Sharing interests
Let 1/n be the likelihood that any given person is interested in a certain topic. Assuming that authors don't differ from the general population insofar as their interests are concerned and assuming further that they only produce works that deal with all the topics they're interested in and that all authors are equally prolific and that the likelihood that any given person is interested in some topic is 1, the likelihood that any given work is on the given topic is also 1/n. (Instead of equally prolific stochastically independently from the topic prolific would of course suffice.)
If n is a big number, the search for works of interest becomes arduous, which leads to the question how it could be facilitated.
Ironically, the way to facilitate it, is to make it more cumbersome by funnelling all works through a group of selectors whose interests also don't differ from the general population and who only select works that interest them and always select two at once and join them in a bundle and letting the general public only have access to these bundles of pairs of works.
Kind of like when you want to download a certain song but can only download a hit compilation, because, when you find a compilation on which all songs are good, you know that the compiler has the same taste that you have and that allows you to profit from his search of interesting songs by looking at his other compilations (TimeLife comes to mind).
The reason for this is that the likelihood for both works being about a certain topic, when the selector is not interested in that topic, is only (1/n)2, stochastic independence between topics presumed, whereas the likelihood of finding such a bundle is slightly higher than 1/n, i.e. 1/n+(n-1)/n3, so that only about every n-th time we found a bundle we'd be mistaken in the presumption that we found a partner in the search for works on the given topic and even if we were mistaken about such a partner, we'd soon find out and have wasted little time.
So, by not searching works directly, but relying on others to bundle works according to their interests, we lay the foundation for networks of like minded people working together sharing their findings.
If n is a big number, the search for works of interest becomes arduous, which leads to the question how it could be facilitated.
Ironically, the way to facilitate it, is to make it more cumbersome by funnelling all works through a group of selectors whose interests also don't differ from the general population and who only select works that interest them and always select two at once and join them in a bundle and letting the general public only have access to these bundles of pairs of works.
Kind of like when you want to download a certain song but can only download a hit compilation, because, when you find a compilation on which all songs are good, you know that the compiler has the same taste that you have and that allows you to profit from his search of interesting songs by looking at his other compilations (TimeLife comes to mind).
The reason for this is that the likelihood for both works being about a certain topic, when the selector is not interested in that topic, is only (1/n)2, stochastic independence between topics presumed, whereas the likelihood of finding such a bundle is slightly higher than 1/n, i.e. 1/n+(n-1)/n3, so that only about every n-th time we found a bundle we'd be mistaken in the presumption that we found a partner in the search for works on the given topic and even if we were mistaken about such a partner, we'd soon find out and have wasted little time.
So, by not searching works directly, but relying on others to bundle works according to their interests, we lay the foundation for networks of like minded people working together sharing their findings.
Labels: 35, formalisierung, gesellschaftsentwurf, gesetze, institutionen, mathematik, präsentation, wahrnehmungen, ἰδέα, φιλοσοφία
Eingetragen Am/um
16:23
