Skip to main content

The Duplicate Audit

Document one file family from your own storage. Every copy, export, screenshot and renamed version of a single piece of work you actually made

Version 1.1 · Supersedes v1.0 · Status: CURRENT

Due: 19 March 2027 · Weight: 15%

Follows weeks 1 to 3. Due 5pm Friday of week 4.

The brief

Pick one piece of work you actually made. Find every file that is a version of it. Document how they are related.

A file family is everything that descends from one piece of work: the original, the copies, the exports to other formats, the screenshot someone took of it, the version emailed to somebody and edited there, the backup, the one renamed when you could not tell which was current. Six members is a small family. Most real ones are larger.

The work is in the classification, not the collection. Copying is not exporting; an export is a conversion with a new format and usually a loss. A snapshot is a copy taken deliberately at a moment; a dead end is a branch nobody continued. Naming which is which forces the distinction to be made on evidence (file dates, formats, content) rather than on memory, which is the part this audit is testing.

Expect at least one relationship you cannot establish. Two files exist, they are clearly related, and nothing available to you says which came from which. Mark it as unestablished and say what evidence would have settled it. That is a finding. Filling the gap with a plausible guess is not.

What you submit

One document containing the genealogy (a diagram or an annotated list, either is fine) and the classification and evidence for each member. Use the real filenames. If a filename would disclose something you would rather not submit, say so and use a consistent substitute throughout; do not silently tidy the names, because the names are evidence.

The spec

The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.

How it is marked

Marked on whether the genealogy is accurate to the files you actually have and whether the relationships are classified correctly. A short family, fully accounted for, beats a long one with guesses in it. Declaring a relationship you could not establish is a stronger answer than asserting one you could not check.