Ask HN: Show me your half baked project

Over the years I've amassed a collection of video files with filenames similar but not exactly like, "Lionel Messi vs Miroslav Klose".

The actual data set is dirtier, can't be regexed, etc. A key property of my data set is that a name features in the filename of different files.

I made this to easily answer the question "Hey, I want to watch a video about Messi. I wonder what I have?"

So, being the computer scientist I am, I thought it'd be interesting to index the filenames. Hence, the indexer whose code I experiment with a lot; dirtier than usual, very "academic" style.

Then I realized all I need is a search function (which I have implemented already!). Life works. Maybe I'll keep with the indexer idea but just for something to play with.

P.S., I have words to say about React but this isn't the thread for it I guess. Been a frustrating experience so far. But well, that's why I used it in such a rough hack in the first place.

Gatekeeper aims to enable small and medium-sized enterprises to have their own on-premise IAM solution that supports all relevant protocols and standards and is secure by default (by offering automated updates and using memory-safe languages etc.).

Features include, for example, LDAP, OpenID Connect, SCIM, and Gatekeeper as an identity-aware reverse proxy. (with fully managed ACME certificate management)

The tech stack is ASP.NET Core + Postgres on the backend. The frontend is written in C# and uses Blazor to run using Web Assembly. If someone is interested in taking a look, we are working on a hopefully helpful Developer Documentation (https://docs.gatekeeper.page/developer/)

The code on GitHub: https://github.com/GetGatekeeper/Server

[0] https://en.wikipedia.org/wiki/Posterous

The idea of this project is to make a small, cozy kind of “book club for code” that enables club members to study and discuss the architecture/structure of open source code. I wanted it to be both a way to study others’ work as a means of developing more intuition about how to structure software, as well as a creative exploration and discussion of code as art.

I know the UI is quite poor right now, and the information about the time of the first session needs to be updated. I’ve pivoted to working on another project (a CLI reader for HN that lets you navigate with a UI like that generated by pager utilities such as “more”) while I flesh out the idea for this one more, as I’m not sure if I can reconcile the two aims for the club I cited above. I have to think more carefully about what I want it to be, and who I want to reach/how I want to reach them.

I really like the idea of doing this, but I just don’t have the mental/emotional bandwidth for it at the moment, which is why I’m working on other, more solitary projects instead. I think it’s important to have discipline and “grind” through things at times, but when I’m doing something in my free time for my own enjoyment and already feel a bit burned out from my other existing roles/responsibilities, I prefer not to add to it by actively building something that might create even more expectations from others. Instead, I try to get back to my first passion, which is just the joy of building programs bit by bit and understanding every piece of what I do in a gradual, methodical fashion.

A related idea I’ve had is making a YouTube channel where I do a similar thing to Code Reading Club, but just break down different applications on my own, bit by bit. I’d include architecture diagrams, doodles, and source code snippets. I imagine it being something like MaiZure’s Decoded: GNU coreutils project[0].

[0]: https://www.maizure.org/projects/decoded-gnu-coreutils/