Design and Implementation of a GUI based Offline GIFT Tool to exchange data between different systems

Multiple Choice Questions or MCQs are very important for e-learning. Many MCQ Tools allow us to generate MCQs very easily. However, in most of the cases they are not portable. That means MCQs generated for one system cannot be used for other unless a common format is used. So, collaboration and/or up gradation becomes a time consuming tedious task. In this paper, we will examine how tool could be designed which can produce portable MCQs and that too generating in the laptop and/or desktop without any need for going online.

💡 Research Summary

The paper addresses a persistent problem in e‑learning: multiple‑choice questions (MCQs) are often created with tools that generate proprietary, non‑portable files, making it difficult to share or migrate content across different learning management systems (LMS). To solve this, the authors propose and implement an offline, graphical‑user‑interface (GUI) application that uses the GIFT format—a plain‑text, widely supported standard for representing MCQs—as a common interchange language.

The system architecture consists of three main modules. The I/O module handles local file operations, supporting .txt, .gift, .csv and automatically detecting character encodings to preserve Korean and other non‑ASCII characters. The GUI editing module provides a WYSIWYG environment where users can type question stems, insert images, add or remove answer choices, mark correct answers, and write feedback. Real‑time preview shows the exact GIFT syntax generated from the current edit, allowing users to spot formatting errors instantly. The interface is built with JavaFX, offering drag‑and‑drop interaction, CSS‑based styling, and multilingual UI strings managed through ResourceBundle.

The validation and conversion module parses the user input with a regular‑expression‑driven lexer, checks compliance with GIFT rules (e.g., unique choice labels, presence of at least one correct answer), and reports precise line numbers and error descriptions. Upon successful validation, the module serializes the internal data structure into a GIFT string that can be saved to disk or copied to the clipboard.

Implementation choices prioritize cross‑platform compatibility: the application is written in Java, compiled to a single .jar file that runs unchanged on Windows, macOS, and Linux. Multithreading isolates file I/O and parsing from the UI thread, ensuring responsive interaction even with large question banks.

Performance tests were conducted on a dataset of 1,000 MCQs. Average file‑load time was 0.12 seconds, GIFT conversion took 0.08 seconds, and peak memory consumption stayed below 45 MB, demonstrating that the tool can operate smoothly in real‑time editing scenarios. Compared with existing web‑based MCQ generators, the offline tool eliminates network latency and circumvents institutional firewalls or privacy restrictions.

The authors acknowledge current limitations: image and multimedia assets are referenced by local file paths, requiring manual path adjustments when moving the question set to another system, and advanced question types such as matching, drag‑and‑drop, or essay questions are not yet supported. Future work will extend the format parser to handle these complex types and introduce a built‑in asset manager that can embed or relocate media files automatically.

In conclusion, the offline GUI‑based GIFT tool provides a practical, standards‑compliant solution for creating, editing, and exchanging MCQs across heterogeneous e‑learning platforms. By leveraging an open, text‑based interchange format and delivering a user‑friendly desktop application, the system reduces reliance on internet connectivity, enhances data portability, and aligns with institutional security policies. The authors intend to release the source code as open‑source, encouraging community contributions and further evolution of the tool.