I think there are a few rare cases where keeping your idea close to the vest makes sense, but overall I think it's delusional to think your ideas are worth protecting. Think of it this way: most, if not all ideas have already been thought of by somebody else. Are you damaged by those people having thought of the idea first, or are you damaged by them having executed it first? If they don't execute, then it's just an…
Of course, once you've answered all those questions you now have more than just an "idea." You have a plan of attack, a strategy, market research, design concepts, historical data, etc. In other words, you actually have something worth protecting. So, are you effectively saying that the answer to the opening question is that startups in stealth mode aren't really protecting ideas, but are protecting the plan of attac…
Where "what you are working on" is presented in broad, nebulous terms ("ebay for dogs", or "online video editing").
The previous post mentions Google as an example of execution over ideas, but the idea for Google was not "search engine", but "search engine that uses the following algorithm ..." And there was good reason to keep details secret.
Design concepts are ideas. Plans of attack are ideas. There are lots of ideas based on specific data and current opportunities, and these are the ones people keep quiet about.
When people say ideas are worthless, they mostly seem to be referring to worthless ideas (e.g. "ebay for dogs", which is more of an idea for an idea ), not the ideas behind a tangible implementation or the ideas for successful marketing.