I wish there were no "software products" at all. They usually just come in the way. I favor open source, and not imposing unnecessary constraints on users. Every piece of software should be swappable, and tying it to a vendor makes it less so.
The problem with "vibe coding"
111–120 of 154 posts
Re: The problem with "vibe coding"
#112One of the genuinely positive things about tools like Copilot and ChatGPT is that they empower people with minimal development experience to create their own programs. Little programs that do useful things - and that’s awesome. More power to the users. I've changed my outlook on vibe coding after seeing the results of this wholesome vibe coding: https://www.youtube.com/watch?v=4pJUXocn7aE . This guy didn't just make…
I didn't look at this video, but be vigilant when seeing one, as I was also surprised by someone demonstrating what they can do with Cursor and I went so far to install the exactly the same version of the app, use the same model and everything (prompt, word capitalization...) I could gather from the video and the results were nowhere near what was demonstrated in the video (recreating mobile web page from screenshot).
I know that LLMs are not deterministic machines, but IMO there is a lot of incentive to be "creative" with marketing of this stuff.
For the reference, this was less than two months ago.
Re: The problem with "vibe coding"
#113Earlier quoted context omitted.
I mean that's not a great counterexample because outsourcing is famously difficult to get right, and from my own experience often results in exactly the kind of terrible code quality GP is talking about.
My point was that a blanket generalisation is bad because it's a blanket generalisation, the details are unimportant. Yes, outsourcing is famously difficult to get right, and yet we've been doing it in every industry for decades now. And the famous examples become famous when something fails (boeing et. all) but there are also success stories out there. The trick, as always, is in the implementation. My other point w…
I am yet to be part of any outsourcing project that was actually a sucess story, where everyone came unarmed without battlefield scars, in a timeframe of the last 25 years that outsourcing started to be a common thing in enterprise consulting.
Re: The problem with "vibe coding"
#114Earlier quoted context omitted.
> ask for an explanation That is risky. The AI might just hallucinate an explanation.
So can any human-based sources. A healthy learning experience with AI takes discipline, just as it does with humans.
The amount of errata in technical books is very small compare to the length of the book. I believe it ranges from none to a page or two.
Re: The problem with "vibe coding"
#115I wish there were no "software products" at all. They usually just come in the way. I favor open source, and not imposing unnecessary constraints on users. Every piece of software should be swappable, and tying it to a vendor makes it less so.
similar thought if you were running a sport franchise as a GM and saying imma make a team where every player is swappable except of course swapping lebron might not be all that feasible :)
Re: The problem with "vibe coding"
#116Sometimes I want to cook a home meal. If I wanted to open a large world class fast food chain, I wouldn't be cooking home meals then. That would be silly. Copilot can help me cook the equivalent of a McDonalds special in terms of software. It's good, I think McDonalds is delicious. But it cannot help me cook a home meal software. It will insist that my home made fries must go in a little red box, and my fish sandwich…
I am in the middle of my third AI assisted project. I disagree ~90%. If you prompt an LLM like an architect and feed it code rather than expecting it to write your code, both ChatGPT 4o and Claude 3.7 Sonnet do a great job. Do they mess up? Regularly. But the key is to guide the LLM and not let the LLM guide you, otherwise you'll end up in purgatory. It takes some time to get used to what types of prompts work. Remem…
I wonder how many lines of code you write in a day. Most PR I’ve seen has a diff in the lowe 2 digits, unless it’s refactoring or formatting.
Re: The problem with "vibe coding"
#117> To me, programs are “works on my machine” code. My main takeaway from vibe-coding is that nobody cared enough to fill that niche and expectation. And it was really frustrating, yet we're getting there through convolutated, inefficient and borderline barbaric means. People are still lamenting after HyperCard. Automation on windows or macos didn't go anywhere. Shortcuts were a better step into that direction but I fe…
Python is not exactly a hard language to learn. But the truth is that people generally don’t like to devote time to what they deem unimportant even if the end result would be useful. I’ve seen smart people go though dense mathematics and physics books, but refuses to take an introductory course in programming.
Re: The problem with "vibe coding"
#118It’s implied that programs are personal but product code goes through peer review, checking the premise of the change as well as the implementation. When someone reviews vibe coded patches and gives feedback, what should the reviewer expect the author to do? Pass the feedback onto their agent and upload the result? That feels silly. How has code review changed in our brave, vibey new world? Are we just reviewing idea…
With the Altair you had switches to input bits of CPU instructions. Then punchcards. Then Telex. Then assembly in a terminal. Then C, Pascal and Smalltalk. Then Java, C++, Python, PHP. Then the mountains of libraries and frameworks, realtime indexers, linters, autocomplete, suggestions. The next step seems that we will write programs in prompts. Maybe we will get a language that is more structured than the current na…
You can’t eliminate formalism in programming. You can only abstract it away by creating a metaprogramming layer for the current one. And prompt isn’t that. Prompting is a blind shot hoping to hit something good.
Re: The problem with "vibe coding"
#119Earlier quoted context omitted.
So can any human-based sources. A healthy learning experience with AI takes discipline, just as it does with humans.
The amount of errata in technical books is very small compare to the length of the book. I believe it ranges from none to a page or two.
That sounds like a similar failure rate for using newer models in technical contexts. Especially for popular technologies where there are enough resources such that someone other than the original creators could reasonably write a book about it.
Re: The problem with "vibe coding"
#120>they empower people with minimal development experience to create their own programs. The author goes out of their way to play up the toy aspect of the LLMs when driven by the inexperienced. No mention is made of them being used by experienced developers, but I get the impression he feels they aren't useful to the experienced. I'm just playing with a small client/server tool (rsync but for block devices), and vibe c…
> vibe coding allowed me to figure out some rough edges of my design document, and experiment with "what would it look like built with threads? As async code? As procedural?” This doesn’t fall under my understanding of the phrase “vibe coding”. In the tweet from Karpathy which many point to for coining the phrase, he says that when vibe coding you essentially “forget the code exists”. I think it’s distinct from regul…
I surely did a hybrid approach, but for a large swath of it I was pretty purely "vibe coding", where I wasn't looking at the produced code at all.
I initially had Gemini 2.5 build some code and then did a code review of it. I tweaked my design document and then had Claude and ChatGPT take a stab at it and at this point I wasn't looking at the code at all. These implementations had some deadlock or early termination problems so I asked for some different async and threaded implementations, but ultimately they went off into the weeds in ways I couldn't really reason about, so I went back to a straight select/procedural implementation and had it start tracking outstanding blocks, which got something working reliably. Then I asked for fancy progress displays, tried several attempts at that. Then at this point I started cleaning up the code some, had it add type annotations, etc...
Big swath of pure or nearly pure vibe coding in the middle there.