Hello, dear readers! 👋
In this issue, among other things:
What is Object-oriented design (OOUX)
Six design principles for the design of generative AI from the IBM team
How color spaces are arranged
Who are design engineers and what they do
What is the "distribution economy" and why will it replace the "knowledge economy"
D-shaped watch concept
AI service for creative image upscaling
…and much more!
Enjoy reading!
🗞 News and articles
An interesting technical article by Eric Portis about the design of color spaces, their capabilities and disadvantages. You will learn from it: what is a color space, what was the first space and why it was three-dimensional, why there is no single standard, why none of the spaces can be considered correct and much more.
The article contains interactive examples and visual illustrations.
Why UX designers should create object maps (and how to start)
Alan Wilson talked about what object-oriented design (OOUX) is and why interface designers should use it.
With this approach, the designer needs to make a map to understand which objects users create, edit and use in the product and how the relationships between them are built. The information obtained will help to create new functions and identify problems in the interface architecture.
Alan explained what the map of objects consists of, and also gave step-by-step instructions on how to create it. As an example, he created a map of objects for a fictional music application.
Design Principles for Generative AI Applications
A group of researchers from IBM has published six valuable design principles for designing visual generative AI tools. They have selected several implementation strategies and successful examples for each principle.
Principles:
Design responsibly. Create a design for the user, understanding their needs and pain points, rather than putting technology or its capabilities first
Design with mental models in mind. Generative AI poses new challenges to users, so it is important to convey to users through mental models how the system works and how individual actions affect it
Calibrate user trust. It is important that people check the results of AI work for inaccuracies or unacceptable results and treat them critically. Be honest about the limitations and weaknesses of the system
Consider the variability of the results of generative AI. Help users choose what they need or adjust the query
Provide opportunities for joint human and AI creativity. Add more controls during generation so that the user can improve the results and create exactly what he needs.
Consider the imperfection of AI. Help users improve the result of generation, point out errors and collect feedback
The article is a brief extract from the research work of the team. The full text can be read in this PDF.
⚡️ Briefly
Keep reading with a 7-day free trial
Subscribe to bezier.design to keep reading this post and get 7 days of free access to the full post archives.