Having done that sort of work early in my career, I think that's a bit much.
My first job in California started out with two 6-foot stacks of octal dumps of operating system crashes, for a mainframe OS that was running 1-2 hours between crashes. I worked through those slowly, with colored pencils and highlighting markers, figuring out what had gone wrong, then putting a fix into the OS, in assembler. Gradually the stacks got shorter and the time between crashes increased. When I left that job, we were going a month between crashes.
Then I changed jobs to work on operating system design and proof of correctness.
This is why I'm so negative on unsafe code in Rust. Memory safety is essential. There are programmers who think they are so good that they don't need bounds checking. They are wrong. The most recent CERT advisory of a security hole for a buffer overflow was 8 days ago.[http://www.kb.cert.org/vuls/id/214283]
My first job in California started out with two 6-foot stacks of octal dumps of operating system crashes, for a mainframe OS that was running 1-2 hours between crashes. I worked through those slowly, with colored pencils and highlighting markers, figuring out what had gone wrong, then putting a fix into the OS, in assembler. Gradually the stacks got shorter and the time between crashes increased. When I left that job, we were going a month between crashes.
Then I changed jobs to work on operating system design and proof of correctness.
This is why I'm so negative on unsafe code in Rust. Memory safety is essential. There are programmers who think they are so good that they don't need bounds checking. They are wrong. The most recent CERT advisory of a security hole for a buffer overflow was 8 days ago.[http://www.kb.cert.org/vuls/id/214283]