Nightly ETL assumes one fixed schema, but a new chatbot begins dumping unstructured transcripts. What Big Data lesson applies?
Select an answer to reveal the explanation.
Short Explanation
The nightly job expected neat columns; the chatbot sent novel soup. Variety just kicked the old ETL in the shins. Flexible Big Data–style pipelines exist for exactly that bruise.
Full Explanation
Source Variety can break rigid ETL and drive need for Big Data–style flexible pipelines. When nightly ETL assumes one fixed schema but a new chatbot dumps unstructured transcripts, the civic lesson is that heterogeneous human communications expand forms beyond the warehouse’s brittle contract. Planning for Variety beats denying new resident-facing channels.
Rejecting all new sources so the warehouse schema never changes fails by concept: freezing innovation to protect rigidity is not Big Data management. Claiming unstructured text is never a Big Data source in civic settings fails by concept: transcripts are classic Variety. Claiming chatbots automatically convert all speech into one relational table fails by concept: conversion is a design task, not an automatic guarantee.
Exam caveat: Variety from unstructured sources stresses fixed-schema ETL—do not reject sources forever, deny civic text as a source, or assume automatic relationalization.
Operational check: Land chatbot transcripts in a flexible zone, define NLP or schema-on-read steps, and update ETL contracts only for curated extracts rather than blocking the channel. Document residual risk for civic operators and revisit after the next production change window.