In-Depth

Q&A: The Making of a Master (Technology) Inventor

A 37-year-old inventor holding 17 patents, with 30 more pending, talks about invention, innovation, collaboration, and patents.

• By Stephen Swoyer
• 02/10/2004

We asked David Kaminsky, a senior architect with IBM Software Group and an IBM Master Inventor, why Big Blue is such a consistent patent performer. At just 37 years old, Kaminsky knows a thing or two about patents: He already holds 17 of them; another 30 are pending.

You’re known as a “Master Inventor” with IBM. That’s quite a title. Could you talk about what that means, and what you had to do to earn it?

IBM started the Master Inventor program six or seven years ago, and the real focus is—it recognizes employees who have contributed to the intellectual property portfolio of the company in two ways: directly through the patent portfolio through innovations that have proven to be valuable, and … through the mentorship process, people who have shown a willingness to help other people navigate the process.

How many patents do you hold, and when was your first patent issued?

Right now, 17 patents, with 30 or so pending. The patents that you saw issued in 2003 really reflected a lot of the innovative work that IBM was doing really around 1999 or 2001. My first patent was issued around 1997, maybe 1998.

And your proudest accomplishment as an inventor?

Obviously, getting the first one! IBM has a plaque and an awards program for both filing applications and for when they’re issued, which is nice. I have a decent number of patents now and IBM has rewarded that fairly generously. But what I think is kind of cool is looking at the ones that were issued and looking at the ones where I’m not a co-inventor, but where I’ve helped the person figure something out. For example, one of the people in Research Triangle Park who has the most patents on site was actually someone I’ve helped bring into the patent process.

You’ve been designated a Master Inventor, but you’re not in IBM Research, per se, so producing patents isn’t necessarily what you do for a living. Could you talk about your work with IBM and about how that provides opportunities for innovation, such that you’ve now been granted multiple patents?

My primary job is I work in autonomic computing focusing on policy-based computing, and I also do a lot of the work on patents and working on the invention part of this. One of the great things about IBM is that you have so many opportunities for invention and innovation just in the work that you do. What I’ve really found is that people here invent from their first day at IBM. The key to invention is to figure out a better way to do something or something that’s never been done before, and IBMers really start doing that from day one. The people that IBM hires have that innate ability to do that.

Personally, I was pretty lucky, right from the beginning: My first mentor (and one of my early managers) had a lot of patent experience, and he specifically took time to make sure that I understood the process.

Help us understand the process. What’s involved with conceiving an idea for an invention and actually getting it patented?

IBM really has fostered a culture of mentorship where they’ll have people who are senior technical people or senior inventors get together with new hires where they can help them understand the patent process. You know, they’ll say, “You have this idea, I’m no help at all, but I know this guy who’s in Rochester or Toronto or whatever who would be a perfect collaborative partner."

So that starts it out. Then we have pretty strong process in the middle where you have the ability to talk with your colleagues and improve the idea, along with a fairly easy-to-use tool for submitting the idea, where you disclose the kernel of the idea and then a peer review board helps people decide which things get filed as a patent and which things IBM chooses not to file as a patent. So you sit down in front of a jury of your peers, and the board will determine if we’re going to file that application. This decision, which is actually quite important to IBM, is given to the technical people.

Finally, the [intellectual property] people are there and work with the attorneys. They take the text on paper, so that it’s a valid legal patent.

About what percentage of patent applications that are submitted are actually approved? Have you any idea?

My understanding is that it’s well over 90 percent. IBM has a pretty good vetting process, so the rejection rate from the patent office is actually quite low.

Could you describe one of your more recent patents for us?

Well, one that I had issued last year, I was working in pervasive computing, and we were looking at ways to have mobile devices work better when they have intermittent connectivity. With these things [connectivity] tends to come in and out, so the browsing experience can be somewhat frustrating, so you’re going through a series of Web pages and all of the sudden you hit a dead spot and can’t get to the next Web page.

So what we did was look at ways to make that browsing experience more effective. When people go to a Web site, they statistically tend to go to a similar Web site. For example, people who go to ESPN might be more likely to go to the Super Bowl site, so using that statistical information about how people use that Web site, the patent talked about using a bundle of Web sites, so you download a bundle of pages, and as they browse, similar pages would be cached for them to view.

It’s an impossibly broad question, but what do you think are the most important qualities for an inventor or innovator to possess?

The ability to just open your mind and combine things in ways that people wouldn’t quite have thought before. People always say that everything’s been invented before and now we’re just putting them together in different ways, but most of the good patents are things where someone says, “I wouldn’t have thought of doing that but now that you’ve mentioned it, that makes a lot of sense.”

The other is the willingness to work together as a team. Most of the good patents I’ve seen at IBM aren’t from a single inventor but are the result of a team of inventors. It’s also important that you’re willing to listen to feedback from other people, where each step they make it a little stronger, a little crisper, and so they come up with an idea that is fairly broad and fairly new.