14
$\begingroup$

A frequent recommendation (e.g., here and here) for retagging or otherwise editing multiple questions is to do it when traffic is slow. Then edit-bumped posts will only push older questions off the first page. However, I have zero intuition of when traffic at CrossValidated is slow, or when it is busy.

When is traffic slow or busy around here?

(I'd assume that this is also relevant for other SE sites and was tempted to ask at Meta, but I guess answers will differ between sites, and I don't know whether to expect a simple query that one could customize for different sites as an answer to the "general" question.)

$\endgroup$
7
  • 1
    $\begingroup$ I think that the rate at which you are tagging old questions is fine. I know that does not directly answer your question, but after 10 hours, perhaps nobody knows. $\endgroup$
    – mdewey
    Commented May 17, 2016 at 20:34
  • 3
    $\begingroup$ +1. I'd love to see a plot that shows the average number of new posts (or answers, or edits, or whatever) as a function of GMT hour. $\endgroup$
    – amoeba
    Commented May 17, 2016 at 20:44
  • 1
    $\begingroup$ @amoeba, if you go to tools -> stats -> site analytics, you can see such plots. $\endgroup$ Commented May 18, 2016 at 2:05
  • $\begingroup$ Really, you can just look at the main page. Sometimes, new (answered / modified) questions just keep popping up; other times, it might be 5 or 10 minutes before the list is bumped. $\endgroup$ Commented May 18, 2016 at 2:08
  • 3
    $\begingroup$ @gung: access to site analytics unfortunately only happens at 25,000 rep. $\endgroup$ Commented May 18, 2016 at 7:05
  • 1
    $\begingroup$ @gung and Stephan, oh I was sure that 20k is the maximum rep unlocking all existing privileges (that's partially why I started regularly giving bounties: I thought I don't need any additionally rep anymore). Turns out that it was a false impression and it still makes sense to reach 25k :-) $\endgroup$
    – amoeba
    Commented May 18, 2016 at 10:04
  • $\begingroup$ Hmm. I don't see the logic here unless it's an indirect way of saying Please don't do that kind of editing unless I am asleep or at least not watching CV so that it doesn't irritate me if I am looking out for recent posts. But when you are asleep could be someone else's peak time. Longitudinism rears its not so pretty head. $\endgroup$
    – Nick Cox
    Commented Jul 11 at 15:03

1 Answer 1

14
$\begingroup$

Based on number of posts:

Time of day: Looks like 3:00-7:00 (all times are UTC) is half as busy as 14:00-18:00

day of week: Saturday is the least busy day, followed by Sunday

combined: 6:00-7:00 Saturday is just over 20% as busy as 15:00-16:00 Wednesday

via this query --

http://data.stackexchange.com/stats/query/51030/posts-by-day-of-week-and-hour-of-day

CrossValidated Traffic

foo <- read.table("http://data.stackexchange.com/stats/csv/51030",
  sep=",",header=TRUE)[1:24,1:8]
foo$Hour <- as.numeric(as.character(foo$Hour))
with(foo,plot(Hour,Monday,type="o",pch=19,
  xlab="Hour (UTC)",ylab="",ylim=range(c(0,foo[,-1]))))
for ( ii in 3:8 ) with(foo,lines(Hour,foo[,ii],type="o",pch=19,col=ii-1))
legend("bottomright",lwd=1,pch=19,col=1:7,legend=colnames(foo)[-1])

More practically, just avoid the very busiest times (15:00 $\pm$ 3 or 4 hours on weekdays) and if you don't do too many per day (I'd suggest you don't take up more than about a fifth of the front page with bumped posts, for example) you should be fine.

$\endgroup$
3
  • 4
    $\begingroup$ Thank you! I took the liberty of turning that query into a plot. Will accept unless something unexpectedly better comes along. $\endgroup$ Commented May 18, 2016 at 7:16
  • 8
    $\begingroup$ I like how Friday on this plot starts as a weekday but ends as a weekend. $\endgroup$
    – amoeba
    Commented May 18, 2016 at 7:41
  • 2
    $\begingroup$ @Stephan Thanks for adding the plot $\endgroup$
    – Glen_b Mod
    Commented May 18, 2016 at 10:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .