Organising tabs without keeping them open

There is a whole category of extension for organising tabs, and nearly all of it is built on one assumption: that you want to keep the tabs and see them better. Groups, trees, vertical sidebars, workspaces, saved sessions.

If you genuinely want to keep them, those tools are good and this article is not for you.

But a lot of people do not want to keep them. They want to close them and cannot, and being handed a better way to look at eighty tabs does not touch that. So here is the other approach: work out what each tab is standing in for, and replace that thing.

A tab is doing one of three jobs

It is the thing you are doing. Not a filing problem.

It is a place-holder for information. You will need this reference again. The tab is acting as a bookmark you have not made.

It is a place-holder for an intention. You want to read this, do this, check this, come back to this. The tab is acting as a to-do item you have not written down.

The second and third are what keep the count high, and they need completely different answers. Which is why generic tab tools help less than you would expect — they treat every tab as the same kind of object, and two thirds of them are not.

Information tabs: bookmark them badly, on purpose

For the reference pile, the honest move is a bookmark folder with a name so generic it needs no thought. Mine is stuff. Everything goes in.

The instinct is to build a taxonomy, and the taxonomy is what stops you filing. If a save requires choosing between Reading, Research and Ideas, you will leave the tab open instead — the decision is small but it is not free, and the tab costs nothing until much later.

You are not building a library. You are building a search index, and a search index does not need folders. Your browser searches bookmark titles from the address bar, which is the only retrieval mechanism you will ever actually use.

Intention tabs: give them a time

This is the pile that does not have a good default answer, and it is where I ended up building something.

An intention has a when buried in it, and no tab can store a when. “I want to read this” means at some point when I am not busy. “Check whether they replied” means in a few days. “Tickets Friday” means Friday.

So write the when down. Whatever you use — a to-do app, a calendar, a notes file — the tab can close as soon as the when lives somewhere other than your head.

The reason I built Latr rather than using a to-do app for this is friction. Typing a task into an app means switching context, writing a title, pasting a URL, choosing a date, saving. That is maybe fifteen seconds and about four small decisions, and I would do it for the important things and leave the other thirty tabs open. Fifteen seconds is not slow in the abstract; it is slow compared to zero, and zero is what leaving the tab open costs.

One keystroke on the page you are already on is fast enough that I do it for the marginal cases, and the marginal cases were the whole problem.

What to do with the eighty you have now

Do not sort them. Sorting eighty tabs takes an hour and you will not finish.

Instead:

Close anything you cannot explain in five seconds. No reading. If you cannot say what it is for, it is not doing anything. This is usually a third of them, and I have never regretted a single one.
Sweep the reference tabs into one bookmark folder in one pass. No naming, no sorting.
For anything with a when, put the page and the when together, then close it.
Leave what remains — that is your actual working set, and it is probably five tabs.

Twenty minutes. The number goes back up, but it goes back up differently: what refills is usually the reference pile, because sweeping at the end of a task is a habit rather than something a tool can do for you.

Then keep it down

One window. Two windows means one of them is a hiding place. Both fill.

Close on completion, not on cleanup. When you finish a task, close its tabs immediately — that is the only moment you still know which ones belonged to it. A cleanup pass a week later cannot tell.

When you catch yourself keeping a tab “so I remember”, that is the signal. Give it a time and close it. This is the single habit that made the difference for me, and it took about a fortnight to become automatic.

Set a default that matches your week. If everything you save lands “in one hour” and you push all of it to the evening, your default is wrong and you are paying for it several times a day. Latr’s defaults take about ten seconds to change.

The measure that matters

Not the tab count. The number of tabs you would be upset to lose if the browser crashed.

Eighty tabs where you would shrug at seventy-five of them is not a tab problem, it is a memory of one. Five tabs you cannot afford to lose is a real problem, and it means the five things holding your intentions are living somewhere that a crash can take them.

Get that number to zero and the other one stops mattering.