Research & Papers

Nanomentoring: Investigating How Quickly People Can Help People Learn Feature-Rich Software

A new study shows over half of software help questions can be answered by experts in less than a minute.

Deep Dive

A team of researchers including Ian Drosos, Jo Vermeulen, George Fitzmaurice, and Justin Matejka has published a study on arXiv investigating 'Nanomentoring'. The research focuses on the potential for experts to provide extremely rapid assistance for learning complex, feature-rich software. By analyzing over 200 real questions from online help forums, the team hypothesized that a quarter of these were 'nanoquestions'—queries simple enough to be answered in less than 60 seconds.

To test this, the researchers conducted a study with 28 experts recruited from software help forums. The results confirmed their assumption, showing that for more than half of the nanoquestions presented, experts could provide advice they believed was helpful in under one minute. The study also explored preferences between text and audio responses. The findings are intended to inspire the design of future tools and platforms specifically optimized for delivering this kind of ultra-rapid, human-to-human mentoring, potentially drastically reducing wait times for help.

Key Points
  • Study analyzed 200+ real questions from software forums, identifying 25% as potential 'nanoquestions'.
  • In tests with 28 forum experts, over 50% of nanoquestions received helpful advice in under 60 seconds.
  • Research aims to define what makes a question quick to answer to guide future help tool design.

Why It Matters

This research could lead to faster, more efficient support systems, reducing downtime for professionals learning complex software.