This morning I had an insight: for years, I’ve been thinking that I have been using read it later apps wrong. I thought to myself that I save it far too many articles, and then never actually read them later. I thought that I should be more disciplined, and save less and read more.
This morning I thought: that’s ridiculous. I am using these apps in the way that suits me. I do not use them solely as read it later apps. I also use them as apps to bookmark articles I think I might want to read later. And that’s OK.
Rather than change my read it later habits, I want to lean into them instead. It seems to me there are four categories of read-later articles:
Articles I might want to read later. Examples: Articles about TV shows I like, humor, political analysis.
Articles I definitely want to read later. Examples: A front page Wall Street Journal article about my employer, an article about a medical breakthrough for a condition I or my wife have.
And articles can also be classified by timeliness:
“Evergreen” articles that will still be readable in months or years. Example: Productivity hacks, oral history of the making of a favorite 1980s movie or TV show.
Breaking news articles that will be unreadable after a day or so.
I wonder how other people handle their read-it-later workflows.
I always confused between read it later service and bookmarking service. I think the line is blurring. Services like Raindrop.io now provides both services but other such as Goodlinks is only for read later I guess
My workflow is nothing special. I use Readwise reader as both. I collected articles (pull) and subscribed to newsletters from Reader (push). I use my free time (mostly weekend) to read some long articles and some shorter reads in the evening, I can use the search function in reader to find any specific articles if I need too.
I also imported all my collections from Instapaper into Reader and stop my subscription
If given the opportunity, I would break my RIL concerns into categories such as:
Time-sensitive
Not time-sensitive (or “evergreen” as you described).
And between these two categories would be
Articles that have some sort of relevancy to a task, project, or responsibility of mine.
Articles that are relevant but not immediately useful; they pack some intrigue.
Articles that are only interesting
I think a crucial step to engaging with all of this would be to have a place where all of these links can be stored plainly. The list would’ve have to be pruned regularly and rearranged.
Articles that are relevant, irrespective of timeliness, belong in the RIL app.
Articles that are relevant but not immediately useful should stay in the general pool, or maybe sent to a subsection within the pool.
Articles that are only interesting should be either read on the spot (or kept in an open tab; skimmed; or something to get them out of the way) or I should just read the headline and the comments and go “Wow, interesting,” and move on.
Theoretically, I’ve found that the PARA system offers a lot to draw from when tackling what data is important in the midst of so much of it. The whole concept of asking yourself questions to gauge your motivations in “personal knowledge management” is insightful even if not literally practiced.
Because I run the risk of being a digital hoarder, I’ve learned that I have to be really, really intentional about what I put into my RIL app (Goodlinks at the moment.)
To that end:
My RIL app is where I put web content that I need to process—it’s not an archive for evergreen content I need to save for future reference. That’s what Devonthink, Notebooks, and Obsidian are for. Once an item has been processed, I delete it from the RIL app.
When I come across something that I think I’d like to toss into my RIL app, I try to be honest with myself about why, and whether I’m really going to read it or do anything with it in the near term. If I hear myself say “Hey, I might want to read this someday,” I take that as my first clue that I shouldn’t put it my RIL app.
I put some friction into my RIL workflow specifically to save me from my hoarder self. Everything that goes into my RIL app gets tagged. (I have a pretty well-defined set of tags I use across all my apps to the extent that I can.) Most items also get logged in my Daily Note with a check-box next to them. I use Zoterobib to generate a citation and paste it into the predefined “Goodlinks” section of my Daily Note. I try to add a bit of context if I can—who recommended it, e.g., or what project it might be related to, how quickly I should get to it, etc.
I know this workflow sounds absolutely nuts, but it’s the only way I can keep from having to declare RIL bankruptcy every 6 months because my queue has turned into junk heap of forgotten content.
“Process” can mean one of the following things:
I read it (and don’t need to do anything else).
I read it and took some notes for future reference.
I read it and saved the link in a markdown note for future reference.
I read and decided I should archive the whole thing in one of my repositories, either as a PDF or markdown file.
I’ve been using the Reading List in Safari to collect the RIL pages - as a developer, I will search for a bunch of info on my current project and add them to the reading list. I can then refer back to them as they are usually short-lived use items and then delete them.
I also collect the items that I want to process later. Usually Sundays are my day to go through the list and get it down to “inbox zero” status - Reading the pages and deciding what I want to extract out of them or perhaps just add them to DevonTHINK as a bookmark or article that I can search for later.
The reading list syncs automatically to all my devices which is great.
I only save articles to read later that I want to read
I read articles either in little periods of downtime (e.g. waiting for a train) or in dedicated reading time.
About once a month I go through and delete articles I’m not going to read to keep my “inbox” low in GoodLinks.
Reference links live in Raindrop. I might not need to read these beginning to end, but I might need to access them in relation to specific topics in the future.
“Interesting” items live in my read-it-later app. Broad focus capture; this is not a todo list to be completed or an inbox to be zeroed, but a concentrated mass of things that tickle my discovery/interest filter, a source of perhaps unexpected enrichment. No anxiety about not reaching the bottom of this pile. When I have time, I draw things out at random and read what I can— anything I read from here is a gift, not an obligation. I do try to refine the quality of what I capture by nudging myself to quickly skim a whole article before saving it— that’s usually enough to help me filter the things that just look interesting at a cursory glance from the things that actually are. Once I’ve read something from here, I also push it to Raindrop.
Essential/time-sensitive items are pushed to a Reminders list. This tends to be much more focused, and receives primary attention. Items may be tagged for specific projects, areas of responsibility or research interests. Again, once I’ve read something here, I also push it to Raindrop.
I’ve been a longtime user of Instapaper. One of its features is folders. I don’t use these much but they can be used to better organize things into read, maybe read, important, etc.
The feature I do use in Instapaper is favorite/star and that automatically saves readings I really like into my Evernote.
I think Instapaper is pretty underrated nowadays for a variety of reasons but I do love its minimalist interface.
Do you know how sometimes when you state a problem clearly, the solution is implicit in the statement?
Yesterday, after writing the first post in this thread and reading a few of the responses, I realized that I need to make a regular habit of triaging articles in my read it later app.
My current read it later app is the beta Reader from Readwise. It had more than 200 articles in it. So I sat down and started flicking through the list.
Readwise offers several built-in categories: the inbox, where stuff goes first; “later,” for things you want to read later; the archive; and a shortlist for special articles. It seems to be made for exactly the type of triaging I was doing.
As I flicked through the inbox, I deleted anything timely that was no longer relevant, and also deleted articles that I was no longer interested in. Articles that were still relevant got moved to the “later” queue. And important, relevant articles got moved to the shortlist.
After a pleasant 45 minutes doing this, I had an empty inbox, and a long list of articles that I really want to, or need to read it.
That’s helpful. I’ve perhaps been too hasty in adding articles to my EagleFiler repository as “possibly useful” without proper review.
I like the triage idea, and the benefit of letting non-urgent articles “sit” for a while is worth considering: so many “I must do this!!” articles are less alluring after a couple of weeks!
I currently use Safari reading list… I perhaps need to consider something that offers more granularity. Raindrop?
I’m also using the readwise reader app. I am slowly figuring out my own read-it-later workflow because in the past I never really did that except for books. I would either read the thing and decide to keep it and then download via a web clipper for long term storage or read and move on. I’d stack the URLS up in safari as tabs and clean them out once a week at my review.
In Reader I am now clipping the articles I want to read as I find them rather than bookmarking them. However, I keep wanting folders to put things into neat categories as I sort the inbox. I am trying to develop a tagging system that will work but am not happy with it yet.
I am using Readwise Reader, and find that the major change it introduces is the ability to quickly jump into notation and annotation with articles I’ve saved there, explicitly, or articles that arrive there unbidden from my RSS feeds.
This change has led me to split my time with Reader into two activities. First, I triage more and triage faster. Because of the feeds, and because I tend to toss things into Reader that interest me, I end up with numerous new items in Reader every day. That’s not new, but since Reader makes it easy to scan and delete, or scan and archive, or scan and keep in the inbox, I triage every day the new batch of articles in Reader from their various sources. In short, I keep up to date with getting rid of the articles I do not want.
The second activity is rapid reading and commenting on the articles that survive the daily triages. So, I’m also staying abreast of the inflow and reading more and more quickly every day. I’m intentionally avoiding growing a bottomless archive that I will never pay attention to. I have so many GB of old stuff collected in DEVONthink and other former read-later tools, that I’m determined not to let that happen with Reader.
So, it’s collect, cull, read the remainder, make notes. Then through the Reader → Readwise → Obsidian flow that takes place in the background automatically, I can take advantage of the nuggets I’ve gathered via the Reader phase of the day when I turn to the really enjoyable part, making connections, reflections, and writing in Obsidian.
The point of a read-later “workflow” is to grease the skids on the discovery phase of learning new ideas. The faster I can get through that phase, the better.
