An idea kept private has not been tested

Jack Dorsey had carried around the idea of a service for posting short status updates for years before he described it to colleagues at Odeo in early 2006. The company was hunting for a new direction, and within weeks the idea was a working prototype. Kept to himself, it would have stayed a note in a sketchbook.

Michael Lopp, a Silicon Valley engineering manager who writes as Rands on his blog Rands in Repose, puts the principle in one line. The author of an idea is the worst judge of it. Inside a single head, an idea faces no competition, costs nothing and has no audience, so it always looks promising. Describing it to someone else is the first real test. Does the listener grasp it quickly and care about it, and can they name a reason it will fail that the author missed?

It helps to be precise about what sharing proves. A friend’s enthusiasm proves very little. A colleague who finds a flaw proves something. A stranger who asks when they can use it proves more, and one who offers to pay proves most of all. Feedback is worth roughly as much as the listener has to lose by being polite.

The usual objection is theft. In practice it is rare. Most capable people already have more ideas of their own than time to pursue them, and the idea is the cheap part; the work of building it is where the value sits. Much of what Lopp writes about management concerns making ideas clear enough for other people to act on, and this line points at the same skill from the other side.

There are cases where the advice is wrong. An invention that could be patented can lose that protection if it is disclosed publicly before a filing, under the rules of many countries. Negotiations, acquisitions and some trade secrets are worth more unshared. Apple builds its products in near total secrecy and does well by it, though Apple shares ideas intensely inside the building.

Timing and audience also matter. An idea described too early to a hostile committee can be killed before it has any shape. Some ideas need a few weeks of private work, enough to become something a listener can react to instead of a sentence they can dismiss. The line is best read as a push toward sharing with people who will engage, not as an instruction to announce everything to everyone.

A workable habit follows. Write the idea as a single paragraph, send it to a few people likely to disagree, and pay most attention to the replies that are uncomfortable to read.