The two of us were part of a small committee assigned to set up a computer for the New Orleans Psychoanalytic Institute. Subsequently, we have been asked a number of questions about the rationale for the task, how we went about it, why particular decisions were made, how they worked out, and what seems in store for the future. We are glad to help, but do want to make it clear that we are by no means experts, and cannot be sure that, of the many specific decisions that could have been made, we always chose rightly. We also should add that we were dealing with the needs of a small organization; the needs of a large institute, which might require a great deal of power and perhaps the use of several machines, might differ in important ways. Keeping our limitations in mind, it still seems good to have set the thing up and we are well-pleased with it; we hope to share our pleasure.
And we have another hope: that the various institutes — and psychoanalysts as individuals—will acquire hardware and software so compatible that we can communicate more with each other—swap information either by telephone lines or by sending floppy disks, or cassettes—or, ultimately, even laser disks through the mail.
A large number of people seem to agree that computers will have a place in psychoanalytic professional life. George Klumpner, of Chicago, Chairman of the Committee on Scientific Activities, has made a fine contribution by catalyzing that interest. During the last several meetings of the Association, he has led gatherings of analysts at the astonishing hour of 7:00 on Saturday mornings to talk about computers. The meeting rooms have been crowded to capacity.
The least tedious way to tell our story seemed to be to invent a sort of question-and-answer interview:
Probably every kind of reason. Leaving aside the irrational—the computer as magic (black and white), arcade game, status symbol—we wanted one because it could be cost-effective for our sort of organization, and because we had people interested in learning to use it as a machine to make life easier and more interesting.
That depends. It certainly won't if used for silly purposes. Or the wrong pieces of information put into it. GIGO is the cliche; garbage in, garbage out. It makes no sense to turn on a computer, wait for it to warm up, then give the orders to find somebody's telephone number. A small Rolodex file is a lot faster. It would make no sense in our institute, relatively small, to do accounting on a computer spread sheet when it could be done more easily by hand. We can do our simple accounting and write our few little checks with a pen faster than a computer could do them. But some day we might get an inexpensive one to at least try. A simple monthly print-out of a spread sheet could help a Treasurer!
On the other hand, a computer can be cost-effective indeed, especially in terms of its use as a word processor. Used properly, in some situations, it can pay for itself within a year. In fact, we believe it already has for us; in preparing for our 1985 Site Visit, we were able to maintain the good humor of our Administrator and avoid having to hire inexperienced, part-time help to do the necessary work—and perhaps inadvertently expose such part-time people to confidential material.
Besides, we believe that the computer has other, though less tangible, advantages—distinct values that would be hard to price.
Work with text and records. For example, if an Administrator is a capable person, he or she won't be charmed by retyping draft after draft of minutes, reports, identical letters, memo's, etc. It is easier to write with a word processor in the first place, and it is infinitely easier to do a new version of any old draft. Many letters that go out of an institute are practically identical: letters of reference, notification, etc. It is ridiculous to type any block of frequently used material more than once. A computer can pop that block on the screen in an instant, providing some easily remembered (or looked up) code for it is decided upon.
Associated records. If there are frequently used names and addresses, it is ever so much easier to merge them with appropriate letters, memo's, etc. While some paper records may be kept, there are a number of advantages to the storing of records of candidates and faculty on appropriate disks, with copies kept in appropriately safe places. (There are various means available to assure that these records' confidentiality can be preserved.)
General educational purposes. Need a new reading list for a class? It is easy to call up the previous ones from the previous years, copy one, then change it if necessary—all in a very short time.
A correct reference? The right address? Easy, if you have organized your computer files sensibly.
Convenience. A small, inexpensive program allows one to call up, with one key stroke, the calendar month, with the correct date highlighted, or next month, or any month for years and years. Touch another key and up pops today's appointment calendar—or any other day's. Still another key brings a notebook, a scratch pad, and if one wishes, it will add the right time and date to each note typed. Another produces a private, coded telephone directory, which will dial the number automatically if desired (and if there are a number of calls to be made, it is faster than using the Rolodex). Still another key brings up a functional calculator.
A glossary, if you have one, will flash whatever you want on a screen: a letter, name and address, the Institute's
name and address properly centered — anything you want (on our machine, we can type “ref” and hit a function key, and the text of a request for a letter of reference flashes on the screen, all ready to be merged with somebody's address. Type “vs” and the function key and you get one of our names with its address. Type “nopi”, followed by the special key, and there you have, properly centered, our Institute's name and address. Any commonly used block of text or numbers can be put into the glossary and associated with some simple code.
And since we have subdirectories for each of our candidates, and each of our faculty members, it is a lot easier to type in initials (or some other easy code decided upon), and follow it with another key to automatically get to any particular faculty member's subdirectory with everything there, or another simple code and one special key to find a given candidate's subdirectory. Then any special file can be called up to be seen on the word processor—supervisory report, summary sheet, whatever.
References. We have put about 3000 references in file form, so that they can be individually popped from one document to another. (We'll be glad to give anybody a copy—providing the equipment is compatible.)
More references. Beyond the alphabetized file just mentioned, a group of our people has been working happily with a group in San Francisco and one from N.Y.U. to put all the major English language references in an easily searchable database form. The project was the idea of Stanley Goodman, and he has been the leader. Early in 1985 we finished entering all the references in the five major English language journals back to 1920. More than 6000 of them. We are now working on the reviews, so that the correct references to the major psychoanalytic books will be available in a few months. The next phase may be for some group or groups to add search words not included in the titles—after that annotations and cross-references. From our joint venture, we will be glad to send you the four floppies now available at cost—$25, including mailing. So, if you have an IBM or compatible machine, you can have the references we now have by contacting Stanley Goodman in San Francisco (please send a self-addressed, stamped envelope). When the Reviews are added, the disks can be had by contacting Vann Spruiell in New Orleans (also with a S-AE).
We are a small and poor discipline, but we can do very well if we make information free, and share it. Only—we need to plan for compatible equipment!
As it “grows” with stored information, the computer will become more and more valuable (keeping GIGO in mind!). We will probably put our library system on the machine, and maybe do simple accounting. And we have plans for a local Newsletter.
Later still, we will add a modem (an inexpensive device which allows computers to communicate over telephone lines). Then we will be able to contact other data bases—for just one example, Dialogue. Dialogue has, on individual data bases, the largest collection of information in one place in the world. It is easy to look up anything in Index Medicus, or Psychological Abstracts, including the new section called Mental Health Abstracts (although, unfortunately, Index Medicus and the Psychological Abstracts do not cover all the psychoanalytic literature, and often abstract it in ways that are irrelevant to Esychoanalysts, so that is why a number of us from several institutes mean to do it properly and without profit—for ourselves!), or summary abstracts and reviews from maybe 500 magazines and goodness knows how many newspapers, or all the books in print, or millions of individual biographies, or—the list goes on and on. To repeat, it is easy to learn to do these things, if they are learned and used one step at a time.
Eventually, we will be able to send or get stuff from the Central Office, or other institutes. But one thing at a time.
Yes there are. It has been exciting to get a group of our people together with groups from other institutes to do mutual work. And there are possibilities for future research projects, perhaps some under the auspices of the Institute, some under private auspices.
One of our group has opened a private bulletin board, accessible by way of an unlisted telephone number and password. Our first major project is a small, closed clinical research pilot study. Later, other research projects could be coordinated through such bulletin boards, as well as exchanges of opinion, schedules, letters, advice—whatever. It all depends on what psychoanalysts, operating together, want.
We are in the midst of a great information revolution and the consequences, which may include some bad ones, are only partially foreseeable. They will be especially important for psychoanalytic research. But the most important of the consequences may be alterations in professional communication. After all, we live in little villages, pinpoints scattered across vast empty space.
Actually, we bought an IBM-PC with one floppy disk, and added a 10-Megabyte hard disk in the other slot. The hard disk stores the equivalent of 30 floppies. The one we bought seems as reliable as the IBM version, uses no more power than a floppy disk (and thus does not require the extra power for cooling that a true XT needs), and works in the same way—and it is much cheaper!
There a number of competing machines as good or better than the IBM XT, and some of the “clones” are less expensive. But our major priorities were: first, technology (the old 8-bit machines, like most Apples, the older Kaypro's, Vector-Graphics, etc., are like 78 rpm records; PC-Dos, and MS-Dos, 16-bit machines, are like 33's; the much more powerful 32-bit machines are like laser phonograph players—and are too powerful and expensive for us now); second, probable longevity of the company (and many have already gone bankrupt); third, expandability (many computers have little or none); fourth, availability of suitable software (a failure, of the early Macintosh and the present Amiga; companies like Wang aim for larger businesses, etc.); last and perhaps most important, availability of a sufficient number of local experts who know the hardware and software (usually teachers, graduate students, and the like, who are glad to make extra money by moonlighting). The aim is to stay away from retail vendors and repairers and do our own “contracting”: the savings can be enormous. But be careful! Just as there are pseudoanalysts, there are pseudoexperts in computer technology.
We ruled out “clones” of IBM, (Columbia and Eagle, for example, both filed for Chapter 11). On the other hand, some highly compatible machines, such as AT&T's, Compaq's, or the new Kaypro's, may be slightly superior mechanically, and other clones are priced hundreds of dollars below the IBM. We ruled out Tandy (Radio Shack), although their new Model 2000 is an excellent machine at a good (retail) price. Aside from the fact that one has little choice but to pay the retail cost, the problem with Tandy comes after you buy the machine. Add-on's, or peripherals, usually have to be bought at Tandy, not at the cost-plus eight or ten percent of mail order companies. It is also hard to find non-Company people to help with repairs. You mostly can't.
Reluctantly, we also ruled out Apple. The larger Lisa has been unsuccessful (too much money spent on “user-friendly” cuteness). The smaller Macintosh is too small, not expandable enough, and also sacrifices too much for early “user-friendliness.” It does not take long for somebody more serious than a game-player to learn the more serious machines—and once learned, they are actually easier and faster to use.
IBM won hands down. It is a reliable, easily repairable machine. There are plenty of parts, expansion boards, peripherals, resident experts, and fellow IBM users. And now, the best selection of software exists for it.
We also guessed that most other institutes and the Central Office would come to the same conclusions.
All that means compatibility—and “swappability.” We can swap many programs and collections back and forth tree, simply by mailing disks. Finally, it seems to us that a common database “language” may well grow up during the next few years.
We chose Microsoft Word as the word processor, because it is pretty much “state of the art” at this time. It uses “windows” (up to eight on the screen at one time, each containing a different document) which allow blocks of information to be transferred from one file to another. It does this by utilizing a “mouse” which aims the cursor and has little buttons to click.
Word will translate Wordstar programs—the old version of Wordstar is the most common word processor, and although it has many virtues, it is hard to learn, does not support the “Tree” structure by which the hard disk is organized, nor does it have windows or mice. (The newest Wordstar system came out after we set up our system. It does have some of the above features, but we still would have chosen Word.)
Word has an excellent spelling checker, automatic hyphenation, a fine glossary system (to store blocks of frequently used text, like names and addresses and standard letters, accessible by the code mentioned above), the best print formatting
capacities available for microcomputers—even the capacity to use the new laser-jet printers (for a recent review, see the June 25th, 1985 issue of PC magazine).
There are other excellent word processors, of course, some with indexing capacities, etc.
Several of us here were able to supply the Institute with public domain programs for nothing. We have PC-File (an excellent database program). And we have Side-Kick (a calendar, scheduler, calculator, etc.). And we have dBaseIII, and are learning to use it. It seems the most promising tool to be used for research purposes in the future.
Later, as mentioned, we will add the modem—providing telecommunications programs. Its cost will be $300-$400.
We bought through a respected mail order house. Looking through the latest issue of computer magazines, the prices of the items we bought currently (January, 1986) would be:
IBM, 256 K, one double-sided 360 K drive, one hard disk with 10 Megabytes—about $2000.
AMDEK amber monitor—$160.
Hercules graphics card (an internal card which provides graphics abilities and vastly improves sharpness of resolution on the screen, and the usefulness of the mouse)—$300.
AST expansion card (adds another 384 K of internal—electronic—power, makes possible the use of an electronic “RAM” drive—extremely useful!—and adds an internal clock/ calendar—also extremely useful!) — $240.
NEC 3550 letter quality printer—now about $1000.
Microsoft Word and Mouse—about $320.
Miscellaneous: $300 includes a heavy swivel arm to hold the monitor off the desk; a stand to hold the central processing unit on one end beside the desk (we strongly recommend both); disks; paper, etc.
As you can see, our equipment provides a fine system which now would cost about $4300. However, if we were buying at present, we might choose one of the better dot-matrix printers. An excellent IBM dot matrix printer has just appeared for only $480.00, mail order prices. It is more flexible in use and produces reasonably good print in near letter quality (NLQ) mode; in draft mode it is fast. For between $900 and $1000 one can buy an Epson LQ1500 which produces letter quality print, beautiful graphics, and is incredibly fast in draft modes. So you could have a system essentially like ours, for under $4000. You might need to plan to pay a moonlighting consultant for several hours to help you, but the price still would amount to between $2,000 and $2,500 less than retail prices would be for an IBM XT, printer, with all the additions mentioned above. Prices vary a great deal from mail order house to house. Shop in the computer magazines—you may be able to find better bargains than those suggested above.
Maybe. But not if the retail store can help it. It is common to receive routine support from decent stores. But if problems are not routine, as happened to one of our colleagues, the store may not be at all eager to help; after paying the store its profit, he wound up having to find his own help. Of course, there is the possibility that equipment would have to be returned to the mail order house, and that surely would be a nuisance. But we believed that the savings would be worth the nuisance risk.
Of course. Early in 1986, an IBM with two double-sided drives and 256 K of RAM has a mail order price of perhaps $1500.00 (and clones can be bought for well under $1000). A monitor costs $140. A respectable dot matrix printer can be had for less than $300. Microsoft Word still would cost about $320. Disks and paper would be about another $100. So, for under $2260.00 you could have a valuable IBM machine (and for a lot less than that, you could have one of the clones). However, the hard disk is so useful, as is the extra internal memory, that soon you would want to add things.
If we had the money, we would want a 20 Megabyte hard disk (for only $200 more); we might decide to get the 20 Megabyte Bernoulli Box instead of the hard disk. The Bernoulli box uses very tough, removable cartridges (actually
mailable), which work even faster than the fragile hard disks which are so prone to “crash” (and, besides having to be repaired, leave you with no data preserved at all). Since the Bernoulli Box has two drives, each with 10 Megabytes, one can be used to back up the other. Blank cartridges cost only about $50, so the potential storage space is limitless. Then, the only reason to use floppy disks at all would be to introduce programs into the machine, or make copies of short things to be mailed. That option would add about $2000 to the total cost if it were used in place of the hard disk, but it would be a good long-range investment.
On the other hand, the price of the Bernoulli Box is also falling. And the first example of an even more exciting method of mass storage has just been advertised. It is a write-once laser compact disk machine which will cost only about $2300, and which will write 100 Megabytes on one little disk (which in blank form such disks should be available for less than $100). To give you an idea of how much 100 Megabytes is, it could include the name and number of every telephone owner in the whole country.
Laser disk technology, along with the ever-cheaper optical scanning devices, which can transfer images of print into digital form without retyping, will in the near future result in whole books and journals available “on-line” or in mailable form. But be warned! It would be better to wait until the laser disk “WORM's” are standardized and well tested.
And were we rich, we would be sorely tempted to buy the new Hewlett-Packard laser-jet printer (about $2500, mailorder). Subtracting, say, $500.00 for a dot-matrix printer, the laser-jet would add $2000.00. The machine puts out print comparable to that of a professional printer — at 8 pages a minute (much faster than letter-quality printers). It is almost silent, and can print graphics. But the prices of laser-jet printers too will fall sharply during the next year. Some models may be available during 1986 for under $1500. But even if we were rich, we would struggle for restraint. The laser jet machines are complex and take a lot of time to learn to use. Does one really need such complexity?
Those who use it certainly are. Our Administrator found that she could put it to use immediately. Learning is interesting, and can be done in increments. One can start using the computer as a word processor within an hour or two. As new projects come along, one learns what one needs from the documentation manuals. Very soon, almost anything can be done faster. One learns, too, to not use the computer for certain tasks.
Even those who don't understand the computer are beginning to see its advantages. Some are beginning to be interested themselves; others have no such intentions. But, as far as we know, there are no real Luddites among us—at least nobody has smashed anything yet. Good luck!