Magazine article Computers in Libraries

Systems Librarian and Automation Review

Magazine article Computers in Libraries

Systems Librarian and Automation Review

Article excerpt

Systems Librarian and Automation Review

Since that time, we have presented this idea (we are being charitable to ourselves here) to various groups. In some cases, we have asked people to fill out questionnaires detailing their own situation for us to analyze.

However, a cursory analysis of the information we do have suggests we may have been over-zealous in the application of regression equations to such a concept. Therefore, in an attempt to capture more valid data, we are offering a much simplified procedure. Less confusion, fewer numbers.

The basic idea is that the degree of computerization of any organization can be measured by the number of central processing units per staff member. This has implications not only for the degree of automation of an organization, but also for the method of support to keep the computers operating. (See the section on Infotrac below.)

How many computers do you have? How many people do you employ to keep them running? The answers to those two questions should yield some interesting comparative statistics, not only between libraries, but also between industries. In fact, one of our hypotheses is that libraries are actually much more automated than many industries. Preliminary analysis would suggest that several libraries have a CPU per full-time equivalent employee (FTE) ratio greater than one!

Think of it! Libraries have more computers than employees. What other industry could boast such a high ratio? Even companies with a computer on every desk would not beat libraries in this measurement.

Why? Primarily because of online catalogs, Infotrac, PC-SIGs, and all sorts of extra CPUs lying around to be used by patrons. Our central reference desk typically staffs two or three people. Yet, there are twenty computers within sight of the desk, not to mention the back offices, where quite a few people do have a computer on their desks.

Formula One

The first formula is simplified from previous attempts to quantify the issues of printers, broken terminals, and the like. We've decided that's all irrelevant for the first formula. We'll amplify in the support formula below. The new simplified formula is: CPU/FTE.

Really. It's almost than easy. The only problem is, what is a full-time equivalent employee? In our organization an FTE works 173.3 hours per month when figuring vacation, sick leave, and pay, but only 160 hours per month when figuring state industrial insurance coverage. So it just depends.

For our purposes, FTE means what you think it does. Full-time employees count as one. Part-time employees get added as fractions of one.

CPUs are actually more difficult to quantify than it would appear. We define CPU as a computer or a terminal, no matter how large or small. Thus a dumb terminal sitting on a desk doing nothing but discharging books is a CPU, because we know there's one in there somewhere.

Conversely, Deep Thought, the monster in the air-conditioned room, also counts as a CPU. To the puristic ZOPL programmers who care about such things, we admit Deep Thought really has four CPUs in its innards, but we maintain that's like a brontosaurus with ganglia in its tail.

For those who take instant umbrage to our slight against the larger computer, we point out that Deep Thought and his brethren actually do gain quite a number of points by virtue of having terminals attached to them. In fact, we thought about allowing even one point for a box that doesn't even have a CRT. Some people might think that he shouldn't count at all. Therefore, the more terminals a minicomputer has, the more credit it gets. It does get all the credit it deserves, but not directly.

There is also no distinction between a dumb terminal and a microcomputer. Why should there be? We're not talking intelligence here; we're talking work. What's the difference between discharging 300,000 books a month on a terminal and running a few spreadsheets and memos on a memo? …

Search by...
Show...

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.