My apologies for the long post here, but I just had a test on this type of info in my psychology class last week >_>
Part of it could be an aspect of memory itself. You've got two distinct types of long-term memory, explicit and implicit. Explicit would be things like "I remember what I was doing on New Year's Eve" but implicit could be things like "I can type a phrase on the keyboard but I can't state the bottom row of the keyboard from memory without looking". The implicit part here could explain why we know how to do something but find it difficult to explain WHY we know it.
As for forgetting, there are a couple different theories as to why we forget things. One theory is that we never really remembered it in the first place. For instance, if you tried right now to draw the back face of a penny, you couldn't do it fully. You'd get basic shapes and characteristics, but you won't remember the details of what exactly is placed where. This is called "encoding failure". Another theory is the decay theory, like a leaky bucket, and if you don't periodically refresh a memory, it will be lost. While this theory is mostly discredited, it's believed to play a small role in forgetting.
There's also the interference theory, where either old information interferes with the forming of a new memory, or vice versa. In this case, perhaps the professor later learned something else that involved the View menu (changing font size, or like theymos said, changing the character encoding), and it interfered with his old memory of fixing the boxes.
Other memory distortions include terms such as "the misinformation effect" (new postevent information influences your memory of the event itself), "source confusion" (misremembering the source of the memory), the idea of "schemas" (current knowledge influencing what is remembered), and the idea of being able to completely fabricate false memories.
All that said, I'd suspect it's the interference theory in this case, but who knows what the actual reason was. Also this:
http://xkcd.com/627/
