3 ms·
The author arbitrarily decides on "number of failures" as their measure of reliability, as opposed to a much more reasonable "number of failures" * "mean cost o
by gcommer 8y ago
The author arbitrarily decides on "number of failures" as their measure of reliability, as opposed to a much more reasonable "number of failures" * "mean cost of failure". There is also no attempt to correct for the confirmation bias of software being so popular, no attempt to present non-anecdotal evidence, and no attempt to provide comparative data for other engineering disciplines.