I had been living in Sicily for a year or two when out of the blue I received a telephone call from the owner of a clothing manufacturing company in Valguarnera, a village situated in the centre of Sicily and with the reputation for being the most backward and out-of-touch village on the island. I had come into contact with this company through a tiny computing company in the foothills of Mount Etna, just above the city of Catania where I lived, with whom I had given technical assistance on the computer systems used to control the clothing company's production line.
As soon as I said my name on the telephone, they asked the unvarnished question "Do you know Xenix? If not we need waste no more time."
Well, I'd worked on Xenix for a few months many years before in England, so I said "yes" and without further ado they invited me to come and see them.
As soon as I got there on the bus, they took me to meet a medical doctor friend of theirs who, having realised the possibilities offered by technology, had commissioned a computer program to be written specially for him for the management of his surgery: the list of patients, all examinations and tests and their results, medicines prescribed and so on, with which he had been working for six years. He had closed the surgery the previous Friday and, when he turned the computer on again the following Monday, a flashing message appeared on the screen saying "Warning! There are errors in the data!" at which point the program froze and refused to proceed. As he had already had an altercation or two with the programmer, he had decided, as they say in the language of his profession, to seek a second opinion.
I asked him for the floppy disks on which he had recorded a backup copy of the program's data on the Friday, moved the current set of data to one side, read the (presumably good) data back in from the floppies, and launched the program. Once again, it gave the flashing message and refused to do anything else.
Following a strange suspicion, I adjusted the date on the computer and set it back to that fatal Friday, started the program and everything went as it should have. The program's windows opened, the patient records could be retrieved and everything seemed to work. I removed the copy of the data that I had retrieved from the floppy disks, moved the current set of data back into place, launched the program again, and once more everything worked fine. I advanced the computer's clock as far as 23:55 on the Sunday and the program continued to work but - hey presto! - as soon as the clock passed midnight, the flashing message popped up instead: "Warning! There are errors in the data!"
I showed this strange behaviour to the doctor who, once he had got over his initial incomprehension, became red in the face and started shouting and cursing the programmer. "In the computing field, this is called a `time bomb'", I told him. "There must be a couple of lines of code which check the date and, when it goes past a certain day, give out this message instead of working".
What should have happened? The doctor should have called the programmer in Catania, who would have come out with a new version of the program (maybe with another time bomb in it, set to some later date), the programmer would have impressed upon the doctor the importance of making backup copies of the data and of treating the computer gently, would have taken a five hundred pound fee and then gone back to Catania.
Once the doctor had calmed down a bit, he dialled the telephone number of the programmer and handed me the receiver. I explained the tests I had just performed and he started shrieking in a strangled voice "But I didn't put a time bomb in the program!" (But who had mentioned time bombs...?)
Even though the cause of the problem had been found, there was still a fine problem left to us. The doctor's last six years' work were recorded inside the program written by this person, the program in its current state refused to work unless the date was anything later than the previous Sunday and, of course, we couldn't count on any help from the programmer in question. Three lucky coincidences came to our aid:
The program had been written in a commercial language called "Informix 4GL", a barbaric and crippled language in which, a few years earlier in Rimini, I had had to write a system to manage reports of defects in a company's products.
The programmer had left copies on the doctor's computer of all the tools needed to develop programs in this language. Even though his program itself was only present in a form that could not be modified by us, these tools gave up hope of being able to decode the program's internal database and so recover the data about the patients.
Lastly, the tiny computer company in the foothills of Mount Etna sold a program to manage doctors' surgeries, which they had written in close collaboration with various doctor clients of theirs and which was also of high quality.
We showed the doctor how set the date on his computers and for a week in his surgery it was always Friday. As Italian prescriptions expire after a week, the patients had to go more and more quickly to get their products from the pharmacy until, by the end of the week, they had to go there the very same day.
In the mean time, we succeeded in lifting his data out of the old program and in transforming it so that it could be loaded into the new one and on the Thursday we gave the doctor a whole new suite of programs running on the same computers as before and with the records for his hundreds and hundreds of patients already inside it.
Since then, I've lost touch with all the people involved in this story. I heard that the doctor had consigned one of his computers to the police with the old program still on it, but I don't think it ever came to court. I forgot (straight away and on purpose!) the name of the programmer from Catania) and the little computing company has since split up, but let's hope the doctor is still happily working with that beautiful program born in the foothills of Mount Etna.