John Larson gives several good reasons why he won't sign NDAs:
Between a first-time web entrepreneur and one who's been for years working on many ventures, there is a huge gap in perspective regarding the importance, rarity, and uniqueness of ideas.
... if you've ever tried to bring even one venture to market, you know perhaps all too well that ideas are just the starting point, and take by far the least work, time, and capital.
If you prize your idea so much (in relation to everything else it will take in order to make it succeed) that you feel the need to put in legal protections from me, it's a tell that you don't have much going for you in this endeavor.
Any agreement that I sign to not disclose or use information shared with me in a casual engagement opens up a whole world of potentially contentious confusion about what is or isn't okay for me to do in the future.
John mentions that he will sign an NDA if there is at least a 10-page business plan being disclosed. That's one possible cutoff, but I feel this is still being too generous.
At GrantTree, we are often asked to sign NDAs by companies that are looking for grants. We very rarely do. The principle is simple: we only sign an NDA when there is a very high likelihood that we'll be working together.
Put it in other words, we will happily sign NDAs with clients - but not with prospective clients. If a company can't disclose enough to us for us to make a decision about whether to take them on as a client, then we simply all have to accept that we won't be working together.
On the other hand, when we work with a client, we often get a lot of internal info from that company, including their accounts, financial position, investment documents, and even some of their strategic plans. This, rather than "the idea", is what's being protected by an NDA - and rightly so.
If you read this far, you should follow me on twitter here.