4 ms·

> Women leaving tech are doing so probably because they can afford to. I am considering leaving tech because selective biases play a larger role in hiring and

> Women leaving tech are doing so probably because they can afford to. I am considering leaving tech because selective biases play a larger role in hiring and advancement than they used to despite all efforts to the contrary. After talking with other people I am not alone in thinking this. I am willing to leave tech even though I cannot afford to because the hit to my finances is temporary but the barriers to my career from remaining in tech are permanent. Let me discuss selective bias in a bit more detail. When I first started programming candidate qualification was centered around the needs to the business. For example, did a candidate understand the minimum criteria of working in a given technology plus the advanced criteria that harmed user interaction, slowed execution speed, or some other measurable concern. Those days are long gone, at least in high level languages. Now the name of the game is programming fashion. Specifically, does the candidate write code according to a current trend a given team or interviewer wants without regard for any performance or measurable quantum. That change is disruptive but not in a healthy way. In the old days the desirable and rewarded developers were those who performed well, such as writing the best code in the least time. The people on the right side of a bell curve. Replacing that with fashionable and non-measurable factors limits candidate selection to the center of the bell curve, which are people who do not perform as well but are better rewarded for it. If I am willing to abandon a long time tech career even though it will hurt and have found others that share my frustration I can only imagine that perhaps this might be some small factor in why some women might want to leave in general. I suspect this problem has increased despite all efforts to the contrary because all efforts to eliminate hiring/reward bias were typically limited to identity factors that carried provable legal consequences.

ethanbond 7y ago
The reliance on measurable outputs from days of yore was probably not the result of some high minded meritocracy which is now gone, but from building things that needed performance - a need which has now diminished greatly for a large proportion of work. This reduction is simultaneously a result of higher level abstractions, as you say, but also a result of simply more tech in more places that doesn’t really need it. There are still plenty of places that must incentivize quality as a matter of survival, just your friends in the Bay Area won’t “oooh” and “ahhhh” when you tell them you work there.
- austincheney 7y ago
  Execution performance is one measurable factor among many that I happened to use as an example. The problem with bias is that it’s self-serving and lingers well past a first impression, like a compounding problem. Any bias exists to cognitively qualify a subjective decision in the absence of stronger criteria. The self-serving nature of bias is that it typically exists as a form of insulation filtering whereby things less like the person, or their proximity of comfort, are discarded for perceptions of compatibility. Compatibility sounds important when building a team, however that factor is unmeasurable and often poorly defined in any existential capacity. Instead this is often an unspoken means of eliminating competition, diversity, and insecurity from a person(s) tasked to make a personnel decision without the soft skills to warrant such. In my experience in the military you don’t need interpersonal compatibility, either real or perceived, to build an excellent team that cares for each other and achieves high performance. Since the problems resulting from bias linger past a first impression the consequences of such are continuous. If, for example, bias is heavily present on a team a candidate who survives selection there will face frustrations moving forward when confronting or challenging the perceptions of the bias present. That person will be limited in career mobility internal to that organization. > There are still plenty of places that must incentivize quality That depends on who defines quality and how. At just about every place I have ever worked developers define quality in a way that is utterly foreign to business concerns or product quality at the end user. Without some highly specific written definition of quality I find it’s an excuse to reinforce like-minded opinions.
  - ethanbond 7y ago
    That’s a great, thoughtful response. I don’t disagree with anything here but thought I’d add one “Yes, and” note regarding your military experience. What I’ve found working in a place with virtually no top-down authority is that interpersonal compatibility matters quite a lot to achieve business outcomes. I figure most organizations land somewhere between a military and my workplace in the top-down authority dimension, so they probably also rely a bit more heavily on interpersonal compatibility than your experience with the military would indicate is actually necessary. I strongly agree that this is a liability, however, due to its potential for maintaining and entrenching biases of various types.

ethanbond 7y ago

The reliance on measurable outputs from days of yore was probably not the result of some high minded meritocracy which is now gone, but from building things that needed performance - a need which has now diminished greatly for a large proportion of work. This reduction is simultaneously a result of higher level abstractions, as you say, but also a result of simply more tech in more places that doesn’t really need it. There are still plenty of places that must incentivize quality as a matter of survival, just your friends in the Bay Area won’t “oooh” and “ahhhh” when you tell them you work there.

austincheney 7y ago
Execution performance is one measurable factor among many that I happened to use as an example. The problem with bias is that it’s self-serving and lingers well past a first impression, like a compounding problem. Any bias exists to cognitively qualify a subjective decision in the absence of stronger criteria. The self-serving nature of bias is that it typically exists as a form of insulation filtering whereby things less like the person, or their proximity of comfort, are discarded for perceptions of compatibility. Compatibility sounds important when building a team, however that factor is unmeasurable and often poorly defined in any existential capacity. Instead this is often an unspoken means of eliminating competition, diversity, and insecurity from a person(s) tasked to make a personnel decision without the soft skills to warrant such. In my experience in the military you don’t need interpersonal compatibility, either real or perceived, to build an excellent team that cares for each other and achieves high performance. Since the problems resulting from bias linger past a first impression the consequences of such are continuous. If, for example, bias is heavily present on a team a candidate who survives selection there will face frustrations moving forward when confronting or challenging the perceptions of the bias present. That person will be limited in career mobility internal to that organization. > There are still plenty of places that must incentivize quality That depends on who defines quality and how. At just about every place I have ever worked developers define quality in a way that is utterly foreign to business concerns or product quality at the end user. Without some highly specific written definition of quality I find it’s an excuse to reinforce like-minded opinions.
- ethanbond 7y ago
  That’s a great, thoughtful response. I don’t disagree with anything here but thought I’d add one “Yes, and” note regarding your military experience. What I’ve found working in a place with virtually no top-down authority is that interpersonal compatibility matters quite a lot to achieve business outcomes. I figure most organizations land somewhere between a military and my workplace in the top-down authority dimension, so they probably also rely a bit more heavily on interpersonal compatibility than your experience with the military would indicate is actually necessary. I strongly agree that this is a liability, however, due to its potential for maintaining and entrenching biases of various types.