# HG changeset patch
# User Steve Losh
# Date 1348701364 14400
# Node ID 0323cd2a813aff25e8a0ea89a07c16c4beacdba7
# Parent 6e03dc4871ae53b5aeff815e7bec3d522834e5b3
Assignment 2
diff --git a/week-01/assignment-02.markdown b/week-01/assignment-02.markdown
new file mode 100644
--- /dev/null
+++ b/week-01/assignment-02.markdown
@@ -0,0 +1,155 @@
+Assignment 2
+============
+
+Problems
+--------
+
+### Problem 1
+
+1. `(π>0)∧(π<10)`: `0 < π < 10`
+2. `(p≥7)∧(p<12)`: `7 ≤ p < 12`
+3. `(x>5)∧(x<7)`: `5 < x < 7`
+4. `(x<4)∧(x<6)`: `x < 4`
+5. `(y<4)∧(y^2<9)`: `-3 < y < 3`
+6. `(x≥0)∧(x≤0)`: `x = 0`
+
+### Problem 2
+
+1. π is greater than 0 and less than 10.
+2. p is greater than or equal to 7, but less than 12.
+3. x is greater than 5 but less than 7.
+4. x is less than 4. The `(x<6)` clause is unneeded because anything less than
+ 4 is also less than 6.
+5. y is greater than -3 but less than 3. This requirement comes from the `(y^2
+ < 9)` clause. Anything larger than 3^2 is larger than 9. Anything smaller
+ than -3^2 is also larger than 9. The `(y < 4)` clause is irrelevant, because
+ we already know y must be less than 3.
+6. x equals zero. This is the only possible number that can satisfy both
+ conditions at once.
+
+### Problem 3
+
+If φ1 through φn are related in some way, start by proving the base case and
+then prove by induction that φ necessarily implies φn+1.
+
+If they're not related (I'm not sure exactly what this question is asking) then
+you have to just prove them all. I'd start with the one that looked most likely
+to be false since if it *does* turn out to be false I can do a lot less work.
+
+### Problem 4
+
+Similar to problem 4: start with the piece most likely to be false, which will
+let you short-circuit quickly.
+
+### Problem 5
+
+1. `(π>3)∨(π>10)`: `(π > 3) ∨ (π > 10)`
+2. `(x<0)∨(x>0)`: `x ≠ 0`
+3. `(x=0)∨(x>0)`: `x ≥ 0`
+4. `(x>0)∨(x≥0)`: `x ≥ 0`
+5. `(x>3)∨(x^2>9)`: `(x > 3) ∨ (x < -3) ∨ (x > 3)`: `(x < -3) ∨ (x > 3)`
+
+### Problem 6
+
+1. Either π is greater than 3, or it is greater than 10.
+2. x does not equal 0.
+3. x is greater than or equal to 0.
+4. x is greater than or equal to 0 (the `(x > 0)` clause is redundant here).
+5. Either x is less than -3, or it is greater than 3.
+
+### Problem 7
+
+Attempt to show one of the pieces is true, starting with the one most likely to
+be true (which will let you short-circuit the rest).
+
+### Problem 8
+
+Attempt to show one of the pieces is true, starting with the one most likely to
+be true (which will let you short-circuit the rest).
+
+### Problem 9
+
+1. `¬(π > 3.2)`: `x ≤ 3.2`
+2. `¬(x < 0)`: `x ≥ 0`
+3. `¬(x^2 > 0)`: `¬((x < 0) ∨ (x > 0))`: `¬(x ≠ 0)`: `x = 0` (assuming we're
+ talking about ℝ)
+4. `¬(x = 1)`: `x ≠ 1`
+5. `¬¬ψ`: `ψ`
+
+### Problem 10
+
+1. x is less than or equal to 3.2.
+2. x is greater than or equal to 0.
+3. x equals 0.
+4. x does not equal 1.
+5. ψ is true.
+
+### Problem 11
+
+D = "The dollar is strong"
+
+Y = "The Yuan is strong"
+
+T = "New US-China trade agreement signed"
+
+1. "Dollar and Yuan both strong": `D ∧ Y`. This one translates pretty easily.
+2. "Trade agreement fails on news of weak Dollar": `¬D ∧ ¬T`. The loses the
+ "weak dollar *caused* the failure" aspect of the headline. Maybe
+ `(¬D → ¬T) ∧ ¬D` would be better?
+3. "Dollar weak but Yuan strong, following new trade agreement": `T ∧ ¬D ∧ Y`.
+ Again, this loses the causation aspect.
+4. "Strong Dollar means a weak Yuan": `D → ¬Y`. If this headline is talking
+ about something that is actually the case, and not a hypothetical, then
+ `D ∧ (D → ¬Y)` is probably more accurate.
+5. "Yuan weak despite new trade agreement, but Dollar remains strong":
+ `¬Y ∧ T ∧ D`. This is hard, because it loses the "despite" part of the
+ headline. I don't think there's a way around that though since there's not
+ "we expected" logical operator..
+6. "Dollar and Yuan can’t both be strong at same time": `(¬D ∧ ¬Y) ∨ (D ∧ ¬Y)
+ ∨ (¬D ∧ Y)` or maybe `¬(D ∧ Y)` or maybe `(D → ¬Y) ∧ (Y → ¬D)`.
+7. "If new trade agreement is signed, Dollar and Yuan can’t both remain strong":
+ `T → ¬(D ∧ Y)` or `T → ((D → ¬Y) ∧ (Y → ¬D))`.
+8. "New trade agreement does not prevent fall in Dollar and Yuan":
+ `T ∧ (D ∨ ¬D) ∧ (Y ∨ ¬Y)` or `¬(T → D) ∧ ¬(T → Y)`. This one is tricky but
+ I think that last one is clearest.
+9. "US–China trade agreement fails but both currencies remain strong":
+ `¬T ∧ (D ∧ Y)`. This one is pretty simple, but again loses that "it's not
+ what we expected" vibe.
+10. "New trade agreement will be good for one side, but no one knows which.":
+ `(T → D) ∨ (T → Y)`. The headline is ambiguous, but they probably mean "good
+ for one side at the expense of the other", in which case:
+ `(T → ((D ∧ ¬Y) ∨ (Y ∧ ¬D)))`.
+
+To Think About
+--------------
+
+1. It's all semantics. In the US "innocent until proven guilty" means that
+ until you're proven guilty, you're "innocent". So in that case yes, `¬guilty
+ = innocent`. But if we're just talking about purely whether they did it or
+ not (disregarding courtrooms), then still yes, `¬guilty = innocent` because
+ if they didn't do it, they didn't do it.
+
+ The only place where there's ambiguity is when you take meaning 1 of "guilty"
+ (proven guilty in a court of law) with meaning 2 of innocent (actually did
+ not do the crime) or vice versa. Sure, when you mix meanings like that it's
+ obviously not going to make sense.
+
+2. The problem in this one is this line:
+
+ "In terms of formal negation, ["not displeased"] has the form ¬(¬pleased)..."
+
+ Which makes the assumption that "displeased" is equivalent to `¬pleased`.
+ That's invalid (`¬pleased` means "I did not have positive feelings about it"
+ while "displeased" means "I feel palpably negative about it"). It entirely
+ ignores the third case ("I had no overall feelings at all").
+
+ To actually capture what "not displeased" is trying to say:
+
+ let pleased mean "I had overall positive feelings"
+ let displeased mean "I had overall negative feelings"
+ let ambivalent mean "I had no overall feelings, positive or negative"
+
+ ¬displeased = pleased ∨ ambivalent
+
+
+