The Era of Calm Technology
Attention is the scarcest resource of our decade, and most software is built to take more of it. The next generation of products will do the opposite.
Attention is the scarcest resource of our decade. Almost every product you use is engineered to take more of it — and to measure its own success by how much it took.
Open any analytics dashboard for a consumer app and you will find the same metrics: daily active users, session length, retention. All of them reward the same behaviour. An app that holds you for forty minutes scores better than one that solves your problem in forty seconds. The incentive is not hidden or malicious. It is simply what gets measured.
We think that era is ending, and that the products people keep for years will be the ones that ask for the least.
What calm technology actually means
The term comes from Mark Weiser and John Seely Brown at Xerox PARC in the mid-nineties, long before the smartphone made the problem acute. Their argument was that as computers multiplied, the scarce resource would not be processing power or storage. It would be human attention.
Their conclusion still holds: technology should inform without demanding focus. It should sit at the edge of your awareness and move to the centre only when it genuinely matters.
A well-designed thermostat is calm technology. It knows the temperature, adjusts quietly, and tells you nothing most days. It does not send a push notification celebrating a seven-day streak of being comfortably warm.
Three things we refuse to build
Principles are cheap. What matters is what you decline to ship.
No streaks. A streak converts a tool into an obligation. It works — that is the problem. It works by manufacturing anxiety about breaking a number, and the anxiety has nothing to do with whether the tool is helping you.
No red badges. A badge is a claim on your attention made before you have decided the thing is worth your attention. It moves the decision from you to us. We would rather you open our apps because you want to.
No artificial urgency. No countdown timers, no "3 people are viewing this", no invented scarcity. If something is worth doing, it will still be worth doing tomorrow.
None of these are difficult to implement. They are difficult to leave out, because every one of them would improve the metrics we are choosing not to optimise.
The measurement problem
If you refuse engagement metrics, you need something to replace them with, or you are flying blind.
The honest answer is that the replacement is worse at being a dashboard. You end up relying on slower signals: whether people still use the thing after six months, what they say when they write to you, whether they recommend it to someone without being prompted by a referral bonus.
These are lagging, noisy, and hard to chart. They are also much closer to the question that actually matters, which is whether the product is good.
Why this is a competitive position, not just an ethical one
It would be comfortable to frame this purely as principle. It is not only that.
Software is now abundant. There is no shortage of apps for journalling, or notes, or focus. What is genuinely scarce is software that does not add to the noise — and people are increasingly aware of which category a product falls into. Deleting an app that nags you is now a normal, deliberate act of maintenance.
Building calm is a bet that in a saturated market, the products that survive are the ones people never think about deleting.
What we are doing about it
Pravshi publishes one app at a time. The first is Hallune, a dream journal that waits for you and never asks you to write more.
One app at a time is itself part of the argument. A studio shipping four products a year cannot give any of them the attention that makes them disappear into someone's life. We would rather have a small family of things that are finished than a catalogue of things that are merely launched.
The best technology is the kind you forget you are using. We build for the moment after the screen turns off.