Wednesday, August 26, 2009


Wow--the code actually ran to completion. It claims it only took 32 minutes, but I guess it just feels longer to me because that was 32 minutes after 20 minutes after 10 minutes . . . But the code worked successfully? on the old tried and true data. Now I just need to figure out what it's doing wrong on the new data.


So I'm still waiting for my code to crash. This is mostly a good sign, as it is mostly working for this set of data, but I really trying to get it to work on another set of data. But I'm running it on the data it should already run on to try to figure out why it isn't running on the new data. But there is a lot of old data . . . .

I'm actually on the mountain "observing." Well, not observing yet but commissioning the instruments so that in the near future someone can observe. But I'm not really doing that either. While I might do some "observing," I've learned that an "observing run" is code for getting you to work a lot because you are there for 16 hours/day with no home to go to. I thought having dinner at 8 pm was odd until I realized that was still at least 4 hours until bedtime. So after midnight sometime we stop, go back to our room to sleep, and then come back in the morning to start all over again.

While the hours are long, it is also rather productive to be around the other people involved in the project, because they are experts in various parts of it all--including the code I am trying to debug. Or at least a similar piece of code that they were tasked with debugging. So I think that is most of the idea behind the "observing" run.

waiting for my code to crash

So I'm trying to debug a piece of code. It is really complicated (I didn't write it) and has lots of subroutines that take complicated inputs generated by earlier parts. So it is hard to run the subroutines separately, so I have to run the whole thing and sit and wait for it to crash. Then I try to fix the cause of the crash, start it again, and wait for the next crash.

So I'm sitting here bored most of the time and wanted to find something productive to do, so I googled " debugging doing something while waiting " or something like that (without those quote marks of course). Of the hits that returned that looked like what I was interested in, five of the six were blogs. Not that they gave any productive suggestions, but apparently writing the blog was filling the time for them. So here I am, writing for the first time in a long time. If any one reads this, do you have any good suggestions? Nothing too complicated since I will (hopefully) be switching back from it frequently while debugging.