2 ms·
>As an initial experiment, we will fund about 5 companies. This is the most telling line in the whole thing. Why? Well, for n=5 to be a useful experiment, the
by whatshisface 6y ago
>As an initial experiment, we will fund about 5 companies.
This is the most telling line in the whole thing. Why? Well, for n=5 to be a useful experiment, the success rate would have to be... way higher than 1 in 100 or even 1 in 10! They're practically asking for people who want to open fast food franchise locations.
- simonebrunozzi 6y agoOr maybe they use these 5 to experiment and iterate on the program, and then fund another 15-20 in 6 months from now, and by then they'll have a nice batch of 20-25 companies, hoping that one of these will become a decacorn or more.
- alextheparrot 6y agoThis assumes each bet has a binary goal state [Success, Not Success] when probing endeavours often are more about manifold mapping rather than optimal sub-topology mining.