4 ms·

Having experienced American bureaucracy, I find the claim that US institutions have been "willing to investigate systemic dysfunction" completely hilarious. I

by t8sr 4y ago

Having experienced American bureaucracy, I find the claim that US institutions have been "willing to investigate systemic dysfunction" completely hilarious. I think what Palantir means is that American organizations have been faster to adopt big data and centralize decision-making. Having worked for multiple large US businesses, I am not convinced that this leads to "less dysfunction". Having a few technocrats with some data make all the decisions centrally is just a reinvention of Planned Economy. It didn't work in the USSR and it doesn't work now.

toxik 4y ago
A plan economy works for smaller organizations, it is very much the norm.
htag 4y ago
From my experience the most common use of Big Data is to 1. Provide confirmation bias for decision makers to feel more confident and comfortable. This happens in the translation layer from Data to Information, and the data can support several possible points of view depending on what questions are asked of it. 2. Provide small adjustments to someone's existing plan. Maybe someone was planning on building fifty connivance stores in a region, but they found data that local market can actually support eighty so they use the data to justify increasing their budget. Maybe they were planning on selling five types of pepsi in their store, but the data shows they should sell six. This type of adjustment is usually used as an excuse to increase capital expenditure, or provide guidance on how to allocate existing resources. 3. (Most rare) Data is used to provide novel insights that give a large compeitive advantage. This is the "Money Ball" and "Big Short" sort of situation. One thing America is very good at is to provide a compeitive marketplace where a smaller participant has a shot if they are good enough. It's exceedingly rare to see an incumbent use big data in a novel way.
- 411111111111111 4y ago
  Its kinda within the definition of Big Data, which is that you have too much data to really inspect all of it. this means that the previous approach had to be inverted: now its necessary to propose a theory, define queries to validate it and then evaluate the results, whereas before you'd take everything in and define a theory from the data set. the new approach is pretty much a textbook definition of a confirmation bias. But there is no other way to really approach these vast amounts of data, so its become the default.

by t8sr 4y ago

toxik 4y ago

A plan economy works for smaller organizations, it is very much the norm.

htag 4y ago

From my experience the most common use of Big Data is to 1. Provide confirmation bias for decision makers to feel more confident and comfortable. This happens in the translation layer from Data to Information, and the data can support several possible points of view depending on what questions are asked of it. 2. Provide small adjustments to someone's existing plan. Maybe someone was planning on building fifty connivance stores in a region, but they found data that local market can actually support eighty so they use the data to justify increasing their budget. Maybe they were planning on selling five types of pepsi in their store, but the data shows they should sell six. This type of adjustment is usually used as an excuse to increase capital expenditure, or provide guidance on how to allocate existing resources. 3. (Most rare) Data is used to provide novel insights that give a large compeitive advantage. This is the "Money Ball" and "Big Short" sort of situation. One thing America is very good at is to provide a compeitive marketplace where a smaller participant has a shot if they are good enough. It's exceedingly rare to see an incumbent use big data in a novel way.

411111111111111 4y ago
Its kinda within the definition of Big Data, which is that you have too much data to really inspect all of it. this means that the previous approach had to be inverted: now its necessary to propose a theory, define queries to validate it and then evaluate the results, whereas before you'd take everything in and define a theory from the data set. the new approach is pretty much a textbook definition of a confirmation bias. But there is no other way to really approach these vast amounts of data, so its become the default.