[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: nm.debian.org

On Tue, Dec 05, 2000 at 08:36:03PM -0600, Donald J Bindner wrote:
> It was pointed out that by definition, you won't be assigned your
> AM until you have waited "Maximum Days" since they are presumably
> handled as a queue (i.e. the person waiting longest gets the next
> assignment).
Nope, that's incorrect.
The average and max waiting times at the times that the people waiting
in the particular cue have waited for.

Let's say there are two in the queue. One has been there for 10 days,
one for 20. So Max is 20, Avg is 15.  You turn up and apply, how long
will you have to wait for? It depends.

If tbm, the NM-crunching machine :), finishes three applicants tomorrow,
then the answer is 1 day. But it could easily be more than 20 as well.

> I believe a bit of the impatience comes from noticing that you
> are waiting "longer than average".  Perhaps it would help if the
> page made it more clear how long you will have to wait.
I'm not happy with the algorithms I am using.  However I don't really
know what is a better one to use.  By definition, even with a perfect
model half of the people will wait longer than average.

> I don't remember if this was available, but it would also be
> helpful if you knew how many people were ahead of you in the
> queue (after you login for example) so you could see the value
> decrease.
You can work that out manually from the nmlist.php page.

I'm willing to change the algorithms to something else if someone can
think of something better.  I was toying with the idea of using a sample
of the last N people who have gone through a gate.  Currently I use
the entire sample of people who are sitting before a gate.

  - Craig

Craig Small VK2XLZ  GnuPG:1C1B D893 1418 2AF4 45EE  95CB C76C E5AC 12CA DFA5
Eye-Net Consulting http://www.eye-net.com.au/        <csmall@eye-net.com.au>
MIEEE <csmall@ieee.org>                 Debian developer <csmall@debian.org>

Reply to: