Human-Computer Interactions: Needfinding


I asked users to perform one of their most common tasks which allowed me an in depth glimpse into their typical interaction patterns. Users of all experience levels were chosen to provide diversity in the information regarding user interactions across a wide knowledge-base.


Fran is 53 years old and unemployed is what we will called a beginner-level user. “What! That is not my username! Why does that even say that!” she exclaimed as she was trying to log into her Gmail account. As she began trying to click upon the user name that was not her own she became frustrated. Fran eventually read the link at the bottom of the screen; had the username had some sort of interaction that allowed the username to switch over to a text field when being clicked upon would have solved her first problem. Next she typed in her password only to be told that her login failed. She got further frustrated as she admitted she had not logged into her Gmail account in some time and did not recall the password. Within 15 seconds she was able to read through the links on the screen and decide that she would need to reset her password. She filled out the reset form and then navigated to her Hotmail account by clicking the back button three times and then finally clicking the bookmark that was located under the address bar of her browser. Fran solves her computer problems by reading through all of the text on the screen as she questions her own goals. The fact that Fran is able to accomplish her interaction to a high degree indicates that the text she reads on the screen is very clear in instructing her through the unknown.

After Fran resets her password she is then automatically redirected to her account settings page. Fran stared at the account settings screen as she felt the discontinuity between what she was looking at and the email page she desired to be staring at. Fran decides to resolve the situation by logging out of Google. Fran clicks the back button three times, which displays a page with an expired session before one more click of the back button yields the Google login screen. I believe that had she been presented with the Google home page she would have been forced to reevaluate her goal of checking her email and been more focused in reaching Gmail. Once she was logged into Google for the second time Fran was presented with her account settings screen. She became frustrated reading through all of the text on the screen while telling herself that she didn’t want to do any of the things listed. Finally she said “I just wanna check my email, right?” And with that she clicked upon the Gmail tab located at the top of the screen.  I believe that this interaction could have been enhanced by enlarging the buttons on the top of the screen based relative to how much she uses each individual service.

Fran has finally made it to her Gmail account; success! She read a couple of emails whose subject lines alluded to some level of employment. She was more likely to choose the emails that were based off of human names, like Natalie or Greg. One of the emails she read was for a job passing out surveys that paid $9.00 an hour for four to six hours a day. Fran is an unemployed RDA with over 10 years experience. The simple fact that this email failed to target her specific career made it spam from a legitimate source, a temp agency she had been placed through in the past.

Next, Fran navigated to Craigslist and clicked on the medical / health link. She stopped at one post titled RDA for Dental Office to check out the compensation and required experience, she determined that the location was too far away. Fran’s interaction could be simplified by implementing the same map service that is used in the housing section to target specific geographical regions. She clicked back and scrolled down the page until coming to one posting that was located closer, it was labeled RDA – Experienced. She expressed how the job seemed more reasonable, and decided to apply. Which turned into a menagerie of copy and pasting the email address into Gmail. And then copying and pasting the title. And then copy and pasting her generic cover letter and then personally tailoring it to the posting. And then finally attaching her resume as a pdf. I feel like this can be handled in a much simpler manner; the user could just email the documents to a temporary Craigslist email address that holds the documents while the users session is active. The user would then be presented with a screen that puts the cover letter and resume next to the Craigslist posting to allow for a greater level of customization. To avoid human spammers (people who mindlessly apply for jobs that are not within the realm of their skill) there will be minimum amount of alterations required to ones default cover letter. This system would also serve as a way for someone to check their skills side-by-side to their resume to see if they are a good fit for the position being advertised.


Hanna is a 30 year old teachers aid working on her masters degree. Hanna is computer literate and has a knack for distilling information into solutions; she is out typical mid-level user. A common interaction that Hanna is faced with is the online portion of her classes. The learning management system (LMS) that  her school uses is called Canvas. She clicks through a popup without clicking the check-box that would cause the popup to cease; the default interaction should be two buttons “Allow action that warranted this popup” or “No”.

Hanna needs to know is what her assignments are and when they are due. She really wishes that it automatically showed her on login what homework she needed to do and when she needed to do it by. She complained about how her assignments for the week were not listed under the assignments navigational tab, which forced her to go to the discussion board in order to determine what she needed to do. She read the professors post which allowed her to pick up on the keywords which acted as clues on where to navigate. This whole time she is being bombarded by flags that are attached to specific links that alert the user to a new item that is located in the sub menu. These flags were flawed due to the fact that some items were being made available too soon and others referenced assignments that were only available after a certain date in the future. The professor also added to confusion by telling students that they should be working on a project, yet that project has not become available. This sub-menus and flags displayed information such as grading, calendar, and discussion boards; the most crucial piece of information that failed to be listed was what assignment needed to be done and when to turn it in.

Emily was vocal in stating that she wanted a simple sidebar navigation. She found the overhead drop-down menus to be a hassle due to the fact that an extra click was quick to lead you to a sub menu page that was unfamiliar and a further abstraction of links. The UI here could have been improved by providing a simple sidebar navigation tool and disabling click through into the sub-menu and forcing the user to navigate through a classic-style GUI.

With the clues from the discussion board, Hanna was able to locate the weeks assignments in a sub menu titled Modules as she proclaimed, “This is exactly what I need to do!” The modules page clearly displayed all of the assignments broken down into easy to understand weekly lists; this was what the assignments page should have been. Hanna was able to tell the difference between what was homework, what was reading, and what was needing to be discussed (on the discussion board) based on the icon. These pictorial representations made the list easy to interpret the assignments and determine what homework needed to be done for the week. These icons displaying the weeks assignments as part of the main user interface would have made Hanna’s interaction much smoother and provided a richer experience.


Randy is a 27 years old developer and our typical expert level user. Randy finds his work flow to be varied and has customized his workspace to suit his specific needs. At times Randy will find himself engrossed in his own little world as he tracks down and squashes bugs. At other times when Randy is sitting around waiting, either for source code to compile or  for repos to finish syncing, he entertains himself by reading the xda developers forum or watching his most recently downloaded torrent.

Randy pushes his computing experience to the extreme by using Arch Linux, which required him to build the system from scratch. That level of intimacy with one’s operating system allows them to understand what is possible and how to integrate it if he wishes to spend the time doing so. Randy uses the application Guake as a way to have a drop-down terminal; Randy uses the terminal as his primary tool for compiling Android ROMs and syncing repos. To accomplish his communications needs Randy uses Pidgeon for instant messaging and Thunderbird for email.

When Nathan was asked what sort of glance-based information dashboard would help him better accomplish his goals he stated progress bars and windows that dynamically tile themselves when their is a prompting/message. The progress bars would show all of his dev tools in real time and his individual download traffic.


Overall this has been a great little exercise in UI/UX design. I have determined that user level directly correlates to the knowledge gap that exists between the human and the computer. It is within this gap that coded solutions and accurate documentation thrive.

Icon Source: Stanford Online; Human-Computer Interactions;; 2012

Leave a Reply

Your email address will not be published. Required fields are marked *