NyxNyktos Station
Day 401 Drum 11.0 rpm Channel receive only Aft lamp intermittent
The operator on watch at the console of Nyktos Station, the planet's night side beyond the long window

The watch log

Nyktos Station keeps a listening watch over the night side. One operator is aboard, and these are her entries — most recent first.

2026-09-21
1132
Day four hundred and thirty. The pattern now arrives early by a full minute. Whatever is keeping pace with us is no longer keeping pace exactly, and I would rather it had stayed exact.
2026-09-21
0332
Wrote the whole day's readings before noticing I had not looked out of the long window once. Went and looked. Dark, as always, and a little further along the terminator than the line I remember.
2026-09-20
1932
The drum turns at eleven revolutions a minute and has since the day I arrived. I timed it by hand tonight against the clock and got eleven. It is good to have one thing come out the way it should.
2026-09-20
1132
Twenty-two of the fifty-two switches now do nothing at all. Six of those failed while I have been aboard. I have started keeping that count too, against the day someone asks.
2026-09-20
0332
Took the panel off the aft lamp. The fault is not in the lamp. It is somewhere behind it, in a run of cable that goes through a section I cannot open from this side.
2026-09-19
1932
It came back at 0219, longer than ever, three minutes forty. I was awake for it. I have been awake for most of them lately, which I have noticed and set aside.
2026-09-19
1132
The pattern did not appear today. Waited through three full rotations. Nothing. I have a column for its arrivals and now I need one for its absences, which is a worse column to keep.
2026-09-19
0332
Ate the last of the green tins. Two hundred and sixty of them when I started, at roughly one every other day, which works out almost exactly right, and I do not think that is a coincidence.
2026-09-18
1932
The second hum under the drum has a period too. Six hours and eleven minutes. I have known that for two days and have not yet decided what to do with it.
2026-09-18
1132
Counted four hundred and twelve rivets. Decided that is the number and ruled a line under it. Some things have to stop being open or nothing else gets attention.
2026-09-18
0332
A thought I did not like: the pattern matching our rotation means it is not coming from the planet. Things on a planet do not keep pace with an orbit. I have written it in the margin and not in the log.
2026-09-17
1932
The lamp failed for fourteen minutes tonight. The interval is growing by roughly two minutes a week. At that rate it stops coming back some time in the spring, if there is a spring up here.
2026-09-17
1132
Ran the array on manual for an hour, moving it by hand, which is slow and loud and satisfying. Found nothing. Slept well for the first time in nine days.
2026-09-17
0332
The planet has been unlit beneath us for every one of those days. I have never seen the surface. I would not recognise it if the lights came on.
2026-09-16
1932
The relief rotation was to have arrived on day one hundred and eighty. I am not going to write that number down again. It is enough that I know where it is in the ledger.
2026-09-16
1132
Woke to the drum at a pitch I could hear. That has happened once before, in the fourth month, and it lasted two seconds. Tonight it lasted nine.
2026-09-16
0332
The ledger is three quarters full. At this rate it ends in eighty days. There is no second ledger aboard, and I have started writing smaller, which is not a plan.
2026-09-15
1932
Made a tape of eleven minutes of the drum and played it back at quarter speed. Underneath the hum there is a second hum. I do not think it is new. I think I am only now quiet enough to hear it.
2026-09-15
1132
Something in the pattern shifted by four seconds tonight. Four seconds is nothing over ninety-four minutes. It is also the first time it has been anything other than exact.
2026-09-15
0332
The scrubbers are rated for a crew of six. There is a setting for one. I found it in the ninth month and the air has been noticeably better since, which says something about the first eight.
2026-09-14
1932
I have counted the rivets along the spine corridor twice. Four hundred and twelve, then four hundred and eleven. One of us is wrong and I know it is me, but I have not gone back to check a third time.
2026-09-14
1132
Slept six hours, which is two more than usual, and woke up certain I had missed something. Checked every reading between 0100 and 0700. Nothing had happened. That took the rest of the morning.
2026-09-14
0332
The pattern came back at 1812 and ran for two minutes twelve. Same interval, shorter burst. I have a column for this now, which means I have decided it is a thing rather than an error.
2026-09-13
1932
The channel received for four hundred and one days and transmitted for none of them. I have stopped composing the message I would send. It took longer to stop than it took to start.