CrowdSurfer: Seamlessly Integrating Crowd-Feedback Tasks into Everyday Internet Surfing

Document Title

CrowdSurfer: Seamlessly Integrating Crowd-Feedback Tasks into Everyday Internet Surfing

Document Information

• Domain: Human-Computer Interaction (HCI), Online Crowdsourcing, and User Feedback Systems
• Keywords: Crowd-feedback systems, Crowdsourcing, Design feedback, Browser extensions, Human-Computer Interaction, Work flexibility, User experience, Working conditions, Feedback quality, Software design

Research Background and Issues

• Key Issues or Challenges:

  1. Traditional methods for users to provide in situ feedback on websites are lengthy and intrusive, leading to decreased participation and lower feedback quality.
  2. While crowd feedback is a scalable solution, it has the following limitations:
     • Lack of immersion in real usage scenarios, which may compromise the authenticity of feedback.
     • Artificially constructed usage scenarios (e.g., simulated personas or environments) increase participants' workload.
     • Traditional crowd feedback is often limited to specific points in time, making it difficult to achieve continuity.
  3. Working conditions for crowd workers (e.g., low pay, inflexible work environments) are often overlooked.

• Significance of the Research:

  • Establishing more efficient feedback channels between users and designers can improve interactive website design.
  • Exploring integrated solutions that enhance both the work experience of crowd workers and the quality of feedback, balancing these two aspects.

• Motivation and Related Work:

  • Existing research indicates that contextualized design feedback can improve feedback quality.
  • The concept of "casual microtasking" suggests that embedding small tasks into users' daily activities is feasible.
  • Crowd workers tend to mix work with non-work activities, but current tools do not fully leverage this characteristic.

Solution

• Research Methodology and Approach:

  • Proposed CrowdSurfer, an integrated Chrome browser extension that allows crowd workers to complete feedback tasks while browsing the internet in their daily routines.
  • CrowdSurfer presents feedback tasks via pop-ups, links to mainstream crowdsourcing platforms (e.g., Prolific), and supports real-time feedback.

• Innovations:

  1. Seamless Integration: Embeds tasks into crowd workers' everyday internet surfing, reducing the disruption of task switching.
  2. Transparency and User Control: Workers can enable or disable the extension at any time, enhancing privacy and work flexibility.
  3. Enables in situ feedback collection while minimizing additional time and invisible labor demands on workers.

• Implementation Steps and Technical Details:

  1. Browser Extension Download: Users register via Prolific and learn to use the tool through a tutorial.
  2. Task Presentation: Feedback tasks are displayed on specific webpage elements, using simple star ratings and text feedback formats.
  3. Data Management and Payment: Feedback is collected and payments are processed through the crowdsourcing platform, with centralized data storage.

Research Findings

• Key Findings:

  1. Feedback Process and Experience:
     • CrowdSurfer simplified the feedback process, reducing workers' workload.
     • Workers reported fairer compensation and greater task flexibility.
  2. Feedback Content Analysis:
     • CrowdSurfer feedback scored lower than traditional survey feedback on dimensions such as "specificity," "actionability," and "relevance."
     • Feedback provided via CrowdSurfer was shorter and more reflective of immediate reactions.
  3. User Behavior:
     • Most users completed tasks passively during normal internet browsing, while a minority actively sought out tasks.

• Advantages Compared to Existing Solutions:

  • Provided more authentic and natural user feedback.
  • Significantly reduced complaints from workers about invisible labor (e.g., searching for and accepting tasks) while improving their work experience.

• Experimental Results:

  • CrowdSurfer users found tasks simpler and more enjoyable, with higher perceived fairness of compensation compared to traditional methods.
  • However, the depth and detail of feedback were lower compared to survey-based feedback.

• Limitations and Future Directions:

  1. Expanding Feedback Scope:
     • Explore the generalizability of the approach across different task types, such as tagging and matching.
  2. Long-Term User Studies:
     • Current research lasted only seven days; longer-term studies are needed to validate the sustainability of the approach.
  3. Personalization and Incentive Strategies:
     • Investigate how to increase the diversity and personalization of feedback tasks to improve feedback quality.
  4. Support for Rarely Visited Websites:
     • Enhancing task exposure for less frequently visited websites could improve feedback collection for low-traffic sites.

Conclusion

CrowdSurfer successfully demonstrates the feasibility of seamlessly embedding feedback tasks into users' everyday internet activities. While the feedback quality is slightly lower, the system offers a new direction for improving the working conditions of crowd workers. Future research should further refine system design and expand to a broader range of task types to serve more application scenarios.