FAQs
The questions that come up every semester, and where each answer lives
Seven questions, and the page each answer belongs to. Nothing is decided here.
Do I need to know anything about Formula 1?#
No. No knowledge of the sport is assumed and none is rewarded: where a fact about racing is needed, the page needing it explains it in the sentence that uses it, and week 1 starts from what a strategist decides rather than from a season. Interest is the only thing assumed.
Why is my finishing position worth nothing?#
Because one race is one draw, and where you finished is as much what the draw did as what you decided. A student who boxed against their own instrument and came second has shown that the safety car came out — the neutralisation where the field bunches up behind a lead car and nobody may overtake — and nothing about the call itself. The race sets out what is marked instead, and argues the case rather than asserting it.
Can I make up a lab I missed?#
There is nothing to make up. A Lab is prepared beforehand and finished inside the session, so once the session ends no version of it is left to hand in. The remedy is in how the Labs are marked rather than in a later date: policies carries the rule for every assessed item in the course, and the labs page carries what it means for these.
What happens if my instrument turns out to be wrong?#
Every instrument is wrong somewhere, and that is not the failure. Not saying where is. The instrument is submitted with a document naming the conditions under which it should not be believed, and the defence is written knowing the result — where your model was wrong is part of the argument there, not a confession buried inside it.
What counts as declaring a limit?#
The domain of validity submitted with the instrument records two useful kinds. A general limitation says where the instrument should not be believed even when that boundary cannot be recognised live. An actionable limit names a clear, observable condition that can be recognised from evidence available when a call is due; it need not be an executable predicate.
Only an actionable limit can support a race departure. The race record needs both the limit declared before the race and the contemporaneous evidence showing that its condition applied, beside the call, timestamp and standing recommendation. A limit noticed afterwards cannot repair the race conduct mark. The defence can still reward an honest, rigorous analysis of an unsupported departure without retrospectively justifying it.
Can I use AI?#
Yes. Undeclared use is the breach, not use — a declaration is a few lines saying what you used, where you used it, and what you did with what came back. Policies carries that rule, and the same page covers citing code you did not write, including code a model wrote for you.
What do I need before the first lab?#
Python and one introductory course in probability and statistics as entry knowledge, and week 2’s supplied codebase installed with its own test suite passing. Setting up your machine explains what that involves, and week 1’s session lists the readiness lines themselves — it is also a drop-in for reaching them with help in the room.