91–100 of 103 posts
Earlier quoted context omitted.
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).
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…
Earlier quoted context omitted.
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 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…
Earlier quoted context omitted.
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…
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.
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 poi…
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.
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 mak…
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.
Earlier quoted context omitted.
> 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.
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.
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 ri…
Interested in what you’re doing with gaming. Are you literally extending game code?
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.
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 mak…
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 thin…
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.