Designing for AI Algorithms Implementation

Today we start with some reading: an article how to design for the implementation of AI Algorithms. The rest of the day is mostly about working on your wireframe prototype!

Learning Objectives:

After this module, you'll know how to designing for AI Algorithms implementation.

Table of contents:

  1. Designing for AI Algorithms Implementation: 2 hours
  2. Working on Wireframe Prototype; horizontal slice: 6 hours
  3. User Feedback: 1 Hours

Now, open your worklog and plan the different tasks for today there!

Questions or issues?

If you have any questions or issues regarding the course material, please first ask your peers or ask us in the Q&A in Datalab!

Tip: Note down any important questions you might have!

Good luck!

1) How to design for the implementation of AI Algorithms

Google features some great articles, we've read some of them before. Now, open your design document and create a note section with the header "How to design for the implementation of AI Algorithms". Then go and read this acticle and note down design principles of how you do it in general; specifically focussing on the differences with regular UX design. Then look at your project using your newly fangled design principles and adjust your UX design; and derivative components, accordingly. Make sure to document what you change.

Upload your document to Github!

2) Working on Wireframe Prototype; horizontal slice

With those changes to the design document we made our laste iterative design loop which included new design techniques. Next, we're going to implement any design changes you made in the prototype. If you're finished with that just keep working untill you consider your prototype finished! If you are saturated or don't have enough time: there's time to work on it tomorrow as well but preferably you can use that time to test it with your peers and lecturers.

3) User feedback

User feedback is essential to any digital application: if the user provides some kind of input to the digital devices, then they will expect some kind of response that:

  1. Their input has been registered.
  2. That their input is meaningful and has some kind of effect. Or not, if it's a faulty interaction. Here's funny demo of succesfull interaction followed by apt user feedback, followed by unsuccesfull interactions with apt user feedback:

The user needs to know what is going on with the device given some interaction he does, otherwise he is not able to anticipate what will happen if he does activity x,y or z. And if he doesn't know what will happen if he does a given activity then the design is not intuitive and the user will drop the product/application.

You can further improve user feedback by making it feel rewarding to the user: this is called juicy feedback.

Rewarding User Feedback: Juicy Design

Juicy design refers to the idea that large amounts of sensory (usually audiovisual or tactile feedback) contribute to a positive user experience which can contribute to positive user experience and continued and increased engagement and invoke a state of flow. Juicy design is a concept which is very popular in game design, therefore we're going to take a short expedition into the world of game design.

Next up!

Coming Datalab we will reflect on designing for AI Algorithms Implementation again and give you an opportunity to ask any questions you might have.

Tomorrow, we will make client demonstration of our full design: a screen captures (video recordings) of our deployed wireframe prototypes which we'll upload to Github and hand-in for assessment!

Recommend Literature:

− Interaction Design: beyond human-computer interaction
− UX Fundamentals for Non-UX Professionals : User Experience Principles for Managers, Writers, Designers, and Developers

Further Reading:

− Designing with Data − The Design of Everyday Things, Donald A. Norman