Review: Timeline Language for Bugs

LESSON

Technical English: Verb Tenses and Engineering Status

011 20 min beginner REVIEW

Review: Timeline Language for Bugs

By the end of this lesson, you will be able to...

  • choose the right time frame for each sentence in a bug timeline.

  • revise a messy incident note into a clear sequence of state, evidence, symptom, and cause.

  • separate timing, evidence, and cause claims without overpromising.

Idea in one sentence: A clear bug timeline uses different verb forms for different jobs, instead of forcing every sentence into one tense.

Core Insight

You now have a small timeline toolkit.

Use past simple for finished events:

The deploy finished at 10:00.

Use past continuous for background action during a past event:

The worker was draining the queue when the node restarted.

Use present perfect for evidence the team has now:

We have seen this error before.

Use present perfect continuous for symptoms still moving now:

The queue has been growing for ten minutes.

Use past perfect for a state that was already true before another past event:

The token had expired before the worker retried.

The review question is:

What job is this sentence doing in the timeline?

That question matters more than the grammar label.

The Small Situation

Imagine an incident note written quickly during checkout debugging:

deploy done 10:00 / alerts started 10:05 / queue growing 10 min / same error before / token expired before retry / checking logs

The note has useful information, but the reader has to reconstruct the story.

What is a finished event? What is still happening? What evidence do we have now? What was already true before the retry? What is still uncertain?

A good update should make those jobs visible.

A Review Map

Use this map before you write the final update.

Sentence job Useful question Common form Example
Finished event What happened? past simple The deploy finished at 10:00.
Background past action What was happening then? past continuous The worker was draining the queue when the node restarted.
Evidence now What evidence do we have? present perfect We have seen this error before.
Ongoing symptom How long has this been happening? present perfect continuous The queue has been growing for ten minutes.
Earlier past state What was already true before that event? past perfect The token had expired before the worker retried.
Live work now What are we doing now? present continuous We are checking the worker logs.

The trade-off is that a clear timeline often uses several short sentences. That can feel less elegant than one long sentence, but it is easier to scan during an incident.

Worked Revision

Start with the rough note:

deploy done 10:00 / alerts started 10:05 / queue growing 10 min / same error before / token expired before retry / checking logs

Step 1: classify the pieces.

Rough piece Sentence job Better sentence
deploy done 10:00 finished event The deploy finished at 10:00.
alerts started 10:05 finished event The alerts started at 10:05.
queue growing 10 min ongoing symptom The queue has been growing for ten minutes.
same error before evidence now We have seen this error before.
token expired before retry earlier state before later event The token had expired before the worker retried.
checking logs live work We are checking the worker logs.

Step 2: order the update for the reader.

The deploy finished at 10:00, and the alerts started at 10:05. The queue has been growing for ten minutes. We have seen this error before. The token had expired before the worker retried, so we are checking the worker logs and token refresh path.

Step 3: avoid unsupported cause claims.

The sentence says:

The token had expired before the worker retried.

That is order.

It does not fully prove:

The token expiration caused the whole incident.

A safer final version is:

The deploy finished at 10:00, and the alerts started at 10:05. The queue has been growing for ten minutes. We have seen this error before. The token had expired before the worker retried, so we are checking whether token refresh failed or the worker reused an old token.

Now the reader sees finished events, ongoing symptom, evidence, earlier state, and current investigation.

A Timeline Review Rubric

Before you send a bug update, read each sentence and ask one question.

  1. Does this sentence name a finished event?
The deploy finished at 10:00.

If yes, past simple is usually enough.

  1. Does this sentence show what was happening during another past event?
The worker was draining the queue when the node restarted.

If yes, use past continuous for the background action.

  1. Does this sentence report evidence we have now?
We have seen this error before.

If yes, present perfect may fit.

  1. Does this sentence show a symptom still moving over time?
The queue has been growing for ten minutes.

If yes, present perfect continuous may fit.

  1. Does this sentence show what was already true before a later event?
The token had expired before the worker retried.

If yes, past perfect may fit.

  1. Does this sentence claim cause?
The token expiration caused the failed retry.

If yes, check whether the evidence supports that claim. Timing alone is not enough.

This rubric is small, but it prevents a common bug-report failure: one sentence tries to report event, symptom, evidence, cause, and current work at the same time. Split the jobs first. Then choose the verb form.

A Good Order for Bug Updates

When you have many pieces, use this order:

visible condition -> finished events -> ongoing symptom -> evidence -> earlier state -> current work

You do not need every piece every time.

For a short incident update, this is often enough:

Checkout is slow. The deploy finished at 10:00, and alerts started at 10:05. The queue has been growing for ten minutes. We are checking the worker logs.

Add evidence or past perfect only when it helps the reader:

We have seen this error before. The token had expired before the worker retried.

The trade-off is detail versus scan speed. During an incident, the best update is not the longest update. It is the update that lets the next reader act without guessing.

Common Confusions

Confusion: One incident needs one tense

Why it is tempting:

The incident is one story, so one tense feels cleaner.

Better model:

One incident has many sentence jobs. Use one tense per job:

The deploy finished. The queue has been growing. We are checking logs.

Confusion: Earlier means cause

Why it is tempting:

Root-cause timelines often begin with an earlier event.

Better model:

Earlier is a clue. Evidence proves cause.

The token had expired before the retry.

This is useful order, not the whole root cause.

Confusion: Present perfect and present perfect continuous are interchangeable

Why it is tempting:

Both connect past and now.

Better model:

Present perfect gives evidence:

The alert has fired in two regions.

Present perfect continuous gives ongoing motion:

The alert has been firing every minute.

Check Your Understanding

Check: Which sentence best reports an ongoing symptom?

A. The queue has been growing for ten minutes. B. The queue grew at 10:05. C. The queue had grown before now.

Think first, then reveal.

Answer: A. It shows a symptom that began earlier and is still moving now.

Check: Which sentence best reports an earlier state before a later event?

A. The token had expired before the worker retried. B. The token has expired before the worker retries. C. The token is expiring yesterday.

Think first, then reveal.

Answer: A. Past perfect anchors the expired token to the later retry event.

Practice

Revise this bug note:

worker retry 10:05 / token expired before retry / queue growing 10 minutes / seen same error before / checking token refresh

Model answer:

The worker retried the job at 10:05. The token had expired before the retry. The queue has been growing for ten minutes, and we have seen this error before. We are checking the token refresh path.

Rubric:

Second practice:

deploy finished 10:00 / alerts started 10:05 / latency rising since deploy / seen same alert before / checking timeout value

Model answer:

The deploy finished at 10:00, and the alerts started at 10:05. Latency has been rising since the deploy. We have seen this alert before, and we are checking the timeout value.

Why it works:

Daily Practice Lines

Say or write these lines during the day:

The deploy finished before the alerts started.
The token had expired before the worker retried.
The queue has been growing for ten minutes.

Then ask:

event, background, evidence, ongoing symptom, or earlier state?

Resources

Key Takeaways

PREVIOUS Past Perfect for Before-and-After Root Cause NEXT Future with Will for Decisions and Predictions