Ask HN: Personal Database Tool?
I'm looking for a tool where I can dump content (images, notes, quotes, names of movies), tag each item with multiple tags ("book" "completed" "5 stars"), and then search for the content by tag ("show me all books I tagged with 5 stars"). Is there a good tool I can use for this?
(
20 comments
| 24 points
| on HN
)
To expand:
>dump content (images, notes, quotes, names of movies)
The images I would keep in relevant logs/wikis, or a timestamped/datetree journal (in it's own file, a header, or a subheader - whatever's more suitable/comfortable). I mostly save images for fun and logging purposes though. The notes I would keep under a (tagged) header that it belongs too, or in the datetree diary if it's personal. Same with the quotes. I keep names of movies I should watch under a (tagged) 'media' header in my tasks.org file (my other org files are wiki.org and archive.org, fwiw).
>tag each item with multiple tags ("book" "completed" "5 stars")
Tags are a built-in feature, but for the '5 stars' portion, you can go a step further and assign a key-value property to an item, say 'STARS:' and a value, like '5'.
>search for the content by tag ("show me all books I tagged with 5 stars")
And this would be done by doing a search, a tag search, with this syntax specifically: `+books&+STARS="5"`
It's a very flexible system!
To me though, one of the killer features of org-mode (compared to some of the other, SaaS product offerings) is that under the hood it's /just/ a plain-text file. You know (unlike the SaaS stuff) it'll be around for 30+ years. To me, it's incredibly valuable when building my own personal information repo/db to know it'll be around as long as I am. Not contingent on a company to remain around or profitable to keep access to my data.
and what are the limits to keeping everything in one file? theoretically i see no downside to it.
My org files (all of which reside in my agenda directory, ~/org) are divided primarily into archives.org (i keep a datetree log here), wiki.org (bookmarks, notes etc), and tasks.org. With sparse trees, narrowing to subtree etc I tend to keep big monolithic files and just filter for the view I want, though many seem to prefer smaller ones (i find it harder to sync more files and directories though, and I’m unsure of the advantages more files provide).
So you have all images of all org docs in one folder? Or one folder per org file? The first would be better for refiling across files, the latter better for portability of the org files.
How many lines does your bigger org than have? My main one has 1.5k. My archive (just a year old maybe) has 15k and many org functions become quite slow, even freezing emacs sometimes.
I'm not sure what you by 'all images of all org docs', but I keep most my org-related images in one file (my archive.org, where I store most of them in an ongoing datetree under a relevant header). Most of the photos I keep in org-mode revolve around logging though - things like bread, beer, garden plants, etc - but I also have some images kept in my notes sections. And yeah, all of my org files are in ~/org. I used to use many files and directories, but it made it annoying for syncing across devices, and having shorter paths is nice anyway.
>How many lines does your bigger org than have? My main one has 1.5k. My archive (just a year old maybe) has 15k and many org functions become quite slow, even freezing emacs sometimes.
My largest one is around 1.5k, after some heavy pruning last week. I respect your 15k archive! I only recently started taking my archiving seriously, and I've really enjoyed it so far. Keeping ALL of my logs (photography, work outs, diet, bread making, etc) under one datetree has brought a lot of Zen to my workflow. Unfortunate though that you experience performance issues at that size. Perhaps looking into `M-x profiler-report` could give you some leads? This will at least tell you what functions are using the most CPU or Mem, and you can potentially adjust accordingly from there. It is likely some package(s) using too many resources.
At the time I used it quite successfully to manage research material across a range of projects; it ultimately hit a wall because it was an electron app that consumed huge amounts of RAM.
However, they are still around and appear to have moved away from electron so you may have more luck now.
However, notes, quotes, movie names etc. can all be listed in a spreadsheet and can be searched, listed, filtered, etc.
It's mostly driven by folders and symlinks, so you could create your own with a few shell scripts.
https://perkeep.org/
You can also do a search here for the name to see other posts on the project.