3 ms·
A bit too generalized for my taste. If you're solving a problem of passing a well-defined benchmark (or closer to home, iterating on a kitten meme-sharing app),
by twelve40 12y ago
A bit too generalized for my taste. If you're solving a problem of passing a well-defined benchmark (or closer to home, iterating on a kitten meme-sharing app), then yes, fail away. If you are solving a problem of putting a man on the moon in the 60s, then you get to coordinate hundreds of thousands of people from education to fundamental science to industry to alloys to god knows how many other things for years. And only then you can launch and really find out if you succeeded or failed - both outcomes have precedents.
- jtbigwoo 12y agoI think the problem is that we tend to treat everything as a moon-landing. There are very, very few times when we should put all our eggs in one basket yet we do it all the time.
- kabouseng 12y agoWell maybe the lesson is to always think how you can verify your solution quicker. To take your example, that is the same line of reasoning Boeing is taking as a subcontractor for NASA, but where SpaceX has shown that they can verify their solution for half the cost that Boeing can[1]. [1] http://qz.com/281619/what-it-took-for-elon-musks-spacex-to-disrupt-boeing-leapfrog-nasa-and-become-a-serious-space-company/ http://qz.com/281619/what-it-took-for-elon-musks-spacex-to-d...