Because programmers have generally been forced to wear additional, invisible hats that are essential to making the code happen in the first place.
Writing code is not hard. Writing correct code is. Knowing what is correct in a setting with paying customers generally involves interacting with those customers. Either directly or worse. The gigantic salaries paid to the most prolific employees is not due to their ability to write code. It is due to their ability to interrogate the shit out of the customer until they finally reveal the true requirements.
That's like saying "building a car is not hard, building a real car that you can use and that passes regulation is".
IOW, writing code is hard in every reasonable context.
Likewise, coding is easy. It's just writing, and any child can learn it. Coding is not programming, and the hard part of the job lives in that distinction.
Any teacher of a low level CS course knows this is completely untrue. In my CS 101 course our problems were relatively straightforward and short, and lots of people struggled mightily to the point of dropping the course. IMO coding requires a particular way of thinking that a large subset of people just aren't good at, and as someone whose done tons of screening interviews at college recruiting fairs where I give relatively easy problems and ask someone to code a solution, I will tell you the idea that anyone can do it is just false.
In it's original context, it's very clear what "Code was never the hard part" was referring to, that of all the steps of product development (ideation, requirements, design, implementation (coding), deployment, operations, maint, etc.), that code was not the hard part. That's the meaning we should be discussing, and I agree with the author, I believe coding very much was the hard part. Since Jeff Dean and Sanjay Ghemawat were recently in the news, I recall some stories about them fixing issues with the search index in the early days of Google where there were random bit flips that were causing the index to be corrupted, and they fixed it and coded a durable solution. This is very much just a code-specific problem that very, very few other people would have been able to solve.
The truly difficult work is knowing when your client/boss asks you to "build a car," do they really just want/need a small Pinewood derby box car as a toy, do they need a four-door sedan that can legally drive on major highways, an 18-wheeler freight truck, or do they actually need/want a bicycle, or a shopping cart, or a railroad box car, or information about how to take public transit that will serve them cheaper and easier than anything you could build in their timeframe and budget.
That's what people mean when they say that code was never the hard part.
[1]: https://missiongraduatenm.org/college-dropout-statistics/#:~...
[2]: https://www.coursmos.com/college-dropout-statistics/#:~:text...
The fact that children cannot do it does not mean that it is particularly hard, and computer science attracts a lot of people who just like video games and have no mathematical ability.
I think that the fact that programmers were/are doing a lot of the job that other positions take credit for is significant. Others can get by on bullshit, but the programmer's job in turning that bullshit into code forces them to make something real out of it through a combination of back-and-forth interrogation and just making it up when necessary. Turning handwaving into a product is a skill that programmers have, and is why AI isn't helping tech-illiterate businessmen create things that work.
Now add the time dimension - keeping code correct as the business and the people in it change.
That's how I explain to people that LLMs will not replace us developers.
Timestamp is: 36:42-39:55
I believe Graeber perfectly predicts the problems, in 2019, with vibe coding creating immediate "value" from production, but failing to produce true value through maintaining the system (like one continually washes a cup to give it value over time).
But it doesn't give value, it prevents value loss.
I don't know why people are telling themselves maintenance work is virtuous. It's waste. It's necessary waste, and doing the work may be virtuous, but the work itself is pure waste. Fighting entropy.
EDIT:
I wish we talked more about the need for low-maintenance patterns and products. In this industry, many of us already recognize this instinctively, but we often misattribute the problem to "complexity". Think of e.g. rather substantial niches and common practices among developers, like static site generators, no-build-step development, or on the backend side, the popularity of header-only libraries in C and C++. All these tend to be labeled as reducing dependencies, but that's just the means - what they do is they minimize independently rotting parts. The build system isn't bad because it's complex - it's bad because you have to constantly babysit it. Conversely, a static site once rendered will open ~forevermore, and so will a piece of C/C++ code that relies on single-header libraries.
Similarly, the popularity of containers is in large part this. All the mess isolated in a self-contained bundle that is preserved against rot, at least for a while. Inside, there's nothing to maintain - it works until it's not needed, or until the "outside world" changed too much, at which point you throw the thing away and get a new one. Etc.
It keeps thing operational. Software changes because requirements changes, the context it is used it changes, or just the iterative nature of it where features are rolled out over time so that users can immediately start getting some functionality if not all that was originally planned (MVP).
Maintenance work is neither virtuous nor waste. Its just nature of the things we build.
I agree however with the part on more talk about low-maintenance patterns and products. IMO its often the trade-off between velocity vs quality/technical debt. So it happens, industry is favoring more and more towards velocity for delivering features that often add little value to the users, just because $$$, competition and maybe the grind culture.
The few people who manage to write good quality production apps with AI are developers whether they like it or not and that number isn't high enough to obsolete existing developers.
You are conflating "not understanding how to build software" with "not knowing how to write code". Most developers I've crossed paths with couldn't build a consistent library, let alone a complete, well written, architecturally sound and useful application. Sure, I also know plenty that don't fall into that category but those are the few.
Problems with deploy? Ask claude, use ssh with key-based auth and he will take care of it :) just saying.
I've been writing code "almost daily" for the last 35 years; been doing it professionally for at least 29 years. I've been around, and my peers consider me a proficient developer. I've built stuff ranging from embedded/os level development to DSL languages, from 3D programming to VBA macros. I wrote software used by me, and wrote software used by millions. In some cases, I've maintained products written by me nore than a decade. By your definition, I must be wrong, truth is I can afford to be wrong - my job is not writing code, is designing solutions. Writing code is often the easiest part, and we're mostly automating it. Thank god.
This already goes over the heads of most non-developers.
I think you're misunderstanding my point. It's not that LLMs aren't a useful tool or that they won't replace some developers. But rather that software development as a specialty won't go away because most people can't build software with LLMs in a way that won't blow up.
(In interesting ways this is programming’s greatest boon and curse: if we treated it more like building bridges or cars, the world would be a very different place.)
Because when I submit building plans, they are manually reviewed and approved (or denied) by registered architects, engineers, and planners employed by the authority for just this purpose.
I had to make some software changes to an old medical device this year. The overwhelming majority of the effort was understanding what the customer wanted and giving them feedback into how that would change the existing system and the risks associated. Then, creating a plan to follow the necessary standard (IEC62304) and creating the associated documentation and getting it reviewed and approved.
The actual code that changed was probably only around 100 LOC but the project took several months. Heck, the code was simple enough that an intern could have done it.
(The distinction I’m making is between the code that operates the medical device and the code that operates the app I make doctors’ appointments in. The latter is subjected to a different - and lighter - regime than the former.)
There are the executives, the lawyers, the product managers, sometimes the designers, who to varying degrees determine this before they land in the requirements the programmer sees. But there are also the libraries and APIs the company pays to handle compliance so that the company and the programmer doesn't. The programmer implements the library (and may not even had a say in or necessarily care which one was chosen).
The level of quality needed (or imposed) will vary. It is a wide spectrum from dealing with banking/transactions (money at risk) to brake controllers and auto pilots (human lives at risk). But there is a lot of this, all over the world.
I work somewhere in the middle (rather slow but extremely heavy industrial equipment, where emergency stop is always a safe if costly option). There are domains where emergency stop is not a thing though: some systems on an aircraft in flight, a pacemaker, etc.
My point is though, that there is a ton of code where stakes are higher than "oops, I guess we will fix it next sprint". And while not all of that have regulatory constraints, sometimes a company realises that the financial cost of issues significant enough that it is worth holding themselves to higher standards anyway.
I mean, this was drilled into me at uni — that software was not likely to escape regulation forever and that you can't know with certainty how all the code you're writing will be used when you're not observing the use.
For example, under what constraint regime should the calculator app bundled with an OS be written? It's just a little bundled toy app. Until someone under pressure uses it to calculate a medicine dose, expecting it to be a calculator like it says.
Perhaps this gives away my age more than anything else.
This captures a sentiment I have often felt when people don't take bugs seriously. Or don't take it seriously that they introduced regressions. You should feel personal responsibility for your bugs. When your shit doesn't work, and people are trying to use it, you are basically hurting them, personally.
But it seems with the increase of AI coding, the industry is going the other direction. Nobody seems to care about bugs introduced by slop coding. Except perhaps the users.
Communicating like this is to try to get them to understand that The Code is barely about the text on the screen and is instead about much more - both abstract in the code (but how on earth do you explain that to someone nontechnical without just sounding like "no trust me my job is really hard, I promise.") and also in all the external stuff - the world The Code lives in (users, ops, support staff...).
Is it a perfect analogy? Of course not, and I'd never explain it like this to someone technical. But they're not the audience.
---
Adjacent: I've always gotten the feeling that even the most well-meaning/trusting nontechnical leaders have always been fairly nonplussed by software complexity and software development.
Deep down, they seem to think it can't possibly be that hard, despite the fact they can't write it themselves. Sometimes, even worse, they have dabbled in writing small, or even medium-sized solo projects. And think - well isn't software engineering just doing that but with other people? And their attitudes can reflect them, sometimes all the time, sometimes just slipping through when under duress like delayed projects etc.
And yet despite their attitudes, they then also find:
- if they try to outsource, they have a bad time
- if they try to proooompt, they have a bad time
- if they try to pay less, they have a bad time
and so the invisible hand of the free market itself forces their hand in paying prodigious salaries and fighting to retain talent. And all throughout they remain internally nonplussed even if they manage to keep up appearances.
This is worth saying precisely because of this difference - people see huge amounts of bad code generated by LLMs and think it is equivalent to carefully written code because the results look similar at first glance and they don’t bother to read it.
That said, it's still reasonable to think that once (and only once) the hard part is done, then the easy part is easy. It would just be wrong to think that you can do the whole thing without having to do hard parts. I think the argument here is that with AI this is more possible—that the tasks are more separable. Even if it's the same one person doing the hard stuff and then passing what they learned to the AI to do the easy stuff.
Two reasons they might not separate well (there are others):
- If in a company's product development it's hard(er) to have one person doing H and another doing E, then you're generally going to have one person doing H and E. More or less, this means a person can only do E easily if they do H beforehand. So hard things are required no matter what.
- People come in whole persons. If people skilled/educated to do H tend to be the same people skilled/educated to do E (can be because of how programming is educated, but also can be because there aren't that many programmers), then you're always going to be plugging programmers who have both H and E into roles and it'll probably be more efficient to plug them into roles requiring both rather than just H or just E. You could, within that population, determine who is comparatively advantaged (and we do do this mildly with senior vs junior or with "architects"), but plugging a person into an E-only role is going to involve that person questioning what happened during the H part beforehand. In part because they're good enough at H to question it, but more importantly because they're implementing the H such that they're aware when their E might not be as easy as it could be.
Both of these seem like they might be less true now with AI (and also perhaps because there are more programmers).
Even in academia, you have meta structures that you constantly need to think about.
Of course, you can keep trying to isolate the "pure" thing from the "accidentals", but it's not going to work when the work gets sufficiently complex
For those cases where it is true, it is pretty much like your example, and the distinction is there in your example just like theirs. You supported their point.
Invoice: $1000
One bolt tightened: $1
Knowing which bolt to tighten: $999
Bit of both probably. I've seen really awful code in my time, so would say "actually coding well" is indeed one of the hard parts.
But knowing what the real problem to be solved is, is indeed important. (Isn't that what sales is? Working with the customer to tease out the real thing they need solving?)
For any product of any complexity you can can essentially never successfully delegate that process to a non-programmer.
Indeed you can not that often even trust that the client employee doing the asking knows what it is they need. Almost certainly someone in the organisation who was not in the meeting is better placed to tell you what is actually required.
This task needs an analyst; that analyst needs to have experience of writing meaningful code.
While it is true that a programmer's job is a lot more than writing code, I also wonder to what extent that businesses will actually be able to tell a good programmer from a bad one. For example, a lot of folks at big companies can honestly get away with being a ticket-taking code monkey because so much of the responsibility has been abstracted away so that they don't actually get punished for not caring about the customer. It's sort of similar to how many schools realistically wouldn't care to distinguish between a teacher who puts in extra effort into their classroom versus one who clocks in and clocks out as long as some bare minimums were being met.
I think strong programmers will get rewarded in the right companies that need them, but it's still an open question as to how many companies exist that have their bottom lines actually depend on a programmer doing a good job at wearing all those extra hats.
I go half the speed of a junior developer, but the code I write lasts five years to a decade with an order of magnitude or two fewer bugs and long-term maintenance burden.
No ... those are not invisible hats ... those are the real hats.
Software is 'Knowledge Distillation' the code is the hieroglyphic artifacts.
Engineers Engineer, Scribes Scribe.
Just so happens developers do their own scribing.
If you have shop where you have the best product-owner in the world, and exact clarity on how you want to build something, how all failures are handled, all the tradeoffs, all the implementation details, all the risks, then product is simple, then your company absolutely can get away with hiring a less than top-tier engineer.
However if you're combining all of those skills/roles into one individual (a staff+ engineer) then of course it's going to be expensive.
Dude.
Writing correct code is the whole process. If you're defining coding without care for correctness, of course you can write it off as not the hard part.
Correct code is that which does what is required of it at whatever level of correctness you are examining, as was said. But it's the whole thing.
There is no sensible distinction to be drawn between writing correct code and merely writing code. The former is the only definition of the job. We shouldn't define down competence.
That doesn’t describe any devs I know, especially not in the old days. Interrogating customers and bringing back requirements was the job of management (who got big salaries, too.)
Wow, some "the killer is calling from inside the house" vibes right there. But I totally agree that the game of telephone has always been _an_ issue - maybe not _the_ issue but certainly a big one.
However, where we agree, is that there is more to engineering than writing code. It's problem solving. Even if the product team, the c-suite, the board, the investors, et al, are all in on a product that they believe customers want, doesn't mean the real problem of bringing that idea to life at scale has been solved.
Thus, I insist on engineering being involved early to put real world constraints on wild ideation ("it would take 3 months for 5 engineers" quickly changes what's a must-have :)) — sometimes, a curious, critical mind can expose things like these without having to do the research or user testing themselves.
Now, throughout my 20 year career, it's been very rare to find a product person who will both understand customers deeply, tie their needs to business value, and be able to formalize the intersection of these in a form of good requirements for design and engineering to eventually build!
So I really believe an engineer's (and design) role there is to serve as a sanity check as they dive into actual building — does this really make sense? If they do not, they run the risk of a project completely failing or perhaps not even shipping once someone else questions the value of continuing to invest in this 3 month project 9 months in. ;-)
I've heard of Sales Engineers as well, embedding programmers directly with sales teams.
But the vast majority of programmers should not be spending any significant amount of their time on this. Their value is in building and scaling well-specified systems.
Writing code has a minimum IQ requirement; a significant fraction of the population will essentially never be able to code by themselves. That labor supply limitation is what's kept programmer salaries relatively high.
Average IQ of Electrical Engineers: 121
https://www.iqcareerlab.com/tools/iq-for-profession/electric...
...which is the top 10% of the population.
https://web.archive.org/web/20120905095856/http://www.ssc.wi...
As hard as it might be for this audience to comprehend, supply is limited simply because it is an undesirable profession. The majority of the population couldn't think of anything worse to spend their time doing. Much as the same reason why they don't want to go work on oil rigs and other such work that is high paying but what most people are unwilling to do.
High compensation is actually not as strong as a motivator as you might think. It can tip the scales if someone is already on the fence, but it doesn't suddenly make someone who absolutely can't stand something to change their mind.
The difficulty of it has absolutely been a bottleneck to the supply of good devs, keeping salaries high.
Xing Y is not Z. Xing `additional adjective` Y is.
Writing prose is not hard. Writing good prose is.
Cooking food is not hard. Cooking good food is.
...
Or are forced to figure them out.
coding is always the hard part, always. whenever I was on any project and we had more work than resource we never hired “people to wear other hats” - we hired people to write code, that’s it. thats the fucking job.
you ever see a leet-code-for-understanding-requirements? yea, me either…
But really, any interview is really about giving you a requirement, and seeing _how_ you understand it, how you clarify it with your stakeholders (interviewers), and then how you address them.
If you can't/won't understand what you're supposed be doing, you're either useless or making useless shit.
What I, and many people who’ve said, “code was never the hard part,” aren’t referring to the skill of an individual. It’s not the hard part of the engineering process of developing software. Programming languages have manuals. Many data structures are well documented. There are frameworks for damn near everything. While the difficulty of producing code varies by the skill of the programmer and the complexity of the problem domain; writing and understanding the code is a tractable and straight-forward problem. I can and have taught many people. People can learn.
What most people are referring to is that the hardest parts of producing software are all the things an organization has to do in the production of it. It’s not writing the code that is the hardest part for an organization. It’s getting everyone to understand the problems, working together, gathering requirements, developing specifications, validating releases, testing, etc. It can often look like herding cats and is probably harder.
And typically (though not always) product managers have better people skills than programmers and consequently they might be more effective at the "gathering requirements" and "working together" bits you mentioned above.
Even much of the "validating releases" and "testing" parts should also be things that a decent PM should be able to wrangle now by themselves with some LLM agents to assist them. Afterall, why bother about code quality of a testing harness. So long as the PM can keep a coherent test case list and have end to end tests that cover them, programmers can leave that to them as well.
Highly experienced programmers seem to have a hard time understanding this.
Why do we have tens of thousands of test cases and we still find new errors constantly? Why is our software so bloated and slow? We do we still have security breaches?
LLMs can be useful tools when guided by experts. I don’t think a PM with a dream is going to cut it in the long run.
It also depends on how you model the problem. You can easily say the hardest part is hiring people if you are the boss. Since the people you hire can do everything else that needs to be done.
Managing and motivating people is the hardest part for the person who is doing it.
If you are hiring you might say it is harder to find good managers than programmers.
It is not measurable who did a good job at what as almost everything requires a group of people doing different things.
You can see how pointless this is becoming as we don’t have a measure for anything.
This kind of problem requires assumptions because it doesn’t hing on anything natural.
For example, if you start by believing salary indicates value then you can go from there.
In the end there are millions of managers, millions of programmers, millions of ux designers etc. It is kind of funny to suggest doing any of these is inherently harder than the other.
Just imagine you are judging a project. You have everything about it recorded. How hard do you think it would be to judge who had more part in the outcome in what way? If 10 people judged it separately, how many would have similar opinions etc.
It is impossible to judge even for a specific case, so it is a joke to consider to find the universal rule for it.
In the end it is ok to believe something but it is also important to not forget that it is a belief
There are manuals for this too, and interestingly enough this has been studied since the Romans at least! Is it then really the hard part?
Whether it’s commercial software company or an internal team writing custom software, the return on that investment depends on many things outside of the code itself.
They’re orthogonal to AI and to the actual hard technical skills needed to execute on a specific strategy. And if the technical skills are lacking, it doesn’t even matter how good an organisation is at collaboration, whereas hard skills plus organisational disfunction are a known successful pattern :)
Many people did look at this through an individual lens and claimed that design skills, domain knowledge are the truly important abilities. I remember reading on HN at least a couple of popular articles claiming that. Actually, they’re all important and having great design skills without matching coding skills is IMO not really possible. The code feeds into the design, the requirements, the architecture and shapes them.
But these are skills that can be taught to individuals.
But teaching an organization that their real bottleneck isn’t how fast they’re writing code; it’s producing production-ready software that people understand and are willing to take responsibility for… that’s much harder.
Many businesses want to treat software development like an assembly line and revert back to Taylorism. It’s knowledge work and there’s no royal road. Good teams get fast when they have the right mix of skills and trust from the organization.
What we’ve been delving into for the last decade has been a decline in the value of labour and work.
“Code isn’t the hard part,” isn’t meant as an insult at individual programmers or to devalue their work. That’s being done by big tech and their AI hype machine.
The real conclusion is that programming is such a high-leverage activity that even technically trivial, low-quality programming is immensely valuable economically. That's not going anywhere, but maybe LLMs are going to make it all that much cheaper. (Which is mostly great! But I really don't look forward to the painful debugging and maintenance that reams of shit code will push down on programmers.)
But also, there still is a ton of programming that is fundamentally difficult. That's not going anywhere either. And LLMs are useful there too, but they're currently nowhere near replacing the expertise needed to do novel and non-trivial technical work.
In an ideal world, making mediocre code cheaper should leave more room for taking on harder technical challenges. In reality, this has always been dictated far more by non-technical factors—culture, leadership, trust, risk tolerance...—than by anything intrinsic to programming. But, at least for now, we can use the LLM hype to motivate the kind of deeper technical work that always made sense but was too uncertain or too open-ended or too long-term for non-technical leadership.
And we should also drop the bullshit "code was never the hard part" framing.
Genuine question and not trying to be snarky here, I am actually curious: what fields or types of programming does this apply too? I think I've read anecdotes online about people in fields I previously (a few years ago lol) thought "oh yea an llm will never be able to help with that" and now see articles about how llm's are doing just that.
Much of the truly LLM-difficult code is probably hiding in the libraries we import. Database engines, compilers, efficient data parser, control theory, signal processing, protocol implement-ions, or anything with a 12000 page German ISO standard that need to pass a $12.000 certification lab. But this also compose of such a tiny fraction of programmers or code in the world.
A-lot of my work lies in that last one... but that's also where that "code is easy, knowing what to code isn't" is the most true; because industrial standards tend to not spare any expense on the word count, while the implementation is a ~2000 row state machine. I've not yet found an LLM capable of successfully parsing this kind of specification documents, but it's possible they will reach there eventually.
But i do feel online debate do clump the software field a bit too much when AI is discussed. JavaScript compose probably 98% of all code the LLMs are trained on since it's powering every website scraped for training. As such, people in web-development seem to have far more praise to LLM capability than i'm able to give.
My personal AI experience has been very mixed in comparison, regularly making up functions of common libraries, hallucinate the description of technical terms, straight up writing un-compilable c-code, or get confused by relatively small code-bases. Useful but not majorly changing my work at the moment (pretty good at comments, test cases, or as google replacement).
Granted, I've only tried models up to Opus 4.8, and not had experience with the newest "tier" of models with Fable, Kimi K3 or GPT 5.6; but the prices on those are also starting to compete badly with my salary at the moment.
When you are looking at something that already exists, where all the requirements are defined, when all the edge cases have been decided, then coding was the easy part. People didn't "burn out" because it was difficult to figure out how to write SQL. People "burned out" because the requirements constantly changed, demand was ever increasing, and edge cases were constantly being triggered.
>If deciding what to build is the hard part, why do so many product managers seem clueless?Why aren't there rigorous 10-step interviews for them?
Classic engineer type opinion where every else is dumb, except for him. So many people have come to see leetcoding as an intellectual badge of honor, when most of us know its cultural rigamarole and the code written on the job will rarely reflect the type of work that will done.
I'm not saying coding is easy, plenty of people struggle with it. But as far as the job goes, unless you are a junior just grinding through JIRA tickets, coding was the easiest (and arguably the most rewarding) part of the job.
Why exactly is it acceptable to not just do your role and expect the underlying requirements to be correct and measurable, especially when there are separate roles whose sole purpose is to do exactly that?
Not even just the customer-facing stuff either, the internal tooling is on fire more and more often. It's not sustainable in the slightest
If they are vibe coding anything, it hasn't had any bad consequences yet. The ads and bloat are human decisions!
Another kind of highly valuable software is a new design which digitizes some process, like the ticket management of a rail company. If you don't follow the internal processes and workings of the firm exactly, or you do not interface with existing systems 100% accurately, your software very soon becomes worthless.
Figuring out how to create a piece of code that solves the exact problem the customer has, while not breaking anything, and fitting into existing operations seems to be a big head scratcher still, and something humans still need to do, at least that was my experience so far with LLMs.
... as long as the buyers could never discern
Low quality ends up taking longer to build over the long run. It gets harder and harder to add or change features.
If you need it to work for at least a month or two (instead of one and done, which some demoware is like). Following a few rules early on will ensure you can keep evolving it — even if it's MVP/demoware/whatever — as long as you know you need to evolve it soon after you build it!
In tech, the era of competing based on quality is long gone. The winning strategy is to get a monopoly/oligopoly and then you can let the quality decay to zero and people will have no choice but to keep paying you money (or to your handful of equally-mediocre competitors).
Now you tell someone else your idea and have to hope they get it. Otherwise you have to argue, rephrase, start all over again.
We are back to the tree-swing project management, but we added another layer
Development is essentially becoming management.
Even with wrangling agents, you’re really just making them right the “correct code”. Something EMs don’t do, or at least the good ones don’t do with their ICs.
Encoding your ideas into a programming language is easy. Understanding that your ideas are bad is hard.
You have clients with multiple devices connecting to your backend simultaneously, while you mediate their interactions with your partner systems. Their versions might not be up to date. It's a distributed system. When was the last time you cracked open a distributed systems textbook?
When was the last time you built a system and stared reality right in face, that is: - can't trust your clocks - pick 2/3 of CAP - exactly-once delivery impossible - the code will need to be altered and released without downtime - hackers will try to exploit you for fun and profit - your manager doesn't want you wasting time getting the above right
Coding is the easy bit.
One thing that is good to have in all those positions is an understanding of the code base and where it can slowly evolve to and how that positions the code base best in the market. I'd say the best way to build this understanding is still to build parts of the system yourself. Not talk to experts or agents about the code base.
We aren't going to understand the systems we whackchitect from now on. It's the endless prodding and begging instead, something that makes me infinitely sad.
This has nothing to do with any intrinsic difficulty with coding and more to do with the industries penchant to rewrite everything every couple of years. The reason you don't see people who went to management become ICs again, is because you have to spend time learning the new, correct™, way to do read and write code.
From the constant API churn for something like React, or even the 20 million updates to write "modern" C++, someone who has had those minute decisions abstracted away will struggle to write code.
This is still difficult. Sometimes the programming language or the programming methods you want to use effect how you desing the system on an abstract level.
The essential complexity can be easily resolved by talking to domain experts. You will get a nice requirements document afterwards. That’s when the engineering and management concerns appear.
Code itself can absolutely be difficult, but it's seldom the most difficult part of a project. Generally the hard part is actually figuring out what you want, and how to achieve what you want, then the actual code is pretty straightforward in most cases.
I've been programming since the the 1980s and I'm not insulted by the idea that code isn't the hard part.
By the end of my career I was being paid for the code I didn't have to write.
One of the nicest complements I ever got from a coworker was that he was astonished how much I accomplished with so little code. Everything I built was designed to be easily and quickly extensible with minimal changes, I planned my structure for future needs.
That is why high paid programmers have been in demand, because learning that takes more than intelligence, it takes wisdom.
Has Claude code et al replaced programmers? Not really. And it will be a long time before it can - because someone needs to still instruct the direction of the code, the base architecture to build upon and that comes with real human experience.
Also I write good code and managers are thrash.
The hard part has always been how to solve x problem. Coding is the last piece of that part, which while not easy is not the hardest.
The art of computer programming books are not about coding they are about computer science, i.e. figuring out how to compute solutions to problems.
Figuring out what to build is definitely not the hard part though.
Edit: Although the current job market is heavily distorted, there used to be distinction b/w developer and engineer in the past. As the mainstream development model shifts from waterfall to iterative models, it became necessary for everyone to be engineering-capable -- only up to a certain point. So, every developer now carries a certain amount of engineering knowledge, but now they started to misunderstand and underestimate the value of engineering, and this is what you would get at the very end.
Well-engineered code is hard. It still is. Code that is reliable, extensible, maintainable, scalable, legible, understandable is hard. Code where the specs and the "why" behind it were pressure-tested via thought and good intuition.
You can guess which companies and people were hit the hardest, when it turned out that all that software could eventually be generated by a machine.
The degrees, the books, that’s not the supposedly easy part. That’s the HARD PART, and it’s actually the “what you build” thing. What you build is the code architecture, knowing how to conjure objects, methods, modules, lambdas out of thin air in a way that faithfully represents a real-world problem. Literally the shape of the resulting code. It’s not product or CS.
The easy part is supposed to be actually typing out the code, putting the methods together, remembering method names and syntax quirks.
Some people say “coding” to mean the process of converting a well-defined plan (requirements, architecture, everything) into executable code. Others use “coding” to also include all the small decisions you make when writing code, like the abstractions you build and how you handle ambiguous requirements.
AI has decoupled “typing the code” from “making the decisions about what to type” because AI can generate code from very ambiguous prompts. So “coding was always the easy part” is meant to use the second definition and emphasize that you may be able to generate code, but having good decisions embedded in the code is still a difficult, unsolved problem with AI.
There are also different definitions of “what to build” with some people meaning the technical details, as you’ve called out, and others (I think especially more business/product roles) meaning the functional requirements of the system.
And that if you make a bot that produces syntax you haven’t produced a programmer.
Some workers are more 1:1 with syntax, like a developer from a gig platform or overseas contractor, the deliverable being code files.
But the staff software engineer deliverable should be software, novel outputs, a face in meetings, someone to blame if it all goes wrong, a member to fill out the ranks, a person to get coffee with and talk about adjacent things that open new doors, in a sales capacity it might mean travel and interpersonal stuff like drinks and talks to convince people to integrate, it’s thoughtful often emotional back-and-forth from frustrated contributors on GitHub that lead to changes and innovations, and other things of value that language models can never do.
I would say further that even if you model emotional intelligence, or the whole world even, unless AI has a face I can punch if it crosses some line and it knows it has a face, what its own face is and why that matters to itself, it will always fall short in terms of empathy. Our sense of empathy and emotion originated in the wilds among fear and terror and the thrill of catching and eating food, discovering others, and understanding the consequences of everything before me in this moment.
Because they're not the ones deciding it, they just provide input for the Heads/VPs/managers who make the strategic decisions. And they are paid pretty well for this.
Writing code is easy; that's why I do it as a hobby on weekends as well. I don't have to sit through five arch review meetings that talk about how we are going to name the class rather than what the API contract is supposed to be.
Sure, writing good C is hard, and writing PHP (used to be) hard because language design is full of inconsistencies. There is so little "hard" code that I wrote for money. Plenty of somewhat harder things that I wrote for myself because it's entertaining. There was a time when I had to spend more time getting a pull request ready than actually writing code in that PR.
Of course he is an AI consultant among other things.
Edit to add: instead of "adapt" (which would imply "embrace AI", which I didn't say), my intent was more along the lines of "be adaptable."
You must be anti ai among other things.
It's not hard, it's just bloody annoying. And that's my favorite use of AI so far. Telling it what the code should do, and it translates that into language X's stdlib incantations.
I also find this to be the optimal level of AI usage. If I let the AI run around the codebase doing other stuff, then I need to spend more time and energy later catching up. Whereas, if I stay "in the driver's seat", ask for tiny changes, and approve them manually, the mental model doesn't never gets desynced.
The way I would define it, coding is to software development as cutting is to open heart surgery. It is the final part of the process after 99% of the decision making and application of experience has already been done. There is still some craft or "surgical technique" to it, but the hard part was surely everything that came before (requirements, architecture, design, etc, etc) that got you to the point where all you had left was a bunch of classes and fully specced out modules (and implied test cases) to code up.
I feel that some people, maybe including a lot of the people working at the AI companies, think that the software development job mostly consists of "coding", and perhaps in machine learning (not much code in an LLM!) it mostly does, but if you are a developer and define coding as the final "sit down and implement it" phase, then surely that is the easy part.
It was a very useful contraction in the right senior circles where there was a pretty good understanding of the meaning and decently reliable assumed consensus.
I eventually stoped using the phrase because it had started leaking deeper into the team and the impact on earlier career or less confident programmers was often no longer positive, it could be misinterpreted in lots of different ways but the most harmful was when it would further decimate confidence and discourage requests for help when something wasn’t obvious to the ultimate author.
As with almost every attempt to generalize in software engineering the repetition or extrapolation beyond the context in which it was intended can have negative side effects, it doesn’t matter if it’s a simple notion like “dry”, or a comment like “code was never the hard part”. None of these phrases survive context loss and still retain efficacy at general receivers.
Real engineering has always been about talking with the multiple parties, organising architecture meetings, taking down requirements, if no infra team available setup the whole CI/CD pipeline, and so on.
Programming could be done in whatever language, or low code/no code tool, solved the business problem.
Now what will remain to humans is a big question.
https://www.nair.sh/guides-and-opinions/communicating-your-e...
I'm not sure I follow the logic here. Wouldn't this mean that product design is hard?
That said, yeah, coding is hard. I suspect that those who claim that coding is not hard are high-level ICs. For better or for worse, as the size of a company grows and as one's career progresses, engineers will often tranform to professional box drawers, expert meeting goers, seasoned report writers, fierce gatekeepers...Anything but deep coders. Over time, they lose touch of the actual building and think that any code can be handled by people under them.
People like Jeff Dean, who still codes and optimizes things like TPU kernel code, is very rare.
Yes, there always will be artisanal weavers and a smaller number of them get paid a lot more money to do this by people who can afford it. Everyone else either started operating a loom or did something else.
Automated looms are now making mass amounts of textiles but the artisans are no longer doing the physical act of weaving. The artisans come up with cool designs, get feedback from customers, solve people's problems and outsource the rest (physical labor).
We are in the loom moment. Are you designing things people want? Are you doing the weaving? These two paths can coexist but they are diverging disciplines with diverging difficulties and diverging value.
Typing code into a computer is rapidly becoming physical labor now. The layer of creativity and problem solving is quickly rising to a level above the code since the code is a fluid now that comes and goes easily.
heh that's quite a statement. How would you even know and what does "credible" mean?
Mass production is certainly here for the majority of programming jobs. The competition for the jobs that are left will be intense, but most simply won’t make the bar. Someone said elsewhere “It’s no different than ordering a pizza: I don’t do the work.” Yes. Pizza is automated now. But no pizza employees are coming for my job. When the AGI comes for my job, we will either have post-scarcity utopia or more likely we’ll all be dead.
Probably is a question of terms: there is two things: -design- and -coding- (but could be anything as: building, drilling, turning, traveling..) so to make things without a good design is the regal way to problems, many times is a problem left by others, and in a world more and more complex and fast-changing can only be more and more worst. Hard to say what part is really the hardest, but starting with a bad design is no good.
And now probably LLM will solve some problems, but for sure these tech will create more and more new ones.
My take on that question is that programmers were in high demand because, for the last 60 or so years, it has been cheaper to write software for general purpose computers to automate things that were done by people and more rudimentary machines, e.g. accounting, manufacturing, music production, etc than it was to pay the people to continue doing those jobs. So, it continues a trend for programmers to be replaced by software, at least until something in the process breaks.
At its core, there is something difficult about programming. Fred Brooks talked about the need for perfection, Don Knuth about there being about 1 in 50 persons who had the mindset for computer science. But we haven't really been paid because it's difficult, we've been paid because we're cheaper than the alternative.
No matter how long you spend on architecture, no matter how carefully you plan your features, if you are an actually good developer there are choices that emerge only from the first draft of the code — things that you could do better, broader ideas that suddenly emerge and change your view of your own work, abstractions that become possible once you internalise the project through writing it, realisations that a requirement is unscalable, unworkable or unsafe, etc.
Nobody ever finds all those things only in the planning stage in any piece of code of consequential size or functionality: if it was easy, we'd all be doing waterfall development like 1970s consultants or using StP like 90s consultants, and none of those other ideas about coding would ever have emerged.
LLMs will just write the code. They will never have the rest of that experience. And I think any coder who doesn't have a visceral feel for what I said above is just bad at it.
"The code was never the hard part" is just edgelord AI evangelists masking denial with a pithy mantra. The code, its capabilities, the tooling choices, it's all indivisible from all the other hard parts.
But then these are often also the people who think they can use AI song or image generators to do the bulk of the work and "add the finishing touches". They also think "taste is all that is left" when the thing that gets us paid is not just our taste, it's our responsibility for and to our work.
But I enthusiastically agree with your point about engaging with your output and having that deep understanding of all the intricacies and, well, you already said it better.
I also can’t shake the feeling that many hardcore LLM proponents are actually, genuinely worse at this than I am, and that’s simply it. In my vicinity, the loudest and most extreme LLM jockeys are all people who I personally don’t think are great engineers, or even that smart. They might be right in the end and I might be wrong, but these people definitely won’t become my role models anytime soon.
The author's main thesis, that coding is hard, conflates what an individual finds easy or hard, with what is easy or hard for an average person.
When I think of code, I like to compare it to old complex physical machines. Think automaton. If you zoom in enough, what you see is indeed somewhat simple. But zoom out and the answer changes.
Then there is maintenance, manufacturing, tooling, etc.
I suspect that as humans living in a physical world, we understand intuitively the vast knowledge and skills required to design and build different kinds of machines. Software tends to hide inside boxes that look similar, but are infinitely variable.
It is uncommon to find engineers doing work they find trivial outside of some consulting niches.
Well, this is also an insult.
Cleanest way I read this is seeing how basically none of it makes any sense for someone coding as a hobby, or in any non-business context.
This is not an article about coding, it’s a promotional piece for business stakeholders.
They think communication, alignment, gathering requirements, and other political bullshit is the hard part because they have never actually solved or had to grapple with a truly hard problem.
That’s fine, but it shows the corporate programmer who is probably in meetings all day has a vastly different reality than those of us who have had to solve open problems with no solution written somewhere because there is none.
- The sorcerer: Have meetings with others until you know just how valuable a solution to this hard problem actually is, characterize it well, pool resources use cases and documentation, and then work with whatever wizard (or university thereof) is known to be able to solve that kind of problem. Have them find a solution to the hard problem and publish it. Use that publication as context, and have your LLMs integrate the solution.
- The wizard: Find hard problems with adequate funding behind them. Solve them. Don't worry about stakeholders or integrations--the problems are hard enough on their own.
It used to be we found ourselves jumping back and forth between sorcerer and wizard. But there are so many hard problems with solutions that are now in the training data for these models. A relevant skill for the sorcerer, besides the skills that are relevant in those meetings, is not solving hard problems head on, but being a sort of remixer of existing solutions to hard problems.
I think this would actually be better, because more hard problems would get solved in the open where they can benefit everybody, rather than ending up as IP-shaped ammo for zero sum games.
> those of us who have had to solve open problems with no solution written somewhere because there is none.
Probably because solving said 'truly hard problem' is niche with little or limited value as few people have tried to solve it (otherwise a solutions will likely have been written).
To be less tongue-in-cheek, your logic is flawed. We simply have too many hard problems than we do smart people, unfortunately.
And the ones we do have seem to like optimizing comfort over risk, such as taking cush consultancy positions.
https://www.kalzumeus.com/2011/10/28/dont-call-yourself-a-pr...
Don’t call yourself a programmer: “Programmer” sounds like “anomalously high-cost peon who types some mumbo-jumbo into some other mumbo-jumbo.” If you call yourself a programmer, someone is already working on a way to get you fired. You know Salesforce, widely perceived among engineers to be a Software as a Services company? Their motto and sales point is “No Software”, which conveys to their actual customers “You know those programmers you have working on your internal systems? If you used Salesforce, you could fire half of them and pocket part of the difference in your bonus.” (There’s nothing wrong with this, by the way. You’re in the business of unemploying people. If you think that is unfair, go back to school and study something that doesn’t matter.)Besides, there was a whole back-and-forth among multiple bloggers and comment sites (including here: https://news.ycombinator.com/item?id=3170766) from back then in response that agreed or disagreed with that patio11 post. Here's one: https://web.archive.org/web/20111126183459/http://www.jacque... Here's another (though a couple years later): https://yosefk.com/blog/do-call-yourself-a-programmer-and-ot... From the second one's conclusion:
> When I introduce myself, I usually call myself a programmer, regardless of my current work on chip architecture and management and stuff. I got into programming for the money, so it's not like I'm overflowing with pride when uttering "programmer". I just think programming is a great career and the right thing to call myself for me.
> There's an alternative approach where you program, but you don't call it that, and you use programming as a starting point from which you transition to some form of being involved in business as directly as possible.
> It sounds a bit roundabout to me – why not just get an MBA instead? – but maybe it's the right path for some (especially considering that some prestigious MBA programs want you to have industry experience before you can even enroll.)
> The important thing is to choose the path that suits your preferences, follow it consistently, and realize where your approach is most likely to succeed. Because where I work, someone applying for a programming position and not calling himself a programmer will not make a good impression.
Sometimes calling yourself a "Software Engineer", or focusing on "$X company revenue definitely attributed to my efforts" rather than the technical details, is the right thing to do. Sometimes it's not. In any case I'll continue explaining to outsiders that "software engineer" is mostly just a fancy term for "programmer", and to programmers to call themselves whatever they think will best give them a chance at working where, on what, and for how much money they desire.
The more interesting part, to me, is the 15 year old recognition that programmers' jobs are to replace other workers. It's definitely not a new thought at all, but it's interesting to reflect on considering how the dynamics are starting to turn the other way around.
Saying this as someone who knows many programming languages... but I don't consider myself a programmer because that's too job-oriented. Like.. I write code for fun and "programming" isn't fun.. it's a living. This approach has let me follow multidisciplinary paths by framing my career away from the code-as-my-product mentality.
Occasionally I see a tech person in SV upset about AI automating away jobs. My dude, your whole job is to automate away jobs.
It's like saying 'Lawyers write documents!' for their job ... no, that's just an artifact.
Architecture, Systems, State, Integration, Algorithms, Pipelines, Platforms, Ops, Design, Communicating with other Eng, Working on a Team, Understanding Product/Product Marketing Requirements ... and of which the 'code' is just a small bit of the written part.
Honestly ...
Because companies resent that they have to take on risk and pay people to extract value from the market.
Ugh, why do we have to pay people to code, maintenance, etc. Lets pay people to extract more value for our bonuses and shareholders. Lets try to only hire heavyweights so that we don't get hung up on that difficult-to-measure coding process, knowledge transfer, messy human-ness. Oooh how nice, we can hire a fewer employees that know how to leverage code agents that free them up to think about that what REALLY matters...
In my career code was the hard part for the first few years. Then I got over the hump and everything else about my job was harder. Today code is the easiest and least interesting part. But I’ve also experienced in 13 years maybe… I’m going to say 2% of the world of professional coding. I bet if tomorrow I was asked to do some kernel optimization or make Postgres better or reverse engineer an emulator for some PLC something something, I would be deep into a land where code is the hard part.
In order to reduce the complexity of eventually coding something, you use modeling and code probes to validate the business model and then write the code.
This has been around since the days of batch programming and evolved through domain driven design principles.
The reality is the business can’t see that so they don’t invest in it and have no patience for it.
Agile wasn’t embraced because it was better. It was embraced because it was cheaper and faster.
Planning and modeling are the levers of complexity.
Shipping a working service or product matters and theres a lot of ways you can get there. The upside of good code is usually in maintenance and extensibility but theres a limit to how much those matter in the grand scheme.
Typing code is indeed not the hard part, programming is.
I still read TAOCP occasionally as a hobby. The combinatorial algorithms and data structures are just fascinating. That said, this argument seems irrelevant to majority of the programming jobs. I doubt most engineers will ever need to implement anything mentioned in TAOCP, thanks for all kinds of powerful abstractions.
Coding is easy just like writing is easy. What makes the difference is what you write.
That doesn't make it inherently easy, but it is the easiest part to automate.
A modern LLM is perfectly capable of maintaining decent code quality and architecture (provided you ask for it) up to a few thousand lines of code, but after that it very quickly loses the plot if you're not designing your documentation right and keeping a hand on the architectural tiller.
Architecture is about staving off chaos for as long as possible given the maximum functionality you expect it to achieve. That's hard enough for a human, with a deep understanding of your business, to do. Architectures that endure is a hard problem, dwarfing the difficulty of writing the actual code. Choosing what product to build is also a hard problem if you expect to meet any success. What it does, what it specifically doesn't do. Sounds easy on paper, and if all you're doing is Sunday prototypes it feels almost trivial. But once you're doing a real product with real consumers, it's a very different thing.
LLMs don't fix that though. I see some truly awful code come out of them. And no, it's not just better context or use more skillz.md.
these two groups are telling me my time is numbered because of llm and ai when in-fact im struggling to see how ai doesnt replace them both
really good software engineers are expected to be experts in their job and that of the rest of the project team, these individuals are primed to be empowered by ai in a really disruptive way.
Figuring out what code to write and what not to write was always hard.
The code would come out easier the more time I spent thinking and designing and talking about it with others.
of course, YMMV
Writing code that's logical and easy to follow, that can be extended in several likely dimensions without major plumbing work, and that doesn't contain "gotchas" for the maintainer, is not at all easy.
Code, is written in a language. Language is opinionated. Things written in that language are also opinionated. LLMs often have horrible opinions.
The point isn't that labor isn't "valuable" it's that "value" here is a moral position. This same thesis could have been said about basically any mechanized industry.
There is no putting the genie back in the bottle. You can't un-invent the nuclear bomb, or the printing press, or the steam engine. We need to find a politically stabilizing way forward, and that means we need political coalitions that don't consume themselves with infighting.
Right now we can't even work together to build housing for young people... how the hell are we going to get through this mess without actually trying to build something bigger by making sacrifices.
Farming was also hard, manual work wise and now machines overtook. It's still hard because of marginalization.
It will happen to Coding, Consulting, Creating work, etc as well.
Isn't the real argument that "writing code is not the hard part"? As in, reading and understanding is the hard part. Figuring out what and how to change is the hard part.
Writing is the last 1% that happens after you have already finished the 99% of talking to people, figuring out what needs to be built, building up context about the codebase and surrounding infrastructure in your heard, planning the actual changes.
As computing systems become increasingly capable and encroach in our territory that distinguishes us and lead to our success as a species, intelligence (whatever that is or isn’t), we redefine the problem and handwave away the new capabilities.
It’s getting increasingly more difficult to do that in knowledge domains with current frontier agentic systems. They’re not AGI, but they start to make it increasingly difficult to move the goal posts for many people’s comfort.
We really need a lot more philosophers, sociologists, and frankly economists working on this problem: in an era where physical needs were mechanized away and increasingly aspects of the knowledge economy are shifting away, what does it look like in modernity? How do we sustain or adapt our current economic models? What new models may be needed? Do we need to continue to enforce this whole work to survive in an environment where much work is disappearing or at the very least shifting around.
No, we’re not there yet. You still need experts to guide things around, but it’s becoming increasingly easier to do more in this space with less humans. That’s not a trivial change in the US where we put most our eggs in this whole knowledge economy basket.
Typing is easy. Coding is hard. LLMs eliminate the typing and aid with all the other parts of writing code (where is the system that I want to mutate, how does it work today, debate the tradeoffs inherent in the potential plans of action).
What remains after factoring out the typing and the time spent assembling an understanding of code is opportunity cost. That's not an insult any more than memory managers are an insult to languages like C.
That’s what people mean by this.
"Code was never the hardest part."
There you go. Doesn't imply that coding is easy.
"Code was never the hard part" is the dumbest thing I've ever heard.
Making something work, has always been easier than making something someone can read. AIs also, seem to benefit from clean abstractions, appropriate code reuse, etc. (coincidentally, the thing they suck most at).
It's the same thing with english, except that english doesn't have a compiler. Comprehension is the only measure of communication if you're talking about a human language. Programming shares the same goal. Both, often also need to do something else useful. Programming, and lawyering, have a lot more in common than people think.
We're creating a culture of sh*tting on codebases so the highest paid execs can cash out when things get tough. It isn't a new phenomenon, but it's one we'll need to endure until enough people lose enough money that the accountants start taking notice and start saying "you should be more careful, or you'll lose your shirt". In the interim, the people that care are working insane hours to try to protect the things they believe in from inevitable doom, and risking being fired to do it. There is a balance to both sides, and the jury is out on whether or not anthropic/openai/alibaba can save us from the future we are creating now with short-term goals.
There's probably a lot wrong with the "coding was never the hard part" take, but this quote right here shows that the author is not willing to engage with the actual idea that folks who say this are espousing. Because if the author was discussing these ideas on good faith, he'd know that the answer is obvious: there's much, much more to a software engineer's job than just coding, and that other stuff is very hard to do well, and people pay for that.
Again, I'm not saying the "coding was never the hard part" folks are right, but I really, really hate straw men.
To be precise, it depends on the domain. The people who could actually write algorithms or core implementations were always a minority. Programmers like me mostly did copy-paste from Stack Overflow or assembled libraries.
It's not that code wasn't difficult—it really was.
In CRUD apps, about 70~80%of the work was building the same thing over and over, so once you got familiar with it, most of it was repetitive practice. But the number of people who could actually create something new was always small.
Most business programs had issues that arose in the application stage, the application layer. In this application layer, only a very small portion involved difficult logic. Most of it was just applied.
The problem is that people often romanticize the lower layers beyond their own, compilers and low level systems, calling that 'real programming,' and in doing so, they make programming seem harder than it is. In reality, the coding that most people make money from is mostly at the abstracted layers. The infrastructure beneath those layers is owned by giant corporations. If you work at one of those giants, that's fine. But beneath them are countless consumers paying those giants, and the coding that targets those consumers isn't that difficult.
In the end, whether coding was difficult or easy depends entirely on which layer you're working in.
What's certain is that coding was difficult, and it still is.
I’ve met a lot of programmers where those concepts where only words and not something they have understood. A snippet of code is either something they have to learn or copy, it’s not something they can fluently manipulate. It’s the difference between having to use a dictionary and sample phrases and speaking the language fluently. The former is a chore, while you don’t even notice the latter.
Software consists of various layers, and everyone has their own specific areas of strength. For instance, because I am an application programmer, I often need to write code that prevents the program from halting—aligning with recent programming trends that involve preserving the computation context using monads. In other words, my strengths lie in the overall architecture and interface design. This fundamentally relates to cohesion and coupling. I excel in this area, particularly when dealing with codebases around the 60,000-line mark. This is a realm where books like Clean Code are quite effective (many people dislike it, but it is actually a well-written book). Put differently, I possess the ability to mass-produce software (regardless of absolute quality). Over the course of 7 years, I have built CRUD applications for 43 companies across 16 different domains (ranging from drones and golf simulators to tax SaaS and supermarket POS systems). Therefore, I believe I have at least an average, solid capability in this regard. The primary area where I actually made money was PLC, so while I may not have deep academic expertise, I certainly do not think I lack capability.
In Korea, the profession known as SI (System Integration) is a field where you enter contracts on a "project" basis. In that environment, I have encountered a wide variety of people. From those experiences, my takeaway is that programming is divided into quite several distinct layers.
To speak of algorithms first: I learned basic algorithms and fundamental data structures in university. However, in the field where I worked, there were many people who struggled to implement those basic algorithms, yet they still built a large number of applications. Why is that? There was even someone who made an amount of money I could never dream of touching in my lifetime. Why did he make so much money when he didn't even know how to implement basic algorithms?
The answer is quite simple. It is because the "implementation model" and the "contract and cost model" are different. The contract and cost model is a self-contained body of knowledge. It is the ability to know exactly where to fit a given piece into the puzzle. Implementation is simply the ability to build that piece from scratch. In fact, mostly due to issues like employee turnover and the organization's future maintenance capabilities, many teams (specifically, organizations with lower implementation capabilities) decide on an open-source library and design their architecture based on its API. In these cases, the primary technical challenge becomes how to connect those components based on the performance of that library.
Yet, people tend to think that only those who can implement from scratch are capable of programming. A person who can take someone else's implementation and piece it together to fulfill their own contract is also a programmer, but people frequently forget this. Depending on which layer you exist in, certain knowledge requires you to implement it yourself, while other knowledge only requires you to understand the contract. I believe this is the core of programming.
Those on the side that must design and build libraries or frameworks naturally have things they must know about implementation, as they are creating the SDKs. However, I have seen quite a few cases where these very people have no idea how their work is actually utilized in the upper layers. And these types of knowledge are highly fragmented.
In my case, I am familiar with quite a few paradigms. On my personal homepage wiki, I can differentiate between OOP, DOD (Data-Oriented Design), and others, and in the context of relational databases, I know exactly where the ORM impedance mismatch occurs. However, this is largely an area intertwined with architecture, and its essence is closely linked to David Parnas's theory of information hiding. In other words: to what extent do we hide the internals, and where do we expose them to minimize the contact surface area and ensure a safe connection?
For example, I can't implement PostgreSQL's B-tree. But I can design a business system using PostgreSQL. I only know the name of TCP congestion control—I don't know how to implement it. But I can build networked applications.
That's what an 'industry' really is. The ability to trust the contracts of other people's implementations and assemble them. The ability to trust others.
In that regard, I agree with many of your points. However, I actually believe the software industry needs more of those "average" people. I think the very definition of an "industry" should premise that average people can maintain their livelihoods simply by dedicating themselves to a single specific field. From that perspective, when building one's expertise within this fragmented landscape of knowledge, it is perfectly natural not to know much outside of your own specific layer.
p.s https://www.makonea.com/en-US/casual/cargo-cult-programming-...
At the core, both involve the same formal thinking, but it’s easier with higher level concepts to have tangible results (it helps with the rote learning) than purely abstract ones.
"A similar idea [to libraries] is the Collected Algorithms published by ACM years ago in an ALGOL-like publication language. I remember when I was a kid in 1968 looking up algorithms for sorting and searching in my first programming job. However, what every manager learns is that reuse under these circumstances requires a process of reuse or at least a policy. First, you need to have a central repository of code. ... Second, there has to be a means of locating the right piece of code ... It does no good to have the right piece of code if no one can find it. Classification in the world of books, reports, magazines, and the like is a profession, called cataloging. Librarians help people find the book. But few software organizations can afford a software cataloger, let alone a librarian to help find the software for its developers. This is because when a developer manager has the choice of hiring another developer or a software librarian, the manager will always hire the developer."
He talks about a few reasons for why this is, one of them being simply that plenty of code for your project will just have to be new / different enough from existing code, so there's not much reuse benefit there. He qualifies that with performance concerns, but that's just one reason why it would have to be different.
Another thing I was reminded of, from the intro of Etudes for Programmers:
"Programming is a craft, and programmers must attain a standard of craftsmanship. Much programming is done in cottage shops -- that is, in small shops with meager tools, much work done by hand, and learning attained from other laborers, by chance, and often not at all. ... the academics have also discovered that a craft cannot be taught well by teachers alone; the guild apprenticeships had considerable merit. In a classic apprenticeship the candidate spent many years doing menial tasks while absorbing fundamental techniques of the trade from more experienced workers in the shop. Gradually the apprentice was given more technical responsibility and, after a formal test of skills, eventually became a journeyman certified competent for all ordinary jobs in the trade. The journeyman traveled the world and, if the muses allowed, one day presented a masterpiece to the guild and became a craftsman of the highest rank -- a master of the guild."
Seeking for the average to be that of journeyman, rather than apprentice, seems like a worthwhile goal for the industry. You sound like a journeyman to me, and I can generally trust other journeyman programmers. We don't all need to be masters to have an industry, and that's probably not possible anyway. You say you can't implement TCP congestion control -- but you can build networked applications. Building networked applications is an ordinary job of the trade, and I'm sure you know how to do it in a number of ways. I also suspect that if you had to create some custom UDP protocol for something, you would be able to implement some form of congestion control for it in not too much time -- even if that involves searching for and copying a known algorithm, or if you're lucky finding and using a pre-built library flexible enough for your custom needs. That highlights two more reasons for why we don't hire librarians over developers. First is that, if you're at a journeyman level, then you can be trusted to handle the rather common scenarios where you have to do something for which there's no perfect matching library or framework or SDK you can just use. Whether you have to roll up your sleeves and go low level, relying on very little, or just need to be able to tie enough separate preexisting things together with some minimal glue to handle the new/different stuff, they're both activities that require a programmer rather than a librarian. (I guess this is just further elaboration of the already mentioned "there will be custom code not eligible for replacement by something reusable" reason.) Second, the task of a librarian to find software has been made much easier since the 90s with things like search engines, massive communication channels to share and find out about stuff, and massive open source repositories. It's so easy now that developers themselves can do a decent job at it among their other duties.
I actually think all these industry issues arise because the ultimate role most programmers aspire to is that of the Master. They want to be the ones designing the rules, frameworks, and architectures that mass produce journeymen. In reality, a department head at a major tech company certainly holds a position worthy of being called a Master.
However, I believe being a Journeyman is a thoroughly valuable and respectable end destination in itself. The reason is that once someone becomes a Master, they drift away from the actual field. They start focusing purely on building tools for the journeymen, whereas it is the journeyman who generally remains closest to the actual consumers.
I read your comment carefully. Thank you for taking the time to reply, and have a great day. It's been a weekend full of things to think about.
Managers and execs aren’t saying this it’s the programmers and coders themselves making the claim that coding was never the hard part.
It is still an insult though. It’s an insult to themselves. It’s the lie all programmers including me tell themselves as reality itself insults us. Coding WAS the hard part.
That’s exactly what we were good at. Now our skills are getting owned by automation. How do we face reality shitting in our faces? We lie. We fabricate a reality that’s more acceptable. We frame our environment in a way that still validates our existence. If AI has invalidated all of my programming skill then I need to find something else to support my identity.
A lot of people enjoy coding, it’s the rest of the job they don’t like. That’s why it feels hard.
If a tool was going to come along and automate away a huge part of a programmer’s job, I think most would want that tool to eliminate the meetings, ambiguity, scope creep, and administrative work… not the coding itself.
My best days at work were days with nothing on my calendar, when I could just put on some headphones and make something. At the end of the day I felt like I accomplished something and had something I could see and use to show for it; I finished the day happy and energized. Contrast that with a day full of meetings, fire drills, and busy work, where at the end of the day I’m mentally and emotionally drained, wondering if I should quit to stock shelves at the local grocery store. Which day sounds “harder”?
I don’t know about where everyone else works, but defining the details of what to build seems so hard that no one actually does it, so it falls on us as we build things. During one project I got a directive from the CIO (which has only happened 1 time in 20 years) to get what I was working on done in 4 weeks. I made all the decisions myself when it came to the details and had something mostly working in 2 weeks (I skipped all meetings and any other distractions during this time). Several months later, the bureaucracy came into play. The principal architect on the project, who I never talked to before the CIO told me to get it done, finished his design and some details needed to be worked out. One such detail was a port list. It took 4+ months of meetings to get that done, and it still required constant tweaking after that for another year. There was a half dozen other things like that in the same project. So yeah, the initial code was pretty quick and fun to write, and then it was followed by 2-3 years of hell, that probably should have been worked out before we started coding. I ended up having to go back and re-write a bunch of stuff to align with the design that was decided on over a year after the deadline the CIO gave me. In most cases, I think the updated code is worse, as the bureaucratic design creates a lot of operational work that my original design avoided entirely, but I digress.
The comparison does not imply that coding is an easy thing to do, just that it's easier than being really, really good at the bigger picture.
What Carmack did wasn't hard because writing C is hard.
nothing has changed since then.
Debugging is hard.
2026 - "Coding was always hard, please don't lay me off."
This is a specific attitude I try to beat out of juniors. You will not be dismissive of the point of this exercise.
I roll my eyes at this when thinking about the poor JavaScript developer that cannot write code without things like jquery or React.
If code were so easy there wouldn’t be so much bloat and slow garbage in the world.
Programming used to be a rare skill. Not so much now. In fact, it hasn't been for well over a decade.
My current employer started aggressively hiring for programmers in India in the early 2010s. They get paid a fraction of what our software engineers do in the US. There are many talented engineers in India, and those who get hired by us either find a way to come to the US for an enormous pay increase, or use us as a stepping stone to quickly find better work. And my employer is seemingly fine with this. They expect these cheap employees to do programming, not engineering.
Programming may not necessarily be easy, but it is cheap. It has become a relatively common skill, driving down its market value. A lot of tech companies were slow to figure this out because times were good, interest was low, and investment was flowing.
AI is waking people up to a truth that has been around for a while now.
Side note: I so forgot about the "Don't make me think" book! Thanks for reminding this exists. I submit this should be part of "Software Development 101", right there alongside SICP.
While the former can certainly be challenging, it’s by far not rocket science. Any reasonably intelligent human can discuss requirements or design a product at a decent level.
Writing code at a decent level is beyond the average reasonably intelligent human. If your mind doesn’t tick a certain way, you will not be able to do it and it will be painfully obvious to anyone that can do it.
I have met many programmers throughout my career, and very few of them want to talk to stakeholders, much less customers (exceptions are freelancers and founders, especially of software development shops). And, “having clarity on the priorities” boils down to “just tell me what to do and don't switch it up every two days”.
Programming is not easy and it takes years to master. Some people became really good at solving complex programming puzzles and 'Code Jams' and focused on it. Unfortunately, the same aspect which made this skill highly visible and highly praised, is what made it easiest to automate.
All those medals, trophies and certificates... not the mention advantages at big tech software job interviews... Came at a cost.
Meanwhile there is a whole group of people who have been honing their skills in software design, architecture, distributed systems, security and other less visible, less rewarded skills who have been ignored by the markets. These people still can't be automated.
It's a large problem space so after a decade or two, the coding aspect feels small relative to all the theory and experience surrounding it.
I met many senior people who didn't take programming seriously as a skill, long before LLMs.
One time, when I was at university, one of my math lecturers was boasting about the superiority of math as a discipline and said to the class "Software engineers... There are no software engineers; they're programmers."
That statement was never true but it's much more obvious now. The fact that a lot of people shared this belief highlights the fact that these other skills were invisible.
There is probably as much engineering (if not more engineering) involved in delivering a complex, reliable software project as there is delivering a complex skyscraper project in civil engineering... It's the same kind of activity; lots of interdependent parts, each with their own constraints and requiring many decisions to be made with lots of tradeoffs. At least with a skyscraper, the customer requirements are relatively very stable.
Sometimes coding is hard. There are two types of hard things: algorithms and architecture. LLMs are good at algorithms, not good at architecture.
Most of the time coding is not very hard, it’s filling in the blanks, implementing the business logic, writing boilerplate code. LLMs are good at this too.
As soon as the context builds up high it just starts doing things straight up wrong
I've been trying to network a simple tactics game with AI and it just is a hydra of sync issues despite clear explanations of what they are and bug reports and desired end-states.
That 14% might have been hard, but it is definitely still the smallest part of being an engineer.
* https://www.microsoft.com/en-us/research/wp-content/uploads/...
"If figuring out what to build is the hard part, why do so many product managers seem clueless" - because people
"LLMs may be good at coding" - they are not;
Because it is hard. :). Writing good code - takes years of practice.
And yet I spent a majority of my career fixing mistakes, including my own
> Those decades spent fighting memory bugs in C or C++, with the scars to prove it, are worthless in the age of Rust, Go, Python and JavaScript.
For most people sure. I’ve had GC kill a service in production, or even just tank p99. And I think a decade of C gives you a massive head start with rust. Less screaming “WTF WHY?” at the compiler anyway.
The phrase "X is the hard part" means that X is the hardest part, not that all the other parts are easy.
It is all the parts of maintaining the software with time and engineering with all the moving parts (not the coding) is the point of why software engineering exists.
Yes, it's fairly axiomic that he was "just at the right place at the right time" because there are dozens of modern "boomer shooters" on steam that are not nearly as successful as DOOM or Quake. You might counterargue that its always harder to do something thats never been done before but that would only be a tacit admission that coding actually was the easy part in that case.
tl;dr nothing has changed.
Claude writes 90% of my code but the bottlenecks always were and still are:
* getting clear requirements from product
* getting the damn code reviewed so I can merge it
Neither of these are fixed. Frankly, overuse of AI has made both of these worse. Claude brained product owners going hog wild with Claude Design are a nightmare to deal with, and the volume of absolute trash quality code being submitted for review is soul crushing.
No, using AI to automate code reviews is not acceptable. Code Review isn’t about a systematic checklist (though they can help) it’s about making sure people understand the actual changes being made to the system because it’s people who are accountable for what happens in production. LLMs can be part of the process of reviewing code but they suffer from the same issues as any other chatbot based tools (hallucinations, context confusion or not enough context, getting bogged down in impossible code paths or other minutia, etc…) so you have to review that review carefully too. Human judgment is still king.
These days my job is primarily reviewing offshore slop and making sure it’s in a good enough state to merge. I’m doing merge and release management way more than actually coding (and it sucks btw because I actually enjoy coding with or without agents). If the quality of the code turned in for me to review and merge is any indication, engineers/system architects are going to have their hands full.
Good riddance to the overpaid coders.
And hello cheap replacements!
The bottom line has improved. And that's good for business. Which was the only thing that mattered. Regardless of any reactionary sentimentalism.
>If coding is easy, how come programmers were in high demand, and have demanded large salaries for years (even before ZIRP)?
There is more to those roles than just coding. In fact the more expensive "programmers" often do not code themselves.
>Why was there so much stress, overwork and burnout even before AI started churning out 5000-line PRs?
Something being time consuming is different from something being hard. There are simple factory jobs that also demand overworking.
>Why did companies seek 10x ninja rockstar coders and subject them to leetcode interviews—surely, a junior fresh out of college could churn out something if it's so easy?
Building software takes time and since velocity is important companies wanted people who could increase velocity.
>If coding is easy, why do we have doorstoppers like Clean Code and The Pragmatic Programmer? Is The Art of Computer Programming a light summer read? Is SICP a coffee-table book? Why do we have bootcamps or even whole college degrees dedicated to it?
So authors can make money? Programming is learned by a ton of kids on their own there is no need for boot camps or college degrees just for the benefit of being able to program.
If coding is easy, was Carmack just at the right place at the right time? Why do we consider Fabrice Bellard a genius?
The earlier you are to a field the easier it is to have your impact recorded. In markets with first movie advantage being earlier also helps a lot. A lot of people were able to program so what made them earlier than others was not just being able to program.
>If coding is easy, why are people angry at AI (or anyone else) copying their code? Why do they act like they've poured their sweat, soul, and copious amounts of time into something so trivial?
Again something being time consuming doesn't mean it was hard.
>If coding is easy, why do many now feel like their identity and professional purpose are being stripped away from them?
When you spend a big percentage of your life doing something it becomes part of your identity regardless of difficulty.
>If coding is easy, why is software so damn buggy?
Because making bug free software takes a lot of time and resources. Those resources have a higher return on investment elsewhere.
>If deciding what to build is the hard part, why do so many product managers seem clueless? Why aren't there rigorous 10-step interviews for them? Why aren't they getting paid more than the developers?
People are clueless because it is hard. Pay is not based off of difficulty.
>If deciding what to build is the hard part, why aren't market researchers, usability experts and—hell, customer success—considered rockstars in a software company? If “understanding the customer” is harder, why are business analysts looked down on as pencil pushers?
Because the company finds it cheaper to outsource? A ton of companies have their employees setting up and using telemetry to understand their customers so it's not a one dimensional thing.
>If implementation is easy and finding demand is harder, why are programmers upset when the salespeople promise a new feature to a customer to close the sale? They've found a genuine demand, something people will pay for!
Programming takes time and resources. These may have a higher return on investment elsewhere than this niche feature. It may make maintaining the entire product take more resources to support a niche feature.
>If coding is easy, why doesn't everyone just build ten variations of a thing and see which pans out?
Again building entire products takes a lot of resources. And 10x the cost of building every product is not going to be competitive in the market.
The divide between product managers and developers mimics the artificial divide between humanities and STEM.
You divide workers into competing groups, then make they outperform each other.
In reality, practically all humans can both become excellent coders and acquire deep product skills as well. We can also learn a wide variety of other skills in a single lifetime. The only blockers on that are social and psychological, you're meant to not believe that is possible.
All this talk about code in this adversarial role with product comes from that, and all of it dissolves under almost any valid critical angle. The engagement with this kind of discussion takes place exclusively in that aforementioned social layer.
TL;DR weak bait
In the rest of the engineering world, the way you do that is by standards bodies developing reliable, tested, certified methods to build things that avoid common problems. Pipes that are certified to a certain PSI or UV exposure. Wires certified to a certain amount of amperage, wetness, heat. Nails certified with a certain metal grade, tolerances.
Those standard parts are then used in a certified building method for a specific application at specific usage criteria. 3x 12d nails in one kind of wood joint. Beams spaced 24" apart, with 3/4" CDX plywood spread load. You don't guess or follow trends. You don't do what you think is "clean" or "beautiful". You solely follow the engineering standards and code. Now you don't have to think much, and your results are highly reliable. The job becomes easy.
Software doesn't have professional engineering and building standards like that. So humans literally just make this shit up as they go, making software however the zeitgeist of HN says "feels good". This results in unpredictable, unreliable software products that are hard to build because nobody agrees on the "right way".
Somebody read a blog post, or a slogan or quip on a Wikipedia page, and decided on their own interpretation of how that generic advice would drive their work. Software engineers only talk to other software engineers, so they don't realize how incredibly unscientific, inefficient, unreliable, and difficult their work is. Trying to make something predictable and reliable is therefore very hard. Not because writing the code is hard, but because the entire software product lifecycle is basically vibes. The uncertainty, variability, and lack of reproducible standard parts makes figuring out how to build something become way more complicated than it should be.
The actual lines of code are easy to read and write. But without the standards common to every other engineering discipline, the rest of the job is a slog.
I'm still not seeing it.
I'm seeing a lot of loud people, a lot of LOC produced, and a lot of people angrily pointing to their sideprojects... but no massive impact outside our bubble. Remember it's 2026, we are 5 years into this hype, the models are better than ever, and all the software we used around us is basically in the state it would have been in had we projected 2021 tech 5 years into the future ("in 2026, there will be... another backend JS runtime!")
I'm afraid all the personal anecdotes of technologists have not translated to real world results... other than negative ones like GitHub now having 0 9s of reliability.
Of course I suppose I'm "coping", as if I wouldn't be over the moon if AI had made me 10x productive... but maybe, just maybe, a lot of people really like talking to chatbots?
The arguments are couched in questions… which all either have ready answers or imply strawman arguments that few are making.
I suspect it’s emotional and indirect because the author understands how poor its arguments are. That leaves open the question of why do the blog post at all, but I guess bloggers have to blog, whether or not they’ve got anything to say. An angry, emotional, vague post probably gets a nice amount of views.
There's still some skill involved, but the skill is mostly in manual testing, and accurately phrasing what went wrong. The AI is better at debugging than people are, and the code isn't great, but perfectly adequate for pretty much everything. And it's even fine at system design these days.
It's interesting realizing how mind numbingly thoughtless my job has become.
Don't get me wrong, I wouldn't mind it if this was skilled work, but it just isn't.
How long does it take? Does it find the issue? In more or less time than the engineers took?
Ask it to review your code for design and cohesiveness issues. How does it do?
A bug that is not immediately obvious, no syntax error, logic is sound, works as expected on local and QA envs. But let it run in prod for 2 months and you have a massive problem.
I'm not dismissing LLM's here btw, just pointing out that there is probably a lot of things that go into the category "you don't know what you don't know". These kinds of issues may or may not be a problem depending on the business you are in.
Most of the time the latest models perform better than I expect.
pub(crate) fn zeroed_safe<T>() -> T { unsafe { std::mem::zeroed() } }
pub(crate) fn read_unaligned_safe<T: Copy>(src: const T) -> T { unsafe { std::ptr::read_unaligned(src) } }
pub(crate) fn box_from_raw_safe<T>(ptr: mut T) -> Box<T> { unsafe { Box::from_raw(ptr) } }
pub(crate) fn isize_to_wndproc_safe(value: isize) -> WNDPROC { unsafe { Some(std::mem::transmute::< isize, unsafe extern "system" fn(HWND, u32, WPARAM, LPARAM) -> LRESULT, >(value)) } }
No, I'm not joking either, this is actual code that Claude wrote (and I'm pretty sure it's the latest models too but I don't actually know for certain, since the "dev" never specified). This idiom is repeated about 238 more times through that one file. I have tried repeatedly to help this person out but they're the kind of vibe-coder who thinks they know best, and who will take your advice and drop it into CLAUDE.md verbatim, and then Claude will go off and do the most literal interpretation of that text and not best practice.
To be clear, I don't hate LLMs. They're really useful for very specific things or where you know the domain very well. But then I see slop like this (and the dependency many people have on them -- I have literally heard my own close acquaintances say that they could not see themselves without a Claude subscription) and it really hammers home that software quality is not at all going to get better because of these things unless something changes.
Edit: grammar/spelling
Use a high effort model and let the llm review it's output and figure it out on its own.
The code still won't be great, but it'll be good enough.
> Making software is no longer very hard. It's becoming a few steps up from burger flipping. Maybe somewhere around line chef.
You did acknowledge that "some" skill was required:
> There's still some skill involved, but the skill is mostly in manual testing, and accurately phrasing what went wrong. The AI is better at debugging than people are, and the code isn't great, but perfectly adequate for pretty much everything. And it's even fine at system design these days.
All I asked was for you to back this up, since the burden of proof is on you to prove your claim, not on me to prove it for you.
Interesting. A little unhinged, but interesting.
Anyways, as I said elsewhere, I don't code for fun, and my employer would be unhappy if I sent you their proprietary code.
You can decide to do some experiments yourself, or you can decide that you don't want to hear it. No skin off my back either way.
No, I didn't. I asked you to provide me an example of code that you had generated via AI that might survive in such an environment. But if you re-read my original post, I also gave you an escape hatch: just code that was correct. Not even formally.
I'd say your refusal to provide any examples of your original claim is pretty telling, and your reasoning is pretty convenient for you, now isn't it?
Edit: also, I'd like to point out that it was you who used the word "software" as a general claim. You never specified what kind of software. Since you continue to expect me to derive the proof for you instead of you providing the evidence, I don't see any need to continue this discussion since nobody is obviously going to learn anything. I'm sure many of us here would love to learn what secret sauce your using that makes software engineering so trivial, but you don't seem willing to actually provide that.
Navigating customer requirements and building something that satisfies both market's needs and company strategy can be an incredibly difficult and frustrating problem to solve. Especially if you need to also oversee the execution of the strategy. So not only you have to predict what they want or know the domain deeply enough to understand what they say they want is not what they really want, you also have to come up with a plan for executing your solution in a corporate environment.
There is a reason that books like "the staff engineer's path" cover topics such as local maximums, communication, establishing support for executing a plan or creating alignment on big efforts. In large corporate environments with multiple international customers, code is most of the time not the hardest problem.
I loved writing GPU shaders or optimizing visualization performance, but most of the time it was wiring up netcode to UI elements that exist.
Ironically as I've moved into focusing on more GPU and kernel programming AI is now lapping me there anyway, however the impact of knowing what sort of algorithsm are state of the art in papers, what is causing memory bandwidth issues etc... does a lot to drive the machine.
If you have a site whose performance steadily gets worse and the rate of new features steadily declines and the rate of bugs steadily goes up, then your site/app will probably not have a great future.
All of those things depend on solid code. If staff engineers who are too busy talking and building consensus such that they aren't connected with the actual programming and situation on the ground, then all the talking and consensus-building won't matter.
On those complex systems in particular the problems start long before any code is written.
A software engineer can create and understand the specs, requirements, design the system, architectural decisions, define everything about that software and data, model everything, failure, performance, operational topics, documents everything, etc. before a single line of code is written, and of course they can write good code. Then there are the coders who patch together chunks of code from Stack Overflow or whatever boilerplate they have in the company's repository. I know every coder likes to call themselves a "software engineer" but there's a world of difference between the two types.
For the first group code was never the hardest part. For the second group there was never any other part.
Some architecture work and design will be done beforehand, but many details will fall into place as the code is being written, thrown away, adapted, etc.
The idea that code is mere transcription - which I see a lot in these AI discussions - is completely false. Code is a form of low-level design and is where the rubber hits the road.
The best requirements, designs, marketing, etc are worth jack if one fucks up the code. The code is the actual product.
Which part of my list was just "a detail" to be dealt with at some point in the lifecycle (but only if you're not too busy shipping features) for you? You're laying bricks before knowing if the wall's supposed to be concrete.
> Code is a form of low-level design and is where the rubber hits the road.
Sure but in keeping with your analogy tires are fungible across most cars and it takes minutes to change one if you picked the wrong compound. It takes years to properly design a tire, not to speak of everything that sits on top of those tires. By the time you actually "meet the road" you already defined to a tee what you want to achieve from every perspective and everything you do is to meet that goal, even if you have to make tweaks. You don't find out if it's a scooter or a roadster tire while working on it.
> The best requirements, designs, marketing, etc are worth jack if one fucks up the code.
Why are you mixing some fundamental things which are essential and can't be changed along the way without massive effort and risk, if at all, with things like marketing?
Do you want to aimlessly write code while chasing a target that moves randomly and conflicting because your plan was to define things "at some point"? You're really making my point with your insistence that it's all about code and every other fundamental thing is "a detail" that just comes along the way.
There's a conflation too that to approach things with this level of thought and care requires waterfall design (it doesn't), so people shrug it off or resist because if you want to think carefully and design thoughtfully you can't also Move Fast and Break Things.
Ironically, we all complain about enshittification.
Fixing bugs, improving performance, paying down technical debt all require this as well, which makes them significantly more difficult to actually implement.
Yes. Once it is true, the code is easy to write. But only because a lot of effort went into making it easy. And keeping it easy is also hard. Without focused effort to keep the code clean and easy to modify, it starts to rot.
I don't think you're really disagreeing with the original sentiment? However, I think you're taking a much broader view of "code" than is intended by the original statement.
In your framing, you're kind of confounding code with architecture. Code is really just the act of making a computer do a thing you want it to do, for some definition of "thing you want it to do". Architecture is more about understanding which things you want the computer to do, and in which ways.
There are a million ways to code a task. That's the "code" the original statement is talking about. Understanding which of those ways is an appropriate way is a separate skill, whether you call it architecture, or something else.
The $480M bug that killed a trading company in 12 secs
Challenger
Probably a bunch of stories from healthcare
This might be the reason why many personal projects are so technical and impressive - it's the itch that's not scratched at work.
Scientific programming, hardware interfacing, embedded, demoscene, game engines, HFT or HPC calls for much different breed of code, and generally way harder to formulate in code w.r.t. these enterprise projects. Trying to make hardware go faster with more efficient code is much harder than optimizing an SQL query and safeguarding it, and these are well understood problems, in general.
If you just mean a large company, I would say they do exist - though it may take some looking for them.
You can find them in companies that have to deal with "real" things (hardware, factories, production lines), or where there is an interest in taking advantage of emerging technology (advertising, e-commerce)
I would call my current project relatively systems-level too, as it's a network proxy. Not quite kernel level but definitely not trivial "if this then that" style coding.
My perspective is that application programming - CRUD, forms, IO orchestration - was always vulnerable, even before AI. Think about APIs for payments, APIs for subscriptions. E-commerce in a box type solutions.
That's why I always pushed to do more systems level work, on more exotic or weird technologies. It's not because I think I'm a better programmer, than someone slinging Spring code or React forms. But because in this industry it's better to be a goat than a cow.
That's the first time I've heard that idiom. What does it mean?
I think 80% of a SWE's job is to communicate with the XFN partners (either gathering requirements, pushing back, managing up, collaborations, etc) and then plan out the actual coding (gather code pointers, look at past code, plan architecture, talk to the team). The last 10% is the coding. And then the other 10% is the maintenance of that and past code (which honestly should be a lot more but incentives are not aligned well).
It's hard for people to understand jobs they don't do. They imagine we spend 8 hours a day clacking at the keyboard.
I do, but the order of the keys makes a difference somewhat.
Unfortunately it is still the job...
Even when code was not the hardest problem, it was still a hard problem.
‘Enterprise’ software includes SaaS, internal LoB software, integration work, and a bunch of other things I can’t name. The LoB and integration work is usually boring from a technical point of view, so articles don’t get written as much, and they likely wouldn’t do as well on HN, compared to something highly technical about scaling something to serve millions of users. There’s probably many more hours of work, and more programmers, doing LoB and integration, but people in a SaaS bubble don’t see what happens elsewhere.
Coding was the job. But there were diminishing returns in that getting better at coding wasn't as impactful as getting better at all the social skills, big picture strategy, and general scheming.
Maybe "code was never the hard part" should be replaced with "coding wasn't the most important part". But I think we all know what it means. Those higher level skills are the things that LLMs can't do, at least for now. Coding? It can do that, at least sort of.
The really hard part of this does in my opinion not so much lie in the aspects that you describe, but rather in doing this without leaving scorched earth with most/all of the stakeholders involved.
In other words:
Doing what you described is in my opinion something that can be learned, and in my opinion a central reason why many programmers consider this to be difficult is that they never learned it, and/or (related to this) were never given the opportunity to be responsible for all of this, so they lack experience.
On the other hand, navigating the whole office politics, and running the political gauntlet that is collateral to it is hell on earth. The only way to survive this is to give a big "fuck you" to everyone, which I more politely described with "leaving scorched earth with most/all the stakeholders involved" above.
Writing code that does this while being clean and efficient is a lot harder. How many slow, buggy programs have been written because the assigned programmer did not yet have the expertise needed to do it right?
"Navigating customer requirements" is mostly everyone speculating on customer needs and pushing the part they own, and whoever happens to get closer to the higher management's ear, wins. Then market decides if that's is a good thing or bad thing. If it's good, normally the person who pushed this doesn't even receive credit for it, because either the command chain too long or the stakeholder's memory too short and postfactum everyone pretends they authored good decisions and opposed bad ones.
It might be tiresome and exhausting, like all intense politics, but it's not hard in any technical sense. Most mediocre people can do it and do it.
For something to be hard and complex you need rules and professionals on all levels who understand and follow the rules and driven by meritocracy alone. That's simply never the case.
True. And there is another angle: ownership. The author also said “having clarity on the priorities” boils down to “just tell me what to do and don't switch it up every two days”. This is like saying that a programming language designer does not own the spec of the language itself but just wants to write hte compiler. I find such altitude counterproductive. Case in point, many companies hire PMs for their internal infra org. I mean, shouldn't the engineers in the infra org know exactly what they design to build? If you don't want to own what to build, you end up letting someone else tell you what to do, except that the person is neither an expert nor even your user.
Programming is the hard part! I want to make this distinction because to me programming is about solving problems and coding is a way to express the solution.
Designing algorithms, architectures, etc can be done without a programming language. Coding is putting it down into some language.
Advice to job seekers I remember in the 2010s was to not call yourself a programmer because that was where the bad "code monkey" jobs were, but it hadn't yet been much of a thing when I first started looking at the end of the 2000s.
https://www.youtube.com/watch?v=fcjBfSiyI0k
For example, one guy I used to work with wrote this really awesome algorithm about 25 years ago, and I'm responsible for maintaining it since he long retired. I can't go into details, but this is the core algorithm in moving billions of dollars between institutions overnight. It's about 10 screenfuls of c that had been converted from the original FORTRAN 77 with dozens of gotos and weird branching statements and about 20 parallel arrays that store indices for pointer chasing. It's almost impossible for a human to follow (I've actually fed it to an LLM and said rewrite this with for loops and no gotos so I can understand it - and it worked!) but it's blindingly fast. The actual problem it solves can be stated in about three sentences, but programming it was hard because of the context management. The reason that my company keeps getting royalties on this is that it's so hard that they'd rather pay us than write it themselves. But I bet an LLM could write it from scratch now.
So maybe programming is no longer the hard part, or at least context management is no longer the hard part and that humans should move up the chain to help manage the context for LLMs so they can get more done efficiently.
Switching topics a little...much of what I think makes coding the hard part was the tension between big-design-up-front and you're-not-going-to-need-it philosophies. Early in my career I worked in health care and that was BDUF and the coding was easy because program managers spent years defining every screen that would be shown to the users, what queries were needed to fill the screen, all of that. We just took the spec and coded it. Coding was easy. But the failure of BDUF was that it still didn't really match what the customer wanted.
Then enter agile and YAGNI, in that limit, coding is easy, just write what the user story says. But then you have to refactor from what was left behind on yesterday's user story. So smart engineers would cheat a little with YAGNI and say, yes we will put an abstraction in because the next user story. I would say that "good engineers" or "good coders" found that balance in abstraction to move fast but make abstractions not overkill. And I think that's what all the wailing and gnashing of teeth is right now: the good engineers aren't needed anymore.
I can tell an LLM to code something, and as I add complexity it's happy to refactor and manage the context so we don't need to worry about "clean code" or "quality code". As long as what the LLM writes meets the spec, then we're happy.
Ah, sorry long rant and ramble. But I just think the "hard part" has been managing context, and the context we're managing context is just changing. Those good at managing context will be good at coding with LLMs, and those that weren't won't be.
> I would say that "good engineers" or "good coders" found that balance in abstraction to move fast but make abstractions not overkill.
i think this is spot on and could be where the concept of "llm's have no taste" comes from. There is art (and science) in determining the right level of abstraction that satisfies the user story in a performant way while leaving the door open for extension.
> As long as what the LLM writes meets the spec, then we're happy.
going back to just meeting the spec vs the art of perfect abstraction is a bitter pill to swallow and I imagine removes a lot of the joy some found in software development.
You can absolutely not do this
I type them in when I'm in front of the computer.
All the actual work though? That gets done in a space where there's no phone, no people walking up and talking to me, no distracting social media, no screens, just quiet-ish and a couple of hours to think.
Why would they cover programming? That's what all the books on programming are for.
Also I see no need for signal processing to make programming a hard problem. Writing correct code is hard. Writing code that makes incorrect code easy to spot and hard to write is hard.
You can be great at local maximums, communication, establishing support for executing a plan or creating alignment on big efforts, and proceeded to still create a ball of mud. Bug ridden, hard to read, hard to debug.
And for a huge amount of code, it's not very hard to make it good enough, because the requirements, once understood, are pretty straightforward and perfect correctness often matters much less than you would think.
This is true, and this is the thing that makes me want to not be part of this dumb system anymore. If leadership on the same company can't align, that shouldn't be my problem, and I hope they get replaced by AIs that can.
Humans suck.
It's absolutely devastating to team morale. We never feel like we're contributing.
What devastated me this week is that I was slogging so hard for the past several months trying to deliver on what I was asked to deliver on, only took 1 week of PTO out of the 4 weeks I have saved up, spent nights and weekends trying to honestly solve multiple high priority yet HARD problems that 500 other engineers in the company couldn't solve, I'm making good progress on a couple of them single-handedly, yet my manager, who just came back from 3 weeks of vacation just gave me a performance review saying I am not meeting the "bar" for my level and need to do more cross-functional work and amplify my "impact". He's going on vacation again next week to watch the eclipse.
Fuck this. I want to travel, I want to enjoy life. I used to chase eclipses, too. I tried my best, all I ever get is "what you are doing is not enough". What the hell IS enough then? I already don't take vacation and don't exercise, I've put on 7kg of weight since I joined, yet you told me THAT is "not enough". Should I stop sleeping and eating?
Change priorities all you want, honestly I really don't care, and I've dealt with customers too, it happens. Just don't tell me I didn't get anything done. Recognize the fact that I tried hard every time you changed your priority, and I only had 2 months out of 8 to work on your latest priority, and calibrate your expectations to 2 months of work, not 8.
Just yet another case of people seeing only the extremes and not the entire spectrum
You nailed it: in a corporate setting, code is definitely not the hardest part and it’s why companies can sometimes make do with a skeleton crew of offshore engineers who make $20-$30 bucks an hour
The way harder part is building the right thing and just designing the thing soundly to begin with
This type of software is where AI absolutely kills it - problems with tons of forum posts, writing code for systems with a ton of various kinds of quality developer documentation
On the other hand, if you’re doing something novel or sending a $10bn machine to mars, you probably don’t want to yolo it with AI
I've read through these comments and, as is typical of HN, virtually none of them refute or even address the points made by TFA.