cross-posted from: https://feddit.uk/post/51641374

The other day I was looking for lyrics to a song. I went to a lyrics website and was hit with a wall of ads, despite my pi-hole! I then went to another site that put me into a redirect loop. It got me thinking about privately self hosting a lyrics site. I started thinking that lyrics are just a type of simple static content, and what would be ideal is an application in which you can upload multiple directories of markdown files. Perhaps the directory should be in a standardised .mds (markdown share) format for instance. It would essentially be a zip file with directories of markdown files and a yml file for indicating how it would ideally be displayed. Perhaps with an a-z, or perhaps text-searchable, or both. The styling would be configurable in the app and independent of the mds files completely. Does this kind of standard for sharing simple text or markdown in bulk exist in any capacity that encourages a known file format? I'm aware that static site generators exist, but they seem to be aimed at the creation of documentation, not at sharing it in bulk. I'm imagining easily downloadable recipe books, wikis, lyrics databases. Does this sound like something anyone would be interested in or am I over/under thinking it?

I'm not convinced that the concept was understood so I've decided to add some more context and perhaps the open source community is the correct place to ask.

To be clear, I'm not thinking about a lyrics solution in particular, nor am I unaware of static site generators. I'm wondering if a standard (and FOSS software built around that standard) for a compressed file format that will contain directories of markdown files that could be dropped into a repository without technical understanding would be sought after and useful.

Standards already exist for lyrics formats. For unsynced lyrics, a plain text file is sufficient; no markdown needed. For synced lyrics, lrc exists. It's a standard already understood by servers and clients, and if you're having trouble finding lyrics, look at lrclib for an API or syncedlyrics for a CLI tool.

You might want to look into Gemini (not Google's "AI" offering, despite the name). It's an internet protocol for sharing text-only communication. There's a whole community out there working with it.

I've thought about similar things than you, something like a P2P database that is easy to use and accessible. For Movies, Shows, News, Books, Music, Music Lyrics, Comics/Manga/Manhwa/Webtoon, Games, Scientific Papers, News Articles, Art pieces etc.

But didn't really get to sensible conclusions. I do believe we're facing a kind of information crisis on the web. It's not just more and more garbage, enshittification and competition, and apps instead of websites and privacy and signup. One little economic collapse and half of all internet companies could disappear, with much of the information lost.

For example IMDB used to be back in the day amazing place to discuss movies. It had amazing "secondary literature" threads on many movies. Then they burned it down. Then you did have some websites even copying the old threads and replicating the forum functionality. But the issue is that the central authority was lost. Back then I deleted my IMDB account in protest (lol). Recently they made user reviews no longer visible without signing up.

Anyway, forking source, data or communities often has problems solution because it dilutes what made it great in the first place. I'd love some clever way to make archives of knowledge easier to share and more compatible with each other. The other issue is that one big commercial company "monopolizing the space" for some application makes it impossible for some open source solution to find adoption.

Lets say you have a new movie or lyrics website or database. The website relies on a community of volunteers to add to it, correct it, make it relevant. But the database is not "open source" it's owned by the website owner and you can't replicate or alter it. Even if you download it, it's a big clunky blob where you need specialized software. Or locally install a webserver and database software and configure it and what have you. Even if it's open source like lrclib (which is rare). I wish there was something more accessible. And forkeable, mergable, like a version control system. So that you can just edit your view of a database and that can be shared and pulled to a maintainer.

Mostly I'd love a kind of "Reader View" webarchive for articles / simple web pages, something localhosted or integrated into the browser and which stores and shares stripped down and highly compressed versions of articles. So when I bookmark something it's also permanently cached. And it completely strips ads and trackers and consent cookies. And everything you cache becomes part of the distributed archive.

I'd also like a browser that allows "static webpage torrent" where you just paste a magnet link to access a website or video or database for some content.