Skip to content

Comment on Show HN: MdSilo – A knowledge silo that runs in your web browserparent

Comments

Just looked at Obsidian - seems interesting and the graph connection seems neat but also a bit shiny object like. Do you find it particularly useful?

- Lively ecosystem of extension developers

- Acts as a GUI to manage markdown files which live on your system

- Works well as a dump of info

- Gives you features such as search and tag handling to your MD files

I feel like I have tried all the notes apps and this is the best for me by far. The editor mentioned in this thread is "Roam like" which is great, but I like something a bit more "lo-fi" in that it's basically just a manager for plain text files. I feel the functionality of "Roam like" is distracting because I think about things like "where should I put this" or "what title should I give this?" Also distracting is that they feel too much like a "no-node" platform on which I find myself building an application. In Obsidian, I name files as numbers based on date and then discover them via a hierarchy of tags and through search. Since the files are already on my system, I can also use other applications to manage or search these files.

Thanks - some helpful points there. As a function of the search and tag handling - do you find that you do use this or its just a nice to have option?

The tag handling and search are critical for me, and they work together. Because I name all files based on the date, I use the tagging system as a way to put files into buckets. I also create a "field" in the markdown file for a description. I can then run a search query which outputs a list which shows filename, tags and, the description field for each result.

The way I use tagging, allows for a hierarchical layout. For example, you could make a tag #layer1/layer2/layer3 and those tags will show in that hierarchy in the sidebar. If I click on layer3 in the list, then I'll get all the files in that tag. If I'm doing a lot of work in a certain bucket, then I might make an "index" file which has an embedded query to list files in that bucket. I could also add more fields to query against if I wanted only a subset of those files (project.)

The tool sits in that Goldilocks spot for me. Compared a "Roam Like" I feel it's a bit hacky and messy while still looking and working well. "Roam Like" feels like what you would get if you aimed more for perfection and more powerful functionality (remember that "Roam Like" is running on a database, while Obsidian is using MD files.) The trouble is, my brain breaks down when trying to be messy in a squeaky clean tool. And everything I do is messy.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.