...

• Check if there are any new notes from your patients.
• It seems that your patient Amy is having a headache. You want to check her medication history especially that of painkiller for this week.
• You think Amy is having her headache because she missed painkiller. You want to tell her about it.
• You want Amy to take ibuprofen instead of painkiller. You remove painkiller from user’s drug list and add ibuprofen starting from today to end of May, 2 pills, every four hours.

Reflection

Our group spent a lot of time and effort designing the look of the user interface and really went into the details of the app. How should a user take a drug? How should the user design a drug? How many messages should the user be allowed to send to the doctor everyday? These are the questions we spend a long time discussing but I think it's worth the time to do it, because those are the key decisions to make for a good user interface.

The paper prototyping and user testing iterations really helped us to pinpoint the usability problems our interface has and helped us improve our paper prototype. It wasn't a easy task, but eventually we made a paper prototype that was intuitive to the user. We took the advantage of the light weightiness of the paper prototype and tried different versions until we're satisfied with our design.

Something the our group came up short is the actual implementation stage. When designing user interface on paper prototype, we didn't pay attention to how we can implement it on computer, and only thought about what was the best interface we can think of. This is the right thing to do, but some of the features that we thought about are not implemented on computer prototype and the final implementation, because of lack of skill and time. For example, the history page is supposed to be a scroll down which we don't know how to do. When implementing our actual app, mistakes and glitches always come up and things are not as perfect as they are on paper prototype. Choosing the color and changing the color can be challenging and lining up different sections of the app isn't an easy thing to do. We chose a very primitive backend (with pain json file) and that also took us a long time. 

Despite a lot of technical difficulty, we still implemented a working app in time and adjusted it according the the feedback from the users. We changed the history color to gray, had a much better calendar widget for selecting dates, and changed the wording in several places to make it more intuitive for the user. In general we did a pretty good job, and one lesson to be remember: paper prototype is one thing, actual implementation is another. If we started our implementation earlier (maybe even at the paper prototype stage), we can have a better sense of what can be done and what cannot be done, and our interface would be better.