I built this site in response to various ask HN posts asking for niche blogs. I know it's [0] getting hard to find stuff you want in the search results or on big platforms.
It is made with next, newcss, and mdx. It should be easily self hostable with few changes so you can have your own curation site.
Edit: I am aware of the firefox issue now. I fucked up and didn't test on it. And it's fixed now.
Everyday I'm listing all the non-commercial blog posts submitted to HN for people to discover new blogs. You can see it here : https://hnblogs.substack.com/ (yeah substack... but free)
I see all the data is stored in one JSON file in GitHub. [link redacted]
Can I ask, why JSON, and why one file?
I'm interested because I'm currently slowly working on an Open Source tool for this use case - community data sets in Git. However I think it's better to have one file per entry. (Easier to edit, much less chances of merge conflicts.) Also while it's currently based a lot round JSON (so we can leverage several useful existing tools like JSON Schema) I'd like to add a YML/TOML/MD-with-meta-data option later as I think that's easier for humans to read and edit (and also because I've already seen several Git repos that follow this format).
I wanted it to be easy to copy and share and since I didn't think curation should add anything that fits the bill, using a single file would limit the number of items people would add to their curation. JSON also has a small limitation in the fields that it cannot have multiline strings which was originally supposed to force to shorten the description.
But yeah you are probably right, I will do something to migrate it later into a directory of files.
You can ping me on an issue there or reply to this comment.
> limit the number of items people would add to their curation
Interesting. I suspect there is a natural limit anyway to how much data one person could curate, given the demands to keep it up to date. It would be interesting to see what a tools UI would encourage/discourage here.
I’m tired of the Web being made up of nothing but four big sites each full screenshots of the other three, so I’ve been planning to do something similar for a while. No blogspam, no content marketing, no SEO nonsense - just a directory of people doing interesting stuff on their own bit of the internet.
It’s why I like the Tildeverse and alt-protocols like Gemini and Gopher so much as they’re defiantly non-commercial attempts to make a space for experimentation and expression that has been lost - or at least obscured - in the horrorshow that is the modern Web.
That's really cool! I remember checking prototype a day or two ago and now it's finished - great job!
I have an idea of making something like "read next" on medium or openring[1] as a service. It's in a very early stage right now, but if you run a personal blog - ping me(info in the profile) I would like to ask a few questions ;)
Since directories are my own personal obsession, I'll pass on my take on the topic: https://www.kickscondor.com/foundations-of-a-tiny-directory/