For 1063 ("Just write this on your wrist or something"), although I respect both the "undermining Hal" theory and the "DFW fucked up" theory, I like this idea much better: we constantly see narrators screwing up, not understanding grammar and words or completely misunderstanding what they're seeing, because of their own imperfections. When Hal is narrating, though, we never see screwups in grammar and spelling, because he's a sort of vocab savant. But Hal's weakness is math, which is why Pemulis is tutoring him (and even encourages him to cheat). So Hal hears Pemulis say something like:
"Function x, exponent n, the derivative's going to be n multiplied by x and x raised to the n minus one power."
And then Hal writes down, from memory:
"Function x, exponent n, the derivative's going to be nx + x^(n1)",
which slightly screws up whatever it was that Pemulis said, because Hal just doesn't know basic calculus and even though he has nearperfect recall (p. 1023: "dictating to Inc, who can just sit there making a steeple out of his fingers and pressing it to his lip and and not take notes and wait and like inscribe [sic] it anytime in the next week and get it verbatim, the smug turd.") Pemulis cannot speak the formula {nx^(n1)}.
Oh, and I just reread the penultimate sentence of n. 123 on 1025: "It's going to be interesting to see if [sic] Hal... if Inc can transpose [sic?] the math here without help from his Mumster."
That, I think, clinches it for me. Pemulis is guessing Hal won't be able to coherently transpose the math he's heard, even though Hal'll remember the words he heard, without help from someone who understands math much better. And presumably Hal doesn't try to get help, since he isn't sure whether transpose is an appropriate word in this context (it is). (He certainly doesn't get help from Avril, because Avril would realize that "[sic?]" implies uncertainty about whether the original word really was thus, not uncertainty about whether the original word was a mistake.)
