Alan writes:

levik i was just wondering instead of publishing one problem per day y not make it 2? Every day under latest there could be two problems and this would speed things up for the problems in the queue. (if there is only 1 problem that has 3 thumbs up then only 1 could be posted under latest.)

So the reason it's currently one problem as the latest, is so that that latest problem receives the most attention from visitors, and hopefully the most comments.

While it's possible to have more than one "latest" problem (like "Latest", and "Previous" - next to latest) and push them live two a day, I personally think that would be overloading the front page with too much content.

To me personally, problems spending a long time in the queue is not a problem, but I can totally see how it's frustrating to someone who submits a problem they want to see up on the site. (I'm even guessing that's what triggered Alan't post)

But if anyone else has thoughts on the subject, I'd love to hear them.

Well, if anything, weekends are definitely our slower time. A lot of people come from google, searching for "logic puzzles" and "brainteasers". I'm assuming these are people surfing at work, since it drastically falls off during the weekend.

I guess I don't really mind if two problems end up being pushed during the weekdays (monday through thursday - friday is also kinda slow), especially if these problems are not particularly difficult, such as riddles and stuff - problems where not much discussion is possible.

The only issue will be knowing when two problems have already been pushed. It's pretty easy to see if the latest problem is from today or yesterday, but it may be less obvious if the next to latest one is. So if any of the scholars out there want to push a second problem sometime in the afternoon, I would ask that you check that only one has been pushed so far.

In other words - use your judgement.

As I said, I have no objections over two problems pushed per day, especially in categories like Riddles or Tricks. The scholars should feel free to do it.
I myself will probably stick to one a day for the most part, but this is more an issue of free time than anything. Most days I only have time to be here in the morning and at night, and I believe that if a second problem were to be pushed, it should be done around lunchtime or in the afternoon.

There's a big string of riddles in the queue right now. I don't have a problem with riddles per se, but given the lack of comments on them right now I'd say that people are starting to burn out on them, myself included. I think these would be prime candidates for going twice per day.

As I have already said, I am all for it. Friedlinguini, you should feel free to push these problems two a day. It's just that I believe the best time for pushing a second problem is in the afternoon, and being at work then I generally am not able to do this.

Presumably you can choose which problem to post rather than just having to post the oldest eligible one (this is sounding scarily like the flooble question again!). Anyway, how about posting a riddle and a 'non-riddle' each day. That way the queue comes down quicker (or at least grows more slowly) and there's always a latest 'real' problem for those not too keen on riddles.

I know that it took about a month for my number sequence problem to go... I saw the queue was around 300 today! (When I submitted a problem)

I like that idea, Percy. The queue does seem to be quite long at the moment. I submitted my first problem 60 days ago and it still hasn't made it to the stage where higher-level members vote on it. And, the first two problems submitted get a higher priority in the queue! It will likely be six months or more before the 3rd, 4th, and 5th problems I submitted are up for consideration.

Perhaps the amount of time between problems posted should be related, not only to the number of problems in the queue, but also to the number of comments posted to the newest problems, or the number of hits received by the site.

I'd just like to clarify on how the queue works.

All problems are assigned a queue weight (recalculated each night). If you have six problems currently in the queue, only the first two will be assigned a queue weight of 1. The queue is reordered each night such that only problems with QW 1 will appear in the front. In effect, all this does is prevent any single user from having more than two problems being voted on at a time. You should be thankful for this. You would probably find it stressful if you had to check on the voting process of three or more problems at a time. Also, the posted problems would become dominated by pcbouhid, who currently holds over half the problems in the queue (nothing against pcbouhid, but Perplexus' goal is variety).

This will not make your problems go through the queue slower--unless you submit problems so close together that later problems are waiting for the voting process of your previous problems. Voting process varies randomly from a day to two weeks.

Another clarification: The two problems a day schedule is upheld by Scholars, not a computer. Every time a problem is posted, it is because a Scholar was there.

The queue length has gone up and down. About a year ago, I think it peaked at about 600, but it's gone down to less than 300 now.

Tristan, you are right to try and explain queue weight but I'm still completely confused and will ask some questions in an appropriate thread.
So how about having more questions per day, evenly spaced to give everyone around the world more access?

Also look at the most recent problems that get pushed from new users. It appears from their last loggins that Alexis and Jax Dakars both missed seeing their problems go live. Then there are all the people who take the trouble to register, only to type "hi" in the CB but give up on posting probs when they see the queue length. Should more not be done to hang on to new users?

Thanks for the explanation, Tristan. I'm still a bit confused about how queue weight is related to queue position, and how that translates into how long it will take between submitting a problem and seeing it posted on the site.

I submitted five problems in November. The first two have approximate queue positions of 16 and 18. The fifth has an approximate queue position of 91. Is there a good way to estimate when a problem might be posted, based on approximate queue position?

And, does submission of more than 5 problems improve a member's chances of seeing a problem posted sooner, hurt them, or have no impact either way?

TOO!

Mindy you should know better! Please find the <forums/commons/QW>appropriate< :-p> thread. Though when your problem is in position 69 (for example), please feel free to contact me on advice on how to make it come quicker.

Yeah but three months? It makes my head hurt.

Does God play dice? I think it's kind of an equivalent question. Levik has assigned me 14064. Quantum mechanics like all physics is just an approximation based on what one is able to observe. I like to think that things are deterministic at a fundamental level, it helps keep me sane.

I haven't made a point of trying to fully understand the QW.
I understand that if more than 1 puzzle by the same poster is in the JM review area, and all may be considered available for publication, then other posts are slotted in ahead to provide for a balance of user posts.

Maybe some form of 'Graphic' might be useful to explain this as many who are likely to continue to ask the QW question are not yet able to see the process at the JM level.

Eg: Date Title Author Score
10/21/05 ABC Y +3
10/10/05 MMM Y -1
10/19/05 DEF P +4
10/22/05 PQQ Y +3
10/20/05 MTZ A +5
10/20/05 ZZZ Y +3
10/19/05 FFF G +6
10/18/05 ............... etc [for 10 entries]
From a more thought out listing as a reference example, maybe those who understand the Queue better might then be able to suggest the precedence of publication with the philosophy/reasons as to the likely order of publication [likely order - because the entry: "10/21/05 ABC Y +3" could be downgraded to +2 by a JM/Scholar before it is published].

Leave this with those who know the system better.