Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We ran the numbers, and there really is a pipeline problem in engineering hiring

blog.interviewing.io

Post reply on HN
Print view View original

641–650 of 668 posts

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#641

Post by goosehonk »

filoleg wrote: post #634
goosehonk wrote: post #589

Earlier quoted context omitted.

I'll never have personal first-hand experience with another kind of degree program, but my engineering program taught thermodynamics, accounting, technical writing, and ethics. I am looking right now at the Stanford CS undergraduate catalog and there are no requirements for technical writing, ethics, etc in here at all. Even the senior year writing requirement can be satisfied by working for Facebook for six months,…

Your comment seemed really strange to me, because I remember that even back at Georgia Tech, we had both technical writing and ethics requirement for anyone in the CS program, so I decided to doublecheck the facts you listed about the Stanford CS program. Where did you get your info from? Because I just checked the Stanford CS curriculum requirements for Bachelors degree, and it clearly has "Computers, Ethics, and Pu…

The senior project is the one I mentioned that can be fulfilled by CS210, working at “our industry partners” in lieu of actual university coursework. The “tech & society” catalog is a joke. Look at the courses. Archeology? Fine as an elective.

My engineering program required an upper-division course from the philosophy department on the development of rigorously ethical systems of thought.

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#642

Post by stale2002 »

commandlinefan wrote: post #621
lordCarbonFiber wrote: post #419

Earlier quoted context omitted.

Ignorning the fact that many engineering orgs can't hit 10% let along 18%; you've already lost if you've restricted your pipeline to "has a CS degree". Statistically your company isn't actually hiring for jobs that need higher level CS theory. It's not assigning malice, but pure negligence . There's a huge population of competent people with amazing ROI for your business and, especially companies with pockets as deep…

> you've already lost if you've restricted your pipeline to "has a CS degree". But, why, though? Are you saying that women as a group are less capable of completing CS degrees than men?

No, women are not less capable of getting a degree.

Instead, the problem is often that they have such a bad time, in those programs, due to the existence of a toxic culture, that they don't want to put up with that crap, and instead leave to get a job in industry that has a less bad culture (because they have other options).

Unfortunately, toxicity on all stages of the pipeline tends to drive out people who would be good engineers, but instead decide to do something else, because they have alternative options that don't suck so much, for cultural reasons.

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#643

Post by filoleg »

vkou wrote: post #588
username90 wrote: post #550

Earlier quoted context omitted.

> but it is obviously possible for any one company to hit 50/50, without compromising on their hiring bar How? If I interview a bunch of people for programming positions and pick the 10 best men and 10 best women, which group do you think will be best? Statistically I have considered several times more men than women, so the male group will almost surely be stronger. The only way to avoid this is to intentionally pic…

No hiring bar in the world is looking for the best 10 people. It's just looking for people who are better than (some arbitrary skill level). More men than women could bench press 160 lbs, but if your hiring bar is 'we hire people who can bench 160 lbs', you can trivially find as many people from either gender as you want.

Would your point still stand, if the baseline of being able to bench press 160lb was just a starting requirement, with a possibility that requirements can change at any point and you might need people who can be trained to bench press 260lb instead?

Also, if you are buying a car with the goal of being able to accelerate from 0-60mph in under 4 seconds, and you have to choose between two cars priced exactly the same, but one can do it in 3.9 seconds and another one in 3.1 seconds, why would you pick the 3.9 car over the 3.1 one, even though the 3.9 one meets your requirements of under 4 seconds just as well (assuming all the other relevant characteristics are equivalent or about the same)?

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#644

Post by filoleg »

fennecfoxen wrote: post #599
Buge wrote: post #508

Earlier quoted context omitted.

If we can't say "Asian" then let's also stop saying "black" and "white".

One is quite welcome to say "Asian" in the abstract. It's the part where one says that "Asian" is an socio-ethnic monoculture in practice that is uninformative. At the very least, name the Asians you're talking about. India and China, for instance, seem to have a lot more representation among US programmers than other Asian nations.

Sure, but the same thing can be said about white people as well. Somehow, there is a disproportionate representation of slavic/eastern european people in top tier tech and fintech jobs, but it seems to fly under most people's radars when it comes to that kind of conerns. So we are back at ground zero with nothing useful coming out of this, now what?

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#645

Post by geebee »

nradov wrote: post #17

My alma mater Harvey Mudd College has had rough gender parity in Computer Science for several years now. It's a small school but other colleges could learn from their practices. https://www.inc.com/kimberly-weisul/how-harvey-mudd-college-...

This article doesn't mention anything about scale. I think we may be criticizing an elephant for failing to lift as many times its bodyweight as an ant.

For a back-of-envelope calculation[1][2], it looks like there were 27 CS majors out of Harvey Mudd in 2017. We can probably add in another 20 from the joint CS-Math degree. So, about 50. Berkeley, on the other hand, graduates almost a thousand students in EE or CS every year. Stanford, MIT, Georgia Tech, University of Michigan, UCSD... these schools all graduate large numbers of CS majors and compete with each other for the top students.

The accounting is tough here because the degrees at Harvey Mudd and Berkeley are different, so I may be over or under counting, but we're talking about a massive difference in scale here.

Let's take one metric - the AP Computer science test. About 27% or the people who took the AP Computer science test are women. Harvey Mudd is small enough that it could probably fill 100% of the starting class with women who got 5's on the AP Computer science test and have near perfect math grades and SAT scores. But if scores are roughly evenly distributed between men and women, eventually, you're going to have a school that has more men with 5's than women - and the only way to reach 50-50 parity will be to select women who got 4s or 3s with men who got 5s. It won't be Harvey Mudd, but that's because HM is tiny and elite. Instead, it'll be Berkeley, or Georgia Tech, or UC San Diego, or U Texas. Would Harvey Mudd be able to maintain this at 1000 CS majors, using the same methods they do now, while competing fiercely for the top students in this environment and at this scale?

[1] https://www.hmc.edu/institutional-research/institutional-sta... [2] https://eecs.berkeley.edu/about/by-the-numbers

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#646

Post by romwell »

CapricornNoble wrote: post #607
romwell wrote: post #534

Earlier quoted context omitted.

Because we treat mental and physical work differently.

If physical work is overwhelmingly dangerous, and physical work is (currently) overwhelmingly male, then wouldn't it be misandrist to treat it differently from mental work, where an equal labor distribution is lauded?

Where did I say anything about danger?

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#647

Post by rinchik »

dang wrote: post #615
rinchik wrote: post #309

Earlier quoted context omitted.

It's not psychiatry, it's psychology. Not even remotely I'm making an attempt to diagnose a psychiatric pathology as in no way I'm qualified to do that. You are giving me more credit than I actually deserve.

I mean "psychiatric" ironically rather than literally—adding "internet" to a word is sort of an instant ironizer. You can read it as "psychological diagnosis" if you like. The point is that making remarks about someone's psychological issues as you perceive them, suggesting that they need therapy and so on, is unduly personal. It crosses a line that is better not to cross, even if you're right, because it frequently…

Ah that's a well written and thoughtful response to my comment. Thank you for that! However I can’t say I agree with the trigger premise though, even if suggestion perceived as personal. Psychological problem suggestions, gentle nudges if you will, are inherently beneficial even if those generate a defensive response. Defensive response is a response! That response is a deliberate thought in itself, an analysis. In modern, developed societies, it seems that the only appropriate time to suggest therapy is when the following sentence includes a suicide hotline phone number, how crazy is that? Therapy or any psychological self-care should be normal, and not in any way different from a regular, scheduled visit to the dentist.

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#648

Post by wsy »

CapricornNoble wrote: post #493
wsy wrote: post #298

Earlier quoted context omitted.

Your examples are not related to software engineering. Please name software engineering related biological differences between men and women, backed by solid scientific studies.

Do we even have a scientifically-backed set of essential biological or cognitive skills related to software engineering? I would probably start with [1] and [2]. Paper [3] concludes that sex-based differences in risk tolerance tangibly impacted approaches to spatial navigation AND program development. There's a few others on spatial cognition and mental information processing: [4][5][6]. Paper[5] in particular links…

[1] and [2] are not studies. [2] isn't even peer reviewed.

[3] is behind a paywall. It appeared in the "2006 22nd IEEE International Conference on Software Maintenance", which is a pretty good indicator that this is not work done by social scientists, i.e., not scientifically backed. I have worked in science, the good papers appear in journals highly relevant for the topic.

[4] and [5] are not about software engineering.

[6] is also behind a paywall, and is based on a study with 24 students.

As I can't access the full-text papers, what is the reported strength of the effect, i.e., if person A is x% better than person B in spatial recognition, how much better is A in software engineering?

edit: here is a study presenting evidence that spatial skill differences in gender are not biological, see https://psycnet.apa.org/record/2013-38414-001. Without paywall: https://www.researchgate.net/profile/Mike_Stieff/publication...

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#649

Post by TrinaryWorksToo »

filoleg wrote: post #635
TrinaryWorksToo wrote: post #630

Earlier quoted context omitted.

I am proof a pool exists.

A pool of statistical outliers is not a "big pool" that the parent comment was referring too. There are quite a few brilliant software engineers I got a chance to work with who had no college education whatsoever, but they are very clearly outliers among very few. I strongly doubt there is some big untapped pool of engineers without degrees who are just being completely overlooked simply for not having a degree.

How could we use Newton's flaming laser sword to settle our argument?

Re: We ran the numbers, and there really is a pipeline problem in engineering hiring

#650

Post by dang »

jeromebaek wrote: post #598
thorwasdfasdf wrote: post #594

I have a 4 year old daughter who I'm trying to get interested in programming. The first step is to teach her reading/writing and math. Every time we start with basic counting, after just 2 minutes, she's constantly telling me "daddy, I don't want to count anymore". I'm not giving up yet. But, we have to accept the fact that certain genders don't like math and don't like programming. Sometimes, I see the same thing pl…

No offense, but you sound like a terrible parent. Keep your unconscious bias in check while you teach your daughter to count. Better yet, don't teach your daughter to count. Get a tutor.

Personal attacks will get you banned here. Please read the rules and make no more posts like this.

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

Post reply on HN