- Independent of the browser.
- Local (no cloud).
- When adding a new bookmark, a readable, entirely offline copy is stored (something like the SingleFile plugin). This is used as a fallback in case of link rot.
- Good tagging and organization tools.
Does anything like this exist?
While I'm not close done curating (the dead/expired/out-of-date links)... I needed to collect it all in one central place, and [linkding](https://github.com/sissbruecker/linkding) is fitting the bill quite nicely. I'm using the tags and description field to annonate and sort the mess of bookmarks. It has a simple to use rest API, uses SQLite, and you can import/export bookmarks using the Netscape bookmarks html format. Best of all, it's OSS you can self-host on a RaspberryPi or even for free on say fly.io.
Content discovery based on topics of interest works really well too.
Is there a bookmark manager which integrates with Obsidian? I think it could be a great combination.
Main itches I was trying to scratch:
* full text search
* no external dependencies
* optimised for self-hosting; simple to deploy and uses minimal resources
- Graph based bookmark views where I can hover over nodes to see bookmarks of the same origin.
- Nodes may be clustered/coloured based on folder, however this might be a step too far/not specifically compatible once there are >n nodes.
- The graph would have some of directionality: - I have some ideas here, however this would require iteration to see what works and what doesn't. A quote from Jobs here seems poignant to me, someone fueled not be designing around users, but solving what is technically complex; "Start with the Customer Experience, then work back to the technology".
- Outgress nodes (A -> B -> C) would show a relationship that these were all bookmarked in the same search journey[1], clicking on any node would bring up a time seried vertically linear panel displaying the nodes.
- I'm sure (read: hope) there's some useful way to also label ingress and neighbour nodes in some meaningful way, I just haven't yet thought enough about it.
- Finally full text search over the store (probably using SingleFileZ) would be a must due to link-rot.1: https://9to5google.com/2022/05/04/what-are-chrome-journeys/
So far I'm sticking with FireFox. When I update my bookmarks, I save them to a JSON file, and then export them to HTML, which I use with my other, non-FireFox browsers.
I also have created a sort of "table of contents" system that allows me to narrow in on similar bookmarks. It's too high-maintenance, but it helps
Example, weekly news links...
02-WEEKLY (heading only)
02-01-WEEKLY (folder containing bookmarks)
02-02-WEEKLY (folder containing bookmarks)
02-03-WEEKLY (folder containing bookmarks)
02-04-WEEKLY (folder containing bookmarks)
02-05-WEEKLY (folder containing bookmarks)
Each of these sections (which are folders) has a different slant: Basic news, Fluff, Food & Health, Science News, Computing News, and so on. Some other categories I have go much deeper, and if I need to revise them it's a major pain.
If eventually I need to do something else, I'll probably adopt or create something with Emacs org-mode.
Example: (I just found this, have no idea if it's actually relevant...) "org-mode for browser bookmarks" at https://www.reddit.com/r/emacs/comments/bshrg0/orgmode_for_b..., which points to "org-linkz" at https://github.com/p-kolacz/org-linkz
There is more along these lines at https://duckduckgo.com/?t=ffab&q=emacs+org+mode+browser+book...
I think the problem is I'm too used to remembering the magic incantations used to search for something I need somewhere.
RSS clients, read-later apps, and bookmark managers are all in the same category in my head. I have no idea why the companies who make the most successful app for one of them doesn't make apps for the others. It's exactly the same kind of person who needs all 3 apps.
There was a brief time, back in the '90s before search engines were any good, when I had a long list of bookmarks - but there were no bookmark managers then, and I wouldn't have trusted one if there were, so I lost them all, over and over again, when moving between machines or browsers, and rapidly dropped the habit of collecting them.
The reason is many: I can organize it easier that way, there is no issue having 50 or 100 links in a note and I can write notes around them if I want.
I like that I don't have to organize anything. I just chuck it in there and let the system organize it for me.
* A spreadsheet
* A markdown file, stored in a git repository