Hello Everyone! These last months I’ve been working on a clean-room FOSS implementation of the CLZ software suite. For those unfamiliar with them, they offer collectionists applications for users, particullary aimed at physical media (comics, movies, tv shows, books, music). I hated when they made the switch to the SaaS model because they separated web and mobile apps and each has a separate subscription fee, for each one of their apps. I decided to start a free alternative to it, almost 1:1: https://github.com/collectarr/collectarr-app
The architecture of the service is centralized and can be soft hosted. The app stores everything locally on the users device, the sync service is supposed to be hosted by the user and allows syncing between devices and the core one stores only item metadata (no personal data that users stores on their entries). That metadata is manually curated and will be expanded by the community with proposals and will be hosted on my server when I’ll have the basic schema for most collection kinds going on. In the near future I’ll also make a sync bridge between Collectarr and Floppy so each physical release watching/reading status can be synced with floppy media entries.
AND YES I’VE USED AND I WILL CONTINUE TO USE AI FOR IT, IT’S A GREAT TOOL. All testing is done personally by me, my main motivation is to have a clz free alternative that I personally use. I would love if you could try this (the desktop app is the most stable currently and I recommend you trying it). I accept any ideas, issues and PR proposals. For an indepth guide there’s also the page that I hosted on github: https://collectarr.github.io/collectarr-page/ But keep in mind docs and the manual keep changing very fast these days because of the nature of the applications, and it takes a while to update all of them, I’m the only contributor so far.
TBH i don’t think most of the criticism you’re getting is based on assumptions or trolling. Did you even read the README and other documentation? There’s a lot of placeholder content and it shows that very little effort was actually made. You say you used AI as a tool, but from the commits, it looks like AI did all the work and there wasn’t even much supervision or review of work being done.
How do you expect others to be able to contribute and form a community to support this project? What plans do you have if you can not afford the AI credits to continue developing this project. What do you expect people to do if you must abandon this project?
No this shows that the project evolves fast and there’s no point in updating the readme every day at each change; I’m getting close to a final release soon and I’ll update the readme accordingly. Again, I’m the only contributor, there’s so much I can do. If you got any questions I’m more than open to answer them; at the end of the day the readme is useless for people anymore, agents can just set you up in seconds with any kind of FOSS project. It’s just a matter of getting you entitled head out of your ass.
Even more, what you do is judge a book by its cover, well done mate, you’re very smart.
You are attempting to distribute software without even following the basics of software development (like creating a changelog). By all means vibecode for personal projects to your hearts content. But if you’re going to have other users than yourself, at least do the bare minimum.
there is a changelog bro… are you blind?
fuckin asshole
Keep editing your comment haha. Maybe on the fourth edit you’ll have really hurt my feelings 😁
If you’re intending this for only AI agents to use this, then yeah I guess a README and documentation isn’t necessary, but the reason you’re posting this here is for people to buy in and start using this in their day to day lives. Nothing you’ve presented gives any confidence that this project will be maintained a year from now assuming you decide to give up the project or Claude changes it’s TOS and refuses to write code for these kinds of projects.
It was supervised; I’m testing all of these changes by hand. Yes there are still a lot of placeholders, I’m still just a man at the end of the day and I’m guiding the AI, not making him do the project however he wants. “What do you expect people to do if you must abandon this project?” Well this can apply to any other project as well, AI or not. Teey can just look at my commit history for other projects and my github activity; it’s not that hard.
The LLM is an “it” not a “him”