Sunday and Monday both ran fairly smoothly.
Much as I hate to see anyone need to take a medical withdrawal, it was nice that one happened on Sunday—it allowed us to verify that Dan’s overnight fix had corrected the bug from Saturday.
Sunday also gave us another example of the advantages of separately posting pool results and tableau. Someone in the Division I Women’s Saber event (why, I wonder, were most of the problems in Portland in saber?) came to us, wondering why it appeared that 83 fencers were promoted to DEs when the format sheet said that only 77 would be promoted. That’s how we caught that operator error, where the DE round was set to the default 80% promotion rate instead of the 75% it should have been. A quick and easy fix, but one we don’t want to have to do at SN.
The most remarkable part of Monday was that I was able to start releasing strips to the armorers for teardown by about 1:30 pm. It used to be routine for teardown to begin (or for the fencing to end) that early, but it’s been years since that was possible. Our strip layout was odd and awkward for a number of reasons (the armorers even stashed some replacement equipment under the BC stage so that it would be more accessible than it was from their armory in the far corner), but it was a great layout for teardown. The first section released (pods A–E) was far enough from where the fencing continued that the teardown process was barely noticeable, and by the time I released the back section (pods H–K), fencing was down to only the replay pod (G) and the finals strip.
We’re getting better at packing up the BC crate, too. We paged some armorers to help get the server case into the crate—it’s astonishing how much easier it is for six people than just three to lift that sucker over the side of the crate. And once we got the server into the crate, the 3-dimensional game of Tetris that is making all our bins and boxes fit went much more quickly and easily than in the past. After congratulating ourselves on our cleverness, we realized that the improvement was mostly due to the reduction in the number of boxes from registration that we had to include—apparently, J.R. and Joe [Sibley—known familiarly among BC types as “Office Joe” to distinguish him from our Joe Salisbury, aka Coffee-Joe] had pared what they ship from one event to the next.
Joe sent me the stats for the live results over the weekend. We had a bit under 7,000 unique visitors, who each averaged 2.46 visits. Joe had set the server bandwidth to his normal default, and both he and Dan and a couple of other tech geeks reacted to the “bandwidth exceeded” message that popped up on Saturday with “Cool!” Joe says we used in a day and a half what is usually a full month’s usage for most of his clients. He bumped up the capacity an order of magnitude or so, so there were no further bandwidth issues.
But we’ve definitely got a hit with the live results.
Update [1/21/12, 10:43 pm]: Joe (that would be Salisbury) suggests I should have mentioned—because it was such a nice big number—that the total page views on the results site as of the morning of January 18 was 105,794, and that there were visitors from more than 60 countries. He’s right—I should have.