Ask HN: Where do you save links, notes and random useful stuff?
10 points
16 hours ago
| 15 comments
| HN
I have 2,600+ notes in Apple Notes and can barely find anything.

My kid just dumps everything into Telegram saved messages. Running a small research - curious what systems people actually use (not aspire to use).

Do you have a setup that works or is everything scattered across 5 apps like mine?

TheKnack
59 minutes ago
[-]
I used to keep everything in Obsidian, but I recently switched to keeping notes in Obsidian and links and articles in Karakeep (self-hosted).

https://karakeep.app/

One of many things that I like about Karakeep is that when you save a link it captures both a screenshot and text from the page, and uses AI to create tags and a summary for the link. Basically it automatically categorizes everything that you save.

reply
weird_tentacles
15 hours ago
[-]
The core idea of Zettelkasten:

1. ONE (shared) dump-pile of all new notes. Your 2,600 pile should do fine

2. REGULAR 'cleaning' of the new notes: a) Each note gets one or many tags (#urban-decay #gaming #assets) b) Each note is trimmed down to its essence, ready to be used for reasonable purposes. (e.g further writing)

3. 'cleaned' notes are moved to your golden store, ready to be found by searching (search "#urban-decay")

You have 1. You need 2. It's slightly work-y, but interesting and ... fun. Rediscovering and polishing forgotten dust-rubies.

reply
a_protsyuk
12 hours ago
[-]
That's a solid workflow. The "cleaning" step is where most people fall off though - how long does it take you to process a batch, and how often do you actually sit down to do it?
reply
gagik_co
3 hours ago
[-]
I’m also a “text myself” kind of person. I’m using my own chat-based notes app called tetrify, which is now adopting the Matrix protocol for sync.
reply
Arathorn
18 minutes ago
[-]
ooh, that's cool - come tell us about it in matrix.to/#/#twim:matrix.org when it's ready :)
reply
CodeBit26
7 hours ago
[-]
My system is basically a 'digital graveyard' if I don't use full-text search. I moved everything to Obsidian because it's just Markdown files on my drive. For links, I use a simple Telegram bot I wrote that dumps everything into a CSV. Low tech, but it’s the only thing I’ve actually stuck with for more than a year.
reply
sandreas
4 hours ago
[-]
I use a selfhosted flatnotes install with a cronjob commiting the changes to a private github repository.

Works pretty well

reply
longitudinal93
6 hours ago
[-]
I pin "Note To Self" in Signal and drop important stuff there. For less important stuff I have a Matrix room on my own server.
reply
throwaway5465
11 hours ago
[-]
Tools: Zettlr for notes. user?weird_tentacles explained the concept of zellelkasten. These are synced to a cloud folder so I have access to them on the move.

Blog: Compiling notes into 'new' knowledge is challenging and interesting. I try to keep on doing what I did in postgrad research.

reply
a_protsyuk
10 hours ago
[-]
Zettlr is underrated. When you're compiling notes into something new - how do you find the right notes to pull together? Do you browse, search, or does the linking do the work?
reply
throwaway5465
4 hours ago
[-]
Memory, notes hierarchy and filename (I tend to keep notes conceptually atomic and not just the date/time as a filename), tag search, free text search, citation backsearch; I have a bibtex library linked but that's mainly focused on maintaining references to published work- I use JabRef but IMHO that's really too heavy for what I use it for.
reply
theMezz315
15 hours ago
[-]
Google Keep CherryTree - which is much nicer than the web site portrays https://www.giuspen.net/cherrytree/
reply
a_protsyuk
12 hours ago
[-]
CherryTree looks interesting - hierarchical nodes. Do you split notes between Keep and CherryTree by type, or is there a different logic?
reply
snowhale
13 hours ago
[-]
plain files in a git repo, one directory per topic, markdown. search is just grep. the friction of organizing is basically zero which means I actually do it. been doing this for ~6 years, it's messy but findable.
reply
a_protsyuk
12 hours ago
[-]
The "low friction = actually use it" insight is real. When grep fails you - topic you don't remember the exact words for - what's the fallback?
reply
choutos
15 hours ago
[-]
LogSeq, with the "brain" shared across devices using Koofr over webdav
reply
a_protsyuk
10 hours ago
[-]

  LogSeq with WebDAV - nice setup. Do you use it mostly for linked notes/graph, or more as a daily journal?
reply
HardwareLust
15 hours ago
[-]
I'm lazy, so I use Google Keep and will probably regret it someday.
reply
a_protsyuk
10 hours ago
[-]
"Will probably regret it someday" - what's the thing you're most worried about losing?
reply
JohnFen
14 hours ago
[-]
I keep all that stuff on a Wiki that I run in my house.
reply
a_protsyuk
12 hours ago
[-]
Self-hosted wiki - what software? And do you access it on mobile when you're out, or is it strictly home network?
reply
LetsAutomate
15 hours ago
[-]
Notion — good for linking related notes
reply
a_protsyuk
10 hours ago
[-]
Does the linking actually pay off when you need to find something, or do you mostly just search?
reply
journal
13 hours ago
[-]
in md files in the file system.
reply
a_protsyuk
10 hours ago
[-]
Do you organize into folders, or just dump everything flat and rely on search?
reply
ZYZ64738
15 hours ago
[-]
...sending myself an email
reply
a_protsyuk
10 hours ago
[-]
Email as inbox - do you ever actually process it, or does it just pile up with everything else?
reply