The articles assumption that only knowledge workers experienced job-as-identity seems suspect to me. Like if you look at deindustrialised regions in the UK and Ireland, or former auto workers in...
The articles assumption that only knowledge workers experienced job-as-identity seems suspect to me. Like if you look at deindustrialised regions in the UK and Ireland, or former auto workers in Detroit, it’s not like these are generations that have died off and there was a lot of this career grief if you talk to them. And sure some of this may be from the material effects of these once high paying jobs going away, or for what that means for their lives and self image outside work, but plenty of it was also that they had their identities in their jobs too.
Yeah, the whole article fell flat on its face because of that presupposition. It completely shatters the central premise that this is somehow a new phenomenon. You're telling me that the artisan...
Yeah, the whole article fell flat on its face because of that presupposition. It completely shatters the central premise that this is somehow a new phenomenon.
You're telling me that the artisan woodworker who created bespoke furniture with careful precision and finely honed skills from decades of experience doesn't use his job as part of his identity, but Kelly, the accounts receivable clerk at Allstate who took the job because it was the first position to call her back does? Kelly takes being laid off because of the new AI accounting tool harder than the woodworker did by being replaced by IKEA?
I understand that a blog post analyzing a bunch of Reddit threads about tech workers dealing with existential dread might not hold up to much scrutiny. But as someone that is going through that...
I understand that a blog post analyzing a bunch of Reddit threads about tech workers dealing with existential dread might not hold up to much scrutiny. But as someone that is going through that same process, I thought the post had some interesting but unpolished ideas that other people may find interesting.
We are going through a large culling at work atm and this did give me a lot of insight, so thanks for posting. I also think it misframes the issue - it's bigger than AI and knowledge work. For me,...
We are going through a large culling at work atm and this did give me a lot of insight, so thanks for posting.
I also think it misframes the issue - it's bigger than AI and knowledge work. For me, for example, a lot of what drives that grief is a major shift in business priorities. I'm not only losing half my colleagues but the work I do is going to have far less relevance. A lot of what I do just doesn't seem to have any point anymore.
Im glad you posted this, it was interesting to read, even though I agree one is it's central arguments falls flat. It has some good points, but i didn't see any that were particularly new or...
Im glad you posted this, it was interesting to read, even though I agree one is it's central arguments falls flat.
It has some good points, but i didn't see any that were particularly new or novel. In particular I agree with it's assertions that the AI change is different because of the shear speed of change, and because the benefits have not yet proven themselves to be concrete.
Haven’t finished reading it, but I like that you posted it. I think we’re possibly in a similar boat. I got into programming because I liked solving problems and seeing the tangible effects of...
Haven’t finished reading it, but I like that you posted it. I think we’re possibly in a similar boat.
I got into programming because I liked solving problems and seeing the tangible effects of code I wrote doing things. Using LLMs for this completely changes how the job feels, in a way where I’m no longer doing what I enjoy, and I’m not even using LLMs all that much.
I certainly don’t think I’ve tied my identity to “Software Engineering” more than a carpenter may have tied their identity to Carpentry (some people I follow like referring to “Software Craftsmanship”, and I think anything you’d refer to as craftsmanship implies a level of care and identity), but I think the article still holds water if you ignore the bit where it claims to only apply to “knowledge work”.
I think grief can come in any job where you’ve tied your identity or pride to the work you do, and then the way that work is done fundamentally shifts out from under you.
Though, there may be some difference in that I doubt there will end up being a market for “handcrafted code” (if the LLM proposition holds and they truly become as widespread as their proponents hold they will) in the way there is a market still for good, handcrafted furniture.
I think they're will be a market for it. Just as there is for hand crafted furniture, it's just utterly dwarfed in comparison to how much factory made furniture is consumed. And I think it will be...
Though, there may be some difference in that I doubt there will end up being a market for “handcrafted code”
I think they're will be a market for it. Just as there is for hand crafted furniture, it's just utterly dwarfed in comparison to how much factory made furniture is consumed.
And I think it will be less about "every character was typed by a human" and more about achieving a certain level of quality. Quality is admittedly subjective, but people who know a craft know when they see it. If I'm looking for some apps to amuse my cats, the code quality is not something I really care about. However the software running the radiation treatment machine for my cancer, that I care about a lot.
Probably true but it's interesting that what "hand crafted" furniture means has shifted over the years to include a lot of machinery and industrial processes and, I think you could argue, is still...
Though, there may be some difference in that I doubt there will end up being a market for “handcrafted code” (if the LLM proposition holds and they truly become as widespread as their proponents hold they will) in the way there is a market still for good, handcrafted furniture.
Probably true but it's interesting that what "hand crafted" furniture means has shifted over the years to include a lot of machinery and industrial processes and, I think you could argue, is still shifting. The Luddites would probably be very unhappy about it.
I assume this has not itself been written by AI, because that would be too ironic. But it is distractingly full of section headings with colons in them, back-references to ideas that were not...
I assume this has not itself been written by AI, because that would be too ironic.
But it is distractingly full of section headings with colons in them, back-references to ideas that were not introduced as if they would be reference targets later ("the backlash thread"), and weirdly dense phrasing ("The adaptability prescription contains an unexamined assumption"), which make writing bad in a way I usually associate with it having been spit out by a text generator. It also generally feels like it is trying to act as if it is at a stratospherically-high level of discourse and is getting at some profound truth for several pages, when the actual ideas presented (that AI-replaced-my-career grief should be recognized as a thing) don't warrant that treatment, another thing text generators like to do.
Does this author just write like that? Is this the place the models are getting their weird style from?
As for the substance of the article, it seems to take it for granted that people ostensibly laid off because of AI being able to replace them actually were, and that people being demanded to use AI tools to do their jobs will not immediately be out-competed by people not stuck trying to use those tools. Until something shows up in those stubbornly flat productivity numbers, we have little evidence that management won't come crying back to these workers.
Commenting on one small part of the article. Solow's paradox is a thing - but there's a couple reasons for it. One is that large firms will lay people off in order to drive the organization to...
Commenting on one small part of the article. Solow's paradox is a thing - but there's a couple reasons for it.
One is that large firms will lay people off in order to drive the organization to adopt technology. Organizational change is really hard and especially hard in large organizations. Executives will preemptively do lay offs if they think the technology is there in order to force the organization to start changing.
Second, it takes time for firms to adapt to new technology in terms of sorting out the ecosystem of capital, labor, and system change to occur. Part of the 90s tech bubble was due to not having the iterative startup processes and venture capital model that we take for granted. The same is showing to be true for AI - it's taking time for capital - VCs aren't used to hard assets - labor - it's taking a long time for workers to adapt - and technology - it's taking time for the technology to be actually useful for enterprises.
Third, it takes time for new firms to replace older, slower firms. Some firms will utterly fail to adapt - see Toys-R-Us, others end up in stagnant zombie state. New firms take time to develop and to replace the larger firms that don't navigate the transition.
The articles assumption that only knowledge workers experienced job-as-identity seems suspect to me. Like if you look at deindustrialised regions in the UK and Ireland, or former auto workers in Detroit, it’s not like these are generations that have died off and there was a lot of this career grief if you talk to them. And sure some of this may be from the material effects of these once high paying jobs going away, or for what that means for their lives and self image outside work, but plenty of it was also that they had their identities in their jobs too.
Yeah, the whole article fell flat on its face because of that presupposition. It completely shatters the central premise that this is somehow a new phenomenon.
You're telling me that the artisan woodworker who created bespoke furniture with careful precision and finely honed skills from decades of experience doesn't use his job as part of his identity, but Kelly, the accounts receivable clerk at Allstate who took the job because it was the first position to call her back does? Kelly takes being laid off because of the new AI accounting tool harder than the woodworker did by being replaced by IKEA?
Nah, sorry. That angle doesn't hold.
I understand that a blog post analyzing a bunch of Reddit threads about tech workers dealing with existential dread might not hold up to much scrutiny. But as someone that is going through that same process, I thought the post had some interesting but unpolished ideas that other people may find interesting.
We are going through a large culling at work atm and this did give me a lot of insight, so thanks for posting.
I also think it misframes the issue - it's bigger than AI and knowledge work. For me, for example, a lot of what drives that grief is a major shift in business priorities. I'm not only losing half my colleagues but the work I do is going to have far less relevance. A lot of what I do just doesn't seem to have any point anymore.
Im glad you posted this, it was interesting to read, even though I agree one is it's central arguments falls flat.
It has some good points, but i didn't see any that were particularly new or novel. In particular I agree with it's assertions that the AI change is different because of the shear speed of change, and because the benefits have not yet proven themselves to be concrete.
Haven’t finished reading it, but I like that you posted it. I think we’re possibly in a similar boat.
I got into programming because I liked solving problems and seeing the tangible effects of code I wrote doing things. Using LLMs for this completely changes how the job feels, in a way where I’m no longer doing what I enjoy, and I’m not even using LLMs all that much.
I certainly don’t think I’ve tied my identity to “Software Engineering” more than a carpenter may have tied their identity to Carpentry (some people I follow like referring to “Software Craftsmanship”, and I think anything you’d refer to as craftsmanship implies a level of care and identity), but I think the article still holds water if you ignore the bit where it claims to only apply to “knowledge work”.
I think grief can come in any job where you’ve tied your identity or pride to the work you do, and then the way that work is done fundamentally shifts out from under you.
Though, there may be some difference in that I doubt there will end up being a market for “handcrafted code” (if the LLM proposition holds and they truly become as widespread as their proponents hold they will) in the way there is a market still for good, handcrafted furniture.
I think they're will be a market for it. Just as there is for hand crafted furniture, it's just utterly dwarfed in comparison to how much factory made furniture is consumed.
And I think it will be less about "every character was typed by a human" and more about achieving a certain level of quality. Quality is admittedly subjective, but people who know a craft know when they see it. If I'm looking for some apps to amuse my cats, the code quality is not something I really care about. However the software running the radiation treatment machine for my cancer, that I care about a lot.
Probably true but it's interesting that what "hand crafted" furniture means has shifted over the years to include a lot of machinery and industrial processes and, I think you could argue, is still shifting. The Luddites would probably be very unhappy about it.
I assume this has not itself been written by AI, because that would be too ironic.
But it is distractingly full of section headings with colons in them, back-references to ideas that were not introduced as if they would be reference targets later ("the backlash thread"), and weirdly dense phrasing ("The adaptability prescription contains an unexamined assumption"), which make writing bad in a way I usually associate with it having been spit out by a text generator. It also generally feels like it is trying to act as if it is at a stratospherically-high level of discourse and is getting at some profound truth for several pages, when the actual ideas presented (that AI-replaced-my-career grief should be recognized as a thing) don't warrant that treatment, another thing text generators like to do.
Does this author just write like that? Is this the place the models are getting their weird style from?
As for the substance of the article, it seems to take it for granted that people ostensibly laid off because of AI being able to replace them actually were, and that people being demanded to use AI tools to do their jobs will not immediately be out-competed by people not stuck trying to use those tools. Until something shows up in those stubbornly flat productivity numbers, we have little evidence that management won't come crying back to these workers.
Commenting on one small part of the article. Solow's paradox is a thing - but there's a couple reasons for it.
One is that large firms will lay people off in order to drive the organization to adopt technology. Organizational change is really hard and especially hard in large organizations. Executives will preemptively do lay offs if they think the technology is there in order to force the organization to start changing.
Second, it takes time for firms to adapt to new technology in terms of sorting out the ecosystem of capital, labor, and system change to occur. Part of the 90s tech bubble was due to not having the iterative startup processes and venture capital model that we take for granted. The same is showing to be true for AI - it's taking time for capital - VCs aren't used to hard assets - labor - it's taking a long time for workers to adapt - and technology - it's taking time for the technology to be actually useful for enterprises.
Third, it takes time for new firms to replace older, slower firms. Some firms will utterly fail to adapt - see Toys-R-Us, others end up in stagnant zombie state. New firms take time to develop and to replace the larger firms that don't navigate the transition.