Study finds AI tutor that prompts user effort leads to better retention.

·2 views

A developer created an experimental SQL tutor to test the effects of withholding immediate answers. Six participants were split into two groups: one using a tutor that required an attempt before giving hints, and another using a standard answer-giving tutor. After a week, both groups took a test without any AI assistance. The group that had to attempt problems first answered all 36 test questions correctly, while the answer-receiving group got 25 correct. The results suggest that prompting effort before providing help may improve long-term skill retention compared to simply giving answers.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Missing email records cause business messages to land in spam

Strict email providers like Gmail and Outlook often filter business emails into spam if domain security records are missing. Three key records—SPF, DKIM, and DMARC—verify an email's origin and authenticity. A developer created a tool to help small businesses generate and check these records without needing full domain access. This prevents order confirmations and receipts from being missed, reducing customer service issues.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Parquet vs CSV: Key factors behind file format performance differences

Data engineers often recommend Parquet over CSV but rarely explain why. Parquet stores data column by column, allowing queries to read only relevant data, while CSV must process entire rows. Columnar storage also enables more efficient compression through specialized encoding methods for different data types. Compression ratios vary significantly based on data schema, from 2x for unstructured data to over 20x for structured data. Real-world examples show Parquet typically reduces file sizes by 4x to 25x compared to CSV.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

New library enables precise autocomplete positioning relative to text cursor

A developer has created Caret Geometry, an alpha-stage browser tool for accurately measuring text insertion points. The library addresses the challenge of positioning autocomplete menus near the caret location as text wraps or scrolls within textareas. It provides viewport-relative coordinates matching getBoundingClientRect()'s coordinate space. The integration requires handling both caret movement and layout changes for proper popup positioning. The tool works with Floating UI for placement decisions while leaving application logic to developers.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

New system adds safeguards for AI support agents making tool calls

A new security system called solvi.Guard has been developed to control AI support agents' access to tools. It prevents agents from executing actions like order cancellations without proper authorization. The system requires explicit customer confirmation and verifies that critical data originates directly from users. All decisions are logged in a hash-chained audit trail for transparency.

0 comments Read more at DEV Community