Assignment 1: The reduction

Choose a target, establish what it actually requires, and reduce the option space to the ones worth attempting — with reasons

Due: 19 March 2027 · Weight: 20%

The brief#

Pick something that has no business running in a browser. Establish what it would actually take, and say which approaches are worth your semester.

You are not building anything yet. You are producing the document that a competent engineer writes before committing three months to an approach, and the thing being assessed is your judgement rather than your conclusion.

Choose carefully, because you keep this target until week 12. Something trivial will not give you an analysis to write — if the honest answer is “reimplement it in an afternoon”, there is no trade-off to reason about and very little to mark. Something impossibly large will leave you unable to measure anything in Assignment 2. You also choose your own new requirement in Assignment 3, so pick a target with somewhere to grow.

Note that the option space is never empty and never small. Every approach in this course is technically available for every target; what differs is which ones are justifiable, and that is what you are writing about.

What you submit#

A written reduction, of any length that does the job. It must name the target, establish its requirements with evidence, walk the option space, and give a reason for every reduction.

Reasons take the form of costs, not verdicts. “Emulating the CPU would be too slow” is a preference. “This is interactive, interpretation runs about an order of magnitude down, and the latency budget is 100 ms” is an argument.

You will be asked to grade this document yourself in Assignment 3, so write it to be checkable. Date your claims, and record what evidence each one rests on.