Why Do We Say "21 Days"?

When we tell patrons how long they can check out materials, we always say "21 days." (Well, okay, sometimes we say "3 days" or "14 days.") Have you wondered why we never say "Three weeks"?

As with everything else in our organization, the answer begins with the words "Many years ago, we used to..." So here we go:

Many years ago, we used to use a primitive checkout system based on McBee cards and a huge, hulking camera device called a Regiscope (which rather resembled a huge Mr. Coffee without the carafe).

Here's how it worked. Every week, we opened a new box of McBee cards with a pre-stamped due date (due dates were always on Saturdays). For each checkout, the staffer would open the back of the book to reveal a pocket bearing typed details about the book: Author, title, subject, copy number, etc. (Actually, this pocket was usually a duplicate catalog card glued down on the sides and bottom, leaving the top open.) I couldn't find a good picture of this arrangement online, so I took one, with a pink bookmark standing in for the McBee card. For procedural reasons (and to maintain the illusion of confidentiality) I must tell you that the name of the institution on the card probably has no relationship at all to the real institution I work for.

Next, the staffer placed the patron's library card underneath the pocket, with (typed) name, address, and phone number showing. All of this was slid under the Regiscope lens. (If you imagine the Regiscope as a giant Mr. Coffee, the open book sat on the hotplate, facing up towards the outflow spout.)

Staffer pressed a button, there was a flash of light and a whirring sound, and the Regiscope took a picture of McBee card, book pocket, and patron library card. Then on to the next checkout.

That was Step One.

As the Regiscope's film spool filled, it was taken out and sent down to Headquarters. At the end of the week the film came out, filled or not. Saturday afternoon, all the films and unused McBee cards got packed up and sent to HDQ. Each McBee card had its own unique serial number: the branch kept track of which numbers had been used and that gave us circulation statistics.

On Monday morning (or Sunday if your branch was open), the whole dance started again with fresh films, a fresh set of McBee cards, and a new due date for the week.

That was Step Two.

Step Three came as patrons returned their books. Staff pulled out the McBee cards and tossed them in a bin. Now the sorting began. (Actually, sorting McBeen cards was a continuous process that occupied circ staff every minute of every day.)

Returned McBee cards were first sorted by due date. Then, circ staff attacked them with McBee needles. I wish I had a video of this process, because it was really something to see; I'll try to paint a word picture.

Start with a stack of McBee cards -- not the puny few in the pictures above, but dozens, even hundreds. Stick the needle in the first hole and shove it all the way down the stack. Pick up the whole stack by the needle and shake. Some cards will fall out: move them to the back of the stack and repeat, this time sticking the needle in hole #2. Shake, move, repeat until you've done this for all the holes.

The trick is, the holes on the sides of the McBee cards were notched in a very clever pattern such that when this whole process was finished, you were holding a stack of cards in perfect serial number order.

That was Step Four.

Eventually, if all books had been returned, you would have all the McBee cards that you'd used for a particular due date. But we know that never happened -- inevitably, some books were not returned by the cutoff date. (I don't remember how the cutoff date was determined...was it three weeks after the due date? A month? Six weeks? I don't know. Someone help me out here.)

Step Five, and this was the key step, involved looking through the returned McBee cards to identify which ones were missing. Those represented books that hadn't been returned.

So now the circ staff sat with the in-order McBee cards and counted them off, noting which ones were missing. The missing serial numbers were carefully written down and sent to Headquarters.

Now the poor clerical staff at HDQ went to work. Remember, they had all the Regiscope films for all the branches, filed by due date. A clerk sat at a machine something like a microfilm reader/printer and scanned the films for the missing serial numbers. When they found one, they printed out the image. (You can see how important it was for the branch to use McBee cards strictly in serial number order. Each Regiscope at the branch had its own independent supply of in-order McBee cards.)

Finally, those images were gathered together and sent to patrons as overdue notices ("Our records indicate that this book was not returned...")

In theory, there was the potential for further follow-up...but I don't know if we ever did any. (Did we send out second or third overdue notices? Anyone know?) The clerks were alert for the names of heavy abusers, and every once in a long while there would be a memo with the name & address of some patron with multiple overdues, so staff could be aware.

And the missing McBee cards? In the fullness of time (I don't know if this was months later or what), replacements were made for any cards still missing, and whole batches were stamped with a new due date. Ah, the eternal cycle of nature begins again....

Besides the intense labor needed to keep this system running, there were plenty of opportunities for mistakes. A McBee card that became mangled or ripped was hard to process and might wind up the entirely the wrong place (oldtimers might remember the saying "do not fold, spindle, or mutilate.") Sometimes staff missed pulling a McBee card out of a returned book. (Pages were trained to open the back of each book before shelving it, to check for McBee cards and for branch ownership -- that's why you might occasionally see a real old-timer absently open the back cover of a book and stare at the back page for a moment. Often we're not even aware that we're doing it.) And gods help the staffer who dropped a stack of in-order McBee cards....

McBee cards went astray in the strangest ways: stuck behind drawers or fallen down cracks in the desk; playing hide-and-seek on the shelves or elsewhere in the branch; falling to the floor of the delivery truck; eaten by the family dog. On at least one occasion, a truck carrying a whole load of McBee cards suffered a door malfunction and lost a few boxes on Ritchie Highway. (This gave us a foretaste of what a hard drive crash would be like.) Patrons returned books without the McBee cards all the time. (And you think it's tough getting them to leave the orange bands on MARINA items!)

Unscrupulous patrons learned that they could bring McBee cards back to the library and leave them hither and yon, thus getting away with stealing books. In my high school, this was a tactic well-known to all the kids. Sometimes the Regiscope went wacky and a single frame (or possibly a whole roll) was ruined.

Plus, the whole thing was so labor-intensive that cataloging, pocketing, checking out, and sending overdue notices cost five or ten dollars in staff time per book. So when we started carrying paperbacks, we didn't bother cataloging them or using the whole Regiscope/McBee process: we just stamped them with a due date and tallied the number checked out for our circulation count. Did we lose a lot of paperbacks? Who knows?

One more note: On the book pocket pictured above, you'll notice a lot of handwritten numbers in various colors. This was our ingenious solution to the problem of tracking checkouts. When a book was returned, circ staff wrote the number of the month on the pocket. Each year had a different color. That way, when info people were reviewing their portion of the collection, they could see how many times the book had checked out (or not) and how recently. A book with very few handwritten numbers didn't check out very often, and so was a prime candidate for weeding. (Catalog cards had an accession date, which was also helpful.)

So now you know way too much about the way we used to check out books. Now we can get back to the question I posed at the beginning: Why do we say "21 days" instead of "3 weeks"?

Well, you see, back in the Bad Old Days all materials, no matter what day you checked them out, were due three weeks from Saturday. When we started phasing in our computer circulation system in December 1983, due dates changed to an exact 21 days from the date of checkout. We needed to re-train both staff and patrons who were used to the old state of affairs. So we started to refer to the due date as "21 days" in order to highlight the difference between the old and new systems.

That was 26 years ago; the transition to the computerized system was complete more than 20 years ago. Interesting how patterns like this, once established, can stay for decades. I suppose that's part of what makes anthropology so fun.

One final thought: Just in case you think of your pre-computer predecessors as being simple and hopelessly unsophisticated, remember that every single person who worked with this bizarre and complex system understood exactly how it worked. It isn't that we were simple; it was just that our heads were all too full of all this nonsense to allow much else. :)

----------

ADDENDUM #1: Skip reminds me of two things I forgot. First, the machine at HDQ that made prints used the old wet-copy process: special paper (which tasted awful) and smeary, blurred copies (which looked awful.)

Also, for the last three years or so, while branches were computerizing, we saved money by not having film in the Regiscopes at all. That's right: we just went through the motions of checking out books, but no record was kept and no overdue notices sent.

I seem to remember that each Regiscope had a little manual readout like the odometer on your father's Oldsmobile: at the end of each day we recorded that number, which gave us that day's circulation numbers. (Imagine getting circ figures on a daily basis.) Since the odometers worked even when there was no film, we were still able to keep circulation statistics during this period.Thanks, Skip!

----------

ADDENDUM #2: Apparently the habit of saying "21 days" is not as widespread as I thought. It turns out that a fair number of staffers say "three weeks" as a matter of course. Good. It shows that habits need not last for 20+ years beyond their rationale. We can grow and change. :)