Portfolio project

Background

Rock climbing as a sport has steadily increased in the number of facilities over the past decade. The proliferation of indoor climbing gyms has increased the accessibility, and thus the popularity, of the sport of climbing in the United States. Since this variant is not outdoors, environmental conditions ranging from rock construction to proper equipment usage can be controlled more in an indoor setting. Because of this recent traction in membership, it presents an interesting new sports-related problem domain in terms of HCI work. Specifically, we can foresee the uses of many technologies in this field including, not limited to: mobile devices, fitness tracking, context-aware ubiquitous technologies, augmented/virtual reality, and many more. With our initial research and gathering of user feedback, we have determined that there is a desire to improve the current condition of indoor rock climbing gyms. In this project, we employed user research methods, such as semi-structured interviews and online research on the current state of climbing technologies, applied analysis of user research through affinity diagramming, presented our findings of design implications, and designed lo-fi prototypes based on the top three ideas on the future of indoor rock climbing technology. We have narrowed it down to the top three prototype ideas and picked the best one to iterate three high fidelity designs. We've also tested it with cognitive walkthroughs with our classmates.

Goal

We have defined three major design criteria about people's motivations, desires, concerns, frustrations, etc. on their current thoughts about modern day indoor climbing:

1. Trackability: The technology solution should be able to track, log, notify, or save rock climbing routes of the past, present and future for some sort of personal provenance inside and outside of the gym.
2. Learnability: The technology solution should be able to teach the basics of climbing, proper technique, correct posture or provide training sessions for beginner climbers, advanced climbers and trainers who use it for instruction.
3. Personalization: The technology solution should provide the ability for greater customization of routes or training sessions on a rock wall according to each individual’s skill set or climbing preferences.

For each design criteria, we have also defined the success and failure of the proposed solution in addressing that criteria:

1. Trackability: Success would be defined as how convenient a solution would be that allows the user to successfully track or save a particular route of interest. This ability could range from allowing users access to see past and current route information, getting notified of when new routes are available in the gym, warning users when certain routes will be rotated out, or save completed/wishlist/personalized routes. Failure would be defined as having a very inconvenient way for members to access personal provenance, having no or restricted access to view route information, failure to notify climbers of route changes in some visible/audible form, or sudden deletion or obstruction of saved climbing lists due to errors or data loss.
2. Learnability: Success would be defined as how easy it is for climbers to successfully learn a given route or training session. This could be some form of visible or audible tutorial that instructs rock climbers how to climb particular routes, beginner climbing basics, or correct posture and form. The instructive format should be as clear and concise as possible such that the climber can understand with ease. Failure would be defined as being really difficult for a climber to understand how to climb or train on a particular route. This could be that the instruction is difficult to comprehend, too short or too long in length (any format), or is inconvenient or disruptive in a climbing gym environment.
3. Personalization: Success would be defined as allowing some method for a climber to set a customized climbing route or training session according to the individual’s preferences. This could be some ability that modifies the gym environment in some shape or form or design customized solutions for climbers. Failure would be described as the inability to set personalized preferences in some form, inconvenience or disruptive use of the technology at hand, or lack of customization options that are not entered in by the user.

Each prototype idea that we come up with must aim to keep at least one or multiple of these design criterias.

User Research Process

The research methods used to gather information and provide the basis for our future decisions primarily consisted of interviews from different stakeholders of the subject. When considering methods that would prove to be the best way to determine a course of action, we decided that going straight to the source would provide the most useful information to us. As such, we decided to interview representatives from 3 distinct groups of stakeholders. We managed to obtain permission from members that fell in each group by reaching out to local facilities such as the Georgia Tech CRC and local rock climbing communities for volunteers.

When formulating the interview scripts for these members, we decided to have common questions for all three groups and specialized questions where each group would have questions that pertained more to their specific experience. Our 3 groups were casual rock climbers, competitive rock climbers, and rock climbing trainers and assistants. Our interview questions asked for background on their climbing experience, including how long they’d been climbing and their routines, their interests and motivations with climbing, and technology that they have seen and would like to see in relation to rock climbing. However, the trainer group and the competitive climber group got additional questions. The trainers were additionally asked about their motivations for training others and their general means of training, while the competitive climbers were asked about their body maintenance in relation to rock climbing.

The interviews allowed us to get a variety of perspectives of the current and aspired future of rock climbing directly from actual rock climbing participants. However, we also decided to supplement this with additional online research, looking into what technologies already existed and if the interview responses corroborated the findings. This led to finding mobile apps that helped climbers find outdoor climbing routes and plan climbing workouts. In addition, this led to finding an LED hold system that allows for more flexible route creation and potential for more training possibilities, as a rising innovation.

For data analysis, our team chose to create an affinity diagram. Each team member was responsible for going through the interview conducted by themselves. As we go through the interviews, we tried to find the interesting ideas and main points from our participants and used them to create sticky notes. In a bottom-up fashion, we created an affinity diagram as a team with the affinity notes (See picture above). We used the yellow notes as our base notes to generate blue note columns by grouping of similar themes. Then using the blue notes, we created two levels of pink notes (due to lack of another color). From those pink notes, we are able to identify our primary problem spaces from our interviews. We were then able to define our design criteria from this affinity diagram as shown in our goals section.

We were then tasked to create user personas representing our main target audience. Our group decided to create two different personas, one for a professional rock climber, and one for an amataur rock climber:

Ideation

First method used for ideation was an open brainstorming session. We wrote our three main design criterias on a white board. We had one person type on a Google Document while the other four members pitched ideas that met at least one design criteria. We wrote down each unique idea pitched that seemed somewhat feasible, meaning it seemed realistic or potentially possible, while also encouraging out of the box ideas. At times, group members rejected ideas that did not entirely meet a design criteria, was deemed impossible with current technology, or was deemed too dangerous for climbers. We came up with a total of 52 ideas in BLACK text.

We met for second ideation method which was SCAMPER done on the same Google Document. Due to the nature of the method, we thought it was appropriate for each member to do around 10 SCAMPERs modifying one variable from an original idea from our brainstorming session in the previous meeting. We positioned it so that the SCAMPER ideas were done in RED text right under the brainstormed idea done in BLACK text. It was more readable this way and allowed us to see which SCAMPER method was used and which idea it was derived from. Since we were all together in the same room and using a Google Document, we were able to talk in real time to explain each of our ideas such that we didn’t overlap. We came up with 48 ideas, which combined with the brainstormed ideas to amount to 100 total.

After forming our 100 ideas, we met again at a different session to narrow our ideas down to three. The first step was creating the groups for clustering. We created our related groups based on the different types of technology discussed during the brainstorming/SCAMPER sessions. In order to identify all of the technologies used, we all sat down and went through each idea one by one, sorting each one based on which technology was used for the idea. If we identified a new type of technology that we hadn’t covered before, we would write it down as a new type of category. After going through all 100 ideas, we were able to create the following related groups:

1. Augmented/virtual reality systems
2. Physical touch screen integration in gyms
3. Web applications for tracking/learning
4. LED/smart rock boards in the environment
5. Machine learning/camera tracking of routes
6. Mobile technologies for ubiquitous tracking/learning
7. Smartwatch/bluetooth sensor connectivity
8. Other technology

It should be noted that mediums of technology that had only come up once or twice were placed under the “other technology” catrogry in order to give them a chance to be considered as one of the three final ideas. You can see all of our 100 ideas at this link and our final top 10 ideas that were reduced down to the top three below:

Prototyping

With our top three ideas, our next milestone was to create low fidelity prototypes for each idea, and then select the best one as a group to create a high fidelity prototype with. We liked the concept of Profile Boards: Mapping of planned routes with LED rock lights. It was derived from one of our original 100 ideas that stated “LED rock board that lets you personalize routes and training sessions based on saved personal profiles on a mobile application”. Here are the screens of the low fidelity prototype below with a sketch of our concept idea:

From there, we created an entire storyboard of how a climber could utilize Profile Boards in a rock climbing gym setting:


We selected the profile board concept to carry onto the high-fidelity prototype phase. Our process involved meeting online via Google Hangouts and going through each of the three prototypes in detail. We discussed how much each prototype fulfilled our redesigned design criteria stated above. We determined that Profile Boards met both Trackability and Personalization much better than the others, and it had some potential for Learnability. We also thought about how well we could transition it from low fidelity prototype to high fidelity prototypes. While all three prototypes involved some way of abstracting and showing the rock wall due to all gyms being shut down at this time, we believed Profile Boards would be the easiest to design for in this situation. We could design interactive prototypes and screens to showcase our thought process similarly to our low fidelity prototype. We reached this consensus by taking a vote virtually and reached a unanimous decision.

We had three primary key activities for our Profile Board Concept based on our design criteria:

1. Designing Routes: The prototype should be able to let the user customize and design rock climbing routes based on the same color. The interface should show the user a blank rock wall and allow the user to select certain rocks as they create their own path on their mobile application. This implies that there is a smart or LED rock wall in place at the gym. This meets the personalization design criteria as influenced by our interviews with an advanced rock climber and the GT rock wall trainer who both stated a greater desire for more integration with advanced rock climbing technology. Important stakeholders would be beginner or advanced rock climbers that would be consumers utilizing this feature, trainers that could design custom training routines for their clients, and gym staff who could automatically generate new routes instead of manually replacing rocks.
2. Saving & Loading Profiles: The prototype should be able to allow users to save personalized routes into a profile on the mobile application. When the user arrives at the gym, there should be smart connectivity such as wifi, bluetooth, or NFC communication that allows the user to connect to the smart LED rock wall. Then, the user is able to select and load a profile onto the wall such that the appropriate LED lights light up the set color and displays the customized route path while turning off rocks that aren’t part of the path. Users should be able to revisit, revise, or delete profiles at will. This meets the trackability design criteria as influenced by many of our Stone Summit member interviews where they believe some sort of tracking or saving feature would be beneficial for personal climbing provenance. Important stakeholders include beginner, advanced, and trainers who would be the consumers that create accounts to design and save profiles to be loaded onto the rock wall, as well as the gym staff or technology consulting companies that manage the rock wall technology.
3. Training Sessions: The prototype should allow users to create custom tailored training sessions such as the design of the route, the progression of difficulty, the amount of repetitions, the time it takes to complete each route, etc. These training sessions would primarily be repetitions or timed exercises. This meets both personalization and learnability design criteria, and it was influenced by our interview with a gym trainer at the GT rock wall who stated having more personalized training sessions are important for the growth of bouldering skills. The important stakeholders in this case would primarily be advanced rock climbers and trainers who would develop personalized training routines for their clients to learn and gain physical experience with.

Based on these design criteria, we created three high fidelity prototypes. Two were interactive prototypes made on Adobe XD, one was a video prototype. Here are some example screens of our best looking interactive high fidelity prototype below, the first one shows the user logged into his rock climbing profile with pre-created routes:

This screen is what a user would see when they edit or create a new route:

Testing

For this portion of our project’s conception, we decided to design an interactive prototype that would promote our chosen idea of personalizing climbing routes for users. Due to the dynamic exchange between user’s actions and prototype’s feedback, a cognitive walkthrough should be the obvious evaluation technique implemented for the analysis of the prototype. This is due to the fact that it would allow the users to assess how the application assists them in the process of accomplishing individual tasks with the intention of achieving a higher level goal.

Overall, the ultimate goal of the application is to create a seamless process that would allow users to effectively create a personal climbing route and project such a route onto the climbing wall of their chosen gym. As a result, the cognitive walkthrough contains specific tasks that the users should accomplish when interacting with the prototype in order to achieve the above-mentioned goal. At the same time, these tasks can be further broken down into even smaller activities that provide opportunities to closely track user-prototype interactions and simultaneously allow the users to evaluate their own experience during such interactions. The feedback obtained from the cognitive walkthrough is taken heavily into consideration when designing the next iterations of the prototype.

From the cognitive walkthrough template that was generously provided by Dr. Mynatt and CS 6750 teaching assistants, a new spreadsheet was constructed in order to accommodate for the relevant tasks that could be accomplished through the prototype in order to achieve the overall goal of personalizing climbing routes. From this idea, three particular tasks were devised: “create a new route”, “edit mapping for existing route”, and “project existing route onto wall and track”. From each of these tasks, smaller steps (e.g. “click on Routes tab”, “click on NEW button”, etc.) were designed in order to allow users to evaluate specific aspects of their experience (e.g. are controls conspicuous, system feedback, etc.) At the time, to allow users the room for further expressing their opinions, text entries, label “Notes”, were also included for each of these individual tasks and there was also room for users to convey their overall impression of the experience.

Results

We had four participants for our 2nd interactive prototype's cognitive walkthrough. Overall, the feedback received for the prototype was positive. Nevertheless, several problems seem to be persistent across different user-application interactions. The first, being the most noticeable, is the discontinuity that disrupts the habituation in how users typically interact with everyday technology. Specifically, one particular function afforded by the prototype is the ability to allow users to name their personalized routes. The problem comes from the lack of implementation available through Adobe XD. As a result, rather than displaying a typical keyboard by which the users can interact with, the prototype has a clickable placeholder that displays an already determined name when clicked by the users. This bypass of a stereotypical typing interaction causes confusion for many of the testers, as evident by the evaluations obtained. This problem could be easily solved by implementing a display of keyboard “pop-up” that would allow the users to connect the interaction with the name placeholder with the typical typing experience, thus negating the source of confusion.

Another problem associated with the current version of the prototype is the implementation of the “play” buttons next to the route name in order to allow users to access route information and wall-projection options. The problem comes from the lack of association between the icon and the activity the button represents, causing some level of confusion for the users. To solve this, the application may allow the user to click directly onto the route’s name to access these functions or implement a more “appropriate” icon (e.g. a “mountain” icon). With the next iteration of the prototype fixing these issues brought up by the users, the application should provide a much more seamless experience in promoting the original idea of allowing users to create and project personalized routes.

Our group had a lot of fun going through the entire design thinking process the whole semester. You can view our shown prototype at this link and our final writeup at this link. Both are available at the top of this page as well!