Nobody’s going to steal your idea

Except for Zuck, he'll steal the shit outta your Hipster Coffee Shop Wifi app.

sure - but signing a generic NDA on a cheap piece of copy paper doesn't cost much...so w/e, if it makes someone feel better then don't worry about it.

The attitude that nobody's going to steal your idea might hold true for a certain audience, but having worked in environments where not only will people steal your idea, but often relationships are formed for that very purpose, I would take that advice with a few grains of salt.

As with everything, context is key -- if you're talking about building a new widget for platform_x, and you're in a conversation with your grandparents, you're probably in the clear (for a specific definition of grandparents).

If you're having that same conversation with the world's most prolific exporter of widgets for platform x, it might be better to listen than to talk.

In contracting environments, specifically in Federal government, the IT ecosystem is an adversarial one, all the work is bid upon, subsequently awarded to a Prime contractor who may or may not sub out some of the work to sub-contractors. It's not impossible for some of the subcontractors to have their wagons hitched to multiple ponies, on the assumption that they'll work with whomever wins.

Idea sharing and NDAs are critical in this sort of environment as the novelty / uniqueness / effectiveness of your particular solution to the problem being bid upon is your key differentiator. Otherwise, you likely end up in a price race, where either the largest contractor or the most ill-adept at bidding will win.

Not always true. We were working on a smartphone game with an original concept and kept blogging about it. Including telling about the idea and showing early gameplay video. It might have been a coincidence, but a virtually identical game appeared before we were finished and became quite successful on Android. So next time I'll be a little bit more careful.

There's lots of reasons for NDAs besides stealing ideas. I doubt Apple is afraid I'd steal the next design for their new phone (although they may worry Samsung might), but they still probably don't want me blogging about it. There's marketing impact.

Also companies often don't like to commit to product before it's committed. I learned an important lesson at one of my first startups that customers will try to hold you to your "proposed" roadmap. Even though we say, "We're just trying to be transparent. This WILL change. We don't know of this we can do." -- people will tend to find things they like and essentially view what you showed them as a promise to deliver. You suddenly have customers who claim you not only didn't deliver a feature, but that you broke a promise. Until I know I can ship, I really prefer not to talk about it publicly -- even with tons of caveats.