« first day (92 days earlier)      last day (1999 days later) » 

8:22 AM
@Glorfindel Would it be possible to find out from your data whether two questions from MO (311062: Sir Michael Atiyah's conference on the Riemann Hypothesis and 311280: What is the definition of the function T used in Atiyah's attempted proof of the Riemann Hypothesis?) wrere in the HNQ?
If I wanted to check it myself using the you posted here github.com/Glorfindel83/SE-Userscripts/tree/master/hnq-research, I would probably have to install some software on my computer, right?
It is not important - I am asking in connection with a question on MathOverflow Meta, which can be considered mostly coming just from idle curiosity: What happened September 23, 2018?
 
 
2 hours later…
10:26 AM
@Martin yep, you'd need PostgreSQL for that. The data there hasn't been updated since September 2nd (I'll get to that in 6-8 weeks). 311062 was not in the HNQ, 311280 only on September 24th, from 11:10-11:30 and again from 13:40-17:00 (times in UTC).
 
Thanks for the reply!
@Glorfindel So you're continuing with collecting the data on HNQ, right?
I suppose you won't mind when I mention in the answer the the other question was not HNQ with a link to your response.
 
10:45 AM
@Martin sure, that's fine.
@Martin yes, now even better: I'm also keeping track of views & vote counts over time.
 
10:59 AM
I'm slowly learning the memes and catchphrases that are being used around here. Only now I've noticed "6-8 weeks" and it made me smile a bit.
I am a bit surprised that 311062 was not in the HNQ list, I remember it was quite popular. Perhaps it was closed before being able to get to the list. (I am below 10k on MathOverflow, so I cannot check timeline of a deleted question.)
 
Closed questions indeed do not appear, and a common misconception is that views count towards hotness, but they don't. It's just votes (and age).
 
In fact, as Stefan Kohl just pointed out in a comment, reaction to that question was big enough to make the meta discussion about it the most viewed question on MathOverflow Meta.
@Glorfindel Yes, that's true. But high number of views: 1) Can be indicator that the question is likely linked somewhere in a quite visible place. (And in many cases, the HNQ list is the natural suspect.) 2) If there are many views, that increases likelihood of getting additional upvotes and increasing score (Matthew effect).
Incidentally, HNQ seems to be in the past few weeks one of the hot topics on Meta Stack Exchange.
 
Yes, that's why I started to gather some more data. A lot of the discussion right now is based on feelings, which is good, but I like to quantify things.
 

« first day (92 days earlier)      last day (1999 days later) »