How I process open browser tabs

TL;DR: it doesn't matter how many tabs you have or how you close them; the value I get from them comes down to intellectually engaging with them fully, which is exhausting and rewarding. The most valuable practice I found is “generating” out of tabs, and for that, sometimes, just the tab title is enough. No tool is going to save you.

I have a few modes of “consuming” these tab piles.

In-depth studying

In-depth studying can be, for example, taking one day to go through one page of a math book. It's sitting down with a tutorial and actually going through it, doing all the side exercises, and then reflecting upon it. It's slow af, but it's gratifying, and of course, I learn things. Over the long term, the value of that learning diminishes, sometimes very rapidly, depending on what I focused on. I learned awk and R repeatedly, at times over months, and it's all gone. What stays are some more profound insights that were uncovered just through sheer focus. This is the “it takes a full day to close a single tab” mode.

Of course, a tab could be a textbook that actually would take three semesters to work through, so there's a wide range in what “in-depth” itself means.

Reading and annotating

Reading and annotating is where I sit down with an article (for example, using Reader) and read it to really engage with it. I don't just highlight interesting passages; I put myself in the mindset of having a conversation with the author, of pitting my ideas against theirs. This is a pretty high-intensity activity, and when I do this over the weekend, it takes me about an hour to process a single tab.

This way of processing information is what I find has the most “return on time reading.” I have a reasonably elaborate Zettelkasten system going, and filing thoughts that come out of articles, along with notes, is very productive; it often leads to blog posts. Over time, I have found how to make highlights and quotes, and crosslinking work for me (it wasn't easy!).

The downside is that usually, for every tab closed, 80 more get opened. I can reasonably process about 5-6 tabs this way during a work week, maybe ten if I'm pushing it. On holidays I would average 5-6 per day because I get more efficient once I get into a particular “scholarly” routine.

Reading for entertainment

This would be just reading a tab for fun. I only do this if I just opened a tab. I rarely go back to an old tab and then just read it for fun, as it's usually just more dopamine-rewarding to open a new tab on HN or scroll Mastodon. This is reasonably fast and generally pretty transient in terms of “return on investment.” Sure maybe over the years, I get something out of it, but I consider it entertainment (which is great!).

I think there is a lot of value in just looking at the title of a tab, quickly scrolling through it, and either discarding it or writing a small paragraph about it. Writing a short entry about why I think the link is interesting, what is interesting about it, how I found it, and what other pieces of knowledge relate to it is the most valuable thing to capture. It puts the rich context I have in my head, at this moment, onto paper.

If that little paragraph is stored in a suitable location, it means that the next time I want to study that topic or look something up, I will find it, along with its link, and immediately get context. That is actually extremely valuable. This filing of links is something I am not very good at, and I definitely want to work on making that part of my workflow better.

It's very easy to think that I will remember the nuances of my current thinking in the future, but that's a delusion. I need to write it down. Otherwise, it's gone.

This not only creates a lasting way for me to find that link again (this is where a crosslinked system becomes very useful because it allows me to crosslink the text I am writing at the speed of typing, really)—it seizes the opportunity to do some generative thinking, creating new knowledge that is entirely mine.

Just the fact of writing that little paragraph explaining why I want to read the link in the future probably gives me more “value” (as in, it will help me generate my knowledge in the future) than actually reading for leisure (see 3), because I actually “created” something myself.

This takes about 2-5 minutes per tab (sometimes faster when groups of tabs relate to the same concept). It is also exhausting work; if I do this for two hours, I'll be ready to plop down in front of Netflix and purge my brain with reality TV.

A concrete example: filing the Fennel homepage

A concrete example: I stumble across the the Fennel programming language. Incredibly interesting to me, but also something I feel would deserve a few months, if not a year, of attention to “really” get it. I can file it away under Lisp / Lua / Programming Languages and my daily log in my obsidian vault, maybe skim the website and make a bullet point list of points I find interesting, and link a HN discussion.

Just for fun, I created the entry for Fennel Lisp by “processing” the index page to show what that is like: https://publish.obsidian.md/manuel/Wiki/Programming/Fennel+Lisp

I mentioned Fennel Lisp in June in my daily notes, so I also linked that. If you step through the vault, you can see how I slowly moved from:

Which is better?

So, is any of these ways of processing open tabs better? I like all of them, and I definitely had to build workflows for 1, 2, 4. I am content now knowing that there is no solution, and feeling like you can process 800 links a day is impossible. Instead I focus on time-boxing “quality time”, and just close all the tabs once I'm done, there'll be plenty of high-value quality time the next day.