JevMade hello@JevMade.com
← Back to guides

JevMade field notes / Advanced question design guide

Organize a complex question with labelled information

Labelled fields can keep the parts of a complicated question separate, such as an invoice number, customer name, amount, and payment terms.

Original by TypeSafe AIExtraction

Listen to this guide

JevMade’s plain-English explanation

Expanded guide unavailable: permission has not been verified.

0:00 /

AI narration

Our summary

Imagine an invoice described with separate labels for invoice number, customer, amount, and payment terms. Jev can read those labelled parts instead of one long sentence. Software calls this arrangement a JSON object; a JSON array is simply an ordered list of items.

Structured instructions can reuse the same labelled field description across yes-or-no, choice, and score questions. A schema is a fixed set of labelled fields. A taxonomy is a category tree; software can walk it one branch at a time while showing Jev the nearby choices.

Labels can clarify a question but cannot repair missing or misleading information. Large category branches may need trimming to direct children and a sample of lower categories. The program must still save items correctly, handle close probabilities, and test whether each boundary works.

Key takeaways

  1. Start with familiar labelled fields before deciding whether software structure adds clarity.
  2. A schema is a fixed set of labelled fields; a taxonomy is a tree of categories.
  3. Walk a large category tree one branch at a time instead of sending every category at once.

Structure can clarify a question but does not guarantee a correct result; examples and boundaries still require evaluation on real inputs.

Official TypeSafe documentation · Source reviewed

Read the original guide Opens the author’s site in a new tab.