Where your station hears: the reception rose is finally on the page
If you run a receiver, the single most useful thing anyone can tell you about it is not how many positions it delivered. It is the shape of its horizon. A hill to the south-west, a block of flats to the east, a clear run out over water — those are the facts that decide whether moving the antenna two metres is worth an afternoon on the roof, and they are invisible in any number that averages directions together.
Until today a station page gave you exactly that kind of averaged number: "typical range, 147 km". It is an honest figure — the median of the daily maxima, deliberately not the record, because one freak tropospheric evening should not be allowed to describe a station forever. But it flattens the one thing you actually want to look at.
Station pages now draw a reception rose instead. Thirty-six wedges, one per ten degrees of the compass, each as long as the farthest aircraft that direction has ever delivered. North is up, the rings are labelled in kilometres, and hovering a wedge tells you its bearing and its distance.
What the drawing will and will not claim
Two rules, and both of them cost the picture some prettiness.
A direction nothing has ever been heard in stays empty. It does not get a short wedge, and it does not get quietly averaged out of existence between its neighbours. The tempting version of this chart connects the tip of every sector into a smooth polygon, which looks like a proper coverage map and is a lie in exactly the places you would most want the truth: over the hill, behind the building, in the gap you are trying to diagnose.
And the rose is the best each direction has ever done, not today's. Range on 1090 MHz moves with the weather and with the traffic; a quiet Tuesday should not erase a horizon that took weeks to measure. Today's figures are on the same page, in their own row, clearly labelled as today's.
It was there the whole time
The honest part of this post is that none of the measuring is new. Our sweep has been recording per-sector maxima since the station pages launched — distance and bearing from your receiver's own coordinates to every aircraft it reported, thirty six buckets, updated every minute.
The page just never showed it. The sweep writes those buckets one positional path at a time, and when the field does not exist yet the database materialises them as an object rather than as a list. The read side checked for a list, found none, and answered "this station has no directional data" — for a station that had three weeks of it. No error, no warning, nothing in the logs: a check that filters when it should have complained. It is the second time this quarter that a mismatch between the shape a writer stores and the shape a reader expects has cost us a feature rather than a crash, and it is worth saying out loud, because the failure mode is always the same — everything looks fine and something is simply missing.
One thing you may need to do
The rose is drawn from your antenna's position, because a bearing needs an origin. If you never entered coordinates when you registered — and most people did not, since nobody knows their own latitude by heart — your station has no rose, and it cannot get one until you add them. The station editor on /feed now nudges about this; a rough position, to the nearest street, is plenty. We never publish it, and we say so on every station page: what we publish is what your receiver delivered, never where it is.
If your coordinates are in place and the rose still looks lopsided in a way you did not expect, that is not a bug report — that is the feature. It is telling you which way to point.