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
Emacs Org-Mode!
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 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!
I'd agree with this suggestion, especially if you're willing to put in a bit of legwork to start and learn some of the basics of org-mode.
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.
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.
Is there any way to use org-mode on mobile? I feel like the best solution would be something where you can take notes on the go too
I use Emacs with Termux on Android. You can use the volume buttons as modifier keys, so it’s not so bad. Other solutions like Orgzly and Beorg exist, but I haven’t tried them personally.
do you have a nice workflow for images and files?
and what are the limits to keeping everything in one file? theoretically i see no downside to it.
and what are the limits to keeping everything in one file? theoretically i see no downside to it.
What do you mean about the images and files, specifically? If embedding images within an org file, it is super easy, just do ‘[[file://image/path.jpg]]’ and run M-x org-toggle-inline-images (i think that’s the command).
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).
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).
Thanks for the reply!
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.
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.
>So you have all images of all org docs in one folder?
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.
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.
Airtable
https://airtable.com/
A few years ago there was an OSS tool called Tagspaces that did exactly this. It was great, and briefly replaced Evernote for me.
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.
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.
I know this is not the answer you are looking for, but Microsoft Excel (or a spreadsheet program) can help you with this, except images.
However, notes, quotes, movie names etc. can all be listed in a spreadsheet and can be searched, listed, filtered, etc.
However, notes, quotes, movie names etc. can all be listed in a spreadsheet and can be searched, listed, filtered, etc.
There a re CLI tagging tools like this: https://tmsu.org/
It's mostly driven by folders and symlinks, so you could create your own with a few shell scripts.
It's mostly driven by folders and symlinks, so you could create your own with a few shell scripts.
I'd suggest to try Notion and AirTable. Notion is slightly more then just personal database, AirTable might be useful if you like working with tables/databases.
Take a look at Perkeep.
https://perkeep.org/
You can also do a search here for the name to see other posts on the project.
https://perkeep.org/
You can also do a search here for the name to see other posts on the project.
Most note taking tools allow all that - organize content, tag, and search. They differ in user experience. Try Evernote, OneNote, Zotero, but there are many more.
Recoll search engine is a mature, robust, and battle tested solution. Set up is fairly straightforward and there a webui available as well.
Check out Bookstack https://www.bookstackapp.com
I could imagine a simple cli with S3 or some other key value cloud storage could be used to accomplish this.
Try using a text file and grep?
Notion.so?