I fear most of you will take this advice the wrong way. We all know that nothing's completely original. You shouldn't be reinventing the wheel with every progr

I fear most of you will take this advice the wrong way. We all know that nothing's completely original. You shouldn't be reinventing the wheel with every program you write. But follow this advice too closely and you'll end up with a clone. A program with all the good qualities of its predecessor, and all the flaws too. Your clone will be competing with other clones for control of the market. And so long as you're competing on the same turf, you'll be at a disadvantage. Big established companies rule the market for time tested software. Their program came out before your clone, and they've had lots of time to grow their business. They have a bigger research budget than you, and a bigger marketing department. Besides, everybody already uses their product. You ever hear of network effects? Start with something that's been done before. But take it somewhere new. Find an emerging market for an established technology. One the big players ignore. Historically, that's where startups have had the advantage.