What the Tabs Are For
Browser tabs are the knowledge system most people actually use. Understanding why they work—and why they fail—says something useful about what knowledge tools should be.
Right now, somewhere between your browser and your good intentions, you have tabs open that you will not close today. Maybe a dozen, maybe sixty. Each one started as something worth remembering: an article, a tool, a thread you meant to follow up on. Some of them are weeks old. You've scrolled past them hundreds of times without clicking. And still you don't close them.
This isn't laziness. It's memory management.
The psychology here is surprisingly precise. In 1927, Bluma Zeigarnik—a Soviet psychologist studying at the University of Berlin—noticed that waiters could recall unpaid orders in complete detail and forget them the moment the check was settled. Incomplete tasks, she found, occupy working memory in a way completed ones don't. The open loop stays live. It intrudes. It demands resolution.
A browser tab is a physical implementation of that principle. As long as the tab is open, some part of your brain treats the intent behind it as unresolved. It doesn't fully let go. The tab isn't just a bookmark. It's a commitment in suspended animation.
The problem is that suspended animation isn't storage. It's slow decay.
Why tabs fail as a knowledge system
A 2021 CHI paper—"When the Tab Comes Due," from a Carnegie Mellon research team—studied what people actually do with browser tabs over time. The finding that stuck with me: tabs accumulate not because people forget to close them, but because they represent genuine intentions that have no better place to go. The tab is a placeholder. The question it's answering is: where do I put this until I'm ready to deal with it?
The researchers also found something weirder. Among the pressures keeping tabs open was what they called "supporting an unrealistic aspirational self"—a version of you who has time to go back, who will follow up, who will actually read it. The tab persists not because you believe you'll close it. It persists because closing it would require admitting you won't.
Most knowledge tools don't answer the placeholder problem. They answer a different one: where do I file this once I've already dealt with it? Notes apps are for recording conclusions. Bookmarks are for filing sources. Neither holds the open, unresolved, "I'll need this for something" state that a tab represents. So the tab stays. Over time you accumulate not a knowledge base but an archaeology of intentions. The tabs from three weeks ago were for a project that's now finished. The ones from last month were from a research detour you never went back to. Some tabs you can't even remember opening.
And because closing them feels like deleting something important, you don't.
What actually closes a loop
The Zeigarnik effect has a lesser-known corollary, discovered later by researchers building on her work: you can interrupt a task without completing it, as long as you make a credible plan for what happens next. Writing down a specific next step—not finishing the task, just committing to how you'll handle it—largely eliminates the intrusive effect. The brain releases the open loop once it trusts that something else is holding it.
This is what David Allen was getting at in Getting Things Done when he wrote about a "trusted system." Not a methodology. A trust problem. Your brain won't stop surfacing open loops until it believes something external has them reliably. The anxious background awareness of the unclosed tab persists precisely because your notes app hasn't earned that trust.
The tab stays open because it's the only thing you trust to keep the intention alive until you're ready for it.
The structure a tab can't hold
There's a hard limit to what a tab holds. A tab holds a URL. It doesn't hold why you saved it, what it connects to, what question it was answering, or what you were going to do with it. Open it a month later and you get the content but not the context—the explanation past-you would have needed to give present-you to make the tab make sense again.
This is the gap that knowledge tools exist to close. Not capture—capture is easy. The hard part is preserving enough context that retrieval is actually useful: what the thing is, why you saved it, what problem it was adjacent to. A URL with those things attached is a record. Without them, it's a tab.
Structured capture—committing to what kind of thing this is before you store it—is the friction that transforms a tab into something retrievable. Not frictionless archiving. Friction with a purpose: declaring intent at the moment you have it, before the context evaporates. That's what earns the right to close the loop. Not better search. Not more storage. Just enough structure to make the capture feel like it actually went somewhere.
The tab you'll close tonight
I'm not arguing against tabs. The instinct behind them is sound: keep the open question alive until you're ready for it. That's how working memory is supposed to function.
The question is what earns the right to replace them. And the honest answer is that a vague note—"look at this later," dropped into a document with thirty other vague notes—doesn't earn that right any more than the tab itself does.
What does earn it is a specific destination with enough structure to find it again. Not a dump. A record. Something that knows what it is.
Most of us don't have that. So we keep the tabs.
Asgeir Albretsen is the founder of Harbor.