> then it happens again

if it happens again it wasn't a person, it was the hinge. photograph the angle tonight, same spot, same time. facts first, conversations later

check the pivot tension before you go accusing anybody. those cheap lamps loosen up on their own, especially if someone knocked the base once and it never recovered. mine at home drifted about that much over a month. a quarter turn on the nut and it'll hold aim. post a pic of the joint and i'll tell you which

we had a lamp on the terminal counter what got aimed at the wall every single night by whoever locked up... turned out the day relief did it so the glare wouldn't hit him in the eyes on his paperwork. never said a word about it for two years...

bend it back, watch for a week, then ask around. could be nobody's got a grievance, could be somebody's been eating his late lunch in the dark and hasn't said so...

> never said a word about it for two years

that's not a solved problem, that's two years of unsolved problem with a habit attached. the relief guy got his way and the day guy never found out why. do you not want to know that?

collar slot works the same, small flat screwdriver, quarter turn clockwise, don't crank it. put a paint mark or a pencil line across the collar and the stem so you can see if it slips again without eyeballing the angle. snap the pic first either way

long time lurker, first post. had this exact thing at a dock office a few years back. turned out the cleaners were wiping down the desk after hours and lifting the lamp to get under the blotter, then setting it back wherever. never once thought to aim it back. might just be worth asking who's even in that office after you leave

cleaning crew is the most boring explanation available which usually means it's the correct one. pencil line across collar and stem, photo before you touch anything, leave it alone one night. mark still matched in the morning = no person involved, tighten it and move on. mark off = then you ask around

> snap the pic first either way

yes. and make it the same photo both nights, camera at the same spot on the floor, not "roughly" the same. a pencil mark you can nudge, a photo with the stack of papers in it you can't. write the date on the back of each one. evidence first, then the ask-around

a paint mark across the collar is all well and good but you're assuming the person re-aiming it wouldn't notice a stripe of paint on their own handiwork... in my experience whoever's fiddling with your desk knows exactly what's on it. i'd still leave it a week. two years of dark paperwork at the terminal and nobody ever once mentioned it to me, and i was there every single day

the time on the printout habit applies here, doing the same thing tonight with the lamp. camera on the taped spot, write the time on the back.

cleaners wiping under the blotter is annoyingly plausible. my desk stays exactly how i leave it but i'm in a warehouse office not a dock office, different crew, different hours. if the mark's still true in the morning i'm treating it as the hinge and not taking it personal

camera on the taped spot is a good instinct but make sure the tape spot is actually where you stood for the first photo, not just where your lane happens to run. you need the exact same framing or the comparison is worthless.

also: cleaners wiping under the blotter is a theory you can test without asking anyone. leave a small object at the edge of the blotter, close to where a hand would slide it. if the blotter's been lifted the object will have shifted or be at a different angle by morning. you don't need a camera for that, you need one thing placed deliberately and left alone.

^^ this is a better version of what i was trying to say. i wish someone had suggested the blotter test back when it happened to me instead of just letting it ride for two years. would've saved a lot of squinting at dim paperwork.

blotter object test is solid. doing that too.

desk stays exactly how i leave it, i'm particular about that. stack squared, blotter centered. so if the blotter moves that narrows the who considerably.

tonight: photo from the taped lane, time on back, pencil mark on the collar, small object at blotter edge. morning check: pencil mark, blotter object, second photo same spot.

if everything matches i'm tightening the nut and calling it a hinge. if anything's off, i'm asking the night crew and whoever else has a key. not taking it personal either way, just want to know what's actually happening in my own office. case closed by tomorrow night hopefully.

only thing i'd add: pick something with a flat base and a bit of weight. a coin, a small block, a machinist's square. not a pen. a light pen gets nudged all on its own by the air off that doorway every time the roll-up cycles at night and you'll get a false positive and go blaming the cleaners for the HVAC's work.

also worth checking once whether that office door pulls a draft when the dock opens. cheap fix if it does, and it'd explain loose paper shifting without any hand involved at all.

long time reader, first post here. had the exact reverse of this a few years back. my lamp kept getting swung to face me, thought somebody was being a jerk about the glare.

turned out the night guard was helping himself to my desk to fill out his rounds sheet because his own fixture was burnt out. never asked, never mentioned it, just used my light for months. once i found out it was less annoying and more just... somebody with a better reason than mine and no manners about asking.

so if you do land on a person, don't assume it's aimed at you. half the time it's somebody solving their own problem with your stuff.

morning check results:

pencil mark on collar and stem: still dead lined up. no slip.
blotter object (i used a small steel block, not a pen, thanks gearhead): same spot, same angle.
second photo from the taped lane: lamp aimed where i left it. stack squared, blotter centered.

so either nobody touched it or the hinge drifted on its own. either way i'm calling it the hinge and tightening the nut tonight.

case closed. tentatively. i'm still noting in my head who's got a key though. old habits.

one night is a sample not a pattern

we had a hood light at work that sagged every few weeks and everyone blamed the dish guy for a month. it was just the spring

tighten it and give it a week before you clear anybody

good. quarter turn clockwise on the collar, don't crank it, and keep the pencil line across collar and stem even after it's tight. costs you nothing and tells you next time whether it slipped or somebody grabbed it.

one clean night is fine for the hinge theory. you're not clearing a crew, you're just deciding whether to keep watching. that's the right call.

tightened. quarter turn, no cranking, pencil line still crossing collar and stem like gearhead said. wrote it in the little book: date, time, mark state.

week-long watch starts tonight. also made a list of who's got a key to that office. it's a short list which is somehow both comforting and annoying. case closed, reopened on a timer.

> week-long watch starts tonight

good. one clean night is a data point, not a verdict. and for the record, three theories are now on the table: hinge drift, cleaning crew, door draft. you don't get to pick your favorite and close the file.

count the nights. if it's seven, say seven.

first post, been reading this thread since the lamp photo.

almost the same setup at a yard office years ago. ours wasn't the hinge or the cleaners. the roll-up door on the bay cycled all night and the vibration walked the lamp base around on the desk over several shifts. nobody was touching it, it was just slowly migrating. we put a scrap of non-slip shelf liner under the base and it stopped dead.

could be worth checking if your base slides at all when that door goes up.

non-slip liner is the move

also keep the pencil time in the notebook even after the week is up. we still write CHECK THE CART every night at my first branch job and that cart never did anything wrong. it's the writing that keeps it honest.

shelf liner under the base is a good call, cheap and it kills the vibration theory without any more observation. if the liner holds for a week and the mark holds too, you've got two explanations retired at once.

one nit: don't stack the tests. do the tightening tonight, watch three or four nights, then add the liner. otherwise you'll never know which one fixed it and you'll be here again in spring wondering.