You must log in or register to comment.
It’s cool that he vibe coded something for his specific needs but the lamenting of being unable to share it because of the fear of a backlash is a bit premature. I wonder how long he (and others like him) will be able to keep his project going when he eventually adds more features that introduce more bugs and the effort/cost of maintaining the project becomes too much. We’ve already seen that for the majority of people vibe coding stuff, the answer is relatively quickly.
I’ve found interesting projects that have been vibe coded and seem to be well-maintained but I think vibe coded projects need at least 6 months to a year of history before they replace anything critical in my stack.
Hey, I’m the editor/author. I may be overexaggerating the backlash a bit, likely because I spend so much time curating content for the newsletter each week that I’m constantly exposed to the worst-of-the-worst reactions.
Actually given that, if you have a weekly with AI as part of it, tagging [AIT] will allow people to filter it out.
Its pretty ancillary to the newsletter itself so its not really required, but just to mention… if it were tagged, any low effort comments like “slop” or similar would then be removed. I don’t really see that on your posts in general though, so more a thought for later should it become a problem.
I don’t think you’re overexaggerating the potential backlash at all; especially on Lemmy. When I read that blurb at the top I was thinking “Bold move, Cotton…”. If you told me you get death threats over it, I wouldn’t be at all surprised.
I’ve vibecoded a number of apps that I haven’t released for about the same reasons you describe. What I have put out have been little helpers with miniscule surface area. But the bigger ones I’ve built and am improving have been tailored to my needs (or for our group) and help immensely. At one time I could have coded them myself (maybe), but I never would have gotten around to it.
What pisses me off the most is when people come down on a FOSS maintainer that uses it to keep up with bugs and security fixes because nobody can be arsed to spend any time helping them, instead they’d rather sit on the internet and criticize other people’s methods of coping with overload.
I have done the same. I have shared my apps though - just not here. I went to the communities they are targeting. Of course I made clear that the apps were vibe coded and lo and behold everyone was fine with it. It’s just this community that is especially toxic about this topic.
I don’t think you’re overexaggerating the potential backlash at all; especially on Lemmy.
I don’t either, like you said, ‘especially on Lemmy.’ The thing is, AI assisted anything is unavoidable. Even in the case of longstanding, well established projects, we’re seeing AI being involved to some extent or the other. Like you, I’ve used AI to create a small handful of things for my server. One would be when I log in via SSH, there is a banner that pulls a random song from Navidrome’s playlists and displays it on the MOTD. For instance, today’s is:
🎵 SONG OF THE DAY
Blues Garage by Marco Tansini
I did make my code available here with the stipulation that it works on my machine, buyer beware, as it were. But I’d never purport to maintain it even tho I released it. For one, I don’t have the patience to wade through mounds of criticism, and requests for other options. That patience left me decades ago.
instead they’d rather sit on the internet and criticize other people’s methods of coping with overload.
Preach it! Dev toils away at creating an app. Dev releases it to the public for free with no obligations whatsoever, to even consider it. Dev gets hammered for a variety of things including AI. Go make your own app and release it to the public and see how that goes. It’s a mote/beam kind of thing.
I appreciate your candor, but do you understand why the backlash happens?
That’s fair. I’m sure we have wildly different experiences around the discussion of vibe coded projects.
I think some of the problem is the term vibe coding. If I want a small app to help something specific I used to look at it and go is it worth writing it for the return most of the time I was just too busy.
Now I sit down and write up a small document outlining what I want why I want it and the outcome typically I’d use AI then to review it and give any suggestions that I may not have thought of and use that for the seed for the application.
That is the document I use with an agent to actually do the code generation. I push that to a local GIT instance and the document lives with the code and then gets iterated overtime and then I typically use the issues list to add improvements and fixes to it again iterating through the issue before I think about letting an agent do it
To me vibecoding is just putting a sentence in and hoping for the best I like a little bit more structure than that.
Again, most of this is for personal use so I don’t share it widely a few friends have access to some of the stuff I’ve written because it helps them. I only ever released one AI first project on GitHub and I probably won’t go any further than that.
I think some of the problem is the term vibe coding.
Spot on. Most engineers at companies are using LLMs nowadays but when we use it in our daily lives it becomes “vibe coded”.
I’ve started using the term agentically built or agentically driven instead.
@sytone @selfhosted Hello, let me contribute with my 2 cents of experience in automated generation of code, I won’t say “vibe coding” as it means nothing and marketing always uses meaningless words to make concepts more appealing.
I started it the last week of July and for my YunoHost based selfhost echosystem I’m building a multilingual framework for WordPress.
There are so many plugins, paid or free, for this CMS but I’m totally blind and most of them don’t cover my accessibility needs. So, I’m experimenting AI use to create what I personally need. This on-demand code generation, lets me be self-determined on what I want, because I guide the machine with words. Text only. And it responds.
Am I a developer? No. Unfortunately I don’t know how exact the code is, if something written in 10 lines could be managed just with two.
But I have something working for my own project, still in development in fact it’s called MARS (Multilingual Accessible Rudimental System), used “rudimental” without hiding anything, to signal that it’s not hardcoded by a very experienced dev. And chose the acronym “Mars” because this planet is the longest and most difficult human ambition, full of obstacle, exactly how my web experimentation journey is.
I can’t deny I’m trying to study ways to share it with other devs, when it’s ready, I have no ideas of dates and I won’t ask for any money back, if it will ever be shared. But my most relevant thought is not to share it as I’ve got no first-hand verification on vulnerabilities.Last but not least, I’d invite to read this article on blindness, AI, and also coding.
I think vibe coded projects need at least 6 months to a year of history before they replace anything critical in my stack.
Reasonable. I do the same with updates. I wait for all the early adopters work out all the bugs, and when I feel comfortable, I update.
Even hand coded projects need more than 6 months before I’ll replace anything in my stacks. Too many projects start with great intention but then the founder gets bored (life gets in the way?) and they go unmaintained. Really I’m looking for projects that enough developers have got involved in that I can be confident it won’t go stagnant. I could be one of those developers, but I don’t have enough time to contribute to all the different projects I use (I barely have time to contribute to the one project I maintain)
as long as developers are aware of limitations and shortcomings of LLMs in coding, have an understanding of best practices in coding, and are upfront about how they use coding agents, i don’t think there is an issue with such projects or using such tools responsibly
