Historical relic

Once in awhile, something arises that triggers an old thought.   In this case, an inquiry from a SIGCIS (Special Interest Group for Computers In Society) member, Matthew Kirschenbaum, an English and Digital Studies professor at University of Maryland :

Hello all,

I am interested in a better understanding of the nature of computational error. My sense is that actual, literal (mathematical) mistakes in modern computers are quite rare; the notorious Pentium bug of the early 1990s is the exception that proves the rule. Most bugs are, rather, code proceeding to a perfectly correct logical outcome that just so happens to be inimical or intractable to the user and/or other dependent elements of the system. The Y2K "bug," for instance, was actually code executing in ways that were entirely internally self-consistent, however much havoc the code would wreak (or was expected to wreak).

Can anyone recommend reading that will help me formulate such thoughts with greater confidence and accuracy? Or serve as a corrective? I'd like to read something fundamental and even philosophical about, as my subject line has it, the nature of computational error. I'd also be interested in collecting other instances comparable to the Pentium bug--bugs that were actual flaws and mistakes hardwired at the deepest levels of a system.

Thank you-- Matt

I couldn't resist, sending a note as follows: 

the ‘famous’ error before the Pentium bug was the “Inverse Log of 2.02” error in the original HP35 handheld calculator.    We wound up replacing a lot of firmware as a result.

The bug is described well down in this article http://www.hpcc.org/calculators/wmjarts.html

For my talk at the ACM History of Personal Computing January 1986, here is the video https://www.computerhistory.org/collections/catalog/102695114

In it at minute 51:00, Tom Osborne, the key creator of HP’s 9100 and HP 35 describes the issue surrounding this “Inverse log of 2.02” error.   This is the only description I’ve ever heard

Well, that generated quite a response, from far and wide.

Here's an English teacher, Tara Fickle at U of Oregon (imagine that, an English teacher watching an engineer give a talk....)

Thanks for sharing this fantastic artifact. I’m only a few minutes in but I can already tell it’s going to be a good ride — your sense of humor combined with your knowledge is admirable. Sounds like the audience enjoyed it, too. Thanks again for sharing!

Julie Cohn, research historian in the Center for Public History at the University of Houston responded: 

Greetings from Houston. What was the “Inverse Log of 2.02” error? I believe that my father was the proud owner of one of those calculators - which he received as a gift in 1972 when he retired from Leeds & Northrup Company. My brother still has it. There was something about duplicate keys or two different keys doing the same function? If this is the same error, and if my memory is correct, my father refused to have it fixed because he believed it was more valuable with the error. My dad (Nathan Cohn) was a friend and colleague of Dave Packard and Bill Hewlett.

Subsequent discussion--Julie's brother still has the calculator, and he sent this picture --apparently the "inverse log of 2.02" created a recall for HP, and we changed something like 100,000 ROMs.   In this case, not only was the ROM changed, but they changed a key cap (incorrectly) for some reason.   THAT made Cohn's calculator MUCH more valuable in his eyes (and I think, in ours as well).