Quote:
Originally Posted by oiuyt  I don't see justification for blaming the delays on the equipment issues. |
I believe that it was a contributing factor, adding to the chaos and confusion of getting things started, but I agree that all delays (or even most) can not be blamed on that.
Some of the reason to blame this is strategic: the lack of strips working on time could have easily gone from a delay to a show-stopper. The fact that we should have had ~ 75% more strips than needed, and we still did not have everything working on time probably got some people a bit panicky and irritated.
But I suspect a large portion of this blame was a face-saving measure. Since I believe that it is my face to save or expose, I will deconstruct some of the delays now.
Two words may be used to understand most of those delays: "Beta Software".
Itemizing how things went wrong:
- It took too long to enter in team rosters. Part of this can be blamed on a moderate software bug, (A problem with the data entry screen), but this was fairly easily worked around, and I suspect cost no more than 5 minutes. A larger part of the problem was that this aspect of the BC was not parallelized; all of this work was done on a single computer.
I thought that this would be acceptable, since teams were required to submit their rosters in advance. I thought that there would be roughly 20-40 additions, corrections & deletions, and the majority of these would be turned into the BC 15 minutes before the close of registration. I thought that this would allow this work to be done no later than 5 minutes after close of registration. I was wrong. I am not sure which assumption I was wrong with, but I suspect it was all of them.
- There was a user error (in this case the user was me) in setting up sabre pools. (This cascaded. See more below.)
- There was a serious software error in how certain pools were set up. All like combinations of # of strips & # of teams in a pool were hard-coded. the case of 10 teams on 5 strips had a problem. (I mis-transcribed #s.) This case happened on pool 3 of foil & epee. It was the best decision to make at the time. (this also cascaded. See more below.)
- The software only allowed the user to print all of the bout slips for all of the pools in a weapon, not one pool or one bout slip. This meant that the above to items cascaded down. It took roughly 55 pages of printing to send out an event. I calculated that this was acceptable on a single printer if this was started 20 minutes before teams were to report to strips. (on most modern consumer grade printers, this should be no more than about 5 minutes of printing.) I did not take into account re-seed issues. I should also have allowed the BC to just print the first 2 rounds to get things started.
The above three items caused a significant portion of delays Saturday morning. I think the BC lack of familiarity with the format & procedures caused other problems. Also, the lack of a controlling document that the BC & head ref could consult to answer almost any question on format or rules caused many small problems.
Other software problems that caused other delays:
- There was an error in how the individuals were sorted in results. This caused confusion, but ultimately the BC got things right for individual pools.
This also slowed down the generation of the individual pools.
- Results were generated in the program, which created a spreadsheet, which could be printed. This multi-step process caused some problems.
Not my software, but a software problem none-the-less:
- It was decided that promotion from the first round of individual pools to the second was to be done per pool (4 from each pool) instead of from the field. Neither I or anyone in the BC knew how to get Engarde to do this. (Fencing Time was considered and rejected. I don't remember why.) This caused the BC additional work.
I believe there were many other organizational and computer issues, and while I bear some responsibility for them, I will let others decide if to post and discuss them here.
W
Linear Mode
