9 ms·

Rolling the ladder up behind us

206 points by techknowlogick 1y ago

Refreeze5224 1y ago
This is the true problem with AI. It's with who owns it, and what they will inevitably use it for. Whether it can do cool stuff with code or equal a junior developer is irrelevant. What it can do is less important than what it will be used for. The owning class will use it to reduce payroll costs, which from their perspective is a cost center and always will be. If you're not an owner, then you have no control over the direction or use of AI. You are doomed to have your life disrupted and changed by it, with no input whatsoever. To quote the article, your six shillings a day can become six shillings a week, and you are left to just deal with it however you can. You are "free" to go find some other six shilling a week job. If you can. And if you think, "Oh, every technology is like this, it's always been this way", you are right. You have always been at the whims of the owning class, and barring a change towards economic democracy, where average people regain control over their lives, it likely always will be.
- dontlaugh 1y ago
  This was the Luddites’ position. Organised labour should one again take up this principled position. Join a trade union!
  - breakyerself 1y ago
    Software engineers needs a lot more organization
  - WalterBright 1y ago
    Do you really want a return to the days when "women's work" was spending every free moment spinning and weaving cloth by hand? When cloth was so valuable there was a profession called "rag pickers"? Where new clothes were rare, hand me downs were the usual, and people wore clothes until they disintegrated? And poor people made clothes out of flour sacks?
    - dontlaugh 1y ago
      That’s not what the Luddites wanted. They praised the technology itself and recognised it would save a lot of time. But they also recognised that they the workers wouldn’t reap any of those benefits, they’d just lose their jobs. Do you really think capitalists would choose to shorten the working day with no loss of pay as productivity increases? If so, you are incredibly naive.
      - WalterBright 1y ago
        The workers did reap the benefits. The Law of Supply and Demand ensures it. 1.1% of the US labor force works at minimum wage jobs. When wages are raised by the government, job losses happen. 16,000 to 36,000 people lost their jobs when California raised the minimum wage for fast food workers. > Do you really think capitalists would choose to shorten the working day with no loss of pay as productivity increases? During WW2, production needed to increase, so hours were increased to 60 hour weeks. The labor force, very patriotic, was all for this. Production increased for a few weeks, and then fell below what was produced in a 40 hour week. Companies wanting to maximize profits are aware of this effect. Also, if you track employee total compensation (not just wages) against productivity increases, the two lines form the same curve. The reason for this is the Law of Supply and Demand pushes those two lines together. The more productive a worker is, the more they get paid, as such workers are more in demand.
        specialist 1y ago
        Now include profit (surplus) in your analysis. Wage grow has not kept pace with corporate profitability. Why?
        WalterBright 1y ago
        Because the correct comparison would be with total compensation, of which wages are only a portion. Total compensation is the costs to an employer to hire some one. It includes: 1. contributions to retirement plans 2. so-called "employer paid" social security contributions 3. sick and paid vacation days 4. employer paid health insurance 5. stock plans 6. 401k plans 7. other payroll taxes Last time I checked, these often added 40% or more over the takehome pay. Also, the pay is based on the value the employee as an individual provides, not the value the company as a whole created. Just like the compensation for pro football players varies in a single team.
  - Cthulhu_ 1y ago
    Even unions are up against it though. in the Netherlands, there have been multiple strikes from e.g. the train operators; they demand a 10% wage increase, after the train company downsized a lot during the pandemic, but who were then unable to upsize again as traffic resumed and who even now don't have the same amount of travelers as they did in 2020. But they're not profitable. They've made a loss for five years in a row, requiring government subsidies to stay operating (because them going bankrupt would make big parts of the country grind to a halt). Ticket prices are so high that it's barely cheaper than driving, and as soon as you travel with two or more people it's cheaper to just own a car. (Or even rent one; I traveled to the other side of the country once for a concert, my own car was in the garage. Did the math, it was cheaper and more convenient (door-to-door) to rent a car, pay the surcharge for long distance, pay the €35 in fuel, etc than it was to buy return tickets for three people) The answer (I think, I'm no economist / politician / etc) isn't unionising and demanding better treatment, because the limits of the capitalist system they operate in have been reached. The answer is to stop trying to make it profitable. Re-privatize it: trains and public transit are a huge and hugely important nationwide economic driver, an essential service that capitalism can't be trusted with because the owners will try and get as much money out of it as possible, the employees get overworked and exploited and will shut the system down (as is their right) out of protest, and the people will have to deal with the consequences. Just in my small bubble, the train strikes led to people not going to the office, missing events, lunch orders that either had too much stock that had to be discarded or that were cancelled entirely, costing that company thousands in income, etc. Don't get me wrong, I'm not anti-union. I'm anti-capitalist though.
    - dontlaugh 1y ago
      You’ll often find trade unionists are also anti-capitalist, some are even socialists or communists.
- WalterBright 1y ago
  > The owning class will use it to reduce payroll costs Of course. That's been going on since the invention of the plow. That's why today we can do more interesting things than turn over the earth with a pointy stick all day every day. > economic democracy, where average people regain control over their lives History shows us that this inevitably means people lose all control over their lives, because the state will make your decisions for you and assign you your job. For example, let's say the color of cars produce by car companies is determined by democracy. 59% vote for the cars to be green. And if you want a red car? Too bad. What if you want a 4 seat car? No dice, 53% voted for 2 seaters to be made. What if you didn't want a car stereo? You're stuck paying for it anyway, as 73% voted for it.
  - schmidtleonard 1y ago
    I have a whole list of things I'd like from my car that the market does not provide because it is more profitable not to. Why do I get the feeling that instead of seeing this as a horror story you would scold me for unreasonable expectations, even though it is the identical mirror form of the horror story you just told?
    - derektank 1y ago
      My take would be to scold you to start your own company that provides the features you want from your car if it's not already being provided by the marketplace
      - schmidtleonard 1y ago
        I'd have more luck petitioning the Central Car Design Committee because > it is more profitable not to ...and if anyone thinks this is ridiculous, I'd ask them if they are for or against repealing all of our current motor vehicle regulations. If "for," they have admitted to being a hopeless libertarian, and if "against," they have acknowledged that important reasonable features can be incompatible with the profit motive of a free market and it no longer seems so strange that I might have a list which is more of the same.
- Cthulhu_ 1y ago
  > The owning class will use it to reduce payroll costs Things are cyclic, nothing new; the previous big scare was (is?) outsourcing, where for the same price as one developer in western / northern Europe or SF you can hire five from eastern Europe or India. But that hasn't affected employability of the one developer, as far as I'm aware. I'm not even thinking of skill level, I'm sure that's comparable (but honestly I don't know / care enough), but both outsourcing and AI require the same things - requirements. I've grown up in this country (the Netherlands) and automatically have intrinsic knowledge of e.g. government, taxes, the energy sector, transportation, etc, so much that I'm not even consciously aware of a lot of things I know. If you spend a LOT of time and effort, you could - eventually - break that down into requirements and work orders or whatever that someone else could process. But it's much more efficient to do it yourself or just hire someone from around here.
  - ge96 1y ago
    > hasn't affected employability of the one developer I wonder, combo that r&d rule, general economics, saturation in the field The norm for new grads seems to be "apply at least 2000 times, get 1 job"
    - tangjurine 1y ago
      the one developer one dev to rule them all, and in darkness bind them sounds like a 10x dev to me
  - worik 1y ago
    > the previous big scare was (is?) outsourcing, I was replaced by outsourcing Seriously. Fuck off
  - bevr1337 1y ago
    > Things are cyclic, nothing new Must be a convenient worldview
  - RainyDayTmrw 1y ago
    I know of two major examples - one personal and one word-of-mouth - where a major US company successfully took their Covid era remote work knowledge, applied it to outsourcing, and successfully replaced most of their US onshore developers with offshore.
- blooalien 1y ago
  This is by far my biggest concern about "A.I." and "smart" robots... Not the technology itself, but what the "ruling class" intend to do with it / how they intend to use it. Their primary concern is not "worker productivity" these days. It's "How can I replace the maximum number of workers (ideally all of them) so that I can keep most / all of the profits / benefits for myself (and the shareholders)?" It's always been about profit, but they now finally see a potential to entirely rid themselves of "those pesky poors" and their annoying paychecks once and for all.
- whall6 1y ago
  The owning class is different today than it was in Ned’s time. Who owns MSFT? You, probably, through your 401k. Nobody is stopping you from creating a coalition of similarly minded shareholders and effecting change. Unless you stuff your cash under your mattress, you are the owning class.
  - worik 1y ago
    > Nobody is stopping you from creating a coalition of similarly minded shareholders Sweet lies.
  - like_any_other 1y ago
    > Who owns MSFT? You, probably, through your 401k. All or nothing fallacy. People have VASTLY different degrees of ownership, and this is reflected in their degree of control. > Nobody is stopping you from creating a coalition of similarly minded shareholders Such coalitions already exist. They have names like BlackRock, Vanguard, and Fidelity Investments [1]. Good luck competing with them. Especially if you abdicate your role in democracy, don't agitate for legal change, and restrict yourself to only market-based means of change (not BlackRock and ilk though - they use every lever available to them, they don't handicap themselves). [1] https://en.wikipedia.org/wiki/List_of_asset_management_firms https://en.wikipedia.org/wiki/List_of_asset_management_firms
    - whall6 1y ago
      This isn’t a false dichotomy because I’m not even representing two separate choices. I get that you alone have probably an insignificant amount of control, but forming a coalition would allow you to make change. I’m aware that there are asset managers. Whose assets do you think they’re managing? Yours! Do you own VOO? Awesome, that’s why Vanguard owns 10% of almost every large cap company. Take control of those shares and go make change. Look up John Chevedden. He owns a minimal amount of shares in companies where he causes huge changes. They call him the “corporate gadfly” [1] Don’t be lured into thinking that you have to invest your money into a predesigned portfolio because it maximizes returns. If you care about control then maximize control. You already have the power! [1] https://www.sec.gov/comments/s7-23-19/s72319-6733874-207512.pdf https://www.sec.gov/comments/s7-23-19/s72319-6733874-207512....
kevmo314 1y ago
> Actually, what are we going to do when everyone that cares about the craft of software ages out, burns out, or escapes the industry because of the ownership class setting unrealistic expectations on people? Nothing, I guess? There's an implicit assumption that software written by humans is a necessity. If the future finds that software written by computers is more profitable then that's just what it is. The universe doesn't owe us value on human-written software. Even food is not able to escape this hole: it's not profitable to manually cultivate food anymore if you wanted to do that as a career.
- th0ma5 1y ago
  This also assumes that non human written code will be of any use to humans and no one has shown that to be possible, it is all humans patching it up so far.
- ChrisMarshallNY 1y ago
  Well, as someone who considers themselves to be a "software craftsman," I have come to the conclusion that the work I do will never be valued, and will always be considered "too expensive." Since I work for free, that's not an issue for me, but that's economically unfeasible for most folks. The issue with an industry awash with cheap dross, is that it becomes prohibitively expensive to produce high Quality stuff. Anyone that tries, will get driven out of business. Some clever folks will figure out how to do "slightly better" stuff, and charge more for it, but a good way to go out of business, is to focus on Quality as a principal axis. That's basic market dynamics. It is what it is, and is neither evil, nor good. It does mean that only "niche" craftsmen, like myself, will produce anything of decent Quality, but will be unable to do so at scale, because we can't get a team together, large enough to do big things. I guess the saddest thing, is that I have really wanted to help teach my techniques to others, but have found that no one wants to learn, so I gave up on that, many years ago.
  - burlesona 1y ago
    > The issue with an industry awash with cheap dross, is that it becomes prohibitively expensive to produce high Quality stuff. This seems to be one of the brutal truths of the modern world, and as far as I can tell it applies to everything. There's always a race to the bottom to make everything as cheaply as possible, and the further the industry goes down that "cheapness" scale, the more "quality" loses market share, the more expensive "quality" must be in order to operate at all, and finally things that used to be just "normal" and not too expensive are now luxury goods. Consider textiles, carpentry, masonry, machine tooling, appliances, etc. etc. This doesn't feel like a good outcome, but I'm not sure there's anything that can be done about it.
WarOnPrivacy 1y ago
industry only ever seems to want to hire people with the word Senior in their title. They almost never want to create people with the word Senior in their title. I suggest that the first ladder that got pulled up is the one on the ground. No one wants to train new entrants to the field. Not training junior workers seems like a natural extension to that.
- whatshisface 1y ago
  I guess universities will eventually backfill that gap with training that focuses on senior skills. Eventually.
  - thmsths 1y ago
    We already require juniors to go through a 4 years university degree. It takes a fair bit of time of real world work to get to the senior level. So unless we expect people to do another 5 years of schooling, I am not sure how this will happen (and even in that scenario I believe there is a difference between hands on, on the job experience and classroom experience).
    - jdee 1y ago
      it happened with architecture degrees in the uk. it went from 3 years to 7 years, as the skill levels and implicit knowledge required increased over time.
- ta1243 1y ago
  In centuries past apprentices would pay their junior positions, in time picking up paid work as the progressed to senior, then eventually taking on apprentices of their own (and be paid)
- teeray 1y ago
  It’s a Volunteer’s Dilemma: why train juniors when you can hire the juniors other companies trained to become seniors?
ChrisMarshallNY 1y ago
> Vibe coding is payday loans for technical debt Stealing it...
schmidtleonard 1y ago
> I don't fear the power loom. I fear the profit expectations of the factory owners.
deadbabe 1y ago
Vibe coding is not going to go away. Just as alchemists always dreamed of turning lead into gold, there is a dream that people have had for decades of writing less and less code: Open source. Scaffolds. Code generators. No-code tools. Coding Agent LLMs. If vibe coding goes away, it will be because it has been replaced with something even more hands off. Perhaps vibe prompting. Or even code farming: just have AI agents constantly build totally random shit and go through and see if any of it is useful for something. Maybe even take a genetic approach where you score the fitness of various software for a certain solution and AI will cross mutate the genes. This will be a change of that rivals our shift from hunter/gatherers to the agricultural revolution.
- blibble 1y ago
  > Or even code farming: just have AI agents constantly build totally random shit and go through and see if any of it is useful for something. so you're saying the venture capitalists will soon be out of a job? what a shame
  - deadbabe 1y ago
    Venture capitalists are basically startup farmers.
mrtksn 1y ago
I knew what was this all about the moment I glimpsed at the title on the front page, my neural nets must have been trained to output that from the few tokens I guess, which makes me think maybe all this AI stuff isn't that different from what's happening in real brains. I bet others too had similar experience. So the AI thing is happening, maybe its not with this particular tech we have today but they are on to something and we probably better embrace it. Humans rolled the ladder up behind on so many things. Very few people will survive the planet Earth without the tools and abstractions we built over the many Millenia's and that's how we all live like kings. Any misery out there in the world is a result of our inability to manage it, not because the resources are scarce. The Luddites, as explained in the article were indeed about the way tech is being adopted. All that tech eventually reached every corner of the world and not every place had Luddites. Programming computers by hand is shitty anyway, good riddance. Finally we are about to have machines that can be programmed without thinking about the intricacies of programming that have nothing to do with the thing we want to achieve. All the tech left behind still does have aficionados, maybe in 10 years we can watch someone program a computer using Java on the Primitive Technology channel.
paxys 1y ago
The part that is always skipped when making the clothing analogy is that for most of human history good clothes were a luxury. At the time of the industrial revolution it was common for a family to spend 15-25% of their annual income just on clothing. A single shirt from that era would cost the equivalent of £2,000 today when considering material and labor. New clothes were something you may be able to afford once a year, if at all. Like the article says, weavers were the equivalent of PhD scientists who could charge whatever they wanted for their skills (and they would cater pretty much exclusively to kings and the 0.1% of the time). Great if you're the weaver, not so for everyone else. So the industrial revolution wasn't a conspiracy to put down the lower classes. The lower classes were in fact the biggest benefactors of the industrial revolution. If software can go the same way, I'd say good riddance. The profession has always kept free from gatekeepers, and that's a good thing.
- stego-tech 1y ago
  > The part that is always skipped when making the clothing analogy is that for most of human history good clothes were a luxury. Literally what OP discusses in their text, right in the first part. Go RTFA. > and they would cater pretty much exclusively to kings and the 0.1% of the time Oh yeah, I totally remember reading about how people in pre-modern civilizations were almost always semi-clad or fully nude due to the expense of clothing. Oh, wait, no I don't, because people still bought clothing and wore it regularly. They just also had economies around mending clothing, updating it, tailoring it, altering it, reusing and recycling it. Rather than building an economic system of destruction for the sake of a handful of profiteers, it was an economy of artisans who provided a staple resource at reasonable rates and quality to support themselves. Because clothing was often tailor-made rather than ready-to-wear, people took care of it - and themselves - for longer periods of time. Techniques were used to keep articles sturdy for longer, rather than disposable machine stitches that fall apart in a washing machine. Experts and artisans are not "gatekeepers", they are skilled craftspeople worthy of respect and deserving of compensation for their skills. To demand anyone be able to do anything of any complexity is to demand a complete elimination of anything that differentiates humans from one another, to create a homogenous mass of genetics with no incentive to grow and evolve. Nobody is "gatekeeping" software developers, or Doctors, or plumbers, or weavers, or artists. Those all take skill, and people with a suitable level of skill can easily pass good credentials checks and tests.
tptacek 1y ago
As a software security person, I don't think the security objections to LLMs are going to pan out. I think LLMs are going to be a strong net positive for security: * The tooling and integration stuff people complain about now ("the S in MCP") isn't really load-bearing yet, and a cottage industry of professional services and product work will go into giving it the same overcomplicated IAM guardrails everything else has; today, though, you just do security at a higher or lower level. * LLM code generation is better at implementing rote best-practices and isn't incentivized to take shortcuts (in fact, it has some of the opposite incentives, to the consternation of programmers like me who prize DRY-ness). These shortcuts are where most security bugs live. * LLMs can analyze code far faster than any human can, and vulnerabilities that can be discovered through pure pattern matching --- which is most vulnerabilities --- will be easy pickings. We've already had a post here with someone using o4 to find new remote kernel vulnerabilities, and that's a level of vuln research that is way, way more hardcore than what line-of-business software ordinarily sees. * LLMs enable instrumentation and tooling that were cost-prohibitive previously: model checking, semantic grepping, static analysis. These tools all exist and work today, but very few projects seriously use them because keeping all the specs and definitions up to date and resolving all the warnings is too much time for not enough payoff. LLMs don't have that problem. LLM-generated code (and LLM tooling) will inevitably create security vulnerabilities. We have not invented a way to create bug-free code; would have been big if true! Opponents of industry LLM use will point to these vulnerabilities and go "see, told you so". But each year we continue using these tools, I think the security argument is going to look weaker and weaker. If I had to make a bet, I'd say it ceases being colorable within 3 years.
- whatshisface 1y ago
  There are also non-llm advances in testing related to AI, like RL fuzzers.
- threetonesun 1y ago
  I'm not terribly worried about code generated security vulnerabilities, but point 3 feels like a cat and mouse game that most companies won't have the resources to stay on top of, so they'll have to outsource it to one of the existing cloud or AI providers. Maybe that's a reality even without AI but it feels like we're heading towards full on extortion from about 4 major companies. Also I don't think you covered my biggest concern with LLM security, a company making an Amazon basics version of your business model and claiming "AI did it". I'm 50/50 on that one though, it's also possible everyone things with AI you can go full NIH syndrome and take back all the software that we've handed off to various SAAS providers.
Cthulhu_ 1y ago
> We're going to run out of people with the word "Senior" in their title Probably not, but job title inflation has made it so that apparently 5 years of experience is enough to be given a "senior" title. I've got like 15 but still feel like a medior at best. Yeah this is humblebragging, whatever it's a throwaway internet comment.
- vultour 1y ago
  This is a funny comment because startups have been churning out sub-3 year "seniors" for years. If you're not a principal engineer after 5 years what are you even doing?
blakesterz 1y ago
There's a photo on that page with this description: "A picture of two patches of wild grass bifurcated by a retaining pond" I was just thinking how I'd describe that different, and how many different ways it could be described.
- Groxx 1y ago
  Perhaps even a thousand words worth of description is possible
  - riskable 1y ago
    A thousand tokens.
Havoc 1y ago
I've kinda seen this already play out in a completely different context. Outsourcing & audit world. It's sufficiently commoditized that you can efficiently outsource. If you're sending the first say 12-months worth of experience work to india then local gang straight out of college end up with a pretty tangible gap. There are key formative experiences missing. The programming equivalent of fighting with manual array memory management. Or shooting yourself in foot with a pointer. Worse they don't realise that they're missing pieces. By necessity you start them off at a higher level and their understanding ends up fuzzy for lack of better word. It's not their fault...to them it looks normal, to the previous gen it's all "how can you not know this" ...still the world keeps turning
stego-tech 1y ago
Xe's stuff is always excellent, and this is no different. What's disappointing is the amount of energy and effort wasted by people to come into the comments and espouse their desire to give up on addressing any of their concerns. "It's just how things are." "That's what generates profit." "If that's what the market demands, who are we to judge?" "Guess we do nothing." We've all heard it before. We get it, so many HN and Tech folks have thrown in the towel and given up on changing anything within their own lives, or within the lives of others. So many more are content just "going along with the ride", abdicating all responsibility to an imagined god in various forms ("invisible hand of the free market" being most common among our peers). For the rest of us who want to actually discuss the merits of the piece, we're sidelined by those who found reward in the path of libertarian ideals and seek to punish or discipline those of us who seek to make any sort of change or improvement beyond direct profit motives. Serious discussions of organization and direction are derailed in favor of blatant troll bait and bad-faith arguments. This, I think, is also what Xe is trying to highlight: those in the position to enact the most positive change are staunchly refusing to do so, which in turn exacerbates the harms caused. Too few people with too much power and too much capital believe themselves to be more knowledgeable and intelligent than anyone else, denigrating any thoughts that may come from the slime beneath them on the ladder. When Xe said "Rolling the ladder up behind us", they make it very clear who they're speaking to: > Look, CEOs, I'm one of you so I get it. We've seen the data teams suck up billions for decades and this is the only time that they can look like they're making a huge return on the investment. They're basically shouting that "the call is coming from inside the house", and that the only place - right now - to address these harms are by and from the people inside said house. Nobody is coming to save us, but history books are filled with examples of what happens to those classes, those societies, those civilizations for whom greed becomes all consuming, standards of living collapse, and meaningful progress stagnates. > Maybe the problem really is winner-take-all capitalism.
- mattgreenrocks 1y ago
  > We get it, so many HN and Tech folks have thrown in the towel and given up on changing anything within their own lives, or within the lives of others. It’s hilarious to read such defeatist takes on the current state of things on a site called Hacker News. The classical definition of a hacker was someone who wielded tech to circumvent power imbalances in whatever form. Now I fear we have plenty of technologists that are happy to use tech to further cement and centralize existing power structures. IMO, this has been brewing for the past ten years or so, where technologists wrongly believe they have an in to this power just because they work for a FAANG. You are labor. Where you work does not matter. So long as you need to work, you are labor. And there’s nothing wrong with that! Doctors are finding out this same lesson. The way out is to build ventures that don’t involve the capital class. Autonomy and independence are worthy goals to strive for.
Veedrac 1y ago
A defense of the Luddites against sewing machines was honestly not something I expected. Support for the Luddites in general, sure I can predict that, but the claim that sewing machines are bad, exploitative, produce "shitty" cloth, serve only the pursuit of profit...
- duskwuff 1y ago
  Power looms are not the same thing as sewing machines. Not even close.
  - Veedrac 1y ago
    Ah, you're right, thanks for the correction.
- int_19h 1y ago
  There's a lot of nonsense there, like: "They were skeptical that the shitty cloth that the power loom produced would be a net benefit to society because it meant that customers would inevitably have to buy their clothes over and over again, turning a one-time purchase into a subscription." This is such an obvious attempt to reframe history to fit the modern discourse. What Luddites cared about was their own jobs, not "net benefit to society".
djoldman 1y ago
There are a lot of issues brought up in this post, but I want to discuss one in particular: technological progress and its economic and societal ramifications. We can say that technological progress occurs when a new method is employed to deliver a product or service with some "more desirable" blend of qualities: it's created/delivered faster, cheaper, with a more desirable mix of resource inputs, and/or results in a more valuable/desired output. Sometimes it's quite obvious when a technology is superior to another as almost all the qualities of it are advantageous: it's made faster, less expensively, and the result is better with such a gap between it and the old way that there's just no denying that the new way is better. Sometimes the new technology is really a mix of qualities. Let's focus on the mix that generally gets the most attention: the new way is faster/cheaper but the output is not of a higher quality. Sometimes this new way of cheaper+faster but lower quality "wins out" and the consumer prefers it. And now the crux of it: why is it so common that discussion concerning these shifts is rooted in everything except the consumer? An imperfect and potentially flawed example: a bunch of consumers have decided that they'd rather pay less for a shirt or shoe that will fall apart faster than more for one that lasts longer. The old way of making a superior product still exists and some consumers still prefer it but most do not. Fundamentally, this is a shift that is rooted in the perception, true or not, of the consumer that the new way is more desirable. Some folks are not happy with the higher prices of the outputs of old ways of doing things or the effects that the new ways have on jobs, the economy, and/or society. Fundamentally, this has to do with consumer preference and that's where any blame should be meted out.
- duskwuff 1y ago
  > technological progress and its economic and societal ramifications I seem to recall that a mathematician wrote a paper about this around 1995 which got a lot of press attention. I'm not sure I agree with all of his conclusions, though. Nor his methods.
- wizzwizz4 1y ago
  > a bunch of consumers have decided that they'd rather pay less for a shirt or shoe that will fall apart faster than more for one that lasts longer They don't know it'll fall apart faster, to begin with… and then by the time they do, they've already spent enough of their clothes budget that they can't afford to buy proper cloth, and it's hard to find that nowadays anyway, so they'll just have to buy more of the low-quality stuff.
youworkwepay 1y ago
The value of vibe coding isn't that it writes good or sustainable code. It's that you can build a sufficiently non-shitty prototype of a concept as a non IT expert to validate the use case and secure proper assistance. From a change leadership perspective, walking in the door with a shitty prototype beats pitching vaporware every day of the week and twice on Sundays. And the fact amateurs can deliver (basic, crappy) "things" without budget accelerates growth. My first projects were 80% copied off Github and some intro tutorials. Know what? They still work. We banked seven figures off of them so far...
- felipeerias 1y ago
  The article hints at that when it describes vibe coding as “fancy UX”, but fails to connect the dots. Essentially, we now have a system that can turn a simple problem description into an interactive tool for that problem. Even if it still does so very imperfectly, it’s easy see already the beginnings of a powerful and empowering new paradigm.
- edmundsauto 1y ago
  In terms of risk, building a prototype and getting a quick win really de-risks a project. Smart decision makers' world is a game of risk - they have tolerance in some areas, less tolerance in others. If you can materialize a quick win from a prototype, it's significantly less risky that sight-unseen work. Coders often don't think in terms of a broad investment portfolio, but that's how I've seen good executives phrase things. AI makes it cheaper and easier to build that prototype - I've been loving it for my own projects, because of how quickly I can deliver that first software.
burnt-resistor 1y ago
3 things. 0. The people must take their governments back from corporate technofeudal overlords. 1. Society must shift away from lionizing unbounded capitalism. 2. Workers must form employee-owned co-ops for more stability and better morale and share profits of their labor more fairly rather than extracting maximum amounts for the ownership class that will fire them whenever it suits them.
secstate 1y ago
I have a relative who works in healthcare. The introduction of AI for medical notes has led her level+1 manager to ask folks to increase their patient numbers in response to the extra time saved not having to manually code charts. The initial response was some grumbling about unionization of the doctor-class. But now they're just kinda going along with it while also complaining loudly. Profit Über alles.
shae 1y ago
What if we train them and they leave? What if you don't and they stay?
- hackable_sand 1y ago
  Yup. It's highly abusive behavior.
worik 1y ago
> . I have a hoodie from AWS Re:Invent in 2022 that I'm going to have to throw out and replace because the sleeves are dying What is the use of the right to repair, if we do not have the imagination to use it? I patch my clothes
lubujackson 1y ago
This post is really all over the place and, I think, misses the main thing here: pretty much every advancement in automation across history has sacrificed some fidelity and craftsmanship. That's the cost of automation. Yet we have continually added more and more automation to our lives because the net value is dramatically higher for the bulk of our civilization. Yes, we lose (or pidgeon-hole) artisans and craftsmen. But even in his "woe the times" example of AI kludging together two game properties into a painting, where the lighting isn't right and furry tail is not ideal, etc., etc... That $500 image is something no one could have created 20 years ago with a computer. This isn't centuries of artisan skills we are talking about. What about the artisan who would have been paid $500 to paint it with actual paints on an actual canvas? You can mourn progress in every direction if you like, but I don't think it changes a single thing about the trajectory of where we are going. Now that AI is coming for my job as a programmer I'm not going to get twisted about it - my job didn't exist when I was a kid and it won't exist (certainly not in the same form) when my kid is an adult. And that's ok, it was a "job of opportunity". And that job has changed dramatically over 20 years. AI is just the continuation of that change. I started in SE because it was fun to make computers do stuff. Programming has only ever been a tool to achieve that goal, and often a frustrating one. AI is just a better tool. If the job becomes "a guy that talks to the AI" that will still be programming because at the end of the day the computer can't do anything without someone telling it what to do. ...until SkyNet.
- AngryData 1y ago
  I mean one of the main complaints of the luddites was decreasing cloth quality despite the looms originally being able to produce higher quality cloth comparable to their own traditional weaving techniques. Powered looms and the dozens of other loom improvements from the previous 100+ years didn't inherently produce poor quality cloth, they were adjusted to produce lower quality cloths after they were already in operationin factories in order to increase profit margins. Powered looms were originally just modified hand looms, not some radically different design from the ground up. The flying shuttle predates powered looms and the luddites.
- edmundsauto 1y ago
  > pretty much every advancement in automation across history has sacrificed some fidelity and craftsmanship. That's the cost of automation Can you help me think through how this looks when considering car manufacturing? Originally hand made cars were things of craftsmanship, but significantly lower quality and reliability than those produced with tiny tolerances on insane production lines. I'd take a modern car over a handmade one just about any day of the week.
egypturnash 1y ago
I'm a pro artist who does a lot of furry commissions and has been getting constant downvotes on this place for expressing my unhappiness ever since this massive abuse of "fair use" began, thanks for being on my side, Xe <3
worik 1y ago
Do we, or do we not, live in democratic societies? If we do then we must ensure that we get the social benefits from new technologies, and we suppress the harm. Does anyone still believe that "self interest" in the aggregate is what is best for society? I think it is clearly untrue So we need to take collective action. AI/LLMs are a "steam engine" moment. We can make life better for everyone. But we do need to change course, together, to get there
nyarlathotep_ 1y ago
The most pressing issue to me currently regarding LLMs is the blatant, flagrant change in tone of the "leadership/ownership" class towards employees. It comes off as little other than thinly veiled contempt for the people that actually make the business function. We've seen, most recently, Jassy come out and say "we're gonna need a lot less people", Zuckerberg doing his "replacing engineers" shtick on several occasions, all the commentary from the Anthropic CEO/Altman etc. Even if none of this comes to pass, I've never witnessed such open "threats" in my lifetime. I can't square this perspective with that of the LLM advocates that run to defend every legitimate criticism of LLM use--unless you're one of the aforementioned or in a similar position, why would you want to defend software that's primary purpose is to displace you? They are literally telling you that's why they're investing in it. I don't get it.
- cedws 1y ago
  It’s masks off for them, they were always sociopaths that see other people as a means to an end.
- int_19h 1y ago
  I want software to displace me when it comes to my day-to-day grind job, so that I can spend this time on something more interesting. But I want that to happen in a way that results in a Star Trek communist utopia, not a hellscape where 90% of humanity is "redundant" to the very few people who own all that automation and its outputs.
_zagj 1y ago
> Honestly though, the biggest impact I've seen across my friends has been what's happened to art commissions Given the author's eccentricities, when I read "art commissions," I immediately suspected they really meant "furry porn art commissions." Then a few paragraphs later: > A friend of mine runs an image board for furry art. He thought that people would use generative AI tools as a part of their workflows to make better works of art faster. He was wrong, it just led to people flooding the site with the results of "wolf girl with absolutely massive milkers showing her feet paws" from their favourite image generation tool in every fur color imaginable, then with different characters, then with different anatomical features. There was no artistic direction or study there. Just an endless flood of slop that was passable at best.
camgunz 1y ago
Thomas Ptacek dismissed SWE AI concerns as hypocrisy, and I actually think he understated the argument. This is exactly what we get for: - using BSD/MIT licenses instead of GPL licenses - accepting absolutely no liability for security holes or data loss - accepting absolutely no responsibility for poorly performing software that contributes to climate change - shipping 17,000 web frameworks, founding one more blockchain startup, founding one more wrapper service around an MIT open source project We're gonna turn around and be like "but what about the security issues" now? We made this bed.
- pdimitar 1y ago
  Who is "we"? A lot of "us" don't have the time to do programming as a hobby -- even if we started off that way. Many of "us" don't accept blanket grouping with privileged others who had thousands of opportunities to improve the general IT area but were too cool to do it and invented yet another LISP interpreter instead. And then they come to HN and use "we" as a big convenient scapegoat, obviously to escape responsibility. Most of "us" are trying to have some semblance of a life, man. Whoever could collect $200k+ a year and be comfortable should have been helping the IT area. Not me who still doesn't have his own house in his 40s.
  - camgunz 1y ago
    Hey same team. Earnestly asking: what should we do? I have some ideas but honestly they're so radical they feel like wild overreactions.
    - pdimitar 1y ago
      Wild overreactions is exactly what we need -- f.ex. the Linux kernel's gross negligence of security is no longer good enough is one thing that regularly comes to mind. New geopolitics are in place, too. Our entire branch should do much, much better than they currently do. One prediction: cybersecurity jobs might see some surge in the EU. Maybe. In general though, I would work on transparently syncing world-wide database with partitions for personal data (photo galleries, contacts, calendars, media collections, game saves, anything at all) that is akin to, but much better than, torrents -- another very good example. In general I am extremely mad at the privileged techies who just wrote philosophical essays about what "we" should be doing... while they themselves spent their riches on expensive (and likely sub-par quality) food in the restaurants of a certain coastal US city.
mikewarot 1y ago
History doesn't repeat itself, but it does rhyme. Most of the discussion here seems to focus on the events of the past repeating themselves in almost the same manner, and ignoring the differences, which are substantial. We're not talking about goods produced in bulk, but rather about a new intelligence that helps produce job shop type goods. That intelligence itself isn't a physical good though, and it can be copied freely and run elsewhere at almost no cost, and copyright won't save it. It's like toothpaste out of the tube once it's on the internet. If we want to keep on track with the analogy to frame weaving machines, the difference is that capital and the availability of power were required to use those machines, which provided the moat required to protect profits.[Econ 101] Models can be copied, and all of the investment in training used freely by anyone possessing a copy. This might have been how DeepSeek was trained, we'll never know. This is clearly different than in the case of weaving machines. So instead of the cost of cloth, the cost of thought is going to drop considerably, and the quality of that artificial thought is (likely) at the lowest point it'll ever be. I can already run quantized versions of Deepseek on my CPU with 32 Gb of RAM. The future really doesn't seem to have an upper limit on what can be run locally. With the ability to run models locally, the costs of running models will always have competition to lower prices. The other force is the capture of prompts and responses, whether covert or in the open, for other use, which must always be kept in mind. The ability to build completely new things, will, I think, help to usher in a new era of innovation and allow anyone, anywhere, to start up companies to service the world. [Econ 101] - In Economics 101, I learned that in a free and fair marketplace, it was impossible to maintain an outsized profit, because competition would always show up to take part of that profit.
- saulpw 1y ago
  > In Economics 101, I learned that in a free and fair marketplace, it was impossible to maintain an outsized profit, because competition would always show up to take part of that profit. This is all "perfectly spherical cow" economic theory which does not apply in reality. Do we really have a free and fair marketplace? Where's the competition that's shown up and been able to get any kind of substantial piece of the outsized profit of Facebook, Apple, Microsoft, Google?
  - 1y ago

This is the true problem with AI. It's with who owns it, and what they will inevitably use it for. Whether it can do cool stuff with code or equal a junior developer is irrelevant. What it can do is less important than what it will be used for. The owning class will use it to reduce payroll costs, which from their perspective is a cost center and always will be. If you're not an owner, then you have no control over the direction or use of AI. You are doomed to have your life disrupted and changed by it, with no input whatsoever. To quote the article, your six shillings a day can become six shillings a week, and you are left to just deal with it however you can. You are "free" to go find some other six shilling a week job. If you can. And if you think, "Oh, every technology is like this, it's always been this way", you are right. You have always been at the whims of the owning class, and barring a change towards economic democracy, where average people regain control over their lives, it likely always will be.

dontlaugh 1y ago
This was the Luddites’ position. Organised labour should one again take up this principled position. Join a trade union!
- breakyerself 1y ago
  Software engineers needs a lot more organization
- WalterBright 1y ago
  Do you really want a return to the days when "women's work" was spending every free moment spinning and weaving cloth by hand? When cloth was so valuable there was a profession called "rag pickers"? Where new clothes were rare, hand me downs were the usual, and people wore clothes until they disintegrated? And poor people made clothes out of flour sacks?
  - dontlaugh 1y ago
    That’s not what the Luddites wanted. They praised the technology itself and recognised it would save a lot of time. But they also recognised that they the workers wouldn’t reap any of those benefits, they’d just lose their jobs. Do you really think capitalists would choose to shorten the working day with no loss of pay as productivity increases? If so, you are incredibly naive.
    - WalterBright 1y ago
      The workers did reap the benefits. The Law of Supply and Demand ensures it. 1.1% of the US labor force works at minimum wage jobs. When wages are raised by the government, job losses happen. 16,000 to 36,000 people lost their jobs when California raised the minimum wage for fast food workers. > Do you really think capitalists would choose to shorten the working day with no loss of pay as productivity increases? During WW2, production needed to increase, so hours were increased to 60 hour weeks. The labor force, very patriotic, was all for this. Production increased for a few weeks, and then fell below what was produced in a 40 hour week. Companies wanting to maximize profits are aware of this effect. Also, if you track employee total compensation (not just wages) against productivity increases, the two lines form the same curve. The reason for this is the Law of Supply and Demand pushes those two lines together. The more productive a worker is, the more they get paid, as such workers are more in demand.
      - specialist 1y ago
        Now include profit (surplus) in your analysis. Wage grow has not kept pace with corporate profitability. Why?
        WalterBright 1y ago
        Because the correct comparison would be with total compensation, of which wages are only a portion. Total compensation is the costs to an employer to hire some one. It includes: 1. contributions to retirement plans 2. so-called "employer paid" social security contributions 3. sick and paid vacation days 4. employer paid health insurance 5. stock plans 6. 401k plans 7. other payroll taxes Last time I checked, these often added 40% or more over the takehome pay. Also, the pay is based on the value the employee as an individual provides, not the value the company as a whole created. Just like the compensation for pro football players varies in a single team.
- Cthulhu_ 1y ago
  Even unions are up against it though. in the Netherlands, there have been multiple strikes from e.g. the train operators; they demand a 10% wage increase, after the train company downsized a lot during the pandemic, but who were then unable to upsize again as traffic resumed and who even now don't have the same amount of travelers as they did in 2020. But they're not profitable. They've made a loss for five years in a row, requiring government subsidies to stay operating (because them going bankrupt would make big parts of the country grind to a halt). Ticket prices are so high that it's barely cheaper than driving, and as soon as you travel with two or more people it's cheaper to just own a car. (Or even rent one; I traveled to the other side of the country once for a concert, my own car was in the garage. Did the math, it was cheaper and more convenient (door-to-door) to rent a car, pay the surcharge for long distance, pay the €35 in fuel, etc than it was to buy return tickets for three people) The answer (I think, I'm no economist / politician / etc) isn't unionising and demanding better treatment, because the limits of the capitalist system they operate in have been reached. The answer is to stop trying to make it profitable. Re-privatize it: trains and public transit are a huge and hugely important nationwide economic driver, an essential service that capitalism can't be trusted with because the owners will try and get as much money out of it as possible, the employees get overworked and exploited and will shut the system down (as is their right) out of protest, and the people will have to deal with the consequences. Just in my small bubble, the train strikes led to people not going to the office, missing events, lunch orders that either had too much stock that had to be discarded or that were cancelled entirely, costing that company thousands in income, etc. Don't get me wrong, I'm not anti-union. I'm anti-capitalist though.
  - dontlaugh 1y ago
    You’ll often find trade unionists are also anti-capitalist, some are even socialists or communists.
WalterBright 1y ago
> The owning class will use it to reduce payroll costs Of course. That's been going on since the invention of the plow. That's why today we can do more interesting things than turn over the earth with a pointy stick all day every day. > economic democracy, where average people regain control over their lives History shows us that this inevitably means people lose all control over their lives, because the state will make your decisions for you and assign you your job. For example, let's say the color of cars produce by car companies is determined by democracy. 59% vote for the cars to be green. And if you want a red car? Too bad. What if you want a 4 seat car? No dice, 53% voted for 2 seaters to be made. What if you didn't want a car stereo? You're stuck paying for it anyway, as 73% voted for it.
- schmidtleonard 1y ago
  I have a whole list of things I'd like from my car that the market does not provide because it is more profitable not to. Why do I get the feeling that instead of seeing this as a horror story you would scold me for unreasonable expectations, even though it is the identical mirror form of the horror story you just told?
  - derektank 1y ago
    My take would be to scold you to start your own company that provides the features you want from your car if it's not already being provided by the marketplace
    - schmidtleonard 1y ago
      I'd have more luck petitioning the Central Car Design Committee because > it is more profitable not to ...and if anyone thinks this is ridiculous, I'd ask them if they are for or against repealing all of our current motor vehicle regulations. If "for," they have admitted to being a hopeless libertarian, and if "against," they have acknowledged that important reasonable features can be incompatible with the profit motive of a free market and it no longer seems so strange that I might have a list which is more of the same.
Cthulhu_ 1y ago
> The owning class will use it to reduce payroll costs Things are cyclic, nothing new; the previous big scare was (is?) outsourcing, where for the same price as one developer in western / northern Europe or SF you can hire five from eastern Europe or India. But that hasn't affected employability of the one developer, as far as I'm aware. I'm not even thinking of skill level, I'm sure that's comparable (but honestly I don't know / care enough), but both outsourcing and AI require the same things - requirements. I've grown up in this country (the Netherlands) and automatically have intrinsic knowledge of e.g. government, taxes, the energy sector, transportation, etc, so much that I'm not even consciously aware of a lot of things I know. If you spend a LOT of time and effort, you could - eventually - break that down into requirements and work orders or whatever that someone else could process. But it's much more efficient to do it yourself or just hire someone from around here.
- ge96 1y ago
  > hasn't affected employability of the one developer I wonder, combo that r&d rule, general economics, saturation in the field The norm for new grads seems to be "apply at least 2000 times, get 1 job"
  - tangjurine 1y ago
    the one developer one dev to rule them all, and in darkness bind them sounds like a 10x dev to me
- worik 1y ago
  > the previous big scare was (is?) outsourcing, I was replaced by outsourcing Seriously. Fuck off
- bevr1337 1y ago
  > Things are cyclic, nothing new Must be a convenient worldview
- RainyDayTmrw 1y ago
  I know of two major examples - one personal and one word-of-mouth - where a major US company successfully took their Covid era remote work knowledge, applied it to outsourcing, and successfully replaced most of their US onshore developers with offshore.
blooalien 1y ago
This is by far my biggest concern about "A.I." and "smart" robots... Not the technology itself, but what the "ruling class" intend to do with it / how they intend to use it. Their primary concern is not "worker productivity" these days. It's "How can I replace the maximum number of workers (ideally all of them) so that I can keep most / all of the profits / benefits for myself (and the shareholders)?" It's always been about profit, but they now finally see a potential to entirely rid themselves of "those pesky poors" and their annoying paychecks once and for all.
whall6 1y ago
The owning class is different today than it was in Ned’s time. Who owns MSFT? You, probably, through your 401k. Nobody is stopping you from creating a coalition of similarly minded shareholders and effecting change. Unless you stuff your cash under your mattress, you are the owning class.
- worik 1y ago
  > Nobody is stopping you from creating a coalition of similarly minded shareholders Sweet lies.
- like_any_other 1y ago
  > Who owns MSFT? You, probably, through your 401k. All or nothing fallacy. People have VASTLY different degrees of ownership, and this is reflected in their degree of control. > Nobody is stopping you from creating a coalition of similarly minded shareholders Such coalitions already exist. They have names like BlackRock, Vanguard, and Fidelity Investments [1]. Good luck competing with them. Especially if you abdicate your role in democracy, don't agitate for legal change, and restrict yourself to only market-based means of change (not BlackRock and ilk though - they use every lever available to them, they don't handicap themselves). [1] https://en.wikipedia.org/wiki/List_of_asset_management_firms https://en.wikipedia.org/wiki/List_of_asset_management_firms
  - whall6 1y ago
    This isn’t a false dichotomy because I’m not even representing two separate choices. I get that you alone have probably an insignificant amount of control, but forming a coalition would allow you to make change. I’m aware that there are asset managers. Whose assets do you think they’re managing? Yours! Do you own VOO? Awesome, that’s why Vanguard owns 10% of almost every large cap company. Take control of those shares and go make change. Look up John Chevedden. He owns a minimal amount of shares in companies where he causes huge changes. They call him the “corporate gadfly” [1] Don’t be lured into thinking that you have to invest your money into a predesigned portfolio because it maximizes returns. If you care about control then maximize control. You already have the power! [1] https://www.sec.gov/comments/s7-23-19/s72319-6733874-207512.pdf https://www.sec.gov/comments/s7-23-19/s72319-6733874-207512....

kevmo314 1y ago

> Actually, what are we going to do when everyone that cares about the craft of software ages out, burns out, or escapes the industry because of the ownership class setting unrealistic expectations on people? Nothing, I guess? There's an implicit assumption that software written by humans is a necessity. If the future finds that software written by computers is more profitable then that's just what it is. The universe doesn't owe us value on human-written software. Even food is not able to escape this hole: it's not profitable to manually cultivate food anymore if you wanted to do that as a career.

th0ma5 1y ago
This also assumes that non human written code will be of any use to humans and no one has shown that to be possible, it is all humans patching it up so far.
ChrisMarshallNY 1y ago
Well, as someone who considers themselves to be a "software craftsman," I have come to the conclusion that the work I do will never be valued, and will always be considered "too expensive." Since I work for free, that's not an issue for me, but that's economically unfeasible for most folks. The issue with an industry awash with cheap dross, is that it becomes prohibitively expensive to produce high Quality stuff. Anyone that tries, will get driven out of business. Some clever folks will figure out how to do "slightly better" stuff, and charge more for it, but a good way to go out of business, is to focus on Quality as a principal axis. That's basic market dynamics. It is what it is, and is neither evil, nor good. It does mean that only "niche" craftsmen, like myself, will produce anything of decent Quality, but will be unable to do so at scale, because we can't get a team together, large enough to do big things. I guess the saddest thing, is that I have really wanted to help teach my techniques to others, but have found that no one wants to learn, so I gave up on that, many years ago.
- burlesona 1y ago
  > The issue with an industry awash with cheap dross, is that it becomes prohibitively expensive to produce high Quality stuff. This seems to be one of the brutal truths of the modern world, and as far as I can tell it applies to everything. There's always a race to the bottom to make everything as cheaply as possible, and the further the industry goes down that "cheapness" scale, the more "quality" loses market share, the more expensive "quality" must be in order to operate at all, and finally things that used to be just "normal" and not too expensive are now luxury goods. Consider textiles, carpentry, masonry, machine tooling, appliances, etc. etc. This doesn't feel like a good outcome, but I'm not sure there's anything that can be done about it.

WarOnPrivacy 1y ago

industry only ever seems to want to hire people with the word Senior in their title. They almost never want to create people with the word Senior in their title. I suggest that the first ladder that got pulled up is the one on the ground. No one wants to train new entrants to the field. Not training junior workers seems like a natural extension to that.

whatshisface 1y ago
I guess universities will eventually backfill that gap with training that focuses on senior skills. Eventually.
- thmsths 1y ago
  We already require juniors to go through a 4 years university degree. It takes a fair bit of time of real world work to get to the senior level. So unless we expect people to do another 5 years of schooling, I am not sure how this will happen (and even in that scenario I believe there is a difference between hands on, on the job experience and classroom experience).
  - jdee 1y ago
    it happened with architecture degrees in the uk. it went from 3 years to 7 years, as the skill levels and implicit knowledge required increased over time.
ta1243 1y ago
In centuries past apprentices would pay their junior positions, in time picking up paid work as the progressed to senior, then eventually taking on apprentices of their own (and be paid)
teeray 1y ago
It’s a Volunteer’s Dilemma: why train juniors when you can hire the juniors other companies trained to become seniors?

ChrisMarshallNY 1y ago

> Vibe coding is payday loans for technical debt Stealing it...

schmidtleonard 1y ago

> I don't fear the power loom. I fear the profit expectations of the factory owners.

deadbabe 1y ago

Vibe coding is not going to go away. Just as alchemists always dreamed of turning lead into gold, there is a dream that people have had for decades of writing less and less code: Open source. Scaffolds. Code generators. No-code tools. Coding Agent LLMs. If vibe coding goes away, it will be because it has been replaced with something even more hands off. Perhaps vibe prompting. Or even code farming: just have AI agents constantly build totally random shit and go through and see if any of it is useful for something. Maybe even take a genetic approach where you score the fitness of various software for a certain solution and AI will cross mutate the genes. This will be a change of that rivals our shift from hunter/gatherers to the agricultural revolution.

blibble 1y ago
> Or even code farming: just have AI agents constantly build totally random shit and go through and see if any of it is useful for something. so you're saying the venture capitalists will soon be out of a job? what a shame
- deadbabe 1y ago
  Venture capitalists are basically startup farmers.

mrtksn 1y ago

I knew what was this all about the moment I glimpsed at the title on the front page, my neural nets must have been trained to output that from the few tokens I guess, which makes me think maybe all this AI stuff isn't that different from what's happening in real brains. I bet others too had similar experience. So the AI thing is happening, maybe its not with this particular tech we have today but they are on to something and we probably better embrace it. Humans rolled the ladder up behind on so many things. Very few people will survive the planet Earth without the tools and abstractions we built over the many Millenia's and that's how we all live like kings. Any misery out there in the world is a result of our inability to manage it, not because the resources are scarce. The Luddites, as explained in the article were indeed about the way tech is being adopted. All that tech eventually reached every corner of the world and not every place had Luddites. Programming computers by hand is shitty anyway, good riddance. Finally we are about to have machines that can be programmed without thinking about the intricacies of programming that have nothing to do with the thing we want to achieve. All the tech left behind still does have aficionados, maybe in 10 years we can watch someone program a computer using Java on the Primitive Technology channel.

paxys 1y ago

The part that is always skipped when making the clothing analogy is that for most of human history good clothes were a luxury. At the time of the industrial revolution it was common for a family to spend 15-25% of their annual income just on clothing. A single shirt from that era would cost the equivalent of £2,000 today when considering material and labor. New clothes were something you may be able to afford once a year, if at all. Like the article says, weavers were the equivalent of PhD scientists who could charge whatever they wanted for their skills (and they would cater pretty much exclusively to kings and the 0.1% of the time). Great if you're the weaver, not so for everyone else. So the industrial revolution wasn't a conspiracy to put down the lower classes. The lower classes were in fact the biggest benefactors of the industrial revolution. If software can go the same way, I'd say good riddance. The profession has always kept free from gatekeepers, and that's a good thing.

stego-tech 1y ago
> The part that is always skipped when making the clothing analogy is that for most of human history good clothes were a luxury. Literally what OP discusses in their text, right in the first part. Go RTFA. > and they would cater pretty much exclusively to kings and the 0.1% of the time Oh yeah, I totally remember reading about how people in pre-modern civilizations were almost always semi-clad or fully nude due to the expense of clothing. Oh, wait, no I don't, because people still bought clothing and wore it regularly. They just also had economies around mending clothing, updating it, tailoring it, altering it, reusing and recycling it. Rather than building an economic system of destruction for the sake of a handful of profiteers, it was an economy of artisans who provided a staple resource at reasonable rates and quality to support themselves. Because clothing was often tailor-made rather than ready-to-wear, people took care of it - and themselves - for longer periods of time. Techniques were used to keep articles sturdy for longer, rather than disposable machine stitches that fall apart in a washing machine. Experts and artisans are not "gatekeepers", they are skilled craftspeople worthy of respect and deserving of compensation for their skills. To demand anyone be able to do anything of any complexity is to demand a complete elimination of anything that differentiates humans from one another, to create a homogenous mass of genetics with no incentive to grow and evolve. Nobody is "gatekeeping" software developers, or Doctors, or plumbers, or weavers, or artists. Those all take skill, and people with a suitable level of skill can easily pass good credentials checks and tests.

tptacek 1y ago

As a software security person, I don't think the security objections to LLMs are going to pan out. I think LLMs are going to be a strong net positive for security: * The tooling and integration stuff people complain about now ("the S in MCP") isn't really load-bearing yet, and a cottage industry of professional services and product work will go into giving it the same overcomplicated IAM guardrails everything else has; today, though, you just do security at a higher or lower level. * LLM code generation is better at implementing rote best-practices and isn't incentivized to take shortcuts (in fact, it has some of the opposite incentives, to the consternation of programmers like me who prize DRY-ness). These shortcuts are where most security bugs live. * LLMs can analyze code far faster than any human can, and vulnerabilities that can be discovered through pure pattern matching --- which is most vulnerabilities --- will be easy pickings. We've already had a post here with someone using o4 to find new remote kernel vulnerabilities, and that's a level of vuln research that is way, way more hardcore than what line-of-business software ordinarily sees. * LLMs enable instrumentation and tooling that were cost-prohibitive previously: model checking, semantic grepping, static analysis. These tools all exist and work today, but very few projects seriously use them because keeping all the specs and definitions up to date and resolving all the warnings is too much time for not enough payoff. LLMs don't have that problem. LLM-generated code (and LLM tooling) will inevitably create security vulnerabilities. We have not invented a way to create bug-free code; would have been big if true! Opponents of industry LLM use will point to these vulnerabilities and go "see, told you so". But each year we continue using these tools, I think the security argument is going to look weaker and weaker. If I had to make a bet, I'd say it ceases being colorable within 3 years.

whatshisface 1y ago
There are also non-llm advances in testing related to AI, like RL fuzzers.
threetonesun 1y ago
I'm not terribly worried about code generated security vulnerabilities, but point 3 feels like a cat and mouse game that most companies won't have the resources to stay on top of, so they'll have to outsource it to one of the existing cloud or AI providers. Maybe that's a reality even without AI but it feels like we're heading towards full on extortion from about 4 major companies. Also I don't think you covered my biggest concern with LLM security, a company making an Amazon basics version of your business model and claiming "AI did it". I'm 50/50 on that one though, it's also possible everyone things with AI you can go full NIH syndrome and take back all the software that we've handed off to various SAAS providers.

Cthulhu_ 1y ago

> We're going to run out of people with the word "Senior" in their title Probably not, but job title inflation has made it so that apparently 5 years of experience is enough to be given a "senior" title. I've got like 15 but still feel like a medior at best. Yeah this is humblebragging, whatever it's a throwaway internet comment.

vultour 1y ago
This is a funny comment because startups have been churning out sub-3 year "seniors" for years. If you're not a principal engineer after 5 years what are you even doing?

blakesterz 1y ago

There's a photo on that page with this description: "A picture of two patches of wild grass bifurcated by a retaining pond" I was just thinking how I'd describe that different, and how many different ways it could be described.

Groxx 1y ago
Perhaps even a thousand words worth of description is possible
- riskable 1y ago
  A thousand tokens.

Havoc 1y ago

I've kinda seen this already play out in a completely different context. Outsourcing & audit world. It's sufficiently commoditized that you can efficiently outsource. If you're sending the first say 12-months worth of experience work to india then local gang straight out of college end up with a pretty tangible gap. There are key formative experiences missing. The programming equivalent of fighting with manual array memory management. Or shooting yourself in foot with a pointer. Worse they don't realise that they're missing pieces. By necessity you start them off at a higher level and their understanding ends up fuzzy for lack of better word. It's not their fault...to them it looks normal, to the previous gen it's all "how can you not know this" ...still the world keeps turning

stego-tech 1y ago

Xe's stuff is always excellent, and this is no different. What's disappointing is the amount of energy and effort wasted by people to come into the comments and espouse their desire to give up on addressing any of their concerns. "It's just how things are." "That's what generates profit." "If that's what the market demands, who are we to judge?" "Guess we do nothing." We've all heard it before. We get it, so many HN and Tech folks have thrown in the towel and given up on changing anything within their own lives, or within the lives of others. So many more are content just "going along with the ride", abdicating all responsibility to an imagined god in various forms ("invisible hand of the free market" being most common among our peers). For the rest of us who want to actually discuss the merits of the piece, we're sidelined by those who found reward in the path of libertarian ideals and seek to punish or discipline those of us who seek to make any sort of change or improvement beyond direct profit motives. Serious discussions of organization and direction are derailed in favor of blatant troll bait and bad-faith arguments. This, I think, is also what Xe is trying to highlight: those in the position to enact the most positive change are staunchly refusing to do so, which in turn exacerbates the harms caused. Too few people with too much power and too much capital believe themselves to be more knowledgeable and intelligent than anyone else, denigrating any thoughts that may come from the slime beneath them on the ladder. When Xe said "Rolling the ladder up behind us", they make it very clear who they're speaking to: > Look, CEOs, I'm one of you so I get it. We've seen the data teams suck up billions for decades and this is the only time that they can look like they're making a huge return on the investment. They're basically shouting that "the call is coming from inside the house", and that the only place - right now - to address these harms are by and from the people inside said house. Nobody is coming to save us, but history books are filled with examples of what happens to those classes, those societies, those civilizations for whom greed becomes all consuming, standards of living collapse, and meaningful progress stagnates. > Maybe the problem really is winner-take-all capitalism.

mattgreenrocks 1y ago
> We get it, so many HN and Tech folks have thrown in the towel and given up on changing anything within their own lives, or within the lives of others. It’s hilarious to read such defeatist takes on the current state of things on a site called Hacker News. The classical definition of a hacker was someone who wielded tech to circumvent power imbalances in whatever form. Now I fear we have plenty of technologists that are happy to use tech to further cement and centralize existing power structures. IMO, this has been brewing for the past ten years or so, where technologists wrongly believe they have an in to this power just because they work for a FAANG. You are labor. Where you work does not matter. So long as you need to work, you are labor. And there’s nothing wrong with that! Doctors are finding out this same lesson. The way out is to build ventures that don’t involve the capital class. Autonomy and independence are worthy goals to strive for.

Veedrac 1y ago

A defense of the Luddites against sewing machines was honestly not something I expected. Support for the Luddites in general, sure I can predict that, but the claim that sewing machines are bad, exploitative, produce "shitty" cloth, serve only the pursuit of profit...

duskwuff 1y ago
Power looms are not the same thing as sewing machines. Not even close.
- Veedrac 1y ago
  Ah, you're right, thanks for the correction.
int_19h 1y ago
There's a lot of nonsense there, like: "They were skeptical that the shitty cloth that the power loom produced would be a net benefit to society because it meant that customers would inevitably have to buy their clothes over and over again, turning a one-time purchase into a subscription." This is such an obvious attempt to reframe history to fit the modern discourse. What Luddites cared about was their own jobs, not "net benefit to society".

djoldman 1y ago

There are a lot of issues brought up in this post, but I want to discuss one in particular: technological progress and its economic and societal ramifications. We can say that technological progress occurs when a new method is employed to deliver a product or service with some "more desirable" blend of qualities: it's created/delivered faster, cheaper, with a more desirable mix of resource inputs, and/or results in a more valuable/desired output. Sometimes it's quite obvious when a technology is superior to another as almost all the qualities of it are advantageous: it's made faster, less expensively, and the result is better with such a gap between it and the old way that there's just no denying that the new way is better. Sometimes the new technology is really a mix of qualities. Let's focus on the mix that generally gets the most attention: the new way is faster/cheaper but the output is not of a higher quality. Sometimes this new way of cheaper+faster but lower quality "wins out" and the consumer prefers it. And now the crux of it: why is it so common that discussion concerning these shifts is rooted in everything except the consumer? An imperfect and potentially flawed example: a bunch of consumers have decided that they'd rather pay less for a shirt or shoe that will fall apart faster than more for one that lasts longer. The old way of making a superior product still exists and some consumers still prefer it but most do not. Fundamentally, this is a shift that is rooted in the perception, true or not, of the consumer that the new way is more desirable. Some folks are not happy with the higher prices of the outputs of old ways of doing things or the effects that the new ways have on jobs, the economy, and/or society. Fundamentally, this has to do with consumer preference and that's where any blame should be meted out.

duskwuff 1y ago
> technological progress and its economic and societal ramifications I seem to recall that a mathematician wrote a paper about this around 1995 which got a lot of press attention. I'm not sure I agree with all of his conclusions, though. Nor his methods.
wizzwizz4 1y ago
> a bunch of consumers have decided that they'd rather pay less for a shirt or shoe that will fall apart faster than more for one that lasts longer They don't know it'll fall apart faster, to begin with… and then by the time they do, they've already spent enough of their clothes budget that they can't afford to buy proper cloth, and it's hard to find that nowadays anyway, so they'll just have to buy more of the low-quality stuff.

youworkwepay 1y ago

The value of vibe coding isn't that it writes good or sustainable code. It's that you can build a sufficiently non-shitty prototype of a concept as a non IT expert to validate the use case and secure proper assistance. From a change leadership perspective, walking in the door with a shitty prototype beats pitching vaporware every day of the week and twice on Sundays. And the fact amateurs can deliver (basic, crappy) "things" without budget accelerates growth. My first projects were 80% copied off Github and some intro tutorials. Know what? They still work. We banked seven figures off of them so far...

felipeerias 1y ago
The article hints at that when it describes vibe coding as “fancy UX”, but fails to connect the dots. Essentially, we now have a system that can turn a simple problem description into an interactive tool for that problem. Even if it still does so very imperfectly, it’s easy see already the beginnings of a powerful and empowering new paradigm.
edmundsauto 1y ago
In terms of risk, building a prototype and getting a quick win really de-risks a project. Smart decision makers' world is a game of risk - they have tolerance in some areas, less tolerance in others. If you can materialize a quick win from a prototype, it's significantly less risky that sight-unseen work. Coders often don't think in terms of a broad investment portfolio, but that's how I've seen good executives phrase things. AI makes it cheaper and easier to build that prototype - I've been loving it for my own projects, because of how quickly I can deliver that first software.

burnt-resistor 1y ago

3 things. 0. The people must take their governments back from corporate technofeudal overlords. 1. Society must shift away from lionizing unbounded capitalism. 2. Workers must form employee-owned co-ops for more stability and better morale and share profits of their labor more fairly rather than extracting maximum amounts for the ownership class that will fire them whenever it suits them.

secstate 1y ago

I have a relative who works in healthcare. The introduction of AI for medical notes has led her level+1 manager to ask folks to increase their patient numbers in response to the extra time saved not having to manually code charts. The initial response was some grumbling about unionization of the doctor-class. But now they're just kinda going along with it while also complaining loudly. Profit Über alles.

shae 1y ago

What if we train them and they leave? What if you don't and they stay?

hackable_sand 1y ago
Yup. It's highly abusive behavior.

worik 1y ago

> . I have a hoodie from AWS Re:Invent in 2022 that I'm going to have to throw out and replace because the sleeves are dying What is the use of the right to repair, if we do not have the imagination to use it? I patch my clothes

lubujackson 1y ago

This post is really all over the place and, I think, misses the main thing here: pretty much every advancement in automation across history has sacrificed some fidelity and craftsmanship. That's the cost of automation. Yet we have continually added more and more automation to our lives because the net value is dramatically higher for the bulk of our civilization. Yes, we lose (or pidgeon-hole) artisans and craftsmen. But even in his "woe the times" example of AI kludging together two game properties into a painting, where the lighting isn't right and furry tail is not ideal, etc., etc... That $500 image is something no one could have created 20 years ago with a computer. This isn't centuries of artisan skills we are talking about. What about the artisan who would have been paid $500 to paint it with actual paints on an actual canvas? You can mourn progress in every direction if you like, but I don't think it changes a single thing about the trajectory of where we are going. Now that AI is coming for my job as a programmer I'm not going to get twisted about it - my job didn't exist when I was a kid and it won't exist (certainly not in the same form) when my kid is an adult. And that's ok, it was a "job of opportunity". And that job has changed dramatically over 20 years. AI is just the continuation of that change. I started in SE because it was fun to make computers do stuff. Programming has only ever been a tool to achieve that goal, and often a frustrating one. AI is just a better tool. If the job becomes "a guy that talks to the AI" that will still be programming because at the end of the day the computer can't do anything without someone telling it what to do. ...until SkyNet.

AngryData 1y ago
I mean one of the main complaints of the luddites was decreasing cloth quality despite the looms originally being able to produce higher quality cloth comparable to their own traditional weaving techniques. Powered looms and the dozens of other loom improvements from the previous 100+ years didn't inherently produce poor quality cloth, they were adjusted to produce lower quality cloths after they were already in operationin factories in order to increase profit margins. Powered looms were originally just modified hand looms, not some radically different design from the ground up. The flying shuttle predates powered looms and the luddites.
edmundsauto 1y ago
> pretty much every advancement in automation across history has sacrificed some fidelity and craftsmanship. That's the cost of automation Can you help me think through how this looks when considering car manufacturing? Originally hand made cars were things of craftsmanship, but significantly lower quality and reliability than those produced with tiny tolerances on insane production lines. I'd take a modern car over a handmade one just about any day of the week.

egypturnash 1y ago

I'm a pro artist who does a lot of furry commissions and has been getting constant downvotes on this place for expressing my unhappiness ever since this massive abuse of "fair use" began, thanks for being on my side, Xe <3

worik 1y ago

Do we, or do we not, live in democratic societies? If we do then we must ensure that we get the social benefits from new technologies, and we suppress the harm. Does anyone still believe that "self interest" in the aggregate is what is best for society? I think it is clearly untrue So we need to take collective action. AI/LLMs are a "steam engine" moment. We can make life better for everyone. But we do need to change course, together, to get there

nyarlathotep_ 1y ago

The most pressing issue to me currently regarding LLMs is the blatant, flagrant change in tone of the "leadership/ownership" class towards employees. It comes off as little other than thinly veiled contempt for the people that actually make the business function. We've seen, most recently, Jassy come out and say "we're gonna need a lot less people", Zuckerberg doing his "replacing engineers" shtick on several occasions, all the commentary from the Anthropic CEO/Altman etc. Even if none of this comes to pass, I've never witnessed such open "threats" in my lifetime. I can't square this perspective with that of the LLM advocates that run to defend every legitimate criticism of LLM use--unless you're one of the aforementioned or in a similar position, why would you want to defend software that's primary purpose is to displace you? They are literally telling you that's why they're investing in it. I don't get it.

cedws 1y ago
It’s masks off for them, they were always sociopaths that see other people as a means to an end.
int_19h 1y ago
I want software to displace me when it comes to my day-to-day grind job, so that I can spend this time on something more interesting. But I want that to happen in a way that results in a Star Trek communist utopia, not a hellscape where 90% of humanity is "redundant" to the very few people who own all that automation and its outputs.

_zagj 1y ago

> Honestly though, the biggest impact I've seen across my friends has been what's happened to art commissions Given the author's eccentricities, when I read "art commissions," I immediately suspected they really meant "furry porn art commissions." Then a few paragraphs later: > A friend of mine runs an image board for furry art. He thought that people would use generative AI tools as a part of their workflows to make better works of art faster. He was wrong, it just led to people flooding the site with the results of "wolf girl with absolutely massive milkers showing her feet paws" from their favourite image generation tool in every fur color imaginable, then with different characters, then with different anatomical features. There was no artistic direction or study there. Just an endless flood of slop that was passable at best.

camgunz 1y ago

Thomas Ptacek dismissed SWE AI concerns as hypocrisy, and I actually think he understated the argument. This is exactly what we get for: - using BSD/MIT licenses instead of GPL licenses - accepting absolutely no liability for security holes or data loss - accepting absolutely no responsibility for poorly performing software that contributes to climate change - shipping 17,000 web frameworks, founding one more blockchain startup, founding one more wrapper service around an MIT open source project We're gonna turn around and be like "but what about the security issues" now? We made this bed.

pdimitar 1y ago
Who is "we"? A lot of "us" don't have the time to do programming as a hobby -- even if we started off that way. Many of "us" don't accept blanket grouping with privileged others who had thousands of opportunities to improve the general IT area but were too cool to do it and invented yet another LISP interpreter instead. And then they come to HN and use "we" as a big convenient scapegoat, obviously to escape responsibility. Most of "us" are trying to have some semblance of a life, man. Whoever could collect $200k+ a year and be comfortable should have been helping the IT area. Not me who still doesn't have his own house in his 40s.
- camgunz 1y ago
  Hey same team. Earnestly asking: what should we do? I have some ideas but honestly they're so radical they feel like wild overreactions.
  - pdimitar 1y ago
    Wild overreactions is exactly what we need -- f.ex. the Linux kernel's gross negligence of security is no longer good enough is one thing that regularly comes to mind. New geopolitics are in place, too. Our entire branch should do much, much better than they currently do. One prediction: cybersecurity jobs might see some surge in the EU. Maybe. In general though, I would work on transparently syncing world-wide database with partitions for personal data (photo galleries, contacts, calendars, media collections, game saves, anything at all) that is akin to, but much better than, torrents -- another very good example. In general I am extremely mad at the privileged techies who just wrote philosophical essays about what "we" should be doing... while they themselves spent their riches on expensive (and likely sub-par quality) food in the restaurants of a certain coastal US city.

mikewarot 1y ago

History doesn't repeat itself, but it does rhyme. Most of the discussion here seems to focus on the events of the past repeating themselves in almost the same manner, and ignoring the differences, which are substantial. We're not talking about goods produced in bulk, but rather about a new intelligence that helps produce job shop type goods. That intelligence itself isn't a physical good though, and it can be copied freely and run elsewhere at almost no cost, and copyright won't save it. It's like toothpaste out of the tube once it's on the internet. If we want to keep on track with the analogy to frame weaving machines, the difference is that capital and the availability of power were required to use those machines, which provided the moat required to protect profits.[Econ 101] Models can be copied, and all of the investment in training used freely by anyone possessing a copy. This might have been how DeepSeek was trained, we'll never know. This is clearly different than in the case of weaving machines. So instead of the cost of cloth, the cost of thought is going to drop considerably, and the quality of that artificial thought is (likely) at the lowest point it'll ever be. I can already run quantized versions of Deepseek on my CPU with 32 Gb of RAM. The future really doesn't seem to have an upper limit on what can be run locally. With the ability to run models locally, the costs of running models will always have competition to lower prices. The other force is the capture of prompts and responses, whether covert or in the open, for other use, which must always be kept in mind. The ability to build completely new things, will, I think, help to usher in a new era of innovation and allow anyone, anywhere, to start up companies to service the world. [Econ 101] - In Economics 101, I learned that in a free and fair marketplace, it was impossible to maintain an outsized profit, because competition would always show up to take part of that profit.

saulpw 1y ago
> In Economics 101, I learned that in a free and fair marketplace, it was impossible to maintain an outsized profit, because competition would always show up to take part of that profit. This is all "perfectly spherical cow" economic theory which does not apply in reality. Do we really have a free and fair marketplace? Where's the competition that's shown up and been able to get any kind of substantial piece of the outsized profit of Facebook, Apple, Microsoft, Google?
- 1y ago