Main idea

A card written six months ago reads perfectly well: the sentence is there, the intent seems clear. What has moved is everything around it. The product has changed, the customers have changed, the market has changed, the team has learned, the technology has evolved — and AI has made feasible what wasn't, or useless what seemed important. The card, meanwhile, still describes a state of the world that no longer exists.

That is what makes picking up an old ticket sometimes costlier than starting from scratch: you inherit a formulation that carries expired assumptions without flagging them. Reopening cleanly means reassessing the topic in the current context, not mechanically picking up the old formulation.

Why it matters

This strips a queue's age of the memory value we lend it. An old stock is not accumulated capital, it is a set of assumptions whose validity was never dated.

It also suggests treating expiry as a property of the item rather than a judgment to be made case by case: what ages fast should leave fast.

Nuances and limits

Not everything expires at the same rate. A reproducible defect on a stable flow stays true for years; a hunch about an opportunity holds for a few months.

And age isn't the only cause: a topic recorded yesterday may already be void if the context that produced it was unstable.

Open questions

• Can you date an item's validity as you write it, rather than discovering its expiry on rereading?