In an ITConversations podcast interview with Tim O'Reilly and John Batelle, Mark Zuckerberg describes a recent conversation with a teenage relative of his girlfriend. Those of us with teenage kids know that they consider email a bit old-fashioned, and this girl explained to Zuckerberg why: because it's so slow. He was puzzled, thinking that email is practically instantaneous; why was it slow? Because, the girl replied, it's slow to create a message. You look up someone's email address, your write out a subject line, you start your message with some sort of salutation, then you write it, then you sign off at end, and so forth.

Obviously, as The Facebook Guy, Zuckerberg is pretty tuned in to how modern teenagers communicate, and he was telling this story to describe the motivation behind whatever Facebook's latest spin on IM is. The story got me thinking back 100 years, though (or 101, now that it's 2011).

JoAnne Yates' excellent 1989 book Control through Communication: The Rise of System in American Management covers part of a topic that I've been interested in for a while: the change in information management that must have accompanied the industrial revolution. The factories making all those new things had to efficiently keep track of the what they made and the parts that went into it if they wanted to make a profit selling those things. Yates' book covers several things that could be considered early content management, and Zuckerberg's story reminded me of one part in particular. To quote her book,

Further changes in form were designed to make internal correspondence cheaper and more efficient to type, handle, and file. Writing in 1910 about what he called "interhouse correspondence," or correspondence between different locations of a single company, one author recommended several changes in form that would make these documents look less like letters and more like present-day memos. His discussion is worth quoting at length, for it sheds light on the underlying reasons for the changes.

In the first place, all unnecessary courtesy, such as "Fred Brown & Co.," "Gentlemen," "yours very truly," and other phrases are omitted entirely. In a business where hundreds and sometimes thousands of interhouse letters are written daily the saving of time is considerable. Next, an expensive letterhead is done away with, and this also is a factor in reducing expense. The blank is made with simply the words, "From Chicago," "From Atlanta," or whatever may be the name of the town where the letter is written, printed in the upper left-hand corner, and underneath the word, "Subject."

The 1910 quote also recommends that internal letters include a serial number and that one letter replying to another should reference its serial number, or as I prefer to think of it, include a link to its unique ID. The book goes on to describe the origins of the memorandum—in later years, "memo"—which dispensed with the flowery niceties of traditional 19th-century correspondence because, in communication within a company, efficiency was more important than politeness conventions. Putting the message's subject, date, and sender and recipient's names in what we would now call a fielded metadata header made the information easier to digest, file, and receive. (Elsewhere, the book covers a 1902 recommendation that for easier filing and retrieval a piece of internal correspondence should cover no more than one topic—a century before DITA and over 60 years before Information Mapping.) The name of Frederick Taylor, who Dan Brickley mentioned in his New Years blog posting, comes up often in Yates' book as a big influence on this thinking in general and Du Pont's operations in particular.

On the one hand, the way kids skip what they see as extraneous information seems to continue this trend. On the other hand, the things that I like about email that the kids don't care about are the things that the Taylorites developed to help manage that content: clearly marked fields of information to make it easier to archive and retrieve the memos.

Either way, it's always interesting to look at long-term trends in information management by looking earlier than 1970, which computer scientists typically consider to be the stone age. I'd love any suggestions about related reading on the topic of information management during the industrial revolution.