing and is your whole reason for existing. In a charity, however, you don't measure out benefits based on how much that person is paying you. Though a charity will have special recognition for exceptional donations, like the donor's name listed on some page, or mentioned as a sponsor of an event.)

Given the dire runway situation they were in, I wonder whether they sent out a recent urgent appeal to their free-tier, as more like a charity, asking for donations? (And if so, was the obvious benefits tier model hurting any charitable goodwill they might've otherwise generated?) Or did they try to push their free-tier members an upsell to their premium tier, like a business? Or neither?


130,000 members x $1/member = $130,000


Huh. You'd think all the organizations that attribute their challenges in hiring non-male engineers to a "pipeline problem" would've spent a small fraction of their recruiting budgets helping to fund Girls in Tech...


For what it's worth, Girls Who Code -- an organization more directly focused on improving the "pipeline" through training programs aimed at K-12 and college students -- seems to be thriving, with over $20M in donations from a variety of tech organizations in 2023: https://girlswhocode.com/2023report/


Our company actively partners with girls who code for that exact reason. Our rather empty post covid office space gets transformed into summer boot camps for middle and high schoolers every year. It is a very productive way to improve the K-12 pipeline.


this deserves to be much higher than a sub-sub-sub-sub comment

the fact one DEI organization failed doesn't mean DEI failed. They could be mis managed just like any other non profit


Agreed, I didn't know the difference and at a glance thought it was Girls Who Code that folded.

It could be that as the "vibe shifts" away from DEI and the funding gets smaller, we'll see a culling where only some orgs survive, hopefully the best ones. "When the tide goes out..."


Thank you. I was confusing these as being the same organization, and I thought Gils Who Code is doing quite well.


That would require an monetary investment into DEI, which has become a negative investor signal for many large companies.

It's a shame, because I've met several developers who benefited from Girls in Tech's work.


monetary investment into DEI, which has become a negative investor signal for many large companies

can you please explain that or point to some articles about it?


https://www.shrm.org/topics-tools/news/inclusion-equity-dive...

https://www.axios.com/2024/04/02/dei-backlash-diversity

etc. There is a backlash underway against any effort to expand workplace diversity beyond the representative fractions circa 1990.


There is a backlash to the heavy handedness of DEI. You can’t freeze out white males simple because they are white males. You need to remove barriers to entry based on sexism and such, but you can’t exclude a whole large class of people because that class had it good in the past. That gets you South Africa.


> You can’t freeze out white males simple because they are white males.

I agree with this. However, I have been working in tech for quite some time now, and I have not seen any place where white males have been frozen out of employment or funding. More typically, in my personal experience, they represent most of the staff and leadership and materially all of institutional funding recipients.

Perhaps my experience is atypical, or my definition of "frozen out" is different.


I have. In April 2019, when I worked at Dropbox, we instituted what we called "Opportunistic Hiring". 20 heads were reserved to hire "opportunistically". What this meant is that when a diverse candidate was hired, the headcount used for that hire didn't come from the team's headcount. It came from the pool of headcount for "opportunistic hires". Unless the "opportunistic hires" pool was already exhausted, then a diverse hire still counted against the team's headcount. The definition of a "diverse" candidate was a woman of any race, and URM men. If this sounds like an overly-complex way of saying "we're prohibiting Asian and white men from 20 headcount" it's because it is. I can provide the exact wording of the policy if you want.

It's not "freezing out" all white males. One, it's also freezing out Asian males. And two, it's only freezing them out of a specific chunk of our headcount. But it still deeply affected me to see a company outright deny employment on the basis of race and gender, even if it was only for part of the headcount.


Thank you for your valuable input. This is one of the most convincing "evidence" I have seen so far regarding the "exclusion" during hiring. I wonder why we don't see such cases written out by people often in these discussions -- most of these are quite handwavy.


Reading your description of the policy, it looks like what it does is say a team can hire from the "Opportunistic" bucket for "free," but does not freeze out anybody else. Is that a correct read?

> 20 heads were reserved

I have not studied Dropbox extensively. In 2019, was 20 hires roughly equivalent to the total overall number of new hires? Or are we talking about a small percentage of overall new staffers? If the latter, how does this anecdote respond to my statements?


It's explicit denial of employment on the basis of race and gender. It's illegal no matter what proportion of the headcount this is.

If I have 100 headcount and I say "20 of our headcount is off limits to white and Asian men" is that ok? If I have 80 headcount and I create 20 "opportunistic" headcount exclusive to women and URM is that acceptable? You're smart enough to understand that these are identical policies.


I'm not reading anywhere in there where it says anybody is denied a job under any scenario. Perhaps if Dropbox had fixed hiring numbers like your hypothetical, but that was not data you submitted.


It doesn't matter if they were hiring 50 people or 500 people. If your policy amounts to "we have N headcount candidates of any background, plus M headcount that's exclusive to X races and Y gender" then that's denial of employment on the basis of race and gender. Regardless of what the values of M and N are - well, unless M is zero.

If a company institutes a hiring policy that amounts to saying, "99% of our headcount is off-limits to Asians" is it correct to say "no Asians are denied a job, under any scenario"? It's technically correct in that Asian applicants could vie for the 1% of headcount that's available to Asians. But how many Asians would have been hired absent this headcount restriction? Probably more than 1%!

Okay, what if it wasn't 99% restricted to Asian, but just 50%? Can a company just say "half of our headcount is off limits to Asians"? Does that lower percentage make it okay?

No, of course not. It's not okay if it's 99%, 50%, 1%, or 0.01%. It's denial of employment to wall-off any proportion of headcount on the basis of protected class.

-------------------

In case you need an explicit scenario laid out to see how candidates are denied employment because of Dropbox's "opportunistic hiring" policy, here it is:

The company has already exhausted its non-restricted headcount. An Asian male applied. He gets rejected automatically because the only remaining headcount is restricted to women and URM. A woman applies. She's allowed to interview because the company still has that "opportunistic" headcount, and unlike our Asian male applicant her gender makes her eligible for this set-aside headcount.

If all headcount (including that "opportunistic headcount") were available to all races and genders, our Asian male would have been able to get hired.


The scenario at the end really illustrates how important it is to not evaluate partial policies without accurate and full knowledge. In particular, we don't know how whether this hypothetical situation ever happened (and if so, what breakdown led to them hiring ~480 white & asian men before even hiring 20 women or URM, assuming a 500 annual hiring budget?).

But that aside, we don't know how management would handle the hypothetical. If I need to hire someone today and all I have are the "opportunistic" headcount available, am I allowed to hire an Asian man or am I required to leave the job unfilled? The use of the word "opportunistic" does make it sound as if this is not a hard requirement in all situations. But again, we are on HN and do not have access to anything like the full guidelines presented to management at Dropbox.


Do I seriously have to spend three comments explaining how a policy of "white and Asian men are prohibited from X% of our headcount" is denial of employment, regardless of the value of X?

> If I need to hire someone today and all I have are the "opportunistic" headcount available, am I allowed to hire an Asian man or am I required to leave the job unfilled?

You are prohibited from hiring an Asian or a while male. You can either leave the job unfilled, or hire a woman or URM candidate. The point is white and Asian males are denied that hiring opportunity. It's irrelevant what management would do in this scenario. Whether they'd hire a woman or just leave the position unfilled is irrelevant to the fact that white and Asian males cannot be hired, explicitly on account of their race and gender.

Again, I really find it hard to believe that people are having trouble to understand that a policy of "X% of our headcount is off limits to $RACE and $GENDER" isn't discriminatory and denial of employment on the basis of race and gender. If you're of the opinion that this discrimination is justified on the basis of advancing equity, by all means you're entitled your own opinion. Just don't try and deny that this is discrimination and ultimately results in people denied work because of immutable characteristics.

I'll post the full policy in a reply.


This is the text of the policy announcement:

> The Problem Statement

> Based on 177 like tech companies in Silicon Valley (market research and EOO-1 Diversity Statistics data), the percentage of diverse engineering talent is sparse. In short, 4.7% are latino, 2.1% are African American, and 19.2% are female. These candidates are being targeted with all of our top competitors with white gloves tactics, strategic outreach, and engagement strategies, while Dropbox has yet to systematically establish any of these practices to compete for this top talent and showcase our uniquely inclusive, dynamic, and thoughtful culture.

> Opportunity to market DBX [Dropbox] more broadly:

> Moreover, diverse engineers are the most sought after group of individuals on the market today. While the average response rate to engage is high (37%) the rate at which they are interested in moving forwards is quite low (11%). We interpret this in two ways:

> First, due to the small pool and scarcity of diverse talent, companies are motivated to keep their diverse talent happy, well-compensated, and engaged; prospects are rarely on the market, and when they are, it is a highly calculated and careful search based on existing relationships.

> Second, traditional sourcing engagement methods (email, LinkedIn) do not adequately showcase what makes Dropbox special, and because these candidates are so highly sought-after, it would serve us to highlight our culture early on, and to take a more long term approach to courting them.

> Opportunistic Hiring

> As the business needs shift and open roles become more narrow, it will become difficult to find a home for diverse candidates that we're able to engage and who pass our bar. We feel like it would be a disservice to use in the long-term if we miss out on hiring critical talent for Dropbox because of current headcount constraints. To this end, we propose that Eng VP's withhold 20 heads to hire opportunistically.

> When a diverse/URM candidate is interested in interviewing, regardless of headcount, we will put them through the process. If they pass the TPS [technical phone screen] we will bring them onsite and evaluate based on their skillset.

>If the candidate goes to HC [hiring committee], we will proactively find a sponsor/team home for the candidate, and that team would receive a preciously withheld headcount for that hire.

As you can see, this really is a needlessly verbose way of saying "we're setting aside 20 headcount off-limits to white and Asian men". The fact that these set-aside headcount doesn't count towards the team's initial headcount does not change this fact: we had 20 headcount that were explicitly off-limits to white and Asian men.

What's even more interesting is that as per the company's diversity report, our tech workforces was 23% female as compared to the 19% cited in the policy announcement. So we actually had an overrepresentation of women at the time, yet we still instituted explicitly discriminatory policies favoring them.


I think you're reading a lot into that policy which is not stated. This looks a lot like "opportunistic" hiring as it is described. In any case, this looks like more of an announcement and less of a policy document describing how to handle the edge cases on which you base your hypotheticals. In particular, did Dropbox hit its headcount target and need to use this "opportunistic" bucket? If so, how did it handle that case? Judging management decisions in the abstract is folly.

> our tech workforces was 23% female as compared to the 19% cited in the policy announcement

> So we actually had an overrepresentation of women at the time, yet we still instituted explicitly discriminatory policies favoring them.

I think you and I have different definitions of what would constitute "overrepresentation" of women in a workforce. (Again, keeping in mind that at many/most tech companies, the majority of jobs are not held by engineers.)


> In particular, did Dropbox hit its headcount target and need to use this "opportunistic" bucket? If so, how did it handle that case? Judging management decisions in the abstract is folly.

This is explicitly spelled out in the policies I sent you: the opportunistic hiring bucket is exclusive for women and URM. If you want to use the opportunistic hiring bucket, the candidate has to be either a woman or URM. What is hard to understand here?

> I think you and I have different definitions of what would constitute "overrepresentation" of women in a workforce. (Again, keeping in mind that at many/most tech companies, the majority of jobs are not held by engineers.)

The "opportunistic hiring" was exclusively for engineers. Also what do you mean by "overrepresentation"? Representation relative to the workforce is what's relevant. In a field that's 80% women and 20% men, having 50/50 representation would mean that women are one quarter as likely to be hired as men. This would require vastly disadvantaging women relative to men.


> What is hard to understand here?

What I said:

> how did it handle that case?

You haven't said a) whether this hypothetical actually happened or b) how it was handled. That leaves all this in the realm of rage-bait hypotheticals.

> Representation relative to the workforce is what's relevant.

You're defining "workforce" narrowly to exclude people who do not currently work in the "field." This is unnecessarily narrow because e.g. there are people who work in tech who do not work at pure-play tech companies in Silicon Valley (this is where the policy pulls its stats). So setting a low bar and barely clearing it.

There's also likely the possibility that they are conflating "engineering" and "tech" roles. Female CS majors are indeed ~19% in the US, but "tech" jobs often include lots of people on product teams who are not engineers: QA, UX, etc. I don't know whether Dropbox counts a UX person as "tech" or not tech. I also don't have good stats on e.g. gender balance in UX, but I would wager that it is not the same as for CS grads.

Incidentally, Dropbox non-tech is only managing 43% women. This includes roles like marketing, sales, finance, HR, etc. where the excuse of the college pipeline is not operant. (For example -- women are the majority of accountants in the US.)


It absolutely did result in people being hired under discrimination. If I have 80 headcount for men and 20 headcount exclusive to women, and 90 people were hired total then 10 hires were made with the deliberate exclusion of men, explicitly on the basis of gender. Creating situations like these is fundamentally how "opportunistic hiring" would shift company demographics. The only way it doesn't have an effect is if nobody ever taps into the "opportunistic" pool. And yes, there were teams who hired above their allocated headcount. And yes it was pretty awkward for the women and URM hired to realize that aspect of how they landed at the company.

> There's also likely the possibility that they are conflating "engineering" and "tech" roles. Female CS majors are indeed ~19% in the US, but "tech" jobs often include lots of people on product teams who are not engineers: QA, UX, etc.

It's for engineers.

> You're defining "workforce" narrowly to exclude people who do not currently work in the "field." This is unnecessarily narrow because e.g. there are people who work in tech who do not work at pure-play tech companies in Silicon Valley (this is where the policy pulls its stats). So setting a low bar and barely clearing it.

The figure of 20% comes from the number of people who put "software developer" on their tax returns: https://www.bls.gov/cps/cpsaat11.htm it's not just counting the "pure play tech companies" whatever that's supposed to mean.

If you put up a job for a software developer, count the number of men and women who would potentially be interested in the job the former is going to be about 4 times bigger than the latter. If you have a different way of measuring this proportion, you can make that argument.

> Incidentally, Dropbox non-tech is only managing 43% women. This includes roles like marketing, sales, finance, HR, etc. where the excuse of the college pipeline is not operant. (For example -- women are the majority of accountants in the US.

HR is 70-75% women according to Google. So if a company is hiring 70-75 women in HR it's not evidence of any discrimination. Same with all the other roles you listed. We'd have to inspect each role one by one, and comparing Dropbox's representation. Furthermore, the discriminatory policies were specific to engineering. So I'm not sure why these other roles are relevant. Did Dropbox discriminate against female accountants? I have no idea. But how does discrimination against male coders somehow make up for potential discrimination in other fields?


> It's for engineers.

You clearly have a lot of privileged background information you have not shared yet (for which I do not blame you). So it's going to be very difficult to pre-but some of this. For example, the BLS report you linked also shows women "computer system analysts" at close to 40%, and this is a job title often given to programmers. But you know that the appropriate compare is the row that's 19%, presumably because you have the inside background knowledge I mentioned.

In any case, the debate is stale because the folks who want to keep tech extremely male have won and these types of efforts are being relegated to history. So tech has gone back to broadly not considering qualified women and URM and this debate is mostly historical.



If there's a field that's 80% women and 20% men, is setting a DEI target of 40% men preventing sexism? Or is it perpetrating sexism?


Nobody can manage anything with this little information. A KPI set at a particular value cannot tell why the KPI has that value. Thus without more information, your hypothetical is not productive.

An illustration by way of example. In a field where other firms have a range of net profit margins of 35% - 45%, one firm sets their net profit target at 10%. Is that because they are not competitive?

Another illustration by way of example. In a field where other firms have a range of revenue per employee of $200k - $300k, one firm has no RPE target at all and its revenue per employee is $15k. Is that worth examining?


> Nobody can manage anything with this little information.

I can! Unless you have some reason to believe that one gender is superior at the job than the other, it should match the representation of the field.

> In a field where other firms have a range of net profit margins of 35% - 45%, one firm sets their net profit target at 10%. Is that because they are not competitive?

Net profit margin is not a protected class. Race and gender are.


> it should match the representation of the field

Why? You have not supported this at all. If the field is staffed by wildly bigoted hiring managers, why does your assertion follow?

Separately, why would this be obviously best for the company? Does matching parity with the field somehow ensure the best quality or cost/performance? Does this still hold at firms that are smaller than the largest (say: under 5k engineers)? Why?


If I'm hiring 20 people in a field that's 80% women and 20% men, why would I expect to hire more than 20% men? Unless I have some evidence to think that one gender is better on average than the other, then we'd expect a non-discriminatory hiring process to hire women and men in equal proportions relative to the field.

If a firm is staffed by bigoted hiring managers that favor men, we'd actually expect closer to equal representation amongst those bigoted firms. Because they're bigoted against women and less likely to hire them, they end up with a more "equitable" workforce. If a firm in our hypothetical field is bigoted against women, and hires women at half the rate as men, then it'd actually end up with a representation of ~40% women and 60% men. It's closer to equitable because of its discriminatory policies.

This is a great example of why equity isn't evidence of nondiscrimination, and gender disparity isn't evidence of discrimination.

> Separately, why would this be obviously best for the company?

Illegally discriminating against pregnant women benefits the company: maternity leave is expensive, and avoiding expecting women saves money. Does that make it a good thing? Remember, nondiscrimination is both morally right and the law of the land (in most liberal democracies, at least), even if you have reason to believe that increasing or decreasing the representation of certain groups is advantageous.

> Does matching parity with the field somehow ensure the best quality or cost/performance?

Hiring the best candidates ensures the best quality and cost to performance ratio. Unless you have some reason to believe that men are better than women, then matching parity of the field is a side-effect of hiring the best candidates.

> Does this still hold at firms that are smaller than the largest (say: under 5k engineers)? Why?

Sure, a small sample size is going to have more variance. A one-woman consulting firm is going to have 100% women employees. That's not evidence of discrimination. But nor is a one-man consulting firm that's 100% male. Smaller firms have a lower sample size and are naturally going to have larger variance due to smaller sample sizes. But it'll all average out.


it will probably be downvoted to hell, but it seems like there isnt many that argue the field of plumbing needs more women, more women laying down sewage pipes etc.

I tried reaching out to some feminist organisations to make some campaigns for this, and they were 100% not interested.

it seems the "increase female %" only covers jobs that are super cleanly in office spaces


I understand your sentiment. In theory there is no difference between a plumber or construction worker vs a CEO when it comes to DEI.

However, in the real world, a female CEO or President is much more likely to inspire girls to work towards those roles, compared to a model female plumber.

They call it "glass ceiling" not "glass floor" for a reason.

Those feminist organizations are definitely hypocritical, no question about that. That does not necessarily mean they are doing anything "wrong", according to how the world runs. I wouldn't blame them for not doing anything.

All that said, I think you bring up a valid point that people almost never talk about. But that's as much as it's worth.


if broadening female involvement in the upper levels is good, it makes no sense that it isnt also in the "lower levels" (though I would say its not proper to describe plumbing as a lowlevel job).

The goal is NOT to "inspire girls", the goal is to make the job sector reflect the population, no?

for feminists it seems like only the nice cushy office jobs needs to reflect the population, not the dirty labour


the goal is to remove barriers. reflecting the population implies that every job should have 50/50 gender parity. i don't think that should be a goal unless there is a benefit, such as in school where children need both male and female role models.

the problem is to identify what the barriers are and how to actually get rid of them. lack of inspiration can be a barrier.


There are many studies that have identified barriers that causes gender segregation. The most interesting aspect that I find in those studies is that the barriers are similar for both male and female dominated industries.

The biggest barrier of them all for both men and women is that young adults will look to similar aged and gendered peers when choosing career paths. Such choices generally provide a feeling of safety which provide protection against setbacks and doubt (something most student and later professionals generally face multiple times at some point in their life). This concept is a major aspect of gender equality paradox, which predict higher rate of gender segregation with fewer barriers.

The most effective way to get rid of those barriers is to create alternative forms of protection that give similar effects. Mentorship programs has shown to be very effective substitute for both women and men, but they are costly and do not fix the initial problem when young adults chooses career paths.

Sweden has public data on gender segregation, and they provide an additional risk factor of gender segregation. Career paths that has natural points for segregation has generally higher rate of gender segregation. Examples of those are teachers who can first segregate on age/educational level, and then further segregate on subject. An other is nurse, doctors and other health care specialists, who can later segregate further based on specialty. Give people multiple chances to self-segregate based on identity and they will do so.


that is very insightful and interesting. the more opportunities there are to choose a direction along your career, the more often you have to consider if you are comfortable being part of a minority in your field.

but i think we can do more with mentorship and go beyond its perceived limits. however getting there takes us to rethink education as a whole.

there are a few aspects: for one i think everyone should have a mentor at least for the first few years of their career. that should be part of our education system. for example, every university graduate should be required to mentor at least one new student. every university student as well as trade apprentice should mentor in highschool. etc... in the montessori education method older students always teach younger ones, so mentoring at a higher level is really just an extension of that.

this would obviously be easier to ask in countries where higher education is free, because then its fair to ask for something back.

from these, mentoring in highschool is probably the most important because that's when careers are chosen. i remember that time, and i remember being all alone with my choices. i really would have loved to be able to get more insights into the potential careers that i was interested in.


well don't ask feminist organizations, ask the plumbers and the customers:

https://www.workiz.com/blog/plumbing/why-how-hire-women-plum...

The second reason comes down to client safety. For various reasons, some women simply aren’t comfortable with male plumbers. Though most plumbers are perfectly nice people unless you’re a particularly spirited clog in a drain, the fact remains that not every woman is comfortable being alone in her home with a man that she doesn’t know. Having access to industry professionals that can provide a little more comfort can be a huge selling point, particularly for young women, single mothers, and victims of domestic violence.

https://www.worldplumbing.org/attracting-more-women-to-plumb...

https://www.commusoft.us/blog/why-the-industry-needs-more-fe...

“We work with women that are survivors of domestic abuse. At a women’s safe house, if some work needs doing and they need to contact a tradesman, it can often be quite stressful for them. We get inundated calls from customers who want to use tradeswomen.”

so it appears there is an objective need for female plumbers, not just as a statistic, not just to create a better workplace, or higher shareholder returns. those arguments are given too. but with single women being more common and higher awareness of domestic abuse, the argument can be made that with 30% of women being single, 5% female plumbers is not enough.

this is a very rough estimate. 30% single women is 15% of the total population, but 23% of the total number of households. (30% single women + 30% single men + 70% each of married men and women adds up to 130 household units)

however if you add 25% stay at home women who are usually the ones who would deal with a plumber since the husband is at work, then they will probably prefer female plumbers too. so i'd estimate that we need at least 25% to 30% female plumbers.

it can be argued that being served by a tradesperson of your own gender is a right, and therefore a gender distribution that matches the needs of the population is not just something that would improve the trade (which can be argued about), but actually a necessity that should be enforced by law.

same goes for teachers btw. boys need male role models, and therefore i believe laws should push for a 50/50 gender parity among teaching staff at all levels.

similar arguments can be made for all consumer facing professions. (police for example). this is no longer a question of qualification or job preference. it's not even a question of gender equality, but a question of how to best serve the population as a whole.


Women were a much higher percentage in the field 1990. Women abandoned the field after the IT-bubble and it never recovered.


once interests rate went up all DEI initiatives dried up - these companies don't really have integrity or beliefs beyond doing what is politically correct at the current time.


Almost as if they are organizations formed around the goal of optimizing profits, and not the general benefit of society.


Not even optimising profits, growth hacking to optimise share price.


Then what use are they?


The idea behind a free market is to use the desire for profit as an incentive for the self-interested to collaborate with others and serve the public interest in doing so; this is why the state promotes the formation of businesses and punishes fraud and insider trading, in addition to violent crime.

... Man has almost constant occasion for the help of his brethren, and it is in vain for him to expect it from their benevolence only. He will be more likely to prevail if he can interest their self-love in his favour, and shew them that it is for their own advantage to do for him what he requires of them. Whoever offers to another a bargain of any kind, proposes to do this. Give me that which I want, and you shall have this which you want, is the meaning of every such offer; and it is in this manner that we obtain from one another the far greater part of those good offices which we stand in need of. It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest. We address ourselves, not to their humanity, but to their self-love, and never talk to them of our own necessities, but of their advantages.

- From An Inquiry into the Nature and Causes of the Wealth of Nations, by Adam Smith

https://www.gutenberg.org/cache/epub/3300/pg3300-images.html


It’s the best system we’ve discovered so far to optimize price discovery, efficiency and productivity.


Why would rising interest rates affect what is considered politically correct?


You can read "politically correct" as "politically fashionable".

When money is cheap it's easy to spend a bit of money on political signalling. However when money is no longer cheap that pure cost centre is the first on the list for cuts.


Borrowing money becomes more expensive so companies will focus on their own needs (or surviving) rather than giving or outreach programs. Unlike Apple or NVIDIA most companies need to borrow money to stay in business.


> Borrowing money becomes more expensive so companies will focus on their own needs (or surviving) rather than giving or outreach programs.

All the (Big tech) companies - not just Apple and Nvidia - have higher revenues and profits now than they did during the Zero-interest regime. They are not hurting for money to fund outreach programs that meet their strategic goals.

What has changed is their hiring outlook. Online services saw unprecedented growth when everyone was cooped up in their homes due to Covid lockdowns, and the tech companies thought the growth would be permanent, rather than a temporary bump, and couldn't hire engineers fast enough to meet the anticipated growth: hence the outreach to non-traditional hiring-pipelines. After the layoffs, they stopped hiring aggressively and the labor market is now a buyers market


> most companies need to borrow money to stay in business.

With respect, this is entirely untrue. Most companies don't need to borrow money. It's the functional cancer of the VC-funded silicon valley meta that needs to borrow money to do anything useful. Most corporations run based on what they earn. This is the only reasonable thing too, or every country would have the egregious debt load that the US has.


>Unlike Apple or NVIDIA most companies need to borrow money to stay in business

What do you mean by this?


Nobody inside the org cares about gender ratios, but they do have to react to people outside of the org who care a whole lot.

It's easier to explain reality than to try and change it.


Idk, I've read b2b contracts that have demands similar to this. They arent explicit, its softer.


I have as well, but I'm more of a cynic. Usually you can trace requirements back to either DEI dependent funding or government contract requirements. Less common is an attempt to market or build positive brand association by making a public commitment. With the occasional case where one individual uses their position in a company to sneak their personal agenda in.

Mostly the behaviour is determined by tangible external benefits rather than any kind of real belief that gender ratios should be acknowledged.


If the problem is solved, the funding disappears.


That would be a complete success, since the funding disappeared anyway


Except in the real world, where the infrastructure to raise money for a previous problem can out compete new infrastructure.


Funding for many, many worthy orgs crashed with COVID-19, and has barely recovered. From the outside looking in, it seems the whole administrative capacity (ecology) of the fund raising world just dried up and needs to be rebuilt.


Perhaps the administrative capacity dried up with the lack of free money once interest rates rose?


That’s a bummer I was hoping it was more of a “mission accomplished” kind of closure.


> they decided to check notes directly compete with _the_ largest ecommerce website in the universe

I actually mis-read the headline that way too. They're doing the opposite - making sure they are only selling hand made items (again) and NOT competing with Amazon anymore.


No, you read it right the first time.

It's a misleading PR move pretending to move back to "hand-made" items with loopholes big enough to drive a planet through.


Thank you, I have been caught not RTFA and going on my priors.


Georgia - also 1 mile from the school. I tried riding by bike with my daughter and the principal called and told me to never do it again as it was too unsafe. She claimed you were only allowed to go to school by car or bus, despite there being a sidewalk.


I moved to the Netherlands last week. I was paying $120 a month for 100 mbps with Spectrum. New customers paid less and got more. I complained and explained that was hardly fair since I had been a customer for 15 years at the same address. They said I was welcome to cancel my service for 3 months and re-activate.

I now pay €67.50 for 4gbps up / 4gpbs down. I'm actually getting those speeds and the day I arrived service was already turned on. I just had to ride my bike down to the post office and retrieve my modem.


This is mentioned all the time but the cost to get a driver's license in Germany is incredible high. And even after spending a lot of money on training, most native Germans fail the exam the first time. Then, like you said - there are rules and very stiff penalties in place for after you get your license.

Americans would simply not put up with this.


The HN rules say the original source should be posted (which may mean a paywalled link) and that you shouldn’t post comments complaining about it:

https://news.ycombinator.com/newsfaq.html

The best thing to do is to comment an archive link or upvote an existing (working) one.


lunchtime doubly so


Thanks for saving me the trouble to post that. Somebody needed to. DA - RIP.


Thankfully, a Hoopy Frood was available this day.


This is how it’s done in the Netherlands. You sign up for a bank account and they check your ID, your proof of residency, they do a video selfie, etc. It was impossible for me to get a Dutch bank account (as an American) until after I signed my apartment lease which was a bit of a catch-22 but not insurmountable.

But once that is all setup they have a system called digiID that is wonderful. Anytime your identification needs to be verified it’s done through your bank. I signed up for renters insurance, they did a little oauth thing with the bank, the bank asked which details I was willing to share (the insurance company asked for just last name, dob, and address) and I agreed. It took seconds.


> Note: The “Holborn” name is to signify that these computers were “Born in Holland”. Hol. Born.

Not entirely accurate as the company was in Hengelo, Overijssel but "Overborn" just doesn't have the same ring to it.


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

Search: