5 ms·
Age discrimination is all over tech. But what's the argument for age discrimination after 40? What's the mindset that someone who is for this discrimination wou
by clnq 3y ago
Age discrimination is all over tech. But what's the argument for age discrimination after 40? What's the mindset that someone who is for this discrimination would have, no matter how immoral or unethical? Is there a defensible view that supports discrimination at all?
EDIT: Sorry for creating a downvote hole for some of you who respond. There are good responses immediately downvoted. It is important for us to understand the arguments behind these biases to effectively fight them. It is not productive to silence this discussion.
Thank you for those who responded seriously and helped me understand the situation better.
- belter 3y agoGetting rid of expensive employees.
- rightbyte 3y agoNo I think it is more due to implied seniority due to age. The older the engineer the less BS he eats from management.
- seattle_spring 3y agoI dunno, my real life experience shows the most insistence on agile/scrum bullshit coming from older engineers. SDETs specifically.
- d0mine 3y agoIs it really that value/price becomes worse with age or the companies are just being short-sighted?
- yaseer 3y agoI think the way you've phrased your question will invite negative responses. Perhaps a better way would be "I'd like to understand why age discrimination exists in tech. What motivates companies to do this?"
- clnq 3y agoMy earlier words weren't liked and got downvotes fast. But, we need to look at all sides of tough topics to really understand them. This includes the best reasons people give for discrimination, even if they go against what's moral, ethical, or legal. We can only substantially argue against these views if we first understand them. Even though age discrimination is very often illegal and probably always (maybe depends on the culture) unjust, there might be some sense to it from a managers point of view. For example, people with more experience often demand higher salaries. This is something we can't overlook when we're trying to reason about the ageism bias. And we can't really combat it substantially and directly if we ignore the strongest arguments for it — steel-manning is a very powerful tool in changing minds. I probably could have phrased it better.
- xtiansimon 3y ago> “I think the way you've phrased your question will invite negative responses.” And? If you see nuance, and rephrasing the OP can help make your response on point, then please share. Otherwise I find this browbeating and condescending. You’re just telling us how you want to hear the argument framed. Geez.
- yaseer 3y agoPerhaps the tone and context were not clear. > And? A set of negative responses tends to make the quality of Internet discussions deteriorate. I thought there was an interesting discussion to be had, so suggested an alternative phrasing.
- christophilus 3y agoYeah. I’m over 40, and I expect a high salary due to my productivity and experience. You could replace me with 3 folks from around the world, and it’s arguably a reasonable business decision.
- cmrdporcupine 3y agoIt's maybe only a "good" business decision if you have very short-termist thinking.
- clnq 3y agoYes, the institutional knowledge lost would cost more than the difference of that salary a year in most cases. But companies definitely do not value institutional knowledge as much as black-on-white figures on their financial statements. Short-termist, but a lot of business decisions are very short-termist, so I think the argument is good. Someone might think that.
- Mistletoe 3y agoUnfortunately most companies, shareholders, and execs care about the stock performance the next quarter and that’s about it. A neat compensation package for a CEO would be “What will the stock price be in a decade?” But you could never get a CEO to commit for a decade because they know like we all do that job hopping is how you make more money and get raises. The system is messed up all the way down. https://hbr.org/2019/11/the-ceo-life-cycle https://hbr.org/2019/11/the-ceo-life-cycle > For instance, when we asked CEOs about the ideal tenure for the role, many mentioned the widely touted seven-year average. When we surveyed directors, they said that CEOs generally should leave the job after 9.5 years—a point at which, many believe, performance typically plateaus. Why these expectations? No one has a compelling or evidence-based answer. They are simply conventional wisdom. The graph in that linked article is particularly telling as it shows performance increases all come after ten years.
- dangwhy 3y agoDo CEO's really job hop though? I am looking at top tech companies and they seem to have been there for a long time and none of them are job hoppers.
- cmrdporcupine 3y agoI honestly think it's mostly a set of connected biases along the lines of: a) "Progressivism", a bias towards thinking of "good tech" as the "novel/new tech". Or believing that success rides on quick adoption of new things, along with b) A bias that believes that that the older you are the less likely that you have engaged with new novel tech. c) A bias that believes that if you're still "just" an engineer and not management, there must be something wrong with you. Younger staff will accept "grey hairs" at the executive level but sometimes be bewildered by it at the individual contributor level. d) A bias against more expensive talent generally. e) A bias (among some) against talent that cannot be easily manipulated f) A bias (among some) against people with children/families/mortgages/elderly parents, etc. that might "slow things down." (This one is maybe the trickiest because it can express itself in ways that are not at all intended to be malicious but end up with bad effects. E.g. organizing team offsites and gatherings which are hard for parents or committed spouses to engage with because of their obligations. And the older we get the more likely we are to have such obligations) Most people who hold these biases hold them unconsciously. I personally look (or sometimes act) younger than I am, and get "surprise" from people when I tell them my age. Likely because they have bias about what that age should/could imply. Age bias is mostly "correctable" in most decent human beings. Where it gets malicious is when it gets in the hands of management who have a cost focus.
- clnq 3y agoThank you for taking my question seriously. These are all illuminating points. To continue on point e... I also think that older people tend to be more difficult to manipulate with long work hours, unreasonable deadlines, or incompetent management. With enough experience, we learn to recognise these things and we set some boundaries. Someone might dislike that and say that the more experienced people don't fit into the corporate hustle culture, without examining the reasons in-depth.
- eitally 3y agoThis is the sort of statement that isn't wrong, but which illustrates the echo chamber that is HN (or tech in general). There are plenty of industries and even more companies where "hustle culture" just isn't a thing, and management don't devolve into a default of employee abuse. Also, besides growing a spine and looking out for #1 increasingly as they gain age & experience, most people also start realizing that their time is finite and they need to actively prioritize where to spend it. For many, this realization comes with relationships & children, something most white collar professionals don't experience until they've been working 7-10 years.
- op00to 3y agoOld people are more expensive. That’s it.
- gedy 3y ago[flagged]
- tyingq 3y agoThey are also less likely to cave into pressure to work long hours. Or to buy into silly kool-aid movements.
- wongarsu 3y agoThey will also push back on your plan to replace the existing system with a microservice data-lake using event sourcing with Kafka, using a service-mesh on a Kubernetes cluster on the distributed cloud. Either because they haven't kept up with the latest tech, or because they have the experience to tell beforehand that your vanity project may sound cooler but increases complexity for no real technical advantage.
- belter 3y agoThe above post should not be down voted. Although as others mentioned, it could have been phrased differently, it was clear the post was about understanding the reasons behind age discrimination, not about endorsing or justifying it.
- Dalewyn 3y agoAge is an undeniable factor when it comes to learning new things[1], whether any of us like it or not. Ultimately, the question to you as an employee is whether you can do the work demanded of you. While it would be nice for everyone to be judged on an individual basis, I can understand there needs to be some level of generalization to maintain a certain level of efficiency within an organization. [1]: https://news.ycombinator.com/item?id=35595846 https://news.ycombinator.com/item?id=35595846
- belter 3y agoYou are providing a case of airline pilot training. It's hard to believe a experienced and high quality software engineer with 20 year of experiences, will not learn a new programming language faster and better, than a freshly graduated 22 year old child :-)
- HPsquared 3y agoNeuroplasticity is supposedly on a constant decline after childhood. Also processing speed, working memory, long-term memory. Offset by increased experience, of course.
- BossingAround 3y agoWhile true that neuroplasticity is probably declining after reaching age of 20-21 (?), that doesn't mean much. People can learn new things during their lifetime. It's not like when you have a 21-year old, they are learning at the speed of light, whereas a 40-year old takes ages to learn the same tech.
- nibbleshifter 3y agoThat can be somewhat fixed. Not if your employer does drug tests though...
- serial_dev 3y agoOne argument (though, very cynical) for getting rid of 40+ yo people in tech: they cannot be so easily manipulated into blindly following company directives, and they are more likely to actually push back against bs rules/guidelines, whereas young people either don't know better, or they are more agreeable.