My Read it later and discoverability systems in 2025
How I’m using Obsidian as a read-it-later app, while using Raindrop as a capture-all inbox, and how I discover new stuff.
Originally published at DanielPrindii.com on July 30, 2025.
Last year, Omnivore closed. This year it was Pocket’s turn.
Pocket was my first bookmarking app. It was a simple app, effective in discovering new things and saving them. Survived a few phone changes. I even found an archive with links saved around 2010.
After all these changes, I started thinking about a simple system, one that can be easily transferable between formats or applications.
You can also see my article I've written when Omnivore shut down.
Raindrop as an early inbox
Raindrop is the first element in my flow. With it, I’m capturing everything that catches my interest: links, videos, newsletters, and podcasts. The app acts like a capture-all inbox.
It’s easy to use on mobile, has a browser add-on and a desktop app.
What I like about Raindrop is the fact that I can have folders (they call them Collections), tags, and I can highlight and take notes. Furthermore, I can have daily backups to Dropbox or Google Drive. Or, I can download them manually.
My Raindrop archive is organised in a few folders:
One for Temporary- for things I’m sure I would rather not transfer into Obsidian
One collection for Archive- for things that most likely will go into Obsidian
One final folder for everything that is archived or read into the app.
I can use tags to prioritise what I need to read or find. I have an urgent tag for time-sensitive items, and one for already read. Raindrop has an AI feature that can suggest tags relevant to bookmarks, so it’s helpful to organise later.
This system helps me keep my Obsidian vault fairly clean, as I can make a first review of what I have saved. And a nice quality of life thing: keeping all saved items in Raindrop, including those exported to Obsidian, eliminates the duplicates because the desktop browser add-on remembers what is saved.
Yes, I do read on Raindrop when I’m processing my archive, but also because the Android mobile app is working a bit better than Obsidian’s. The app is fetching the full content, so you don’t have to switch between sites.
Raindrop has an unofficial plugin for Obsidian that will sync notes and highlights. It doesn’t do full content sync like Omnivore used to, but that’s better than nothing.
Obsidian as organiser and archive
Two elements are transforming Obsidian into a good read-it-later app: its web clipper and the core plugin Bases. The web clipper supports custom templates, allowing you to control the metadata you save and how it is saved.
I’ve organised my saved items in a dedicated folder, called Clippings, where the items are grouped in subfolders based on the save date. You can create a custom template for nearly every site.
My standard template has the following:
The Obsidian Discord is a good place to discover new ways to capture pages. There is also a community-run GitHub with templates.
The core plugin Bases, for now, available with a Catalyst license, brings database power to your notes. Bases works based on every note’s metadata, and you can create nice-looking databases with your notes. For now, there are available two views: table and cards.
As the plugin is in beta, I’m expecting many things to change from one release to the next, but simple use cases work perfectly.
For my saved items, I run a simple database based on the ToRead and Read checkboxes. Of course, you can sort and filter based on the other metadata information like Author, Publish Date, Created Date or any other information you have there.
Discoverability of new ideas
In our algorithmic age, where AI is starting to create a lot of garbage, finding something new and interesting is hard. I’m currently testing Sublime, Linxpy, and Scour.
Sublime says they are “a second brain with soul”, and it’s a recommendation engine based on your bookmarks. Every saved item goes into a card (consider it to be a note with metadata and actual content), and you can organise these cards in collections (folder, really). They have a canvas option where you can play around and visually organise your bookmarks.
The discovery part comes from the public library- a curated list of cards that you decide to make public, and from the recommendations engine. For example, I saved a card with information about the Viking Leicestershire, and the first recommendation is a card with a fragment from Neil Price's book “The Children of Ash and Elm”.
They can import from Kindle, Readwise and X Bookmarks, and have an iOS app. I'm still waiting for the Android version.
Linxpy are a “human-machine symbiosis for professional curated bookmarks, powered by a serendipity engine”. Currently, they are in the alpha stage. You can add links, tags, and create clips, which are a list of links around a topic. These clips can be shared via other mediums.
It is a work in progress, but it has great potential. They have seven guiding principles, inspired by the works of Paul Otlet, a Belgian visionary thinker, documentarist, and peace activist:
Life is about the joy of discovery and exploration.
Humans harbour a fundamental need to connect.
Value, meaning, and creativity reside in the connection.
Sharing is a social activity.
You can’t predict what people are interested in. You can guess.
Accepting or denying advice, recommendations, or suggestions determine the direction of growth.
Fun is the best motivation to work on a project.
Last on the list is Scour- a personalised content feed. From Bluesky posts, Medium, Substack, Reddit, RSS feeds and more, Scour can find a feed and include it in its recommendations. You start by adding some feeds and interests, and it will generate a list of recommendations. In the feeds screen, you can select between your Subscribed feeds and All Feeds-which will include recommendations. It has only a web app, but you can export your interest feeds as an OPML and include them in your favourite RSS reader.
In the end
My mix of Raindrop, Obsidian, and discovery apps can look a bit complicated, but I’m seeing it as a way to stay up to date with what I save and at the same time keep my archive organised.
For any system to actually work, you’ll need to dedicate time to process and clean.