"The Death of Software Engineering as a Profession" ... is greatly exaggerated.
Could be the title of the piece.
I agree: throughout my own career as a programmer (I prefer the more blue-collar sounding term—it better fits my skill set) I have also seen large changes in the industry that certainly made waves, did not capsize the profession.
At the same time, the profession I retired from was by no means the profession I entered into in the '90s. I confess I liked the older profession better.
Its just the title, I have read the post texts before posting, he actually says its here to say dispite mainstream claiming coding is dead every other five year.
For today’s lucky 10,000– @JKCalhoun is most likely making reference here to American humorist Mark Twain, who, in response to news stories claiming he had died, was reported to have quipped “reports of my death have been greatly exaggerated.”
Software engineering isn’t dead and won’t be dead from LLMs, I have no fear of that, but coding by hand may become as usually unnecessary as coding in assembly is now.
Besides what Sam Altman wants everyone to believe, there isn't a lot of evidence what you're saying is true. My experience with LLMs hasn't borne it out, and I also don't think it's okay -- I LIKE writing software!
I posted this link after reading all the way. Jason actually makes a good point - its just that this title is loud. Blog post itself isn't claiming the death of software engineering at all. If anything, it just shows that every five or ten years someone claims software engineering is dead.
Its not dead at all and it wont die either.
Why? chagpt, or figma or v0 can spin up a few pages of brochure site, even some blog posting level web apps, basic cruds you know. But I don't think it will replace full software engineering.
I work with a large codebase, thats almost 30 years old, multiple framework ( backbone, react, angular) and then java, python for backends. All from different phases and everything is stitched together to make it work, and have a well profit making business going on. There is no model or chatxyz that can dig throug all these connected apps and services and replace our engineering team. It helps us here and there- yeah a lot.
Isn't this a bit revisionist? I started to become interested in programming around late 90s and I don't remember anyone floating the idea that OOP, libraries or IDEs will make programming obsolete as a profession. If anything, pre-2023 most programmers considered their job as one of the hardest ones to automate.
> I started to become interested in programming around late 90s and I don't remember anyone floating the idea that OOP, libraries or IDEs will make programming obsolete as a profession.
The version of this hype that I remember from circa 2004 was UML[1] was going to make most programming automated. You'd have an architect that would draw out your problem's architecture in a GUI[2], press a button to automate all the code to build that architecture, and have a programmer fill in a couple dozen lines of business logic. Boom, program done by two or three people in a couple weeks, let's all go home. It uh, didn't work out that way.
You can read a lot more about all this by following the various links to concepts & products from Rational's Wikipedia page, https://en.wikipedia.org/wiki/Rational_Software (the Rational Unified Process page in particular brings back some memories). It wasn't badly intentioned, but it was a bit of a phase that the industry went through that ultimately didn't work out.
It's interesting that even with the rise of transformers, UML still isn't popular. I wonder if that, or some other visual way of representing specs, might make a comeback.
There was definitely a widely held belief in the late 90s, early 00s that programming was commoditized to the point that it would be fully offshored to the lowest cost of labor. This happened in some areas and failed. It still happens now and then. But I remember hearing some of that based on OO and libraries making it so unskilled people could just put together legos.
I remember that. I studied CS in that period and some professors were convinced that software development was going to become an unskilled job, analogous to bricklaying, and that our goal as future CS graduates should be to become managers, just like someone that studies a university degree about making buildings is intended to become an architect and not a bricklayer.
I never believed it, though (if I had, I would probably have switched degrees, as I hate management). And while the belief was common, my impression is that it was only so among people who didn't code much. The details on how it would happen were always highly handwavy and people defending that view had a tendency to ignore any software beyond standard CRUD apps.
In contrast, if I had to choose a degree right now, I'd probably avoid CS (or at most study it out of passion, like one could study English philology or something, but without much hope of it being a safe choice for my career). I think the prospects for programmers in the LLM era look much scarier, and the threats look much more real, than they ever did in that period.
The bigger issue is that so many people have jumped into CS because programming (not the same thing I know) has become seen as this thing that will earn you big bucks.
Of course, some level of computer skills is important in most professions at this point. But logic suggests that CS (and programming) compensation will level out at a level comparable to similarly skilled technical professions.
It's a bit too generalizing that it failed and happens "now and then", offshoring is a multi-billion industry employing millions of people.
And the "unskilled people putting together legos" is also very much a thing in the form of low/no-code platforms, from my own circles there's Mendix and Tibco, arguably SAP, and probably a heap more. Arguably (my favorite word atm) it's also still true in most software development because outside of coding business logic, most heavy lifting is done by the language's SDK and 3rd party libraries.
>I don't remember anyone floating the idea that OOP, libraries or IDEs
Oh man, you had it lucky. Object databases were going to replace SQL multiple times, XML would eat the world and I strongly remember a UX person taking one look at Ruby on Rails at maybe 1.0 and declaring he would not be needing us programmers anymore.
> Object databases were going to replace SQL multiple times
Someone should honestly write a history of Zope. Can't say I hated it, because it was my first paid job as a programming working with it, but it was special, very special, let's call it that. 20 years from that moment I do think though that it had its good moments, even bright ideas in some places, too bad that they didn't scale.
It's an idea that surfaces every few years - back in 1981 I can remember reading about "The Last One" - named because it was supposed to be the last computer program that would ever need to be written:
It had definitely been a thing even from the 80s, though everyone always seemed to think it would be the next generation of languages that did it. For example, in 1981, James Martin wrote a book called "Application Development Without Programmers".
I find the idea that IntelliJ being a job killer hard to believe, just like when some of my colleagues used to think Dreamweaver would wipe out frontend development - or 'HTML slicing', as we called it back then.
Yeah, that's weird; IntelliJ was more like "this is how amazing and friction-free Java development and refactoring can be". Enabling more ppl to be 10x programmers, not putting ppl out of work.
I remember reading 25-30 years ago about how 4GLs and object libraries were going to democratize software creation. Don't recall it being sold as an apocalypse for coders, though.
> before LLMs I definitely thought coding would be the last thing to go.
While LLMs do still struggle to produce high quality code as a function of prompt quality and available training data, many human software developers are surprised that LLMs (software) can generate quality software at all.
I wonder to what extent this surprise is because people tend to think very deeply when writing software and assume thinking and "reasoning" are what produce quality software. What if the experience of "thinking" and "reasoning" are epiphenomena of the physical statistical models present in the connections of our brains?
This is an unsolved and ancient philosophical problem (i.e. the problem of duality) of whether consciousness and free will affect the physical world. If we live in a materialist universe where matter and the laws of physics are unaffected by consciousness then "thinking", "reasoning", and "free will" are purely subjective. In such a view, subjective experience attends material changes in the world but does not affect the material world.
Software developers surprised by the capabilities of software (LLMs) to write software might not be so surprised if they understood consciousness as an epiphenomenon of materiality. Just as words do not cause diaphragms to compress lungs to move air past vocal cords and propagate air vibrations, perhaps the thoughts that attend action (including the production of words) are not the motive force of those actions.
> I wonder to what extent this surprise is because people tend to think very deeply when writing software and assume thinking and "reasoning" are what produce quality software.
It takes deep thought and reasoning to produce good code. LLMs don't think or reason. They don't have to though because humans have done all of that for them. They just have to regurgitate what humans have already done. Everything good an LLM outputs came from the minds of humans who did all the real work. Sometimes they can assemble bits of human generated code in ways that do something useful, just like someone copying and pasting code out of stack exchange without understanding any of it can sometimes slap something together that does something useful.
LLMs are a neat party trick, and it can be surprising to see what they do and fun to see where they fail, but it all says very little about what it means to think and reason or even what it means to write software.
I'm still of the opinion that coding will be the last thing to go. LLMs are an enabler, sure, but until they integrate some form of neuroplasticity they're stuck working on Memento-guy-sized chunks of code. They need a human programmer to provide long context.
Maybe some new technique will change that, but it's not guaranteed. At this point I think we can safely surmise that scaling isn't the answer.
I’m more inclined to believe that no jobs (as in trades, professions) will go, but programming will be the most automated, along with design and illustration.
Why? To this day still they’re the showcase of what LLMs “can” do for (to) a line of work, but they’re the only ones with all the relevant information online.
For programming, there’s decades of textbooks, online docs, bug tracker tickets, source code repositories, troubleshooting on forums, all laying out how a profession is exercised from start to finish.
There’s hardly a fraction of this to automate the tasks of the average Joe who does some paperwork the model has never seen, who’s applying some rough procedures we would call “heuristics” to some spreadsheets and emails, and has to escalate to his supervisor for things out of code several times a day.
Beyond training data availability, it's always easiest to automate what you understand. Since software engineering is a subset of the discipline of AI/LLMs, it has been automated to the extent that it has. Everything else involves more domain knowledge.
I'm not sure what happens when you replace coders with 'prompt generalists' and the output has non-trivial bugs. What do you do then? The product is crashing and the business is losing money? Or a security bug? You can't just tell llm's "oh wait what you made is bad, make it better." At a certain point, that's the best it can make. And if you dont understand the security or engineering issue behind the bug, even if the llm can fix this, you don't have the skills to prompt it correctly to do so.
I see tech as 'the king's guard' of capitalism. They'll be the last to go because at the end of the day, they need to be able to serve the king. 'Prompt generalists' are like replacing the king's guard with a bunch of pampered royals who 'once visited a battlefield.' Its just not going to work when someone comes at the king.
> You can't just tell llm's "oh wait what you made is bad, make it better." At a certain point, that's the best it can make. And if you dont understand the security or engineering issue behind the bug, even if the llm can fix this, you don't have the skills to prompt it correctly to do so.
In that case, the idea is that you'd see most programmers in the company replaced by a much smaller group of prompt generalists who work for peanuts, while the company keeps on a handful of people who actually know how to program and do nothing all day long but debug AI written code.
When things crash or a security issue comes up they bring in the team of programmers, but since they only need a small number of them to get the AI code working again most programmers would be out of a job. High numbers of people who actually like touching code for a living will compete for the very small number of jobs available driving down wages.
In the long term, this would be bad because a lot of talented coders won't be satisfied being QA to AI slop and will move on to other passions. Everything AI knows it learned from people who had the skill to do great things, but once all the programmers are just debugging garbage AI code there will be fewer programmers doing clever things and posting their code for AI to scrape and regurgitate. Tech will stagnate since AI can't come up with anything new and will only have its own slop to learn from.
Personally, I doubt it'll happen that way. I'm skeptical that LLMs will become good enough to be a real threat. Eventually the AI bubble will burst as companies realize that chatbots aren't ever going to be AGI, will never get good enough to replace most of their employees, and once they see that they're still going to be stuck paying the peasant class things will slowly get back to normal.
In the meantime, expect random layoff and rehires (at lower wages) as companies try and fail to replace their pesky human workers with AI, and expect AI to be increasingly shoehorned into places it has no business being and screwing things up making your life harder in new and frustrating ways
Seems like groundhog day, although I'm not sure I remember anyone telling me that software engineers were on borrowed time until relatively recently and I'd largely ignored it.
Yet one thing does seem different for anyone who just missed the dotcom crash, is that the roles available have fallen off a cliff while the numbers looking for roles seem to be up, at least in the UK. The UAE is even worse. I've spent 20 years hiding from recruiters and now they're all leaving me on read. Karma, maybe.
Same thing happened after dotcom. Same thing happened after 2008. Same thing will happen now during AI and the trump recession.
After every downturn ends, there comes a sudden hunger for engineers, and companies can’t seem to get enough. Some companies will even hire engineers just so other companies don’t hire them. Be ready.
I think the 2008 downturn (when it comes to this profession) was a little bit less severe compared to just after dotcom, or so it felt here in Eastern Europe.
But, to be honest, I do miss the "just after dotcom crash" period, there were lots and lots of interesting things people were working on back then, and the majority of it didn't involve making money. Maybe this is just nostalgia, because I first became a paid programmer in 2005, but that's how it feels 20+ years from that time.
I changed jobs in '08, '10 and was contracting by '11. Didn't notice any downturn from 2008 at all.
I'd only once in 20 years been turned down for a job I'd applied for. Every other job I applied for I was offered. I've applied for over 600 roles in the past year and barely had a handful of interviews. That certainly feels different.
I was, for a long time, scared of my future due to the low/no-code, automation, LLMs, outsourcing, etc. Until, at some point, I realised something simple - the risk factor for my job is not determined by how good new tools are, but only by how lazy people are about learning and adopting them. And here history gives another lesson - we never learn, eternal cycle of mistakes will continue.
The outsourcing is the only real threat in your list. In the past, we have eliminated jobs in the primary (eg farming) and secondy (eg manufacturing) industries through automation and outsourcing with the goal of moving workers into higher level industries, including tertiary industries (eg software). If we are outsourcing tertiary industry jobs as well, what does that leave us?
The US outsources something like 300k jobs annually, with over half of these being IT jobs. Adding 10k IT jobs per month could change the employment numbers and economic outlook we've been seeing lately. It seems like we're in a race to the bottom. I do think AI will make things worse, economically at least, with the reduction in jobs. But this could be offset by policies promoting on-shore employment.
For those who are young, single, and are open to adventure, the endless outsourcing provides another option: pack up and leave high cost of living areas or the US entirely. There are plenty of places where you can get a fraction of a US salary while living a quality of life beyond what the US offers. US companies have return to office mandates so they can fire US workers. But if you're a US citizen living abroad and willing to accept 1/4th the salary of a person in California (and living in a place with 1/10th the living costs), companies get the best of both worlds: an employee they feel they can trust while also undercutting wages. Yeah, it sucks for people still in the US, but it's been great for me.
"pack up and leave high cost of living areas or the US entirely."
The lower cost domestic areas still can't compete with the labor cost of low cost countries. Most low cost countries have tradeoffs when it comes to rights and freedoms enjoyed in most developed (high cost) counties.
Incredible comment. I live on biz side of insurance but use tech/automation skills all the time. My industry should have solved so many problems so many years ago.
But it didn't because of exactly what you said: "how lazy people are about learning and adopting them"
Except in other industries it is usually the lazier people who got the power to fix the things as they were like 30 years ago. While in IT they got washed out quickly.
Have we forgotten that COBOL was going to eliminate the need for programmers (even though FORTRAN and ALGOL had both somehow _increased_ the need (not to mention LISP, which for some reason mostly no one did at the time (though it still got more attention than APL)))?
But then BASIC came along and really finished the job. Now _anybody_ could program, without needing all that specialized training. It was just a skill, not a job!
Jason has nailed it. If he were older, his list of vignettes might be longer, but the point would remain the same.
"The dream of the widespread, ubiquitous internet came true, and there were very few fatalities. Some businesses died, but it was more glacial than volcanic in time scale. When ubiquitous online services became commonplace it just felt mundane. It didn’t feel forced. It was the opposite of the dot com boom just five years later: the internet is here and we’re here to build a solid business within it in contrast with we should put this solid business on the internet somehow, because it’s coming."
Just to react to the "i automated myself out of a Job" part: happened to me at my first job, as we automated more and more our deployment, we could take more and more clients, and I ended up spending 90% of my time fixing routing issues, onboarding clients, integrating their ETLs or inhouse software, or fixing their "chmod -R 777 /" and other mistakes. Which wasn't an issue when it was 30%, or even 50% of my job to be clear, but became extremely boring and soulcrushing at the end.
I'm still happy i automated stuff, that was the interesting part of the job,
Sounds like your company should have hired some (cheaper) ops folks to handle that stuff, and given you bigger projects to take on.
My current company is doing exactly that type of thing. We have tons of data imports from different clients that have to run regularly, and my team started out writing the ETLs and troubleshooting stuff, but eventually we were able to write out playbooks for how slightly less-technical folks can do this stuff. We're now trying to automate more tools and processes for those folks, so they can take on more clients without needing more people.
Software Engineering isn’t a profession. Software Development is. Software development as a profession may wane and morph, due to advancements in technology and other creations.
I don’t know any engineer who has ever said “engineering is a dead end”. Because that’s an obviously nonsensical statement. So, engineering stands on its own for time immemorial.
And no - I’m not nitpicking over terminology. Learn engineering.
> “Coding is over, with Object Oriented programming one person who is much smarter than any of us could hope to be will develop the library just once and we will all use it going forward, forever. Once a problem is solved it never needs solving again.
Every time a new tool is made, more complicated things come within reach. And then you need a guy who can use the new tool. Who is that guy going to be? He's gonna be the guy who already knows how to use the previous tool.
Have to say, love the typography on this site. Did some quick digging and the headings font is an Adobe font called Forevs, with a similar Google font being Fraunces.
This should be taken in the context of software demand versus production. Demand has risen steadily over these years and efficiency gains in production just unlocked more demand because it lowered the cost. I think LLM coding will also unlock more demand but only up to a point until unit cost is so low it doesn't make sense to have a human in the loop anymore. Then we software devs are SOL.
"We need people who understand how computers work more than ever."
In small numbers, yes. In current/large numbers, maybe not. Do college students need to understand language, grammar, or the subject to write B grade papers? No, they can just prompt an LLM to do it for them. Same thing for basic CRUD apps and websites. We will always need people who understand computers, but it seems likely that the proportion of the overall IT employees that need to know how it works will approach a horizontal asymptote.
Sure, they can do it for them - but the purpose of college is not to write papers. The papers are so that the students can demonstrate that they understand the subject and that they have "learned to learn". If an LLM writes it for them then they haven't proven anything other than that they can prompt an LLM. Which is great if your college degree is for "LLM prompting", but not much else.
I hope people that use LLMs to generate papers fail in other tests, else the value of a degree will be reduced to nothing - it's already suffering from a lot of "inflation" due to lowered standards and oversupply. (The lowered standards are because graduation rate became a metric and a target)
"but the purpose of college is not to write papers. The papers are so that the students can demonstrate that they understand the subject and that they have "learned to learn"."
That depends on the perspective. In theory, that is the correct view. To many, the degree is just a piece of paper used to gatekeep jobs.
It's not just a piece of paper - it's a piece of paper that shows you can complete a 4-year course of study in a particular field. I'm not sure how useful a piece of paper that says you spent 4 years typing prompts into a LLM could be without a huge change in how education works.
I'm not sure how you think those two things are different. Right now, we have people graduating a 4 year course of study in a particular field and spending a large part of that 4 years using LLMs to do a lot of the work.
How would the person promting the LLM know if it were producing "B" grade results? That's the biggest issue right now: you have to know more than the LLM to verify its output or you're likely to get garbage.
That's also my biggest concern is that corporations will decide that crap results are good enough as long as it cost them very little to produce.
The title of the blog post is "A Series of Vignettes From My Childhood and Early Career". The heading for the first section is "The Death of Software Engineering as a Profession". It makes sense to me to use the title of the blog post, not one of the section headings.
I’ll be honest, I’m confused by the ending. It says this is a series of jabs at LLM hype, and I understand that it’s intended as a series of jabs against those who would say “software engineers will be out of a job”, but I’ll be honest I think only a loud minority are saying that yet it’s treated like it’s a majority position.
All the stories listed seem interesting, but none of them seem all that relevant.
I feel like most people understand that this is a seismic shift in abstraction layer, but intelligent people will still be in demand to manage the machines at whatever level is currently highest. The motor car didn’t kill taxi drivers, unless those who drove a carriage refused to learn how to drive a motor car.
Perhaps I’m not expressing my point very well… but this feels like both an argument against something almost no one is saying seriously, and it uses examples that also aren’t that applicable to the current situation other than having the commonality that people have said before that software engineers will die out. Make me wonder… How many times did people think an invention would kill off a job incorrectly, until one day it actually did?
Intelligent and well educated people will always be in demand somewhere. Until we’re in some post money utopia, we’ll just have to roll with the punches. In the meantime, HN readers like ourselves will simultaneously upvote any article that says humans are super necessary down at lower levels of abstraction and are way better at coding than LLMs, whilst quietly also coding less and less by hand and crawling up that abstraction layer themselves. That’s just human nature.
Forgive me but a lot of the examples seem like strawman.
> The dream of “multimedia” became commonplace and everyone just accepted it as normal. I’m not aware of any industries that collapsed dramatically due to multimedia.
But "multimedia" was never purported to be something that would lead to collapse of any segment of the industry, much less industries. If anything, the multimedia hype was purported to increase IT work which it did for some years.
> In 2000 a coworker took me aside and showed me his brand-new copy of IntelliJ IDE. “It’s over for us,” he said, “this thing makes it so programmers aren’t strictly necessary, like one person can operate this tool and they can lay the rest of us off.”
I've a hard time believing this. Literally nobody I've met was ever mistaken that IntelliJ would mean the doom of software engineering work. It's a great IDE and all IDE including IntelliJ required engineers to write code with them. Nobody was foolish enough to really think one engineer or one manager or one salesperson can "operate" IntelliJ and generate all the code to meet business requirements.
> And then he showed me the killer feature “that’s going to get us all out of a job:” the refactoring tools.
I'll bet there was no such "coworker". No sane person would think "refactoring" could mean "magically understand business requirements and write code"? All of this sounds like strawman setup so that the author could go on to making their next point like the bit where he challenged his "coworker" and asked if refactoring tools can write new code.
Don't get me wrong. The rest of the post is on money though. I just think the post would do better without these fake stories to set up strawmans only to take them down. Feels a bit forced!
Just because people make stuff up on certain corners of the Internet doesn't mean everyone does. Occam's Razor applies here: some guy who has been in the industry 25 years doesn't have any good stories to illustrate his point (how did he reach the conclusion then?) so he makes one up, writes a whole blog post and then someone else posts it here-- what's the endgame?
Is it not obvious? The endgame is to publish a blogpost that sounds interesting. That itself may be the reward and the endgame for the author.
Someone who has worked for 25 years would definitely have stories. I've worked far longer than that. I've got stories too. But none of the stories are as ridiculous as a "coworker" thinking that of all things IntelliJ would lead to the doom of software engineering or multimedia would lead to the collapse of an industry. I mean, what the heck! How does one even go from "multimedia" to "collapse of an industry". There is no logical connection. On the other hand, "multimedia" creates more work in the industry.
So my application of Occam's Razor tells me, surely the stories are made up to set up strawman that they can take down one by one to write an interesting blog post.
But I'll reiterate that it's a good post. I can like a post and find issues with it at the same time. I find all the strawman a bit forced which detracts from the reading experience.
"The Death of Software Engineering as a Profession" ... is greatly exaggerated.
Could be the title of the piece.
I agree: throughout my own career as a programmer (I prefer the more blue-collar sounding term—it better fits my skill set) I have also seen large changes in the industry that certainly made waves, did not capsize the profession.
At the same time, the profession I retired from was by no means the profession I entered into in the '90s. I confess I liked the older profession better.
Its just the title, I have read the post texts before posting, he actually says its here to say dispite mainstream claiming coding is dead every other five year.
Yea I think Calhoun also read the article before posting, and agrees with you; they meant the full title should be
> "The Death of Software Engineering as a Profession" ... is greatly exaggerated.
As in "is greatly exaggerated" is part of the title
They did not mean the title should be
> The Death of Software Engineering as a Profession
and that they think the thesis is greatly exaggerated.
For today’s lucky 10,000– @JKCalhoun is most likely making reference here to American humorist Mark Twain, who, in response to news stories claiming he had died, was reported to have quipped “reports of my death have been greatly exaggerated.”
[0] https://quoteinvestigator.com/2024/06/07/report-death/
Software engineering isn’t dead and won’t be dead from LLMs, I have no fear of that, but coding by hand may become as usually unnecessary as coding in assembly is now.
And I think that’s ok.
Besides what Sam Altman wants everyone to believe, there isn't a lot of evidence what you're saying is true. My experience with LLMs hasn't borne it out, and I also don't think it's okay -- I LIKE writing software!
I posted this link after reading all the way. Jason actually makes a good point - its just that this title is loud. Blog post itself isn't claiming the death of software engineering at all. If anything, it just shows that every five or ten years someone claims software engineering is dead.
Its not dead at all and it wont die either.
Why? chagpt, or figma or v0 can spin up a few pages of brochure site, even some blog posting level web apps, basic cruds you know. But I don't think it will replace full software engineering.
I work with a large codebase, thats almost 30 years old, multiple framework ( backbone, react, angular) and then java, python for backends. All from different phases and everything is stitched together to make it work, and have a well profit making business going on. There is no model or chatxyz that can dig throug all these connected apps and services and replace our engineering team. It helps us here and there- yeah a lot.
Isn't this a bit revisionist? I started to become interested in programming around late 90s and I don't remember anyone floating the idea that OOP, libraries or IDEs will make programming obsolete as a profession. If anything, pre-2023 most programmers considered their job as one of the hardest ones to automate.
> I started to become interested in programming around late 90s and I don't remember anyone floating the idea that OOP, libraries or IDEs will make programming obsolete as a profession.
The version of this hype that I remember from circa 2004 was UML[1] was going to make most programming automated. You'd have an architect that would draw out your problem's architecture in a GUI[2], press a button to automate all the code to build that architecture, and have a programmer fill in a couple dozen lines of business logic. Boom, program done by two or three people in a couple weeks, let's all go home. It uh, didn't work out that way.
You can read a lot more about all this by following the various links to concepts & products from Rational's Wikipedia page, https://en.wikipedia.org/wiki/Rational_Software (the Rational Unified Process page in particular brings back some memories). It wasn't badly intentioned, but it was a bit of a phase that the industry went through that ultimately didn't work out.
[1] https://en.wikipedia.org/wiki/UML
[2] https://en.wikipedia.org/wiki/File:Component-based-Software-...
It's interesting that even with the rise of transformers, UML still isn't popular. I wonder if that, or some other visual way of representing specs, might make a comeback.
There was definitely a widely held belief in the late 90s, early 00s that programming was commoditized to the point that it would be fully offshored to the lowest cost of labor. This happened in some areas and failed. It still happens now and then. But I remember hearing some of that based on OO and libraries making it so unskilled people could just put together legos.
I remember that. I studied CS in that period and some professors were convinced that software development was going to become an unskilled job, analogous to bricklaying, and that our goal as future CS graduates should be to become managers, just like someone that studies a university degree about making buildings is intended to become an architect and not a bricklayer.
I never believed it, though (if I had, I would probably have switched degrees, as I hate management). And while the belief was common, my impression is that it was only so among people who didn't code much. The details on how it would happen were always highly handwavy and people defending that view had a tendency to ignore any software beyond standard CRUD apps.
In contrast, if I had to choose a degree right now, I'd probably avoid CS (or at most study it out of passion, like one could study English philology or something, but without much hope of it being a safe choice for my career). I think the prospects for programmers in the LLM era look much scarier, and the threats look much more real, than they ever did in that period.
The bigger issue is that so many people have jumped into CS because programming (not the same thing I know) has become seen as this thing that will earn you big bucks.
Of course, some level of computer skills is important in most professions at this point. But logic suggests that CS (and programming) compensation will level out at a level comparable to similarly skilled technical professions.
It's a bit too generalizing that it failed and happens "now and then", offshoring is a multi-billion industry employing millions of people.
And the "unskilled people putting together legos" is also very much a thing in the form of low/no-code platforms, from my own circles there's Mendix and Tibco, arguably SAP, and probably a heap more. Arguably (my favorite word atm) it's also still true in most software development because outside of coding business logic, most heavy lifting is done by the language's SDK and 3rd party libraries.
>I don't remember anyone floating the idea that OOP, libraries or IDEs
Oh man, you had it lucky. Object databases were going to replace SQL multiple times, XML would eat the world and I strongly remember a UX person taking one look at Ruby on Rails at maybe 1.0 and declaring he would not be needing us programmers anymore.
> Object databases were going to replace SQL multiple times
Someone should honestly write a history of Zope. Can't say I hated it, because it was my first paid job as a programming working with it, but it was special, very special, let's call it that. 20 years from that moment I do think though that it had its good moments, even bright ideas in some places, too bad that they didn't scale.
It's an idea that surfaces every few years - back in 1981 I can remember reading about "The Last One" - named because it was supposed to be the last computer program that would ever need to be written:
https://en.wikipedia.org/wiki/The_Last_One_(software)
It had definitely been a thing even from the 80s, though everyone always seemed to think it would be the next generation of languages that did it. For example, in 1981, James Martin wrote a book called "Application Development Without Programmers".
I find the idea that IntelliJ being a job killer hard to believe, just like when some of my colleagues used to think Dreamweaver would wipe out frontend development - or 'HTML slicing', as we called it back then.
Yeah, that's weird; IntelliJ was more like "this is how amazing and friction-free Java development and refactoring can be". Enabling more ppl to be 10x programmers, not putting ppl out of work.
I remember reading 25-30 years ago about how 4GLs and object libraries were going to democratize software creation. Don't recall it being sold as an apocalypse for coders, though.
Yeah, I can confirm, before LLMs I definitely thought coding would be the last thing to go.
> before LLMs I definitely thought coding would be the last thing to go.
While LLMs do still struggle to produce high quality code as a function of prompt quality and available training data, many human software developers are surprised that LLMs (software) can generate quality software at all.
I wonder to what extent this surprise is because people tend to think very deeply when writing software and assume thinking and "reasoning" are what produce quality software. What if the experience of "thinking" and "reasoning" are epiphenomena of the physical statistical models present in the connections of our brains?
This is an unsolved and ancient philosophical problem (i.e. the problem of duality) of whether consciousness and free will affect the physical world. If we live in a materialist universe where matter and the laws of physics are unaffected by consciousness then "thinking", "reasoning", and "free will" are purely subjective. In such a view, subjective experience attends material changes in the world but does not affect the material world.
Software developers surprised by the capabilities of software (LLMs) to write software might not be so surprised if they understood consciousness as an epiphenomenon of materiality. Just as words do not cause diaphragms to compress lungs to move air past vocal cords and propagate air vibrations, perhaps the thoughts that attend action (including the production of words) are not the motive force of those actions.
> I wonder to what extent this surprise is because people tend to think very deeply when writing software and assume thinking and "reasoning" are what produce quality software.
It takes deep thought and reasoning to produce good code. LLMs don't think or reason. They don't have to though because humans have done all of that for them. They just have to regurgitate what humans have already done. Everything good an LLM outputs came from the minds of humans who did all the real work. Sometimes they can assemble bits of human generated code in ways that do something useful, just like someone copying and pasting code out of stack exchange without understanding any of it can sometimes slap something together that does something useful.
LLMs are a neat party trick, and it can be surprising to see what they do and fun to see where they fail, but it all says very little about what it means to think and reason or even what it means to write software.
I'm still of the opinion that coding will be the last thing to go. LLMs are an enabler, sure, but until they integrate some form of neuroplasticity they're stuck working on Memento-guy-sized chunks of code. They need a human programmer to provide long context.
Maybe some new technique will change that, but it's not guaranteed. At this point I think we can safely surmise that scaling isn't the answer.
I’m more inclined to believe that no jobs (as in trades, professions) will go, but programming will be the most automated, along with design and illustration.
Why? To this day still they’re the showcase of what LLMs “can” do for (to) a line of work, but they’re the only ones with all the relevant information online.
For programming, there’s decades of textbooks, online docs, bug tracker tickets, source code repositories, troubleshooting on forums, all laying out how a profession is exercised from start to finish.
There’s hardly a fraction of this to automate the tasks of the average Joe who does some paperwork the model has never seen, who’s applying some rough procedures we would call “heuristics” to some spreadsheets and emails, and has to escalate to his supervisor for things out of code several times a day.
Beyond training data availability, it's always easiest to automate what you understand. Since software engineering is a subset of the discipline of AI/LLMs, it has been automated to the extent that it has. Everything else involves more domain knowledge.
I'm not sure what happens when you replace coders with 'prompt generalists' and the output has non-trivial bugs. What do you do then? The product is crashing and the business is losing money? Or a security bug? You can't just tell llm's "oh wait what you made is bad, make it better." At a certain point, that's the best it can make. And if you dont understand the security or engineering issue behind the bug, even if the llm can fix this, you don't have the skills to prompt it correctly to do so.
I see tech as 'the king's guard' of capitalism. They'll be the last to go because at the end of the day, they need to be able to serve the king. 'Prompt generalists' are like replacing the king's guard with a bunch of pampered royals who 'once visited a battlefield.' Its just not going to work when someone comes at the king.
> You can't just tell llm's "oh wait what you made is bad, make it better." At a certain point, that's the best it can make. And if you dont understand the security or engineering issue behind the bug, even if the llm can fix this, you don't have the skills to prompt it correctly to do so.
In that case, the idea is that you'd see most programmers in the company replaced by a much smaller group of prompt generalists who work for peanuts, while the company keeps on a handful of people who actually know how to program and do nothing all day long but debug AI written code.
When things crash or a security issue comes up they bring in the team of programmers, but since they only need a small number of them to get the AI code working again most programmers would be out of a job. High numbers of people who actually like touching code for a living will compete for the very small number of jobs available driving down wages.
In the long term, this would be bad because a lot of talented coders won't be satisfied being QA to AI slop and will move on to other passions. Everything AI knows it learned from people who had the skill to do great things, but once all the programmers are just debugging garbage AI code there will be fewer programmers doing clever things and posting their code for AI to scrape and regurgitate. Tech will stagnate since AI can't come up with anything new and will only have its own slop to learn from.
Personally, I doubt it'll happen that way. I'm skeptical that LLMs will become good enough to be a real threat. Eventually the AI bubble will burst as companies realize that chatbots aren't ever going to be AGI, will never get good enough to replace most of their employees, and once they see that they're still going to be stuck paying the peasant class things will slowly get back to normal.
In the meantime, expect random layoff and rehires (at lower wages) as companies try and fail to replace their pesky human workers with AI, and expect AI to be increasingly shoehorned into places it has no business being and screwing things up making your life harder in new and frustrating ways
Seems like groundhog day, although I'm not sure I remember anyone telling me that software engineers were on borrowed time until relatively recently and I'd largely ignored it.
Yet one thing does seem different for anyone who just missed the dotcom crash, is that the roles available have fallen off a cliff while the numbers looking for roles seem to be up, at least in the UK. The UAE is even worse. I've spent 20 years hiding from recruiters and now they're all leaving me on read. Karma, maybe.
Same thing happened after dotcom. Same thing happened after 2008. Same thing will happen now during AI and the trump recession.
After every downturn ends, there comes a sudden hunger for engineers, and companies can’t seem to get enough. Some companies will even hire engineers just so other companies don’t hire them. Be ready.
I think the 2008 downturn (when it comes to this profession) was a little bit less severe compared to just after dotcom, or so it felt here in Eastern Europe.
But, to be honest, I do miss the "just after dotcom crash" period, there were lots and lots of interesting things people were working on back then, and the majority of it didn't involve making money. Maybe this is just nostalgia, because I first became a paid programmer in 2005, but that's how it feels 20+ years from that time.
I changed jobs in '08, '10 and was contracting by '11. Didn't notice any downturn from 2008 at all.
I'd only once in 20 years been turned down for a job I'd applied for. Every other job I applied for I was offered. I've applied for over 600 roles in the past year and barely had a handful of interviews. That certainly feels different.
Now's a good time to look for the less exciting but stable companies in recession-proof industries. Doesn't even mean you need to look for lower pay.
I was, for a long time, scared of my future due to the low/no-code, automation, LLMs, outsourcing, etc. Until, at some point, I realised something simple - the risk factor for my job is not determined by how good new tools are, but only by how lazy people are about learning and adopting them. And here history gives another lesson - we never learn, eternal cycle of mistakes will continue.
The outsourcing is the only real threat in your list. In the past, we have eliminated jobs in the primary (eg farming) and secondy (eg manufacturing) industries through automation and outsourcing with the goal of moving workers into higher level industries, including tertiary industries (eg software). If we are outsourcing tertiary industry jobs as well, what does that leave us?
The US outsources something like 300k jobs annually, with over half of these being IT jobs. Adding 10k IT jobs per month could change the employment numbers and economic outlook we've been seeing lately. It seems like we're in a race to the bottom. I do think AI will make things worse, economically at least, with the reduction in jobs. But this could be offset by policies promoting on-shore employment.
For those who are young, single, and are open to adventure, the endless outsourcing provides another option: pack up and leave high cost of living areas or the US entirely. There are plenty of places where you can get a fraction of a US salary while living a quality of life beyond what the US offers. US companies have return to office mandates so they can fire US workers. But if you're a US citizen living abroad and willing to accept 1/4th the salary of a person in California (and living in a place with 1/10th the living costs), companies get the best of both worlds: an employee they feel they can trust while also undercutting wages. Yeah, it sucks for people still in the US, but it's been great for me.
"pack up and leave high cost of living areas or the US entirely."
The lower cost domestic areas still can't compete with the labor cost of low cost countries. Most low cost countries have tradeoffs when it comes to rights and freedoms enjoyed in most developed (high cost) counties.
It leaves us nothing :)
Incredible comment. I live on biz side of insurance but use tech/automation skills all the time. My industry should have solved so many problems so many years ago.
But it didn't because of exactly what you said: "how lazy people are about learning and adopting them"
Except in other industries it is usually the lazier people who got the power to fix the things as they were like 30 years ago. While in IT they got washed out quickly.
Have we forgotten that COBOL was going to eliminate the need for programmers (even though FORTRAN and ALGOL had both somehow _increased_ the need (not to mention LISP, which for some reason mostly no one did at the time (though it still got more attention than APL)))?
But then BASIC came along and really finished the job. Now _anybody_ could program, without needing all that specialized training. It was just a skill, not a job!
Jason has nailed it. If he were older, his list of vignettes might be longer, but the point would remain the same.
Ahh - so many gems!
"The dream of the widespread, ubiquitous internet came true, and there were very few fatalities. Some businesses died, but it was more glacial than volcanic in time scale. When ubiquitous online services became commonplace it just felt mundane. It didn’t feel forced. It was the opposite of the dot com boom just five years later: the internet is here and we’re here to build a solid business within it in contrast with we should put this solid business on the internet somehow, because it’s coming."
Yes. And it continues on.
Just to react to the "i automated myself out of a Job" part: happened to me at my first job, as we automated more and more our deployment, we could take more and more clients, and I ended up spending 90% of my time fixing routing issues, onboarding clients, integrating their ETLs or inhouse software, or fixing their "chmod -R 777 /" and other mistakes. Which wasn't an issue when it was 30%, or even 50% of my job to be clear, but became extremely boring and soulcrushing at the end.
I'm still happy i automated stuff, that was the interesting part of the job,
Sounds like your company should have hired some (cheaper) ops folks to handle that stuff, and given you bigger projects to take on.
My current company is doing exactly that type of thing. We have tons of data imports from different clients that have to run regularly, and my team started out writing the ETLs and troubleshooting stuff, but eventually we were able to write out playbooks for how slightly less-technical folks can do this stuff. We're now trying to automate more tools and processes for those folks, so they can take on more clients without needing more people.
Not sure what to make of the post.
Software Engineering isn’t a profession. Software Development is. Software development as a profession may wane and morph, due to advancements in technology and other creations.
I don’t know any engineer who has ever said “engineering is a dead end”. Because that’s an obviously nonsensical statement. So, engineering stands on its own for time immemorial.
And no - I’m not nitpicking over terminology. Learn engineering.
What profession are engineers like mechanical or chemical in?
These days, software development
There is one and only one important question... have companies been hiring as many juniors as in the past recently?
And is that a transient phenomenon or a secular shift?
I thought this was Jason Schreier and got confused for a minute.
> “Coding is over, with Object Oriented programming one person who is much smarter than any of us could hope to be will develop the library just once and we will all use it going forward, forever. Once a problem is solved it never needs solving again.
This kind of happened, and it was a good thing.
I don't see how that requires OOP, as one can write a library of a series of standalone functions too and distribute that. The two are orthogonal.
Every time a new tool is made, more complicated things come within reach. And then you need a guy who can use the new tool. Who is that guy going to be? He's gonna be the guy who already knows how to use the previous tool.
I still love programming. Even more so after trying out llm coding in some projects.
Have to say, love the typography on this site. Did some quick digging and the headings font is an Adobe font called Forevs, with a similar Google font being Fraunces.
This should be taken in the context of software demand versus production. Demand has risen steadily over these years and efficiency gains in production just unlocked more demand because it lowered the cost. I think LLM coding will also unlock more demand but only up to a point until unit cost is so low it doesn't make sense to have a human in the loop anymore. Then we software devs are SOL.
> New problem sets not covered by the garden path come up all the time.
I believe the appropriate term is “happy path”, not “garden path”.
The idiom "garden path" typically refers to a misleading path, as in "He led me down the garden path" (to take advantage). See https://en.wikipedia.org/wiki/Garden-path_sentence.
Yes, I know. This is why the term “garden path” does not fit in that sentence.
I was yes-anding you, not no-butting.
I see. You could have made that more clear by prepending something like ”For those who don’t know,” to the start of your comments.
shrug I thought it was pretty clear from the context that I was expanding upon your point, but I'm not interested in starting flame wars about this.
Well, I misinterpreted your comment. Much like you seem to misinterpret my comments as starting a flame war. Oh well.
I really like the font this author is using for the headers. Reminds me of the fonts used on covers of 1970s scifi novels.
The ongoing issue is the maintenance.
This can't be solved without fully trusting the LLM period.
Just don't autopilot on important code you want to own. That's good start.
One day "real programmers" will be gold.
Once we understand demand set the price we'll understand why our "career" is dead. Thank me later.
I don't think the career is going anywhere unless the career just consists of typing. We need people who understand how computers work more than ever.
"We need people who understand how computers work more than ever."
In small numbers, yes. In current/large numbers, maybe not. Do college students need to understand language, grammar, or the subject to write B grade papers? No, they can just prompt an LLM to do it for them. Same thing for basic CRUD apps and websites. We will always need people who understand computers, but it seems likely that the proportion of the overall IT employees that need to know how it works will approach a horizontal asymptote.
Sure, they can do it for them - but the purpose of college is not to write papers. The papers are so that the students can demonstrate that they understand the subject and that they have "learned to learn". If an LLM writes it for them then they haven't proven anything other than that they can prompt an LLM. Which is great if your college degree is for "LLM prompting", but not much else.
I hope people that use LLMs to generate papers fail in other tests, else the value of a degree will be reduced to nothing - it's already suffering from a lot of "inflation" due to lowered standards and oversupply. (The lowered standards are because graduation rate became a metric and a target)
"but the purpose of college is not to write papers. The papers are so that the students can demonstrate that they understand the subject and that they have "learned to learn"."
That depends on the perspective. In theory, that is the correct view. To many, the degree is just a piece of paper used to gatekeep jobs.
It's not just a piece of paper - it's a piece of paper that shows you can complete a 4-year course of study in a particular field. I'm not sure how useful a piece of paper that says you spent 4 years typing prompts into a LLM could be without a huge change in how education works.
I'm not sure how you think those two things are different. Right now, we have people graduating a 4 year course of study in a particular field and spending a large part of that 4 years using LLMs to do a lot of the work.
How would the person promting the LLM know if it were producing "B" grade results? That's the biggest issue right now: you have to know more than the LLM to verify its output or you're likely to get garbage.
That's also my biggest concern is that corporations will decide that crap results are good enough as long as it cost them very little to produce.
Many workers don't care what grade they get. The PMs just want the shit delivered.
Eh, maybe. I think it could actually be the opposite: we might need more people than ever to clean up the crap that LLMs spit out.
Exactly and arguably we will always - unless AGI
The title is “The Death of Software Engineering as a Profession", why was this changed? Not really fair to the author.
I actually posted with that title. Not sure what happend.
@dang could perhaps help?
The title of the blog post is "A Series of Vignettes From My Childhood and Early Career". The heading for the first section is "The Death of Software Engineering as a Profession". It makes sense to me to use the title of the blog post, not one of the section headings.
That's not the title, that's the subheading.
I’ll be honest, I’m confused by the ending. It says this is a series of jabs at LLM hype, and I understand that it’s intended as a series of jabs against those who would say “software engineers will be out of a job”, but I’ll be honest I think only a loud minority are saying that yet it’s treated like it’s a majority position.
All the stories listed seem interesting, but none of them seem all that relevant.
I feel like most people understand that this is a seismic shift in abstraction layer, but intelligent people will still be in demand to manage the machines at whatever level is currently highest. The motor car didn’t kill taxi drivers, unless those who drove a carriage refused to learn how to drive a motor car.
Perhaps I’m not expressing my point very well… but this feels like both an argument against something almost no one is saying seriously, and it uses examples that also aren’t that applicable to the current situation other than having the commonality that people have said before that software engineers will die out. Make me wonder… How many times did people think an invention would kill off a job incorrectly, until one day it actually did?
Intelligent and well educated people will always be in demand somewhere. Until we’re in some post money utopia, we’ll just have to roll with the punches. In the meantime, HN readers like ourselves will simultaneously upvote any article that says humans are super necessary down at lower levels of abstraction and are way better at coding than LLMs, whilst quietly also coding less and less by hand and crawling up that abstraction layer themselves. That’s just human nature.
The reports about our death were greatly exaggerated.
> This is indeed a set of passive-aggressive jabs on the continuing assault on our senses by the {{$_BUZZWORD_DU_JOUR}} hype lobby.
FTFY
[dead]
[dead]
Forgive me but a lot of the examples seem like strawman.
> The dream of “multimedia” became commonplace and everyone just accepted it as normal. I’m not aware of any industries that collapsed dramatically due to multimedia.
But "multimedia" was never purported to be something that would lead to collapse of any segment of the industry, much less industries. If anything, the multimedia hype was purported to increase IT work which it did for some years.
> In 2000 a coworker took me aside and showed me his brand-new copy of IntelliJ IDE. “It’s over for us,” he said, “this thing makes it so programmers aren’t strictly necessary, like one person can operate this tool and they can lay the rest of us off.”
I've a hard time believing this. Literally nobody I've met was ever mistaken that IntelliJ would mean the doom of software engineering work. It's a great IDE and all IDE including IntelliJ required engineers to write code with them. Nobody was foolish enough to really think one engineer or one manager or one salesperson can "operate" IntelliJ and generate all the code to meet business requirements.
> And then he showed me the killer feature “that’s going to get us all out of a job:” the refactoring tools.
I'll bet there was no such "coworker". No sane person would think "refactoring" could mean "magically understand business requirements and write code"? All of this sounds like strawman setup so that the author could go on to making their next point like the bit where he challenged his "coworker" and asked if refactoring tools can write new code.
Don't get me wrong. The rest of the post is on money though. I just think the post would do better without these fake stories to set up strawmans only to take them down. Feels a bit forced!
Just because people make stuff up on certain corners of the Internet doesn't mean everyone does. Occam's Razor applies here: some guy who has been in the industry 25 years doesn't have any good stories to illustrate his point (how did he reach the conclusion then?) so he makes one up, writes a whole blog post and then someone else posts it here-- what's the endgame?
> what's the endgame?
Is it not obvious? The endgame is to publish a blogpost that sounds interesting. That itself may be the reward and the endgame for the author.
Someone who has worked for 25 years would definitely have stories. I've worked far longer than that. I've got stories too. But none of the stories are as ridiculous as a "coworker" thinking that of all things IntelliJ would lead to the doom of software engineering or multimedia would lead to the collapse of an industry. I mean, what the heck! How does one even go from "multimedia" to "collapse of an industry". There is no logical connection. On the other hand, "multimedia" creates more work in the industry.
So my application of Occam's Razor tells me, surely the stories are made up to set up strawman that they can take down one by one to write an interesting blog post.
But I'll reiterate that it's a good post. I can like a post and find issues with it at the same time. I find all the strawman a bit forced which detracts from the reading experience.
The newspaper industry arguably collapsed due to internet multimedia.