[AdrianoKF]

I mostly use QOwnNotes (https://QOwnNotes.org) for Markdown-based note taking, synced to my private Nexzcloud instance. If it ever goes away, I'll still have the plain markdown files plus metadata in a SQLite database.

For bookmarks and long-reads I've mostly made the transition from Pocket to Zotero (https://www.zotero.org), originally intended for literature management. One feature I wish it had is support for more complex semantic relationships between items to be able to create a more comprehensive knowledge graph (e.g., A cites B, X is related work to Y, a URL is the implementation of an abstract algorithm in a research paper). Anybody know of a suitable plugin maybe?

[laybak]

I have been working on Rumin (https://getrumin.com/) as a "build your own knowledge graph" tool that works well with existing apps and websites.

It is currently in private-ish alpha, but thought you would find it useful. Happy to do a 1-1 too to show you my workflows :)

[dethos]

I also use QOwnNotes to take and manage my notes and I've been looking for a solution to store and organize bookmarks for some time, never thought that Zotero would be a good fit for that.

Could you share, based on your experience, what are the good parts of Zotero for this use case and where it falls short?

[AdrianoKF]

Sorry, I missed your reply..

Here's my workflow in a nutshell: I use the Zotero browser extension to add items to Zotero, the extension takes care to automatically attach a snapshot of the web page or paper to the newly created entry in Zotero. For academic papers or blog posts, I'll try and extract additional web links (e.g. Github repos) and add those to the entry as well. Within Zotero I have a nested structure of collections/folders to roughly structure the entries by type (papers, blog posts, reports, books, ...).

I make liberal use of tags to classify the content and keep track of my reading status (Reading List and Read), for which I have saved searches in Zotoro, so I can easily find something to read, whenever I have a bit of time.

For academic papers that I've read in-depth, I'll extract the most relevant references and add them to the library as well and cross-reference them using the "Related" feature. This feature is what I was mentioning in my original post - it only allows for a single generic "related" association between entries. If I could customize the type of relation, I could model almost arbitrary knowledge graphs here.

[dethos]

Thanks for the explanation.