In four years of reading Cassini updates, they have never once mentioned worrying about software errors due to cosmic rays. And this is without any atmospheric

In four years of reading Cassini updates, they have never once mentioned worrying about software errors due to cosmic rays. And this is without any atmospheric protection at all; we have 100 miles of atmosphere. They have mentioned that their solid state relays trip 2 or 3 times a year due to cosmic rays. I'm not sure how comparable those are to DIMMs, but it does suggest that the author's claim of one error per day is a bit off...