Revision cramming

Exams start tomorrow and finish two weeks later. I have so much random knowledge floating around my brain at the moment:

Computer System Design

Differential Manchester encoding is as normal Manchester encoding (with a regular transition in the middle of each bit-cell, used to regenerate the synchronising clock at the receiver end), except that 0 and 1 are indicated respectively by the presence and absence of a transition at the start of the cell.

Introduction to Software Engineering

Adding more people to a late project makes it later. Software myths #7: "By eliminating operators, human errors are eliminated." Software requirement specification should be: Correct, Nonambiguous, Complete, Consistent, Understandable, Modifiable, Traceable, Verifiable

Human-Computer Interaction

Soft Systems Methodology: Developed from the world of systems theory, this holistic approach seeks to understand the wider context in which the perceived usability problem lies. It identifies Clients, Actors, Transformations, Weltanschauung, Owners and the Environment in its analysis.

Logic For Computer Scientists

Binding priorities: The unary connectives (¬, AG, EG, AF, EF, AX and EX) bind most tightly. Next in the order come ∧ and ∨; and after that come →, AU and EU.

An occurence of x in Φ is free if it is a leaf node in the parse tree of Φ such that there is no path upwards from that node x to a node ∀x or ∃x. Otherwise, that occurrence of x is call bound. For ∀x ψ, or ∃x ψ, we say that ψ — minus any of its subformulas ∃x χ, or ∀x χ — is the scope of ∀x, respectively ∃x.

Automata and Formal Languages

Let G = ⟨V, Σ, S, P⟩ be a context-free grammar, and let ω ∈ Σ*. Then a derivation tree for ω in G is a labelled ordered tree satisfying the following: The root is labelled by S. Internal nodes are labelled by elements of V.singset. Leaves are labelled by elements of Σ ∪ {ε}. If A is a label of an internal node and X₁,…,X_n are labels of its children from left to right, then A → X₁,…,X_n is a production in P. Concatenating the leaves from left to right forms ω. We can show that ω ∈ L(G) if and only if there is a derivation tree for ω in G.

Systematic Software Design

The weakest precondition for the program P to establish the postcondition R, wp(P, R), is a predicate describing the set of all states from which execution of P is guarantees to establish R.

Data Structures and Algorithms

Since keys can be chosen to break any fixed hash function, the solution is to randomly choose a hash function from a carefully designed class of functions at the beginning of execution in a way that is independant of the keys that are going to be stored. Let H be a finite collection of hash functions that map a given universe, U, of keys into the range {0, 1, …, N-1}. H is universal if and only if for each pair of distinct keys k_i, k_j ∈ U, the number of hash functions h ∈ H for which h(k_i) = h(k_j) is at most |H|/N.

Knowledge-Based Systems

The General Boundary G with respect to a hypothesis space H and training data D, is the set of maximally general members of H consistent with D: G ≡ { g ∈ H | Consistent(g,D) ∧ (¬∃g' ∈ H)[g'>_gg] ∧ Consistent(g',D)

The Specific Boundary S with respect to a hypothesis space H and training data D, is the set of minimally general members of H consistent with D: S ≡ { s ∈ H | Consistent(s,D) ∧ (¬∃s'' ∈ H)[s>_gs'] ∧ Consistent(s',D)

Two nice things about tomorrow's exam (Introduction to Software Engineering) though: It's the least important exam (the module was 70% coursework), and through looking at the last five years of past papers, it appears that they set the same paper every year.

They've had exactly the same first two questions on the last two papers. There are a few questions that have been on four or five of the last five papers, and the rest of them crop up in about half the papers. And none of them are even good questions; they seem to mostly require us to have memorised certain acronyms, anecdotes or statistics related in the course notes rather than requiring us to, say, understand the course material, think about it and apply that knowledge to a question or situation.

Sigh. I realise I should be happy about having an easy exam, but I just find it very depressing how this module's been such a joke all the way through. And the icing on top? It's partly run by my favourite incompetent lecturer. Hopefully they'll be less-than-incompetent with the marking.

Wish me luck...