10/2 @amWhy (rather than 10/1)

@quid how do you define closest? Suppose there are two users with two dates, let's say, 1st & 3rd October and the winning date is 2nd October then who is the closest amongst these two users.

@paulplusx be might use hours, etc. as a tiebreak, if we know about when the million was crossed. But maybe another precission is needed: the dates (of course) are the dates according to SE-time (also known as UTC)

@quid Yes, I suppose hours will do. It's highly unlikely that the question will be posted right in the middle of the day.

@quid Where can we see the current number of questions?

@KarlKronenfeld Oops... I knew that, by it seems my typing fingers did not! Thanks for the correction.

So if October 1 is still free, that would be my guess.

Is this the correct summary at the moment?

@MartinSleziak What information are all of you using to make guesses?

Any information available :-)

A lot of stats can be found in SEDE. I have also looked a bit in the site analytics.

But I did not spend too much time with that - it's more a guess than number crunching. (I'm pretty sure that Nate Silver or somebody in that business would be able to find out a lot more from the numbers that are available.)

Ah thanks.

So 'clearly', a significant fraction of questions are due to homework.

One needs also to factor in average number of deletions daily.

@amWhy Oh no. What a complicated challenge.

Ah nice, two stars for Sep. 29.

@KarlKronenfeld I merely starred it at the time, so it would hopefully appear in the right column. Then it was pinned by quid for the same reason. Pinning (star outlined in black) keeps it ontop...

10/5; furthermore, the packers will lose.

@AlexanderGruber grumble, grumble, shakes head, keeps her cool, sticks her tongue out :p

@MartinSleziak thanks noted.

@AlexanderGruber thanks noted.

@user21820 question tab under the heading. math.stackexchange.com/questions

@paulplusx yes, an isssue is though to know when it passed exactly. But that can be more or less achieved via going back in the newest list.

I think there could be cases of "too close to call" at least without going through hoops. Yet then there'll be just two winners.

@quid That would still be an accomplishment. So if the millionth question is reached on October 6th, e.g., then @AlexanderGruber (who denigrated my beloved Packers), and I would both be winners.

One possible scenario is that just before UTC 00:00, a question is posted, bringing to number of questions asked to one million. The number of questions asked may reach one million, then fall back below one million (say, due to deletions), then build up to surpass one million questions, shortly thereafter, likely, the following day. Is reaching one million questions for the first time what we are betting on?

@amWhy That's a nice question :D. It's like stable 1 million vs unstable 1 million.

Also, there has to be someone to continuously monitor and record the number of questions every 5-10 mins right after it reaches somewhere like 999,990.

I'm not sure, @paulplusx, how frequently the number of questions asked is updated. I doubt it is continuously updated in real time.

Let's ask @quid! quid knows everything!! ;-)

@amWhy Yeah :D

@amWhy In any case even if the update interval is like 1-10 hours, it shouldn't really matter because we are betting on days. Yes, it would cause a problem if there is a tie (which probably will be).

I feel I am contradicting my own statement: first saying we need to monitor every 5 mins and then saying it doesn't matter if the update interval is 1-10 hours :-/

The whole problem is the tie breaker. Nothing else.

@amWhy @quid Please see the comment I'm replying to. Do you know how frequently the "questions asked" count is updated and revealed on the "questions" tab?

@amWhy Off hand I don't know exactly. But every few minutes. Not realtime.

I don't know if it is a fixed interval even; could also depend on load.

@quid Well it seems then that my question is legitimate: are we betting on the day when the number of questions first tops 1 million, (and may subsequently drop below 1 million), or when it has reached 1 million, and remained 1 million or more?

@amWhy I think we go for the former; the latter has the issue that we can never really know that.

@quid That is probably the correct route, as it is easiest to determine, for the reason you mention.