Today is 9/9/99. Ma told me that, at work, everyone was scurrying to "fix" their personal computers so they won't blow up today. Twits. Nobody over there is running software or batch programs that are looking for "99999999" for termination code. Not these days. Not since many, many moons ago when the computers were coal-fired and steam-driven... In fact, I think that was even before my time... And nobody-- I don't care what they said on the TV-- uses 9999-- or 99999 for Julian dates-- in any configuration for a test date. Nobody.

Twenty years ago I ran into a government Medicare program that looked for the letter "T" to signal End of File condition on their "trailer" tapes. Medicare hospital providers submitted these trailers-- thousands of them-- every month. The trailers, all indicating different services rendered to beneficiaries, each type of trailer designated by a letter of the alphabet, had to be processed and attached to Medicare beneficiary records so that Medicare benefits could be tracked and benefits remaining calculated. For some reason, Medicare had mandated that the indicator for the End of File condition should be the words "THE END" as the first characters of the last record.

Now each trailer record began with a single letter, "A" or "B" or whatever-- there were a lot of "S" trailers indicating "Service" dates, and "U" trailers indicating "Uncovered" hospital stays. Anyway, since each type of trailer had to be processed differently, the original programming specified that the first character of each record be read to determine which sub-program would handle the trailer. And, because this check of the first character was clean and simple to program, the original programmer just decided that if the first character was "T," the program was at the end of file and could quit. Well, that was just fine-- until Medicare forgot the original program specs and began submitting "T" trailers.

For a while, the Claims Department didn't notice. They just thought it was interesting that "T" trailers were so rare, and they were glad that the "U," "V," and "W" trailers "weren't showing up so often." I think there were no "T," "U," "V," or "W" records recorded on any Medicare claims for about two years. Finally someone got suspicious, though, and actually called Medicare to ask about the trailers.

I'm the programmer who was assigned to the problem. It took me about 10 minutes to find the "glitch." I thought it was amusing. My manager thought it was downright funny. Claims and Medicare took a dimmer view. Inside of twenty minutes I had changed the program so it wouldn't quit until the whole phrase "THE END" showed up in the record. But then, before the amended program could be put into production, I had to prove that the problem didn't exist anymore.

In those days, almost all data processing was done on magnetic tapes. To test a program for production, a programmer had to submit a processing request to Operations indicating what resources needed to be allocated for the test, and how many "scratch tapes" would be required. If Ops deemed that sufficient resources were available, and if the test had a high enough priority code, the test would be scheduled, and the programmer could submit his test: a deck of JCL cards. Ops "owned" all the "scratch tapes," and the assignment was entirely up to Ops discretion. Sometimes one had to wait until "scratch tapes" were available. Ops would mount "scratch tapes" manually when the test ran.

Medicare trailers were always submitted on 14-inch magnetic tapes, one each month. Even though there were thousands of trailers, the records themselves were only 20-100 characters-- bytes long, so a lot fit on one tape. After processing the trailers, Ops got to keep the tapes. They turned them into "scratch tapes." For purposes of testing the new trailer program, Medicare sent me a whole stack of trailer tapes.

Because I was the programmer in charge, and I had to document the testing, I stipulated that Operations return the tapes to me after processing. This made me very popular in the Programming Department. After all the testing had been done, I was the lucky "owner" of 12-- count 'em! 12!-- "scratch tapes." And if you had your own "scratch tape," well, life was sweet...

And, if you were a good programmer, and you didn't get an 0C7 (data exception), (or screw something else up) your programs would run very nicely until they met all the proper End of File conditions, closed all their files, exited the programs, released all the system resources, and the "/*" at the end of your JCL deck told the system you were done. Those were the days... did I ever tell you about the time Ops removed all the "//*" comment cards from the production decks--

My god. I sound like a senile old programmer. They're going to come cart me away. I must be more tired than I thought.

I do have too much to do. Two jobs and Projects. Like this:

I wouldn't be surprised if you don't recognize them. They're characters from the old Blake's 7 BBC series. It was on from 1979 to 1981. I've been illustrating the online edition of Undercurrents: Another Aspect of Blake's 7 by James I. Ide. And it has been a most interesting project.

First of all, I've had to learn how to draw all over again. I was never really, really good at it to begin with (I'm a much better calligrapher), but now I have the option of editing the drawings with the computer. That means that if I screw a drawing up, I don't have to chuck the whole thing and begin from scratch (or, more likely, forget the whole thing). I can scan it into the computer (where it's going anyway) and fix it. It's difficult trying to draw/fix lines with a mouse, though... I wish I could afford a tablet... and a new computer.

Character portraits are pretty easy. I start with color stills, which are easy to obtain, and I can use the computer to crop and resize the pictures for dramatic effect. But actually illustrating the story is another thing entirely. And I may not be up to the challenge. I don't know if I've got enough imagination. Or talent. Or technical expertise. (I don't care what everyone thinks: if you can't draw without a computer, you won't be able to draw with one. All the technology in the world goes for naught without talent and technique.) I know what the characters look like, but when the picture I want to draw exists only in my head, I have the very devil of a time putting it on paper.

But I have been getting better as I go. I am learning...

Or maybe not.