5 exercises — choosing the most precise word: use vs utilise, deploy vs implement vs execute, deprecated vs removed, and quantified vs vague descriptions.
0 / 5 completed
1 / 5
A technical spec says: "The system will utilise an event queue to process messages asynchronously." Which is the better word choice and why?
Option B is correct. "Utilise" has a specific meaning: to find a practical use for something that was not originally designed for that purpose ("we utilised the spare CPU cycles for background tasks"). When you mean simply "to use", write "use". "The system will use an event queue" is shorter, clearer, and more precise. "Utilise" is a common verbose upgrade that adds syllables without adding meaning. This is one of the most frequent precision errors in non-native technical writing. The Google Developer Style Guide explicitly recommends "use" over "utilise" in all standard technical contexts. Reserve "utilise" for the specific sense of repurposing or making creative use of available resources.
2 / 5
A DevOps engineer describes a deployment step. Which sentence uses the most precise word?
Option B is the most precise. In deployment contexts: deploy = move an artefact (container, binary, code) from a source to a target environment and make it live. implement = build or put in place a system or feature (not an action performed on an artefact). execute = run a process or command (the container is executed once it is running, but you deploy it first). release = make a version available (often to users or for download), not specifically about infrastructure placement. Technical precision map: "implement a feature" → "deploy a build" → "execute a binary" → "release a version". Using the wrong verb signals imprecision to technical reviewers.
3 / 5
A code review comment reads: "This function is very slow." Which rewrite demonstrates precise technical word choice?
Option C is the most precise. "Slow" is a vague evaluation; "O(n²)" is a precise characterisation. The rewrite adds three layers of precision: (1) the algorithmic complexity (O(n²)), (2) the condition under which it degrades ("at scale"), and (3) a concrete measurement ("4 seconds at 10,000 records"). This gives the author actionable information. Options A and D are still vague (extremely, poorly). Option B uses jargon ("latency profile") that sounds technical but adds no specific information. Word choice precision in code reviews: quantify (4 seconds, O(n²)) rather than qualify (slow, fast, poor). Precise language shortens the back-and-forth needed to understand an issue.
4 / 5
An API changelog uses: "The old endpoint has been deleted." Which word is more precise in an API versioning context?
Option B is more precise for a removal action. In API versioning, the terms have specific meanings: deprecated = still available but scheduled for removal; users should migrate away. removed/deleted = no longer available. "Removed" is the conventional term in API changelogs for endpoints that are gone in a new version (it is the term used by Stripe, GitHub, and AWS in their changelogs). "Deleted" is also accurate but sounds more like a database operation than an API versioning event. "Eliminated" is unusual in this context. Critical nuance: if you write "deprecated" for something that has been removed, users will not migrate urgently — they will assume it still works. Deprecation and removal are distinct events in an API lifecycle.
5 / 5
A backend engineer writes: "The job ran and did the task." Which rewrite uses the most precise vocabulary?
Option C is the most precise. It names the job type (batch job), confirms the outcome (ran successfully), and quantifies the result (12,400 records processed). This is the level of precision expected in status updates, monitoring dashboards, and operations logs. Option A replaces vague words with slightly less vague words ("performed the operation" — which operation?). Option B is more formal but still says nothing specific about what was accomplished. Option D is colloquial and provides no technical information. Precision in status reporting: always include what type of job, the outcome (completed, failed, timed out), and a measurable result (records processed, files written, tasks queued).