if it’s a software problem, maybe, hardware problem not a great idea - what if the noise was indicative of a critical component on the verge of failure - hard to find a software fix.
“The noisy amplifiers were noticed during ground testing before the device was launched. “Everybody knew and worried about this,” says instrument scientist Doug Caldwell. But in the end, he says, the team thought it was riskier to pry apart the telescope’s electronic guts than to deal with the problem after launch.”
Launch garbage, sort later?
It’s a good argument for not building the darn thing in such a way that it has to be “pried apart” to fix a problem that pops up just prior to launch.