For years I've used things like self-hosted wiki (mediawiki, TracWiki etc), Atlassian Confluence, and lately just a Github wiki to note down things I or team mates might forget and having a place to keep a structured knowledge.<p>It all still feels clumsy, no longer want self-hosted services and they are all somewhat hard to edit&navigate.<p>Are you happy with your solution for sharing with the team mates&storing your knowledge for easy retrieval? What is it?
I started using <a href="https://histre.com/" rel="nofollow">https://histre.com/</a> a few months ago and I love it and would strongly recommend it. I can easily save information in notes, which has the option of being connected to a link or not. Then I can easily access the notes I created using tags I create myself (I love that it uses autocomplete!). Histre supports easily sharing your knowledge with others. You can share a single note with anyone, whether they are a histre user or not. You can publish a note just by tagging it with #pub (see <a href="https://histre.com/pub/valeriemettler/" rel="nofollow">https://histre.com/pub/valeriemettler/</a> for an example). You can also create a team and have a shared notebook where the entire team can have one central place to share and access useful information on any given topic. I recently got my team at work to start using it and they absolutely love it!
I'm in the same boat, I decided to turn all the features ive hard coded into my wiki (dokuwiki) over the last few years into a standalone fresh, from scratch, file manager with wiki journal capabilities.<p>I sort of outlined some key points here: <a href="https://tumblr.macleodsawyer.com/post/187317505027/file-manager-alpha" rel="nofollow">https://tumblr.macleodsawyer.com/post/187317505027/file-mana...</a><p>* Making this for myself, but might open source it. Should be able to run independent without server, or on a server for cloud capabilities accessed through a browser since it uses a flat-file structure with a json registry of file names and metadata for collections and records.
I am perpetually changing this up, but the one constant is workflowy. Things like a cars Vin number, small notes like shell one liners, all the way to large categorized outlines for projects. It takes some getting used to, but the key lesson is for a great many things - text as individual files per subject is not the right metaphor.<p>I wish they would support an official api, and it kills me that the native desktop app is an electron thing, but all the ways they have not fucked it up in the last 10 years is what makes it worth looking at.<p>I have a nagging fear that as a VC funded (yc I think) company they will have to sell out eventually.
Let me links to a previous discussion on this topic: <a href="https://news.ycombinator.com/item?id=17892731" rel="nofollow">https://news.ycombinator.com/item?id=17892731</a><p>Are you talking about knowledge storage for you personally or within a work context?
I keep a git repo of a 'commonplace book', which I prune to plain text, and keep scraps of various texts - short tutorials, poems, a password vault, and usage examples. The same repo gets loaded at home and work, and so those examples are available to each other.<p>One subdirectory is more magical, 'data/cheatsheets/<filenames>. If I run a bash function 'cheat $1' with the filename, it will just dump the file into a pager like the less command.<p>When I fizz out on how git works, I run 'cheat git' and get my own notes. I can edit it when something ends up being useless to me, and keep the most important stuff at the very bottom. Eventually it gets drilled into my head and I learn it.<p>This scales to one person very easily. The value is the subjectivity. With some actual editing, this would also scale to a small team of maybe 4 people.<p>Of course, most of the stuff in my repo would not be appropriate for sharing, but the model would extend in general.
I have a folder called "TIL" (today I learned) and under that I have one folder for each technology/topic. Every time I learn or do something that I need to remember, I create a new markdown file in the relevant folder (or create a new folder).
Most of this I publish to the internet, at <a href="https://TIL.secretGeek.net" rel="nofollow">https://TIL.secretGeek.net</a><p>To publish it all I have to do is push.<p>I have a private repo too for TIL's I don't/can't share (and all passwords go in a password manager)<p>At work we have have a custom in-house wiki that stores all of our shared internal notes.
I have a folder `~/Projects/KB/<Subject_Area>/'
with the usual subfolders:<p>- Documents: my notes, Anki flash card files<p>- Books: Related to that subject area<p>- Sources: one off PDF's, or PDF exports of interesting articles on the web<p>- Papers: that I read related to that subject area<p>- Code: which contain git repos of code that is related to any of Books/Documents/Sources/Papers.<p>The idea is all related to one 'Subject_Area` in one directory. I have a large private KB directory hosted on `Gitlab` and backed up to the cloud/2 local disks.
It's a semi regular question and search for personal knowledge base or personal wiki on HN gives a few previous discussions with lots of pointers to software.<p>Three months ago "Ask HN: Do you keep a personal knowledge repository?" with 100+ comments <a href="https://news.ycombinator.com/item?id=20007108" rel="nofollow">https://news.ycombinator.com/item?id=20007108</a>
One of my customers told me that they use my Helpinator for this. It allows to generate a static HTML knowledge base that they update from time to time and release to co-workers. It's mosty HR inner company instructions stuff. It's not that I recommend everyone to use it that way, but it's a fun experience to know that someone uses your tool in a way you never thought of.