Ola Hulleberg

Sustainable Web Design Guidelines

This bachelor's thesis focuses on the environmental consequences of the Internet and the need for sustainable web design practices. The Internet requires significant computing power and electricity, which contribute to climate change. The goal of this bachelor's thesis is to minimize the carbon footprint of the Internet by examining the awareness, practices, challenges, and barriers faced by both the industry and academia.

The bachelor's thesis presents research conducted for Yavuz Inal on behalf of the Department of Design at the Norwegian University of Science and Technology. It is divided into three studies: Awareness and Practices, Challenges and Barriers, and Guidelines. The studies in this bachelor's thesis investigate sustainable web design by examining developers' awareness and practices, identifying current challenges and barriers to implementation, and proposing practices that should be included in dedicated guidelines for sustainable web design.

By studying these aspects, the bachelor's thesis aims to provide practical guidelines for sustainable web design to help stakeholders prioritize sustainability in web solutions and collectively reduce the carbon footprint of the Internet. In addition, the research results have been published and will be presented at relevant Human-Computer Interaction (HCI) conferences, and disseminated along with the guidelines on a website developed by the group.

TemAI

Recently, I had the opportunity to explore exciting AI technologies in collaboration with Sivert Gullberg Hansen. Our project involved training a recognition model using YOLOv8, a popular and powerful object recognition algorithm. The goal was to develop a model that could identify and recognize different creatures in a video game.

To achieve this, we used a combination of technologies and tools. Python, a versatile programming language, was used to implement the model itself and the training framework. We leveraged the OpenCV library for image processing and manipulation, and DXCam to capture screenshots from the actual video game.

The training of the model involved several steps. First, we gathered a large amount of training data, consisting of screenshots from the video game that included various creatures and contexts. Then, we used annotation tools to label the creatures in the images and create a training dataset. We then trained the model by feeding in the training dataset and adjusting the weights through multiple iterations of training.

The project has expanded my horizon in the field of AI and machine learning, and has opened up further exploration of the use of models in context-aware decision-making in video games. The collaboration with Sivert Gullberg Hansen has been highly rewarding, and I look forward to seeing where this project can lead us in leveraging AI technology for exciting applications.

Office Coffee

In our project, Office Coffee, we used both front-end and back-end technologies to create a comprehensive solution. The front-end part of the application was developed using React.

To ensure that only authorized users can access the application, we integrated the Gmail API to verify email addresses associated with approved domains. We documented the API using Swagger, which allowed us to describe API endpoints, parameters, and response structures in a standardized and organized manner.

Additionally, we placed great emphasis on security in the project, especially when it comes to handling authentication and authorization. We used JWT (JSON Web Tokens) to ensure that only authorized users had access to certain pages and functionalities. We ensured that tokens were properly stored and verified to maintain a high level of security in the application.

We developed a secure RESTful API for the project. This allowed users to seamlessly create, read, update, and delete coffee data in the system. We built functionality that enables users to update information about the latest brewed coffee and handle various aspects of coffee brewing.

Overall, the project allowed us to enhance our skills in React development, team collaboration, and integration of front-end and back-end technologies. It provided us with valuable hands-on experience in creating a full-stack project where we addressed requirements and challenges based on real-world scenarios.

Smart Mirror

Imagine having a mirror that not only reflects your appearance but also provides you with a unique and personalized experience. We have created something revolutionary in web development and smart technology - a digital personal smart mirror that will transform the way you start and end your day!

By combining the timeless concept of a mirror with the latest technology, we have created an innovative product aimed at providing students at NTNU Gjøvik with an exceptional experience. Our smart mirror is not just a mirror; it is a gateway to efficiency, organization, and personal growth.

Picture waking up in the morning and seeing your personal calendar and scheduled tasks right on the mirror. You get an instant overview of your day without having to unlock your phone or start your computer. No more time-consuming searches for information - everything you need is right there in front of you.

By eliminating the need to constantly check your phone or computer, our smart mirror allows you to focus on what truly matters - your personal development, productivity, and well-being. You avoid unnecessary disruptions and distractions, achieving a more balanced and purposeful everyday life.