That works for some things, but not the others. For instance, right now I can almost mathematically prove that we're building the wrong thing, starting from 2 very "common sense" and reasonably easy-to-believe "axioms". I have no trouble convincing fellow engineers that we're not building the right thing, but that's more harmful than useful. I've managed to convince some director-level people, but trying to bring this up to VP-level yielded clear warning signs that not only will I achieve nothing, but it's going to be a career limiting move. The Upton Sinclair quote is quite true - "It is difficult to get a man to understand something, when his salary depends on his not understanding it."
In the meanwhile, my company spent in excess of two years, and increasingly more resources trying to build the same thing and being not an inch nearer to their true goal - I wonder how soon till the outside world starts to notice. The CEO is frustrated because the strategy is sound and he doesn't get why it's not getting done. In the meanwhile, trying to tell to the VP that the tactics is wrong (and how it's wrong, and how to fix it) gets you the axe. It's messed up, but all I can do right now is say "it's not my job to fix this mess" and play along with it/ learn interesting technologies & cash my cheques. :(
The downside is when it doesn't pan out, you can have a bit of catching up to do.
That is so true as I am catching up now on a Saturday. You also learn to not roll your own solutions and check a second or third time for a solution on Github or open source community.... and check again if you don't find it.
I've made the mistake recently of being too detailed an open about the joe and it's come back to bite me since we have "gone agile". I try to be as vague as possible now. The less they know the better. We do have Code reviews by the "dev lead" since most of the other developers are junior, but luckily, I am the "dev lead".
It is a balancing act. You do not want to waste company time with oddball ideas, and some of your ideas will be oddball. You need to seek feedback on them so that they can be improved. You need freedom to try them out, and fail. I think this article is good for both managers and developers. Developers should know that this pattern is real, and there are things you can do to protect yourself from going down this path. For me, its basically not always asking for permission to try stuff out... take a risk and believe in yourself, but also get feedback to filter out oddball ideas that waste time.