Bjarne Stroustrup is the designer and original implementer of C++ and the author of "The C++ Programming Language" and “The Design and Evolution of C++”. His research interests include distributed systems, design, programming techniques, software development tools, and programming languages. He is actively involved in the ANSI/ISO standardization of C++.

Dr. Stroustrup is the College of Engineering Chair Professor in Computer Science at Texas A&M University. He retains a link with AT&T Labs – Research as an AT&T Fellow. Member of the National Academy of Engineering. ACM fellow. IEEE Fellow.

1. What is "Innovation" to you?

I basically agree with Edison: “1% inspiration and 99% perspiration.” There are few great ideas, and many good ones. Even the great ones require much work to validate them and to make them into useful tools for someone. I think he called it “invention,” though, but the main point is that you need a good idea (or several) carefully refined and embodied in some form of gadget, tool, or system to make a real innovation. An idea by itself isn’t much. Think how far the idea of “atoms” have come since the early Greeks. Think how far computing has come since Turing’s paper. And those were two of the most revolutionary ideas in history – the 99% perspiration is probably an underestimate.

Obviously, I associate “innovation” with technology, rather than pure science or art, though I have no doubt that the notion of an idea needing serious thought, development, and experimentation to become more than “just a good idea” applies universally.

2. Who are your favorite innovators?

OK, let’s get back to earth and look at ideas and innovations of a more manageable magnitude. Consider the relatively few and simple ideas that became Unix, such as “have each program do one thing well and combine them using streams of characters.” The Unix pioneers – McIlroy, Thompson, Ritchie, Aho, Kernighan, Feldman, Morris, and many more – created a system and a set of related ideas and tools that live on today inside most of our software-based systems.

3. What do you consider the most promising innovations of the last 3 years?

That’s hard to say. As a rule of thumb, a major success exists in embryonic form 15 years before it becomes a major success. The one with the most ramifications to programming and software systems is multi-cores. We now have to get serious about quite fine-grained concurrency, and we have never been very good at that.

I have a camera, a cell phone, an MP3 player, and a laptop. It is obvious that some synthesis of these three will happen. We see it happening: last week someone in Amsterdam showed me a talk I gave in Canada in August on his iPod. I want such a gadget that is good at each of the tasks, rather than just a compromise that is mediocre at each and relatively large. For example, as long as a camera phone features 3Mpixels and a lousy lens, it’s unacceptable as a camera.

These two examples are huge. They are not individual innovations but sums of many and drivers of further innovation. Design – aesthetic concerns – will be important. I’m very keen on solid, functional, and beautiful designs. A beautiful and functional system contains innumerable small innovations and refinements. These are easy to overlook because they so quickly become taken for granted. For example, for decades, shower-curtain rods were straight, running parallel with the rim of the shower area to ensure than water didn’t splash out. This led to most people – literally – having too little elbow room until some genius though of having the rod curve outwards. Brilliant! Water that splashes out is caught be the curtain and runs back into the tub – and life is just a little bit better for a few hundred million people.

4. What does it help to become a successful innovator?

A solid technical education, a sense of what is practical, persistence, impatience with dogma, a willingness to take (calculated) risks. In many cases, an aesthetic sense that deems existing solutions inadequate and guides innovation. You can’t innovate in the abstract; every innovation is a response to problems.

I think idealism often plays a part. Individuals who are just out for themselves are too easily diverted in short-term money-making schemes or corporate climbing.

5. Is there a price to pay to be an innovator? Which one?

To be an innovator in a technological field, you have to have a serious technical education and work hard at developing your ideas. You become a “geek.” That’s great and often involves desirable personal traits, such as trustworthiness, stamina, and a skeptical attitude toward unproven ideas in general. Unfortunately, those are not universally appreciated traits – especially among technophobes – so it can carry a social cost. You must also devote serious time and effort to “technical details” that are often not ppreciated by managers or people in general – even if their lives or livelihood depend on them. On the other hand, survey after survey have shown engineers to be among the most stable and satisfied groups in society, so maybe the negative aspects are overrated.

6. What are the rewards to being an innovator?

Satisfaction of having made a positive contribution to the world, and sometimes status and wealth. Most of the successful innovators I have met also built up a network of friends and colleagues that can sustain them for life. I think that the “lone wolf” image of an innovator is misleading. Many of the most successful innovators are at the center of a network of exceptional people.

7. What are in your opinion the top 3 criteria for successful innovation?

Curiosity, persistence, and – of course – luck.

8. What would you recommend to young people who wish to pursue innovation?

Get a good degree in a technical subject – science or engineering – and then get a bit of practical experience trying to put ideas into working products. Also exposure to an aesthetic field, such as literature, architecture, furniture design, image composition.

9. In your opinion how can we create a culture that supports and sustains innovation?

That’s harder than it appears. Most cultures highly value and encourage regularity, predictability, providing the right answer to conventional questions, respect of authority, not wasting time on “extraneous matters”, etc. This is especially true in educational settings. Major innovation more often arise from asking unconventional questions and working hard to find elegant answers.

We need to tolerate and encourage people to “take a walk in the forest” (as A.G. Bell expressed it) repeatedly in their education and career. We are too keen on giving people well-specified tasks and having them set definite short-to-medium-term goals. Sometimes, there has to room to do a task, a project, a course “just for fun”, etc., and there must be rewards for coming up with something unusual. A culture goes stale fast if the greatest awards typically go to people who never take a chance and never look up from assigned tasks.

10. What do you think stops/slows down innovation?

Lots of specific tasks; tight deadlines. Lack of general direction; lack of deadlines. Lack of rewards, trivial rewards. Emphasis on huge, life-altering, monetary awards for the very few. Emphasis on individuals moving from technical work into management.

10+1. Do you think becoming an innovator can be taught? If yes, how?

I think I have seen it done. In Bell Labs technical managers and senior researcher often spend serious time mentoring a new researcher. I think “innovation” is more suited to one-to-one discussions than to courses; also, what is required for successful innovation in one field and in one company isn’t necessarily the same as is required elsewhere in industry or in academia. The selection of a mix of topics to work on – such as, shor term, long term, risky, and low-hanging fruit – can be crucial. I doubt that the more personal aspects of innovation, such as calculated risk taking, perseverance, and curiosity can be taught, at least not to adults.
 

• Java through the Eyes of a C++ Programmer
• Java through the Eyes of a C++ Programmer Part 2
• XML for C++ Developers
• High Performance XML Parsing in C++
• Java, Mono, or C++?
• It's the Java vs. C++ Shootout Revisited!
• What Makes a Great Technology Innovation?