Texas Instruments Educational Solutions
I attribute my time at TI to software pioneer Jerry Weinberg, with whom TI consulted to re-engineer their development process. He recommended hiring a usability manager, HR found my resume on Monster, and a few months later I was building the human centered design competency within TI Educational Solutions. I managed a team of interaction and visual designers, and integrated UX into the product development process which was engineering and feature-function driven at that time. I also spent some time working in marketing, managing the scientifically-based research program initiated with federal education policy.
Methods: Business Process Re-engineering, Planning, Ethnography, Observation, Surveys, Interviews, Contextual Inquiry, Personas, Storyboards, Scenarios, Design Workshops, Wireframing, Paper and HTML Prototypes, Typography, Icon Design and Development, Visual Design, Cognitive Walkthrough, UI Specification, Heuristic Evaluation, Collaboration, Ideation, Experience Mapping, Usability Testing, Experimental Design, Product Strategy, Managing, Mentoring.
Challenge: re-invent the school graphing calculator
While still a market leader, the TI-83 graphing calculator is a ‘DOS’ type interface, a throwback in our GUI world. We wanted to use the ‘power of visualization’ to create a teaching and learning tool to help every student understand mathematics.
Working with leading educators, our team of architects, product managers, developers, and UX professionals started with stories and scenarios to envision the future state. We arrived at the concept of dynamically linked representations — allowing students to visualize the connections between algebraic, graphical, geometric, numeric and written forms of mathematics. Students can graph a function to see a parabola, for example, then drag the parts of the parabola around on the screen to see how the equation changes. Instead of the typical ‘drill and kill’ of working problems, students can interact with the problems using dynamic worksheets from the teacher, wirelessly transmitted to their handheld. They can connect the handheld to their computer at home and get the document, work on it on a large screen, then email the homework back to the teacher or download it back to their handheld.
I came into the project after the previous UX designer left for another position — I had a short hiatus from UX, managing the Research Program, contracting with educational researchers to ‘prove’ our technology improved student learning. A lot of conceptual work had been done already, with some proof-of-concept applications, but there was a lot of concern about implementing a mass-market solution. I worked with the team to take early concepts and sketch out implementation alternatives, getting the product managers to align on a framework solution. A major theme of my work at TI was helping them visualize concepts in various peoples’ heads, to work towards consensus and product implementation.
I defined the high-level UI architecture, and behavior rules, much like the Mac, Windows, and smartphone OS UX guidelines. This was done for the overall product framework, and for each individual mathematical ‘app’ — graphing, calculator, notes, statistics, spreadsheet, geometry, and data collection. Each app had a different product manager, so I had a task of reconciling each manager’s issues to maintain a consistent look and feel across the applications. I had a good relationship with the development team as well, acting as liason between development and strategy on an individual engineer level.
Since we were developing our own handheld GUI, and didn’t have an existing set from the embedded OS we were building on, I created and maintained pixel-specific guidelines for all the GUI controls and dialogs. Since the handheld was quarter VGA, I spent a lot of effort scrubbing every pixel, to maximize the amount of content available to the student work, and minimize the ‘chrome’ of the GUI. I remember having a breakthrough once, freeing up enough pixels to add another line to the menu, eliminating a scrollbar!
The interesting challenge of the TI-Nspire project was cross-platform work. An Nspire document could be created and shared between the Windows, Mac, and our proprietary handheld. I took pains to respect each OS’ idiosyncrasies; Mac users especially noticed if a Cancel button was on the wrong side…
Since the handheld was grayscale and quarter-VGA, this led to experimentation to determine some default colors which were distinctive between grayscale and color. The other major challenge was working to create an equivalent to both free mouse pointer and directional input cursor (left-right-up-down keys) on the handheld. Getting ‘Enter’ and ‘Cancel/Escape’ working right took some time.
With a beta release, and after first shipment, we spent a considerable amount of time doing customer research — running workshops with educational leaders, field studies observing classroom use, international pilot research, surveys, usability tests — and iterating on the design.
After release the work was just starting. We had a novel product and had the new product introduction challenges. I introduced the concept of ‘customer experience mapping‘, or the customer journey, to the business. We created a cross-functional team of training, packaging, sales, support, development, product marketing, and branding to ensure that our brand message was consistent across all the customer touchpoints. This led to some rude awakenings for some, like pulling out all the paper inserts we inserted into bulk packaging during an Out-of-Box-Experience activity (e.g. “we don’t even use this form any more…”).
The final product has received a lot of critical praise and some awards, including a WIRED magazine 8 of 10 rating.