By 1990 most people had yet to experience email, but PCs existed in sufficient number to warrant asking, "How might we revolutionize publishing?"

One answer came from Jerry Borrell. I worked with him and his team at Sumeria to conceptualize a framework for a digital issue of Macworld magazine. Our CD-Rom was one of the first commercial uses of Apple's newest technology, QuickTime.

While it seems quaint by today's standards, the ability to see a step-by-step video tutorial for a memory upgrade was a clear harbinger of YouTube and Instructables.  The world now insists that you show them how, right now.

Meanwhile, multimedia tools found their way into unlikely hands. One pair of hands belonged to a leading cardiac surgeon at the Mayo Clinic. In-between his consultations and surgeries he cobbled together an impressive proof of concept.  It was compelling because he knew the customer really well (he was the customer). Subject matter experts began to prototype their own vision of an interactive future. 

As a consultant with Clement Mok's design studio, I worked directly with the client and the studio to sort out the information architecture and nurture each idea to its fullest. Why not include 3D, virtual patients that test a doctor's ability to observe, develop a hypothesis, run tests and diagnose? 

The result was groundbreaking. 

Igor, to the laboratory! Macromedia brought their first version of Director to life in 1987. Four years later Apple and IBM announced their partnership, Kaleida Labs, and ScriptX was to be their digital esperanto. M-Factory was tinkering away on mTropolis, and it escaped to market in 1995. There were so many new tools and philosophies to buoy interactive CD-Rom creators, but something ominous bubbled up from the depths.

Almost overnight the specter of the internet frightened away investment in a content delivery mechanism that couldn't be updated. There seemed to be little hope until a Duke emerged, literally. Java was released in 1995, proclaiming, "write once, read anywhere." This did much to quell the fears of content producers, but where were the new Java tools?

In 1995, I joined Autodesk because of Hyperwire, the first visual, icon-oriented Java authoring tool. It launched to rave reviews. Anyone could build an interactive slide show with sound in minutes and publish it to the internet with a click. 

The fact remained, however, that CD-Roms spat data at 600kb/second when modems only dripped at 33.6kb. A webpage took over two minutes to load, and the data to run your snazzy slideshow could take much longer.

While Hyperwire was ahead of its time and didn't last, it paved the way for today's frameworks and libraries of standardized objects. Even then I could see a future when vast teams of developers and designers would spin data, code and experiences into interactive gold.

After the concrete has been poured is the wrong time to suggest a different location for the rec room. And it's the same for software development. You can manufacture something correctly, but if you didn't make the right product you might as well not have done anything at all. If you're building a small app, trashing a few weeks' work to start over might not spell disaster. But what if the stakes are higher? What if it's not one product but 100s? What if you don't have thousands of customers, but millions?

30 years into delivering a legacy of the world's leading 3D design and engineering tools, Autodesk faced a new challenge. Through years of acquisitions, the company had expanded from a single product in 1982, to well over 200. Teams had become highly specialized and somewhat insulated from each other. The lack of coordination was evident in the experiential fault-lines between product experiences, and several top-down and grassroots approaches had not produced lasting solutions.

In the wake of one such effort, I was invited to design a different approach for Autodesk - a cross-company design workshop. I wrangled interviews with executive staff members and captured, in their own words, why "cohesion" was so important. With design leaders across the globe we formed a two-day experience that brought together Product Management, Engineering, Design, Project Management and Marketing into a single, guided conversation for the first time.

The Cohesion Workshop generated the guiding principles adopted across the company. They were further refined over a series of follow-on workshops held in key offices including China, Singapore, Tel Aviv and Toronto. We also conducted virtual workshops using Mural and Google Docs with Hangouts. This proved the value of the workshop format, and has lead to a sea change in the way innovation is approached within the company. 

Just when you think you're done, you find the journey has just begun. It's no small task to rally a large company around a common set of principles, but it's an even larger task to develop the processes to put those principles into daily action.

Over the course of several months, I tested and iterated on various workflows. The goal was to balance the positive attributes of a "startup-like" creative studio model and the rigor and discipline of an enterprise solution provider. Creating too much process is as problematic as creating too little process. After a few design sprints, making changes based on feedback from our partners in product management and development, we had a flexible process and a robust, user-centered design practice in place.

User-centered design brings perspective to an organization. It helps you see where you actually stand in relation to your customer's needs, and it can help you point yourselves in the right direction. User-centered design also creates a reliable feedback loop to help evaluate the success of each step. In other words, through user-centered design, "we see in order to move, we move in order to see.*"

*William Gibson, Author and coiner of the term, "cyberspace."