[FPSPACE] msnbc.com (Boyle/Oberg): NASA frees rover from
Martian sand trap
zirconic1 at earthlink.net
Wed Jun 8 17:27:46 EDT 2005
From: Dennis Newkirk <issguide at comcast.net>
Sent: Jun 7, 2005 10:24 PM
>Another interesting remark was that the imaging software used for
driving the vehicle was not loaded at launch time, and in fact wasn't
loaded until shortly before entry.
I think that there was quite a bit of software that was not ready at launch and was loaded en route. Check out the two _excellent_ PBS Nova documentaries on the rovers for some mention of this. You can find them at www.wgbh.org in their store.
I'm not sure that this is a good approach, however. It means that the vehicle is sent off without the software having been fully tested. Now maybe it gets fully tested by the time the software is finally loaded. But what if it requires more time? You cannot delay the landing, you can only delay a launch. So it seems prudent to get this stuff done _before_ launch.
I wonder if this is one of the things that mission managers have vowed not to do again. The rovers have been amazingly successful. But a lot of corners were cut and they did a lot of things in the wrong way--by their own admission. After the landings, one of the senior mission managers commented (paraphrasing) "We will not do things like this again."
More information about the FPSPACE