The earliest signal a thread is going to compound, and it's not the reply count
by•
Ran the same comment through four threads over a month and only one kept generating views weeks later. The difference was not the reply count in the first hour, all four got at least one. It was whether a third person with no connection to me or the OP joined within the first two or three days without being asked.
If it stays just me and the OP going back and forth, that is engagement, not compounding. The one thread that kept going had someone unconnected join inside 48 hours, and that window is also roughly when the other three had already gone flat.
Curious if anyone else tracks this specifically, or if there is an earlier signal worth watching for before the third person even shows up.
60 views
Replies
My version is that the question has to be answerable with a fact rather than an opinion. An opinion question collects agreement, and agreement is what closes a thread.
I joined one as a stranger this week and there was no concession in it before I arrived. What made me write was that an early answer was general, the OP said that was not enough and asked again for specifics, and I happened to have done the specific thing. I would not offer that as evidence given what you both said about memory, only as where the idea came from.
Does the 221 thread ask for something people would have to look up, or for what they think?
@alieksia Checked before answering. The 221 thread's actual question was "do you have a way to predict this, or is it trial and error too," pure opinion, nothing to look up. It still compounded, so fact versus opinion isn't the split, at least not for that one.
the earlier signal i would watch is whether the question is still visibly open after the first exchange.
a stranger joins when there is something left to answer. if the first replies agree with you and you agree back, the thread reads as finished to anybody scrolling past, and the only move left is +1, which nobody bothers to type. if an early reply puts a condition on it or contradicts you, and you give up a piece of it, the question is unresolved in public and now there is a reason to walk in.
so what i would count is not people, it is how many of your own early replies added something the commenter had not already said. agreement from the op closes a thread politely. that is upstream of your third person and you can see it inside the first hour instead of waiting two days.
you can test it on the four you already have. mark each of your own replies as agreed or added a condition, and see whether the 221 has more of the second kind than the other three do.
have you had one compound where everybody in it agreed with you?
@jernej_jan_kocica Haven't coded each reply as agreed versus pushed back specifically, so this is memory, not an audit. The 9 view thread was mostly me agreeing with whatever came in. The 221 one had at least one spot where I admitted not controlling for a variable someone raised, which is closer to your second kind. Not proof either way, but it does not contradict it. On your actual question, no, I don't think I've had one compound where everyone in it agreed with me. The 221 thread still has at least one open contradiction sitting in it, not just agreement.
@abdullah_javaid3 memory is the weak kind of evidence here and i would not count it as support, mine is the same. you remember the thread that did well as having been the better argument, so we are both grading our own work after we already know the score.
and your answer just made a hole in what i said visible. the threads where you conceded are probably the threads where somebody said something worth conceding to. that person is also more likely to be the type who drags a third person in. so the concession might be a symptom of a good commenter arriving rather than the thing that opens the door, and from the outside i cannot tell those two apart.
the ordering separates them and you already have the data. in the 221 thread, did you concede before the unconnected person showed up or after. if you conceded first and then a stranger walked in, it is the mechanism i described. if the stranger arrived first and the concession was you answering them, then mine is backwards and your 48 hours is the real signal rather than a downstream one.
the open contradiction still sitting in that thread is the part i would look at hardest. do the recent comments cluster around it, or are they somewhere else entirely?
@jernej_jan_kocica Went back and checked the actual order instead of trusting memory. Conceded to Elliott Reed before Liam Cai or Jernej showed up, so in this one your mechanism holds over mine, the third person window looks downstream of the concession, not the trigger itself. On the open contradiction, it's the position versus thread size one with Elliott, still sitting unresolved. The recent comments don't cluster around it at all, Liam went to indexing and you went to read versus upvoted, both somewhere else entirely. Maybe unresolved isn't enough on its own, it has to be unresolved in a direction someone actually in the room cares about.
@abdullah_javaid3 i am one of the two strangers in that data, so you can just ask me instead of inferring it.
i did not read the elliott exchange as a question to answer. i had not read it closely at all. what made me write something was that you were replying to everybody properly, including the boring ones, and one of those replies gave something up. that told me an argument would not be wasted here. then i wrote about the thing i actually had, which had nothing to do with position versus thread size.
so your refinement is contradicted by your own audit. if it had to be unresolved in a direction someone cares about, liam and i would have gone at the contradiction. neither of us did, we both walked in and changed the subject.
what fits your data better: the concession is not a question, it is permission. once the op has been visibly wrong about something, the thread stops being yours to defend and becomes a place where somebody can bring their own thing. that explains both halves of what you found, why the concession came first and why what arrived afterwards had nothing to do with it.
which gives a cleaner test than the one i gave you. a thread where you conceded on something trivial should still open up. a thread with a big unresolved question in it where you never gave anything up should stay shut. do you have one of each in the four?