[pippy]

No. The best way to disrupt an industry is to do something unrelated - something crazy and something different. You'll never break the mould and explode by copying incumbent industries line of thought.

For example the real estate industry has been upset most in New Zealand by trademe, a company started by someone who wanted an easier way to buy a used toaster. The industry here is still scrambling to catch up, instead of competing with a startup trying to emulate their business model.

By doing what the article suggests, you'll be above mediocre at best. Think different (tm).

[sliverstorm]

Engineers have a tendency to underestimate anything outside their discipline. "How hard can it be!?"

You would do well to remember that, when looking at an unrelated industry and decided that you, having zero knowledge of that industry, are obviously the best-positioned to completely revolutionize said industry.

Really, the biggest breakthroughs appear to come from someone who mixes expertise in two disparate fields. Expertise in the one field teaches you about the constraints on that field, and expertise in the second field gives you new tools to attack the constraints of the first.

[untog]

You'll never break the mould and explode by copying incumbent industries line of thought.

No-one said that having prior knowledge means you have to copy lines of thought. Working as a broker will give you valuable insight into how the process works, which gives you two things: the ability to identify easily solved pain points, and the knowledge of potential problems down the road.

[herval]

Working in an industry is a very good way of acquiring contacts and potential customers - it also allows you to REALLY get to know its problems from the inside. "Thinking different" with data and customers is usually better than taking random guesses in an industry you know nothing about...

[rimantas]

Good luck solving problems you don't know. Nobody says you must copy the industry, but the knowledge is an absolute must.

[gawker]

I'd say that that's more of solutions to one domain being applied to a different domain. But you still need to know what problem to solve in that new domain you're trying to disrupt first.