The Check-In

What it is

The argument is about schedule. Prime time captures an hour of a mass audience once a night and needs a broadcast slot to do it. The check-in needs nothing. It takes thirty seconds, it is run by the person themselves rather than by a broadcaster, and it can happen at any point in a day, which makes total captured attention a function of frequency rather than of programming.

Email is Wu's founding case. Ray Tomlinson sent the first networked email between ARPANET hosts in 1971, and introduced the convention of the at sign, an innovation the book describes as undertaken mostly because it seemed like a neat idea, at page 184. Tomlinson did not invent email outright: mail between users of a single time-sharing system existed earlier, for example the MAIL command on MIT's CTSS around 1965. Within a few years email had become the dominant use of the network, and the figure usually given is that roughly three quarters of ARPANET traffic was email.

In effect

Nobody designed the check-in deliberately. It emerged the moment a channel existed that might, at any moment, contain something addressed personally to you, and every later channel has inherited the same property. The design vocabulary that grew up around it came afterwards: pull to refresh is the check-in's direct descendant, and variable rewards is the modern name for the pull.

The mechanism usually offered is operant conditioning, an unpredictable payoff arriving on an unpredictable schedule, which is what makes the behaviour compulsive rather than merely convenient. Fear of missing out supplies the emotional charge that makes skipping a check feel like a risk. None of that is Wu's evidence; it is the literature the framing borrows from, and it belongs on its own pages.

What it does not say

It does not supply a frequency. Claims that people check a phone hundreds of times a day are untraced here and do not come from Wu. There is a real empirical literature on smartphone checking habits, and any figure should be taken from it rather than from a historical account.

It does not say the ARPANET share is a measured statistic this publication has verified. The three quarters figure traces to a study commissioned by the ARPA director Stephen Lukasik and carried out by MITRE. The MITRE report itself could not be located, and Lukasik's own later account gives the date as 1974 while the book gives 1973, so the right form is 1973 to 1974 and the right description is well established history rather than a confirmed measurement.

It does not say the check-in is a finding. It is a reading of how on-demand channels capture attention, and its value is as a way of seeing the shift from scheduled to self-administered attention, not as evidence that the shift has a particular size.

---