Main Study

Our research question is to find an adequate task corpus with fitting metrics to describe photo applications and is therefore exploratory. During our pre-study we gathered needs of users and created tasks on these results and findings literary findings.

Tasks

Our tasks are the following:

1. Sort the following in photos.
2. a. You've taken some nice photos and sorted them into files. Click on one file. You only want to keep the best five pictures. b. Now you want to print your favourite photo. c. Send your three favourite pictures to a friend per e mail.
3. You've been to Sydney. For being able to find this Foto quickly, you want to tag it.
4. Find the photo where a woman is sitting on a bench and watching the sea.
5. Find all pictures, that contain a cat.
6. Cut a picture. You can choose any photo.

Our photo corpus contains 60 photos.

Metrics

We are using the following metrics:

• TCT
• observations and comments of participants (Thinking Aloud).

Approach

To make sure our tasks and metrics are able to gather whether a photo application fulfils the user's needs, we are testing them with Google Fotos.

We also want to ensure that our tasks work for various UIs. For this, we are comparing the tasks on a desktop and mobile UIs.

20 participants are pseudo-randomly assigned to an UI. Before a participant fulfils the tasks, he can take as much time as he needs to get familiar with the fotos and the application. This should help to reduce effects of pre-experiences.

The following variables are the resulting of this experiment design:

• independent: fulfilling the tasks on of the UIs
• dependent: TCT
• confounding: a user's prior experience
• controlled: usage of the same hardware, fotos and tasks

To get a deeper understanding of the user's experience, the participants are asked to answer the SUS and Attrakdiff - survey.