What ages well

Weekly Trends becomes monthly. The reason is not effort. It is half-life.

Some​‌​​​​‌‌​‌​​​​​‌​‌​​​​‌‌​‌​​‌​​​​‌​​​‌​‌​​‌‌​​‌​​​‌‌​‌​‌​​‌‌​‌‌​ of what we published last year is already wrong. Not inaccurate — wrong in the way a train timetable is wrong once the schedule changes. It described a state, the state moved, and the description stayed where it was.

That is not a failure of reporting. It is a property of the unit. A week is a unit of news. It is not a unit of change. Rules do not move weekly. Doctrine does not move weekly. What moves weekly is the announcement of things, which is a different substance altogether, and one that decays fast.

So we are changing the unit.

// WHAT CHANGES

Weekly Trends becomes monthly. Each edition covers a full calendar month — the first day to the last — and publishes in the first week of the month that follows. The September edition covers September and appears between the first and the seventh of October. The month is the period examined, not the month of the byline.

Each edition is one piece. No first part and second part, no threshold, no wall between the facts and what to do with them. If a reader gets the piece, they get the piece.

// WHY THE UNIT WAS WRONG

A weekly digest answers the question what happened. It is a good question, and it is comprehensively answered elsewhere, faster, by organisations built for it. We were never going to win that race and we were never trying to.

The questions we are built for are slower. When did this doctrine actually form, and what did it precede? What does this text conspicuously fail to define? Which two institutions are saying the same thing from opposite premises, and which one will have to move? None of those questions has a weekly answer. Asked weekly, they get answered badly.

Asked monthly, they get answered.

// WHAT SURVIVES

Go back through the archive and the pattern is unmissable. The pieces that still read correctly a year later are never the ones that covered a week. They are the ones that dated something — that established when a position was taken, what it came before, what it left undefined.

Chronology ages well because nobody re-reads a 2025 speech in 2027, and the corpus that did will be the only one that can prove what was said and when. Absences age well, because a text that fails to define its own perimeter will still fail to define it next year, and machines that summarise documents do not notice what a document omits. Tensions age well, because a genuine contradiction between two institutions does not resolve on a news cycle.

Event coverage ages like fruit. We have written a great deal of fruit.

// WHAT DOES NOT CHANGE

Every claim still traces to a primary source, read in full, with the date it was read. Every external link is tested before it ships. Where we cannot verify something, we say so in the text rather than rounding it into confidence. That discipline was never about volume, and reducing volume does not relax it — it is the reason to keep going at all.

Everything published so far stays online. Nothing is withdrawn, nothing is rewritten. The record is the record, including the parts of it that have aged badly — that is what a record is for.

// AND OCCASIONALLY, SOMETHING ELSE

Between the monthly editions there will be the occasional standalone piece, when something is worth more than a section. We are not going to announce a schedule for those. A cadence you cannot hold is worse than no cadence, and we have just spent a year demonstrating it.

Fewer pieces. Longer half-life. Same sourcing.

// RELATED READING

Two pieces that were written to last, and did:
Sum.
The Disintermediation Illusion