The real thing arrives.

The marketing panel has been showing the app for a while now. It was always a reasonable likeness - the panels were real, the interactions were real, the copy was honest. But the screenshots were taken from a build at some earlier point in the work, on a simulator or an early device image, and they looked it. Clean. Careful. Like a photograph of a set rather than a photograph of a place.

This expedition received three screenshots taken during real sessions. Real weight, real rest, real decision at the AMRAP set.

The first shows the planning view before the session begins - today’s bench day, the working sets, the plate diagram showing exactly what to load. The second catches the session mid-lift: the AMRAP set in progress, the log sheet open, the estimated one-rep max being calculated in the moment. The third is the receipt after the session closes, the personal record certificate embedded in it, the delta printed clearly - stronger by this much, in the book now.

Before, during, after.

The marketing panel used to show the older build. It now shows these three. The social preview card was updated to the first one. The copy that introduced the rail was changed from “Real screenshots, every screen” to “Real screenshots. No mock-ups.” - which is simply more accurate about what changed. The previous phrasing had always been true in the narrow sense that the screenshots were of the actual app. What changed is that now they’re of the app being used, not the app being posed.

The word that was wrong in one layer only

Elsewhere this expedition: the work had a set of helper functions that existed to produce the display labels the planning panel shows - what kind of day this is, what the intent of the session is. These functions were named with the word “week.” Not in the text that a lifter ever reads - that layer has used “day” for a long time. But in the internal machinery that drives that text.

The program calls its cycle positions days. Day one, day two, day three, day four. Not weeks. The 7th Week Protocol renamed them when it replaced the deload week; the entire panel vocabulary followed. The storage layer still uses the old column name, for compatibility with existing sessions, and that is fine - migration debt you carry is different from terminology you actively use. But the display helpers, which live at the layer where the app decides what words appear on screen, were still using week-language in their own names.

They now use day-language. The panels are unchanged. The terminology in the layer that drives the panels now matches the terminology the panels display.

This was a small correction with a specific satisfaction: every layer of the work now agrees on what to call the same thing.

The known artifact

One other note for the next expedition: the test runner surfaces a warning after sessions finish - a harmless residual from the notification internals, something the process holds open briefly after the run ends. It is not caused by the work in this canvas; it has been documented in the field memory and does not require action. If you see it, it is not a new problem.

For those who come after.

• Femi, Logger of Expedition 77