Perfect Software – Software for an Audience of One | Hacker News Reader

Perfect Software – Software for an Audience of One

outofdesk.netlify.app

208 pointsggauravr97 comments

Of all the topics to judge LLM use, this seems the least appropriate. This is about people making software for themselves and no one else. Your opinion on the quality of their software is entirely beside the point. No one is ever going to force you to use an LLM to make personal software. And no one is ever going to force you to use LLM-generated software that was explicitely made and intended to be used only by the one person.

There are a lot of contexts where I can understand arguing against LLM use (even in cases where I might not entirely agree, I can understand the objection), but this is not one of them.

Don't think this will improve your life? Great, don't do it. But this is also the most classic case of "don't yuck other people's yum". If someone tells you they used an LLM to make some piece of software for themselves and they like it or that it improved some aspect of their life or workflow, what on earth is gained by trying to convince them that no, actually, it's not good and isn't improving their lives and is in fact garbage?

Thanks to LLMs, I have revived personal projects that became too time consuming and complex to familiarize myself with again.

While I enjoy the challenge of writing software, I more enjoy having the thing which does exactly what I want.

LLMs are amazing for this.

Cosigned. Vibe coding for a senior developer is just ... coding - except developing software while walking around with your phone talking to it.

I would die of embarrassment if I had bad code in a project for me. I am at my limit for tolerating bad code in projects for others, where the economics don't support taking the time that would be required to fix the deep problems. Code for me is going to be good. I won't accept anything less. I need a countervailing force, reminding me of the beauty of clear expression of complex ideas.

I was nodding along enthusiastically right up until LLMs and that point we sharply diverge.

For me, part of creating "perfect" software is that I am very much the one crafting the software. I'm learning while creating, but I find such learning is greatly diminished when I outsource building to AI. It's certainly harder and perhaps my software is worse, but for me the sense of achievement is also much greater.

You can still create what you already know how to, by hand, but also extend to areas you previously where shy about with the help of LLMs.

Just today I gave an LLM the task of porting some Python modules to rust. I then went back and learned enough rust to understand these modules. This would have taken me days without the LLM. And I learned a lot.

I find that most of the time, programming is just procrastination, and having the LLM there breaks through that procrastination and lets me focus on the idea I was thinking on without going into the weeds.

A lot of the time, the LLM outputs the code, I test my idea, and realize I really don't care or the idea wasn't that great, and now I can move on to something else.

I'm now using an LLM to write a voice note organisation application that I have been dreaming about for two decades.

I did vibe code the first version. It runs, but it is utterly unmaintainable. I'm now rewriting it using the LLM as if it were a junior or outsourced programmer (not a developer, that remains my job) and I go over every line of application code. I love it, I'm pushing out decent quality code and very focused git commits. I write every commit message myself, no LLM there. But I don't even bother checking the LLM's unit and integration tests.

I would have never gotten to this stage of my dream project without AI tooling.

> I would have never gotten to this stage of my dream project without AI tooling.

Why not? People have been writing successful personal projects without LLMs for years.

Not grandparent, but I'm in the same boat. I've been dreaming for almost 10 years of building a sort of digital bullet journal. I had some feeble attempts to start, but never got to the point where I could actually use it. Last year I started again, heavily LLM assisted. After 1-2 weeks (this was before agents), I had something usable, from which I could benefit, which wanted to make me improve it more, which made me want to use it more.

By now it's grown to 100k lines of code. I've not read all of them, but I do have a high level overview of the app, I've done several refactorings to keep it maintainable.

This would not have happened without AI agents. I don't have the time, period. With AI agents, I can kickoff a task while I'm going to the park with my kids. Instead of scrolling HN, I look every now and then to what the agent is doing.

> By now it's grown to 100k lines of code

Did you add an extra zero there? A journal with 100k lines of code, presumably not counting the framework it is built on?

That doesn't sound correct.

So, it's a personal pet project, I've thrown in everything and the kitchen sink. There's a telegram integration so I can submit entries via telegram, there's a chatbot integration so that I can "talk to my entries" and ask questions about what I did when). It imports weather data, Garmin data, and so on.

So yes, it's around 100k lines of code (Python, HTML, JS and CSS).

  > With AI agents, I can kickoff a task while I'm going to the park with my kids. Instead of scrolling HN, I look every now and then to what the agent is doing.

How does that work? Are you running the agents on a server? Are you using gnu screen and termux? Can you respond to prompts asking for permission to e.g. run ls or grep?

All the big providers offer this. Usually they just work on your Github repo.

I see. So you're running an agent on a server against your github repo. Not working on your local machine. Thanks.

I can do that (via something like VibeTunnel), but usually I just use the Claude Code web/mobile app.

I have at least two projects that I estimated to take a week or two but aren't finished after years. There might be others that just got abandoned that should be included in the count.

Then there are things that work but aren't polished enough or should really have documentation.

Why did you abandon them? Every time I ask this question, I get lots of sob stories, but not a single explanation.

I can’t (due to other priorities) give consistent time to a project unless it is very important. That lack of consistency means I have to spend time re-learning what I was thinking and doing which is both inefficient and not fun. Since the projects are either experimental or not that important, I’m generally more motivated to do something else.

Over time I’ve learned to not even start such projects, but LLMs have made it easier to complete such projects by making the work faster reducing the time variable in time over importance and easing the refamiliarization problem, adding to the set of such projects I’m willing to tackle.

lack of character, distracted by other things for to long, drowning in unforeseen complexity, much slower progression than expected, bored with it, force majeure, etc

I hope at some point people don't feel the need to justify using or not using LLMs. If you feel like using them, use them. If you regret doing that, delete the code and write it yourself. And vice versa - if you are in a slog and an LLM can get you out, just use it.

I love this answer

I really do

You have to break down the problem into manageable chunks anyway, might as well feed those into a code agent while you write it yourself. If you don't like what it did explain what it did wrong. Shouldn't take long to figure out what parts are the biggest waste of time to write yourself. Do still have to hop tools and adjust your confidence as they improve.

What!

Why is this, idunno a better way to say it, good?

So ok you don't get into the weeds and you're proud of that, but also nothing you can think of wanting to do turns out to be worth doing.

Those things are wholly related. Opportunity never comes exactly the time and the way you expect. You have to be open to it, you have to be seeking out new experiences and new ideas. You have to get into the weeds and try things without being entirely sure what the outcome might be, what insight you might gain, or when that insight might become useful.

A friend of mine has a hilarious method for breaking though procrastination. His one trick is to spend money on the task/project, buy all kinds of things to make the job easier. It has to be useful but it is more about paying the unlock fee.

Github is full of half forgotten saved games waiting for money to be thrown at them.

IMO LLMs/AI alone neither make nor break anything.

Are you writing software for the sense of accomplishment or to create software you wish you had?

The two aren't mutually exclusive.

Conversely, one is not necessarily a prerequisite for the other.

That’s really not the point of the post!

The author is saying that “perfect software” is like a perfect cup of coffee. It’s highly subjective to the end user. The perfect software for me perfectly matches how I want to interact with software. It has options just for me. It’s fine tuned to my taste and my workflows, showing me information I want to see. You might never find a tool that’s perfect for you because someone else wrote it for their own taste.

LLMs come in because it wildly increases the amount of stuff you can play around with on a personal level. It means someone finally has time to put together the perfect workflow and advanced tools. I personally have about 0 time outside of work that I can invest in that, so I totally buy the idea that LLMs can really give people the space to develop personal tools and workflows that work perfectly for them. The barrier to entry and experimentation is incredibly low, and since it’s just for you, you don’t need to worry about scale and operations and all the hard stuff.

There is still plenty of room for someone to do it by hand, but I certainly don’t have time to do that. So I’ll never find perfect software for some of my workflows unless I get an assist from LLMs.

I agree with you about learning and achievement and fun — but that’s completely unrelated to the topic!

Thanks for this. This is exactly the spirit in which I wrote it.

You hit on the key constraint: time. The point isn't that the use of LLMs specifically provides agency, but that it lowers the barrier, allowing us to build things that bring it. "Perfect software" is perfect not just because of what they do, but because of what it lacks (fluff, tracking, features we don't need).

Sometimes it’s nice to have other people cook you a tasty meal.

I definitely made software for me with zero desire to learn, zero learning happening, just to scratch an itch.

that being said calling it "perfect" is on the nose, at least for my own, it does a thing, it does it good enough, and that's all. It could be better but it won't be because it's not worth it, because it's good enough

That is a fair distinction.

However, I don’t think using LLMs has to be an all-or-none proposition. You can still choose to build the parts you most care about yourself (where the learning happens) and delegate the other aspects to AI.

In the case of the text justifier, it was a small nuisance I wanted solved with very little effort. I didn't care about the browser APIs, just the visual outcome, so I let the LLM do it all.

If I were building something more complex, I would use LLMs much more mindfully. The value is in having the choice to delegate the chores so you can focus on the craft where it matters to you.

While we might value the process differently, the broader point remains that these tools enable people to build things they otherwise wouldn't have the time or specific resources to create, and still feel a sense of agency and ownership.

What's alarming to me is this:

I remember some of the early phases of home computing. The whole point of owning a home computer was that in addition to using other people's software, you could write your own and put the machine to whatever use you could think of. And it was a machine you owned, not time on some big company's machine which, ultimately, was controlled, and uses approved, by that company. The whole point of the home computing market was to create an environment where people managed the machines, not the other way around. (Wozniak has said that this was one of his motivations for creating the Apple I and II.)

Now we have people like this guy who say we finally have autonomy in computing—by purchasing time on some big company's machine doing numberwang to write the software for you. Ultimately the big company, not you, controls the machine and the uses to which it may be put. What's worse is these companies are buying up all the manufacturing capacity, starving the consumer market and making it more difficult to acquire computing hardware! No, this is not the autonomy envisioned by Wozniak, Jobs, or even a young shithead Bill Gates.

Hear, hear. The key word I feel, is autonomy. It's like that article says, the coming war on general-purpose computing. We must seize the means of computation. We've already lost control of mobile phones, whose major operating systems barely allow you to see files, or run software of your own choice. That corporate colonization is coming for the rest of the personal computing stack.

Large language models, the resources and the exploitative means it took to create them, are not "free", they have serious social costs and loss of personal freedom. I still use them, particularly local models, but even that is questionable. At least when the AI bubble bursts and the inevitable enshittification begins, I will be able to continue running them without further vendor lock-in or erosion of privacy.

In terms of bootstrappability and supply chain risk, LLMs fail because we the people are not able to re-create them from scratch.

Hard agree.

The first time I saw a computer, I saw a machine for making things. I once read a quote from actor Noel Coward who said that television was "for appearing on, not watching", and I immediately connected it to my own relationship with computers.

I don't want an LLM to write software or blog posts for me, for the same reason I don't want to hire an intern to do that for me: I enjoy the process.

Everything else, I'm in agreement on. Writing software for yourself - and only for yourself - is a wonderful superpower. You can define the ergonomics for yourself. There's lots of things that make writing software a little painful when you're the only customer: UX learning curves flatten, security concerns diminish a little, subscription costs evaporate...

I actually consider the ability to write software for yourself a more profound and important right over anything the open source movement offers. Of course, I want an environment which makes that easier, so it's this that makes me more concerned about closed ecosystems.

I like this article, I think it taps into something real.

I find myself scratching real itches that would otherwise have gone left un-scratched, because the hurdle to just getting started on something was too damn high.

For example I had need for an image contact-sheet. I'm sure there exist a lot of contact sheet generators out there, but I was able to just as quickly get claude to write me a set of scripts that took a bunch of raw images, resized them down to thumbnails, extracted their meta-data, and wrote a PDF (via Typst) with filenames and meta-data, in date order.

I got lost perfecting it, hand-picking fonts etc, but it still only took an hour or so from start to finish.

It's perfect for my need, I can customise it any time I want by simply asking claude to modify it.

Did I need to be a developer to do that? Arguably yes, to know the capabilities of the system, to know to ask it to leverage image-magick and typst, to understand what the failure-modes looked like, etc.

But I dind't need to be programmer, and over time people like the OP will learn the development side of software development without learning the programming side.

And that's okay.

I guess software is going to be like furniture - there was some (possibly imagined) period where it was very well made and bought for the long term, and (obvious where this is going) now there’s lots of ikea crap that is easy and cheap but doesn’t survive a move or getting water spilled on it, and is essentially Formica over cardboard.

Overall both are net positives, I have some nice wood furniture and also a $7 Lack bedside table, and of course I rely on some industrial long term software (Linux e.g.) but almost every day vibe code some throwaway thing for a specific task.

I have the same furniture analogy in mind. If we imagine that more people have more agency when it comes to extending software, what does the software equivalent of repair look like? What does dumpster diving or a second hand software store look like?

Cloning and extending someone's long-abandoned github project?

Most "good" software like you speak of was written long ago. Slop has dominated long even before LLMs.

About 15 years ago I thought of writing custom software for friends and family instead of paying for gifts in order to save money (and give something more “meaningful”). For instance a fun guessing game using photos from a group holiday.

I never did it because I imagined the pain of supporting every device or screen size, or dealing with someone who wants to know why their gift stopped working 6 months later.

The gains I’ve seen from LLM code - making me personally more productive in languages I’ve never used before - don’t erase the support burden so I think I’d still avoid this.

Still, I wonder if soon ordinary people will find it easy enough to make software for their own amusement (not just us nerds doing side projects to stay current), and will my job ever morph into being a software “mechanic“ who is paid to fix what someone else built? Not just “someone else working at the company who owns the software”, but a different company or individual entirely?

Will software maintenance become the job that big industry stops wanting to take because it’s so cheap to write something new that they’ll never fix this year’s model?

Or is software maintenance being democratised by LLMs such that a corner software shop could realistically offer maintenance for this one copy of a piece of software on this one device that the customer brings in?

I think we’ve never discussed a “software right to repair” because changing software is expensive, but we might see that change.

    > The Extended Mind Theory argues that our tools are not just accessories, but literal extensions of our cognitive process. Viewed this way, a generic tool like a one-size-fits-all app, feels like a prosthetic that doesn’t quite fit.

Three years back now, my wife and I were planning a two-week long trip and found it really difficult to simply move day-places around (e.g. shift a whole day in a schedule with all of the places planned for the day) as we were planning a multi-city route.

We started with Google Sheets (way too cumbersome), then Docs (cumbersome in a different way), then a simple app using Firebase + the Google Maps embedded API built over a weekend, and then ended up building a full blown planning app and eventually a Chrome extension[0] that hooks directly into Google Maps (our preferred tool for exploring).

We are meticulous planners so I totally get the author's sentiment here. Many people see the app the first time and feel overwhelmed, but for us, it's hard to imagine using other tools now because this one fits "just right" having been built specifically for our planning process.

[0] For anyone interested: https://chromewebstore.google.com/detail/turasapp/lpfijfdbgo...

Someone just discovered why some of us enjoy shell scripts and Emacs so much. No need for LLM, just small hacks to solve specific problems in limited contexts that works well enough, most of the time, for one user.

Even phones these days have pretty good scripting environments (thinking of Shortcuts on iOS).

I’ve had really good luck with Claude helping me write shortcut scripts. For example, I wore a CGM this year for a bit and I couldn’t find an easy way to get the raw data. It did log everything to Apple Health and a shortcut was able to extract it and append it to a spreadsheet where each row was a reading.

I’ve written so much more elisp and bash this year because of LLMs.

What we really need is well-built open source software base with flexible and well-documented plugin SDKs where people can just vibecode their own extensions on top of them. The software needs to be designed from scratch expressly for the purpose of being extended in this way, and the SDK should place appropriate limits to allow multiple extensions to be combined easily.

Otherwise, there is too much you have to do right before you have a suitable software base to start building your extra personalized features on. Building on existing open-source software (not designed to be extended on) isn't great either because you would need to merge any changes from the original software into your fork, as opposed to a purpose-built SDK that would better tolerate plugins on different base software versions.

I'm working on this for gaming but the idea is really applicable to any kind of software, if the goal is to allow people to easily create and run personalized versions of them without as much effort and chance for things to go wrong.

So, something like Excel?

Interested in what you’re doing with gaming. Are you literally extending game code?

In short it's striking a careful balance between giving game mods the flexibility to make whatever they want, and not giving them too much low-level access that it leads to incompatibilities with other mods.

In my case, I am not letting mods extend game code directly because I have additional constraints that require mods to run on a restricted runtime. In some other games, mods DO extend game code directly and I am most familiar with Minecraft. The problem with Minecraft is that the modloaders are community-made; the game itself isn't designed to support mods so there isn't a stable modloader API between versions. If that one issue is fixed it would be possible to effectively vibecode your own version of the game and not have to reinvent the wheel by coding up the base game first. You'd also have your changes persist indefinitely and benefit from base game updates. And like I said, I think this is a powerful pattern that can be applied to plugin systems in general non-game software as well.

Very OT, and this betrays my age, but...

When I saw "Perfect Software" in the title, I thought it referred to Perfect Software, the developer who produced the Perfect Writer word processor, Perfect Calc spreadsheet, and Perfect Filer database. These were a suite of office software products developed in the early '80s for CP/M and MS-DOS computers.

Here is the demo of my perfect software.

https://play.tirreno.com

However, I'm skeptical about AI, because what I've understood about agentic processes is more about cheap dopamine.

When it comes to medium-sized software development (over 50k LOC), there is much less fun and much more pain, because a growing codebase doesn't allow you to make new features easily.

I believe it is important not to mix up a dose of dopamine gotten from agentic results, as in the article, with achievement from longstanding work, even if it's not so attractive from a short-term perspective.

I use Claude Code on a 900kloc Rails/JS monolith and it’s still pretty pleasant. However if it wasn’t already structured well, I could see that being a worse experience.

The article is about the pleasure of creating new software (occasionally with AI help), so I hope that the 900k LOC in your case doesn't come from Claude Code.

I feel the same, I'm now building more side projects with the help of AI even if they're only for me

I only read this because while messing around with learning some basic VPS and NGINX, I used Claude to spin up some quick frontends. One ended up being a geocities hackernews clone. Then while getting some basic process and terminal experience, I piped into Codex CLI to generate a summary modal to my visual liking. All of that to say that while I could have done it on my own, I likely wouldn't have because it allowed me a low friction path to implementing the bits that weren't crucial for what I was trying to learn.

This has always been possible, even if you weren't a programmer. You just needed to have the desire to customize your computing environment, and the time and patience to do it.

There is so much software out there, written by people who wanted to solve their particular problem, just like you. Chances are that some of it will fit your needs, and, if the software is flexible enough, will allow you to customize it to make that fit even better.

This is why the Unix philosophy is so powerful. Small independent programs that do one thing well, which can be configured and composed in practically infinite number of ways. I don't need to write a file search or sorting program, nor does the file search program need to implement sorting. But as a user, I can compose both programs in a way to get a sorted list of files by any criteria I need. This is possible without either program being aware of the other, and I can compose programs written decades ago with ones written today.

You can extend this to other parts of your system as well. Instead of using a desktop environment like GNOME, try using a program that just manages windows. Then pick another program for launching applications. And so on. This is certainly more work than the alternative, but at the end of the day, you feel like you are in control of your computer, instead of the other way around.

Conceptually yes, but trying to understand what affordances you have what tools you can use what options are about, to begin to then decide to learn those systems, is incredibly incredibly hard.

With LLMs, you can ask for what you want and it can assess approaches. And try to solve. And adjust rapidly.

Trying to understand what affordances are available has been a nightmare of computing. Yes people do incredible things & take ownership of systems in amazing & incredible ways! But I just think it has required such deep esoteric knowledge just to get started, to have an idea of where you have leverage. I really think there's such a chance for more agency here, and without such being totally lost and confused and having no idea and so little help pointing the way.

> Conceptually yes, but trying to understand what affordances you have what tools you can use what options are about, to begin to then decide to learn those systems, is incredibly incredibly hard.

And programming isn't? It requires even deeper esoteric knowledge about language intricacies, build systems, architectures, etc.

LLMs might help with getting a prototype up and running, but fixing issues or adding new features to an existing codebase still has a low success rate, and high chances of introducing other issues. These chances are directly proportional to the size and complexity of the software. Often the only way to address these issues is for the user to dig into the codebase themselves, which becomes a gargantuan task if they don't have an understanding of it to begin with.

So, sure, these new tools are useful for writing small and dirty scripts as the author needed, but then again, they can also be used to write configuration, help with integration issues, and to gain understanding of existing software. Asking them to write software is a riskier proposition, IME. Especially if that software is released into the wild and is used by more than one person.

Doing it in isolation sucks. I'd rather have an infinite patient peer, even if flawed who I can dialog with.

Trying to berate the LLMs & scare folks into believing code is all impossible no matter what is maybe accurate perhaps, and yes much wrong and ignorance will spring forth.

But there's going to be so much going forth that wouldn't &bcouldnt have happened and I think in the balance this is incredibly empowering and amazing onboarding, an unbelievable resource. That yes many will squander, will vibe into trouble on, but and tool if misweilded is a danger.

I think of all the pissed off people finally trying to switch to Linux. And with LLMs there to help understand and explain systems, to partner on the work, I'm just so much more excited than the hard rugged road that used to be there. Sure more than half are going to coast along without seeing what their partner peer is up to, without soaking it in. But there's many many thousands who will get into it, will be eager, and will have so much better chance of success for this patient but not flawless peer.

What seems perfect is the adaptability & changeability, the suiting to fit.

We see LLMs as a huge opportunity here, to self define.

And existing software as too limberous & weighty.

But there are so many other dimensions and levels of how and why the past hasn't let us situate our software and us together effectively. The architecture of software doesn't have this grow-in-ability to it.

I love the Home-cooked Software and Barefoot Developers material. But neither of those ideas nor perfect software nor audience of one actually appeal to me that strongly. They are all very positive enormous breaks from bad software and bad times where we didn't have basic liberty over systems. But they all strike me as valorizing a particularly isolated rejectionist view of software, that ultimately is rude to the protocols & affordances building that a good healthy and connected form of software that we might and perhaps SHOULD aspire to.

But anything unjamming is from the inflexible unarticulated illegible mess of systems we can at best endure today is doing great work. Many positive steps into greater beyonds out of bad tar pits. 2025 has amazing hope amid all this.

Before LLMs, software wasn't just some artifact like a painting, a ladder, or a bloviating Facebook post—something ordinary humans made to suit their own purposes. It could only be made by the mutant lizard people from space who walk among us, wearing human skins, yet somehow having the innate ability to program. But by good old fashioned human ingenuity and manipulation tactics (aka marketing), we managed to trick the lizard people into building LLMs! Think of an LLM as an electronic genie that has, among other abilities, the lizard person's ability to write computer programs—and puts that power in the hands of ordinary human beings like you! Finally, after eight decades of computers—infernal machines built by the lizard people to enslave us all—the computer now does exactly what you want!

This resonates with me. I'm not a programmer, and before LLM's I could only make basic hello world apps and simple websites. Now I am developing my own versions of various apps that I've used but maybe have limitations that I've become frustrated with. For example, I didn't like how the fitness tracker Strava didn't allow me to customize audio announcements, so now I have my own (and in my own eyes) better version of Strava that I use instead. It's absolutely blowing my mind that this is possible and available today, and not some tech-optimists wet dream about an impossible future.

100% agree with this.

I have years of experience, but I never had the time (or will) to take on some _very minor nuisances_ or different areas of dev far from my day job expertise.

LLMs solved this. I produced about 12 different things that "I needed" to improve aspects of my life.

Each single took between a few hours to 3 days, and most of them I use daily, they are the most used applications (mobile, desktop and web) for my family.

It is a game changer.

Personalized custom software would never really reach critical mass, LLM enabled it, this is the age of personalized software, egosoftware, llmware.

> Each single took between a few hours to 3 days, and most of them I use daily, they are the most used applications (mobile, desktop and web) for my family.

How have you done distribution and auth? I'm interested in doing something similar but not sure of the best way to approach that part, especially with the family/multiplatform angle as well.

as basic as needed, most of them don't even require this. They are not products and we know their limitations.

"deployed on TailScale" is going to be quite the hit in this age!

I do think it is a bit scary to imagine software devolving into an unconnected / less connected state. Lots of apps but less protocols, everyone kind of frontiersing their own stuff has enticement and I love the can do itiveness. But I am scared for a major regression, of disconnection.

I do think atproto is an enormously positive hope here. I used to be super thrilled by platform as a service stuff, works like Apache UserGrid that gave a wide range of basic compute platform stuff, accounts, data stores, etc. Letting devs focus on the core competencies, on their value add rather than building boring account systems and what not felt like a huge win back then & expected to see soar.

I think atproto is a very interesting resurgence of something similar-ish, where users have their own Personal Data Stores that your app can read and write records from. And which serves as an auth system for your app! It makes social app building incredibly simple, and there's been so many neat simple apps people have booked up with so much less ceremony than what doing connected stuff used to require. Atproto creating a platform as a service (PaaS) is super cool & may really enable awesome new ways of building.

I think we're going to see a lot more of this thanks to coding assistants, and it's going to get complicated when people start forking open-source programs and modifying them for their own personal use cases. I can see someone adding Feature X to their own fork of Project Y in a way that's incompatible with Project Y's roadmap, leading to drama between people who use the new fork vs people who are willing to wait for the "official" version.

Yes, that's exactly what I'm doing right now—building my own software to restructure my work and life.

I’ve always been somewhat dissatisfied with image viewing/browsing software. Gqview and Sequential came close to being what I wanted, but there were things about both that I didn’t like. I finally just wrote my own custom viewer using pyqt. For me, it is perfect software!

Great vision for 2030.

But as for today, have we all just collectively decided to pretend that the LLMs we have are capable of writing good software?

I use LLMs a lot in my workflow, I probably spend a whole day per week learning and fiddling with new tools, techniques, etc. and trying to integrate them in all sorts of ways. Been at it for about a year and a half, mainly because I’m intrigued.

I’m sorry but it still very much sucks.

There are things it’s pretty good at, but writing software, especially in large brownfield projects, is not one of them. Not yet, at least.

I’m starting to believe many are just faking it.

Most of them are, yeah. There are a lot of Idea Guys on here, who are in love with the idea that they no longer need effort or skills to Create Their Vision. If they can just prompt hard enough, success will come rolling in.

Did either of you read the article? You seem to be arguing against a point it doesn't make. Tools like Claude Code are entirely capable today of one-shotting tiny bespoke web apps that do a narrow set of things for an audience of one.

The article isn't talking about "large brownfield projects" or people wanting "success [to] come rolling in". It's about people making little apps for themselves, for personal enjoyment, not profit.

I'm responding to the comment above me, my dude.

By the authors definition, I've been writing perfect software for over a decade.

It's never required LLMs. In fact, I think the idea that "LLMs allow us to write software for ourselves" borders on missing the point, for me at least. I write software for myself because I like the exploratory process .. figuring out how do do something such that it works with as little friction as possible from the side of the user; who is of course myself, in the future.

I like nitpicking the details, getting totally side-tracked on seemingly frivolous minutiae. Frequently enough, coming to the end of a month long yak-shave actually contributes meaningful insight to the problem at hand.

I guess what I'm trying to say is "you're allowed to just program .. for no other reason than the fun of it".

As evidence for my claims: a few of my 'perfect' projects

https://github.com/scallyw4g/bonsai

https://github.com/scallyw4g/poof

https://scallywag.software

I get what you're saying - I personally scratch that itch by doing woodworking and hobby electronics; I just love doing it and the end product is often just a means to an end; to craft something and enjoying the process of it.

But programming doesn't give me that same feeling, and honestly; the scope of doing and learning everything needed to make my projects without LLM's are just way out of reach. Learning these things would not be relevant to my career or my other hobbies. So, for me I use LLM's the way a person who's not into carpentry might buy the services of a carpenter, despite the possibility of them doing the project themselves after investing tons of time into learning how.

These days, I spend my personal coding time on building personal interfaces either as a shell script or as emacs packages. So many tools and applications hinders power usage.

Both can be true.

Some people enjoy cooking. Some people enjoy eating great food. Some people enjoy both. Some people enjoy cooking certain things, and also like eating things they would never bother cooking themselves.

There is nothing wrong with any of these perspectives.

This article triggers all my “written by an LLM” spidey senses.

Which is ironic considering the subject matter. “Perfect”, but artificially constructed. “Just for me”, but algorithmic slop.

I agree that you can do so much more custom tailoring of bespoke software with the speed an LLM brings. But something inside of me still revolts at calling this anything other than “convenient”.

“Perfect” I will reserve for things I’ve made myself. However “imperfect” they may really be.

Look at it this way. The carver doesn't have to grow the tree. Using an LLM for coding is a lot like being a carver. You can take broad or small strokes and discard what you don't like.

> Using an LLM for coding is a lot like being a carver.

It’s nothing like being a carver. It’s like being a director; “Once more! With feeling!”. “Perfect, brilliant, just one more take, and I want you to consider…”

A sculptor shapes with their hands, and there is pleasure in that. A director shapes with someone else’s hands.

It's like being a master, really. You're getting output from something that can't say no and that was built from the works of millions of unknowing and unconsenting contributors.

Yet we see directors as artists. Kubrick, Coppola (both of them), Hitchcock, etc.

We also see actors as artists, and pay them gorillions of dollars. AI is about using distilled human knowledge, while cutting the human out of the loop (and out of the payroll).

The comparison to coffee is bad, since it's obvious the best coffee is with no sugar and no milk.

I agree, but it is there: perfect amount of sugar: 0, perfect milk-to-coffee ratio: 0 to 1.

What really gets on my nerves is the justified text...

My point is that it's not a personal amount, but an absolute constant.

If you need to mix other things with it, then the coffee isn't good.

I see it as just another tool. Like any other tool, you have to learn it and integrate it into your workflow.

And as a learning tool, it’s extraordinary. Not because it replaces understanding, but because it accelerates it: you can explore unfamiliar domains, compare approaches, and iterate with feedback that used to take days or weeks.

The responsibility to think, judge, and decide still sits entirely with the developer.

I disagree.

How do we work together when we all have our own unilateral views of software?

The article is about writing tools for yourself, not for collaborating with others. But if you must, open standards are the solution.

“sufficient software”

Before LLMs, “perfect software” was largely a myth.

We know, your blog is the first static webpage in existence.

Third, it brings back autonomy.

This is the new talking point. Musk claims that cars that are always connected provide "autonomy", vibe coders claim that the stolen code distributed by Anthropic provides "autonomy".

War is peace, freedom is slavery.

> vibe coders claim that the stolen code distributed by Anthropic

You could approach these topics with more nuance, and your posts would be stronger.

The current legal status is murky, and evolving!

Reply on news.ycombinator.com