Mostly correct advice, for most cases, to startups in stealth mode:
... stop being in stealth mode. Stop asking for advice. Stop doing your start-up. You're not ready.
- Execution is more important than the idea;
- Someone else has the exact same idea anyway;
- Completely unique ideas generally don't make it;
- The most likely cause of failure is your incompetence, not the competition;
- You desperately need real feedback;
- First mover advantage is just silliness.
Another side-effect, perhaps less important, of stealth mode, is that people you meet in networking events will think you're a startup newbie (and they'll probably be right). You won't get the same level of introductions, because who wants to risk their reputation introducing someone who doesn't even know the basics of startups? Instead, you'll get lots of advice about why you shouldn't be in stealth mode.
My first startup was involuntarily a stealth startup. We didn't tell anyone other than friends because we didn't think it was that important, and we wanted to build a sense of excitement around the launch. It was one of the main reasons the product stank at launch.
If you don't get user feedback from the right people during product development, your product will suck. That was lesson from startup number 2. It's not enough to get feedback, you need feedback from paying users, as soon as you possibly can, or else your product development bus will drive down the wrong road, and it's damn hard to turn it around later.
There's always a caveat, though, always a context where this doesn't work. I believe it works for most web startups, but in other industries, stealth mode may be de rigueur.