7 ms·

Key point: > there really aren’t enough women to meet demand… if we keep hiring the way we’re hiring. Namely, if we keep over-indexing on CS degrees from top s

Key point: > there really aren’t enough women to meet demand… if we keep hiring the way we’re hiring. Namely, if we keep over-indexing on CS degrees from top schools, and even if we remove unconscious bias from the process entirely, we will not get to gender parity. And yes, there is a way to surface strong candidates without relying on proxies like college degree. I am all in favour of reducing the industry’s reliance on fancy name schools for hiring.

xyzzyz 7y ago
> I am all in favour of reducing the industry’s reliance on fancy name schools for hiring. Soon after the industry seriously stops doing this, the fancy name schools will quickly get female-dominated, and whatever the industry starts using as a signal of candidate quality that hadn't been male-dominated before, will start to be. When proposing new policies, always solve for equilibrium, don't ignore second order effects, and never assume the world is steady state system.
- majos 7y ago
  > Soon after the industry seriously stops doing this, the fancy name schools will quickly get female-dominated Why would this happen? Is your thinking that men will respond to the lower incentive of a fancy school degree by pursuing other options, but women won't? If so, why?
  - Pharaoh2 7y ago
    Presumably because the gap exists in the current system, and the current system is optimized for the current scoring function. Changing the scoring function will cause the system to rebalance into a new steady state. Although, there is no clear reason why the new steady state would not be female dominant or evenly split, but if I were to take a bet, I would bet that the new scoring function will still hold existing bias because the problem starts much earlier.
  - xyzzyz 7y ago
    Pretty much, yes. You are asking why would the response to this incentive affect people of different gender differently. I don’t have any clear answer to this question, other than pointing out that in the current system, we already have an existing imbalance, so there is already a difference in how incentives affect different people, and unless we determine the exact mechanism here, random changes are unlikely to change it. By the way, this is called Campbell’s law. https://en.wikipedia.org/wiki/Campbell%27s_law https://en.wikipedia.org/wiki/Campbell%27s_law
- wolco 7y ago
  I would think the opposite. More males are in trade schools programming than females. Females make up a higher percentage number of college students compared to males.
  - nostrademons 7y ago
    I think that was his point - "the fancy name schools will get female-dominated".
- DoofusOfDeath 7y ago
  > When proposing new policies, always solve for equilibrium, don't ignore second order effects, and never assume the world is steady state system. I've thought along similar lines, but it was never clear to me if such a policy really performs "better" in real-world, politicized policy issues. Do you know of any supporting examples?
brightball 7y ago
Several of the best developers I know have no degree at all. I know one guy who did a boot camp 2-3 years ago and since then has self taught himself so much that he's one of the most knowledgeable people I've been around. He was recently told in an interview that he otherwise passed with flying colors that, "someone with a boot camp education couldn't be seriously considered." It's mind blowing to me that anyone in charge of hiring for technical positions can be dense enough to think that your education stops when your career starts.
- x0x0 7y ago
  You're going to find any stupid choices in a wide enough pool. That's not representative: I know boot camp grads at FAANGs. At smaller companies -- like mine -- it's not that we won't hire out of boot camps. We do, and have. But we're very conscious of search and training costs. Search costs can be somewhat fixed with money by using internal recruiters; training costs are horrific. Bootcamp grads are nearly useless in a professional environment without enormous amounts of senior / staff eng time.
  - zebraflask 7y ago
    If the boot camp is used as a get up to speed mechanism by someone with a previous technical background of some kind, even tangential, I guess it could work out sometimes. I tried to look for a junior dev recently and was flooded with boot camp resumes from career changers. The code samples were all more or less cut and paste type stuff where you could tell immediately that even if the candidate were a fairly bright and motivated person, the degree of handholding needed to ramp them up would be just impossible.
  - bsder 7y ago
    > Bootcamp grads are nearly useless in a professional environment without enormous amounts of senior / staff eng time. To be fair, this is also true for standard undergrads. The difference between a cooperative engineering student and a standard undergrad is quite staggering--9-12 months of working experience causes a vast differential.
baddox 7y ago
That does seem to be the key point, but it directly contradicts the catchy headline "there really is a pipeline problem." This reads like the author is complaining that there's no way to improve hiring diversity without....significantly changing the hiring process. Well, yeah?
bsder 7y ago
> I am all in favour of reducing the industry’s reliance on fancy name schools for hiring. 1) The fact that people aren't hiring from schools lower on the list tells me that we have an oversupply of CS majors. If I need an employee, I'll do what I have to to get one--this indicates that most companies don't really need CS people. 2) "60% of software engineers at FAAMNG hold a degree from a top 20 school" isn't automatically damning. Some of those schools simply produce a LOT of students. I seem to remember something like University of Illinois produces almost 20% of all engineers graduated annually. Silicon Valley also may simply have a local bias because Berkeley and Stanford are literally next door. 3) If I had a chance to direct a talented female in terms of college study, I would point her at engineering, not CS. Engineering has its own bias problems, but seems to be a lot less toxic than CS. It also seems to have better geographic spread of jobs than CS right now.

xyzzyz 7y ago

> I am all in favour of reducing the industry’s reliance on fancy name schools for hiring. Soon after the industry seriously stops doing this, the fancy name schools will quickly get female-dominated, and whatever the industry starts using as a signal of candidate quality that hadn't been male-dominated before, will start to be. When proposing new policies, always solve for equilibrium, don't ignore second order effects, and never assume the world is steady state system.

majos 7y ago
> Soon after the industry seriously stops doing this, the fancy name schools will quickly get female-dominated Why would this happen? Is your thinking that men will respond to the lower incentive of a fancy school degree by pursuing other options, but women won't? If so, why?
- Pharaoh2 7y ago
  Presumably because the gap exists in the current system, and the current system is optimized for the current scoring function. Changing the scoring function will cause the system to rebalance into a new steady state. Although, there is no clear reason why the new steady state would not be female dominant or evenly split, but if I were to take a bet, I would bet that the new scoring function will still hold existing bias because the problem starts much earlier.
- xyzzyz 7y ago
  Pretty much, yes. You are asking why would the response to this incentive affect people of different gender differently. I don’t have any clear answer to this question, other than pointing out that in the current system, we already have an existing imbalance, so there is already a difference in how incentives affect different people, and unless we determine the exact mechanism here, random changes are unlikely to change it. By the way, this is called Campbell’s law. https://en.wikipedia.org/wiki/Campbell%27s_law https://en.wikipedia.org/wiki/Campbell%27s_law
wolco 7y ago
I would think the opposite. More males are in trade schools programming than females. Females make up a higher percentage number of college students compared to males.
- nostrademons 7y ago
  I think that was his point - "the fancy name schools will get female-dominated".
DoofusOfDeath 7y ago
> When proposing new policies, always solve for equilibrium, don't ignore second order effects, and never assume the world is steady state system. I've thought along similar lines, but it was never clear to me if such a policy really performs "better" in real-world, politicized policy issues. Do you know of any supporting examples?

brightball 7y ago

Several of the best developers I know have no degree at all. I know one guy who did a boot camp 2-3 years ago and since then has self taught himself so much that he's one of the most knowledgeable people I've been around. He was recently told in an interview that he otherwise passed with flying colors that, "someone with a boot camp education couldn't be seriously considered." It's mind blowing to me that anyone in charge of hiring for technical positions can be dense enough to think that your education stops when your career starts.

x0x0 7y ago
You're going to find any stupid choices in a wide enough pool. That's not representative: I know boot camp grads at FAANGs. At smaller companies -- like mine -- it's not that we won't hire out of boot camps. We do, and have. But we're very conscious of search and training costs. Search costs can be somewhat fixed with money by using internal recruiters; training costs are horrific. Bootcamp grads are nearly useless in a professional environment without enormous amounts of senior / staff eng time.
- zebraflask 7y ago
  If the boot camp is used as a get up to speed mechanism by someone with a previous technical background of some kind, even tangential, I guess it could work out sometimes. I tried to look for a junior dev recently and was flooded with boot camp resumes from career changers. The code samples were all more or less cut and paste type stuff where you could tell immediately that even if the candidate were a fairly bright and motivated person, the degree of handholding needed to ramp them up would be just impossible.
- bsder 7y ago
  > Bootcamp grads are nearly useless in a professional environment without enormous amounts of senior / staff eng time. To be fair, this is also true for standard undergrads. The difference between a cooperative engineering student and a standard undergrad is quite staggering--9-12 months of working experience causes a vast differential.

baddox 7y ago

That does seem to be the key point, but it directly contradicts the catchy headline "there really is a pipeline problem." This reads like the author is complaining that there's no way to improve hiring diversity without....significantly changing the hiring process. Well, yeah?

bsder 7y ago

> I am all in favour of reducing the industry’s reliance on fancy name schools for hiring. 1) The fact that people aren't hiring from schools lower on the list tells me that we have an oversupply of CS majors. If I need an employee, I'll do what I have to to get one--this indicates that most companies don't really need CS people. 2) "60% of software engineers at FAAMNG hold a degree from a top 20 school" isn't automatically damning. Some of those schools simply produce a LOT of students. I seem to remember something like University of Illinois produces almost 20% of all engineers graduated annually. Silicon Valley also may simply have a local bias because Berkeley and Stanford are literally next door. 3) If I had a chance to direct a talented female in terms of college study, I would point her at engineering, not CS. Engineering has its own bias problems, but seems to be a lot less toxic than CS. It also seems to have better geographic spread of jobs than CS right now.