Portfolio

Here you can find design and research projects of mine. Just in case: you can as well have a look at software I wrote

A Beginner’s Guide to Finding User Needs

Responsibilities:
  • Teaching Concept
  • Writing
  • Design

A text written for all those who want to design products according to the needs of human beings. It covers recruiting, interviews, observation, data analysis and reporting. It is written in a hands-on manner and has lots of examples.

Since it has its origins in teaching design students, it does not require the learner to spend much on costly equipment or services. The text itself is free  (CC-BY-4.0)

visit this project

User Research: eLearning

Responsibilities:
  • Research Concept
  • Data Collection
  • Report
affinity
To find out how and why students use digital media for their learning, students of different courses were interviewed and observed. I analyzed the data using affinity diagramming to determine the needs of the students
how students share files diagram
Showing how students actually use IT for learning
suggestionsSlide
Report

The research helped to focus project resources on topics relevant for student's learning: distribution of learning materials in the form of files and communication were key concepts for the students. Project resources could be allocated to these core topics that might have been  neglected otherwise.

Teaching: Human Centered Design Research

Responsibilities:
  • Teaching Concept
  • Teaching
  • Contact to Mentors

For three terms I taught basic skills for Human Centered Design Research, including interviewing, observation, data analysis, idea generation (based on the earlier research, naturally!), paper prototyping and user tests.

I taught lightweight methods suitable for the student projects. To teach new skills I used educational scaffolding including worked examples, exercises and  »how-to-cheat sheets«.

Design Studio Exercise
Design Studio Exercise Design Studio is a fast-paced, visual ideation workshop method. Because it is strongly timeboxed and combines ideation and constructive critique it was well suited for my style of teaching
Affinity diagramming example
Affinity diagrams were the suggested method for analyzing data. And it introduced my students to the inevitable solving of design problems by using many many sticky notes.
finding areas for improving the teaching
Improving the class Not only 3rd party persons got better designs: The class enjoyed a regular makeover too. Combining in-person and anonymous feedback as well as my diaries I analyzed the resulting data at the end of each term before I planned for the next. Methods were introduced, others discarded and my explanations got hopefully better and better

Students used a wiki for documenting and presenting their projects and for exchanging information.

Each term we cooperated with another company or foundation which provided the design brief and mentoring:

During the three terms the class has been constantly evaluated and improved using an action-research approach. I collected data via feedback rounds, questionnaires, interviews and evaluation of deliverables. For analyzing the data I used affinity diagramming. The research triggered several changes like an introduction to tools for collaboration, improvement of teaching material, the creation of templates and the shift to interactive workshops for learning skills.

On the wiki which we used for organizing the class are examples of students’ works as well:

See also: https://blog.mozilla.org/ux/2013/08/bauhaus-wrap-up/

User Research and Usability Evaluation: Medienwiki

Responsibilities:
  • Usage Analysis (Interviews, Web-Analytics)
  • Touchpoint Analysis
  • Design Concepts
  • Usability-Tests
  • Programming (CSS/Javascript)

The medienwiki is an important part of the eLearning strategy in the Media Arts and Design course of the Bauhaus University. I prepared a redesign of the platform.  I analyzed the students use of the system with qualitative methods and web analytics. Core problems were identified using this mixed-methods design:

  • It is difficult to add images though it is often needed.
  • Information is hard to find
  • many functions are not used (and visible in the Interface) while others are often needed but hard to locate.
  • Students must be reminded by teaching staff to add a license to their content

Results:

mapping the content
Information Architecture From several card sorts by current users I derived a meaningful structure for the content.
working interface
Skin A customized mediawiki skin, based on mediawikis standard theme – but with a design according to the university’s CI

paper prototype for the upload dialog
Paper Prototype The upload dialog was iteratively developed using wireframes and testable paper prototypes first. A coded prototype would have been difficult to implement since the results of different internal stages where complex and it would have made iterating hard.
working  implementation
Upload DialogWorking implementation of the upload dialog. After the paper prototype proof-of-concept I programmed the dialog against the mediawiki APIs and tested it again with several users. I adjusted, added or removed functions based on their test performance and feedback until we reached a satisfying stage. I ran a usability test  on the working implementation (measuring completion rates and using a  think aloud protocol). According to the results, the usability was greatly improved. Testers added license information to their files, probably as result of support of the interface

Interaction Design and Research: Hold-And-Move

Responsibilities:
  • Gesture Concept
  • Data gathering
  • Prototyping
  • supporting data analysis
  • academic writing
Hold and Move in Action: The thumb 'holds' the background, the index finger moves the piece

In this work, we designed, implemented and evaluated a new gesture for mobile devices. The work was done with Alexander Kulik We published the results on the ACM MobileHCI '12.

The hold-and-move gesture allows seamless switching between navigation (panning/scrolling) and manipulation on touchscreen devices: While a single touch still pans and scrolls (e.g. a list of songs) an additional finger will move items (e.g. a list entry when resorting a playlist).

Without hold-and-move, this switch from scrolling to dragging items is usually achieved using a cumbersome ›long-tap‹ gesture.

The hold-and-move gesture runs on off-the-shelf multi-touch devices and does not need on-screen widgets. It is build upon natural principles of interaction: Holding something in place with one hand, doing manipulations with the other – just like one does when repairing a small device or when writing on a sheet of paper.

pan and scroll by touching with one finger
pan and scroll by touching with one finger
Hold with one finger…
…and drag the item with the other finger

The research was published in the proceedings of the ACM MobileHCI '12 (read the preprint-Version).

For examples, go the Hold-and-Move Webpage . It even has interactive demos you can try on your smartphone or tablet!

Usability and Interaction Design: Netspeak

Responsibilities:
  • Usage Analysis (Web-Analytics, Interviews)
  • Prototyping
  • Usability-Tests

Netspeak is a web-service which offers the search for suitable word in the context of an existing sentence. The service demands interaction by writing in query syntax. In the redesign, a graphical interface was developed and evaluated.

Sketches
(testable) Mockups
implementation
implementation (jQuery+API) In a usability test users had shorter task completion times and gave higher subjective ratings when using the redesigned interface.

RKWard

Responsibilities:
  • Interface Concepts
  • Expert Reviews
implementation

RKward is an interface for the popular data analysis programming language R. Like R, RKWard is open source software.

I review the user interface and suggest improved designs. A challenge is to come up with solutions that allow advanced users to set needed options while not overwhelming beginners (like students, who learn stats using RKWard) with all the details you can adjust in R and its libraries. Also, the designs must take the ease of implementation in mind, since the team is not big and time can be scarce.