Engineering archaeology

Here’s an ac­count by a re­tired en­g­ineer of what hap­pened when his old com­pany wanted to stream­line a pro­cess in the fac­tory where he used to work.

Peo­ple only knew how to keep the fac­tory go­ing from one day to the next, but all the doc­u­men­ta­tion was lost—the fac­tory had been sold a cou­ple of times, and efforts at digi­ti­za­tion caused more to get lost. Even the name of the fac­tory had been lost.

For­tu­nately, en­g­ineers keep more doc­u­men­ta­tion than their bosses al­low them to. (Trade se­crets!) And they don’t throw the doc­u­men­ta­tion away just be­cause they’ve re­tired.

I’ve been con­cerned about in­fras­truc­ture ne­glect for a while, and this makes me more con­cerned. On the other hand, in­stead of just view­ing with alarm, I’d like to view with quan­tified alarm, and I don’t have the fog­giest on how to quan­tify the risks.

Also, some of the in­for­ma­tion loss is a re­sult of a search for effi­ciency. How can you tell when you’re leav­ing some­thing im­por­tant out?