Rendered at 04:07:46 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
cxr 9 hours ago [-]
Since this presupposes Markdown of the flavor that comes with YAML frontmatter, and that frontmatter is doing all of the heavy lifting, this is more like a YAML database than a "Markdown database".
In any case, nowhere is it indicated why this approach (no matter what you call it) should be preferred over e.g. recfiles.
xamde 8 hours ago [-]
Or just embed structured data in any text file document, not limited to Markdown.
I've recently switched to SilverBullet[1] as my day-to-day note taking and task tracking app, because of it's very powerful ability to embed little queries and scripts directly in a page to build lists/tables/links that based on data from other pages. This looks like an interesting way to leverage markdown libraries in different ways. Nice!
Same! Still getting used to it, but the benefits of being able to host my notes on my Tailscale network are huge. Having a built-in Lua engine to customize the interface and content is a really interesting concept, though I'll admit I haven't explored it beyond adding an AI assisted voice to text button. Still, it's been fun using it!
Specifically, my observation was lots of people use a collection of markdown files (or other data formats) in a git repository to crowdsource data. This ties in really nicely with a lot of static site generators.
I was interested in seeing if a tool could help data reuse and help people who aren't comfortable with git contribute.
adsharma 4 hours ago [-]
For what benefit though?
Much better off storing this in an embedded graph database and use cypher.
MVCC, Indexes, Strong typing, WAL, changefeeds.
The simplicity of markdown and JSON are deceptive. SQLite is a solid choice, but then you deal with graph-relational impedance mismatch.
I told AI to improve on that format and told me having to read all files just to get the frontmatter data is optimizable, sooo... to the trenches B)
AI > The problem with Frontmatter+YAML is O(n) read complexity. To know if a file is a "Person" or a "Meeting Note," the parser must read the entire file into memory, parse the YAML block, and then discard the content. For a vault of 10,000+ files, this becomes a massive I/O bottleneck
skybrian 5 hours ago [-]
Maybe store it in a Sqlite database and present records to the user as frontmatter + Markdown for import/export/editing?
adsharma 4 hours ago [-]
Many competent embedded graph DBs out there. More suitable than graphs on SQLite projects.
lsb 6 hours ago [-]
As we have more powerful AI plus agents that can run on your phone, keeping everything as a pile of markdown plus some compute on top makes a lot of sense
Your filesystem is very much not a database, and you risk data corruption and loss by using markdown files as a database, especially if you have multiple programs reading and writing to and from the same files. Things will start getting weird and you will start rebuilding an ACID compliant database engine in your application layer.
In any case, nowhere is it indicated why this approach (no matter what you call it) should be preferred over e.g. recfiles.
https://ddot.it
Disclaimer: My project
[1] https://silverbullet.md/
Specifically, my observation was lots of people use a collection of markdown files (or other data formats) in a git repository to crowdsource data. This ties in really nicely with a lot of static site generators.
I was interested in seeing if a tool could help data reuse and help people who aren't comfortable with git contribute.
Much better off storing this in an embedded graph database and use cypher.
MVCC, Indexes, Strong typing, WAL, changefeeds.
The simplicity of markdown and JSON are deceptive. SQLite is a solid choice, but then you deal with graph-relational impedance mismatch.
I told AI to improve on that format and told me having to read all files just to get the frontmatter data is optimizable, sooo... to the trenches B)
AI > The problem with Frontmatter+YAML is O(n) read complexity. To know if a file is a "Person" or a "Meeting Note," the parser must read the entire file into memory, parse the YAML block, and then discard the content. For a vault of 10,000+ files, this becomes a massive I/O bottleneck
Your filesystem is very much not a database, and you risk data corruption and loss by using markdown files as a database, especially if you have multiple programs reading and writing to and from the same files. Things will start getting weird and you will start rebuilding an ACID compliant database engine in your application layer.
That said, I love markdown. It's great. Simple.