Alan Nelson's Daily Commentary for 5 September 1999

Kwajalein Island, Republic of Marshall Islands

I now have a task assigned to me for the next several days. It is not an exciting one but it will fill a need that must be met. In looking over some of the data from the radiosondes, it had been discovered that the program was outputting range data in the temperature column of the summary data files. Plus azimuth and altitude of the sonde rather than wind direction and wind speed. So all the radiosonde files since August 8 need to be looked at to see if the are correct or not. If not, then I have to go back to the raw data files and reprocess summary files.

It took a couple hours to find the people who knew which machines have summary data and which machines have raw data and which machines have the software capable of generating new summary files. It turns out that the raw data files reside on a very busy machine but the software resides on a machine that I can usually have access to. I can connect to the raw data files over a network that runs throughout the trailer at the "Ops Center" (short for operations center). I have also learned where the program wants the raw data file when it calculates a new summary file. I discovered that I can't just read a file from wherever it resides, it must be in this program's favorite directory. So there will have to be a lot of file movement.

I need to figure out whether it will be easier to just run the software that already exists, or write a program that will read raw data files and write new summary files. Most of the project's data has already been written by this software and so writing a new program isn't an obvious choice. But this whole problem came out of the fact that the software must first load a data file, rewrite the data file, exit the program, re-enter the program, load the newly written file, and then output the summary file. Now there is a bad piece of progamming code! Doing that once for every sonde you release is just a minor nuisance. Doing it for a whole month's worth of sondes will be a major nuisance!

When the TV crew was here, they expressed some concern that they didn't want to get any more shots of people working at computers. Turns out that virtually everyone here is working at a computer! The Ops Center has 10 computer monitors crammed into a trailer. When aircraft are flying, I think that 9 of them are constantly in use. Tomorrow I think I'll be on the 10th one. This picture was taken while the planning meeting was going on and so the trailer is pretty empty! Dave is busy storing a radar image for each hour during the day. I think it will be fun to see the movie of all the hourly images from KWAJEX put back to back!

I went to the chapel service again this morning and was pleasantly surprised to hear Song's (one of my radiosonde trainers-she's in the picture from August 28) 11-year-old daughter, Amber, sing a fantastic solo during the offertory. There are not many outlets for a great 11-year-old singing voice on this island. So far, she has done very well on her own!

Alan Nelson