3 ms·
You know this might be the perfect opportunity to make history. let me explain. The thing that struck me the most about the post was that there was no one thing
by todayiamme 16y ago
You know this might be the perfect opportunity to make history. let me explain. The thing that struck me the most about the post was that there was no one thing to blame and there was no one way to fix this. That's the thing.We are looking at the problem from the human frame of reference.
If someone can make a data analysis program that takes ALL of the data generated everyday and crunches it together for every conceivable variable and then uses machine learning to find common cost patterns vs. time and policy (the time and date of execution of policies and their direct impact can thus be judged). Then we might as well have magic in our hands.
This is hard data that no one can argue against (some will still manage...). If we can make something like this might as well be the first time in history that policy has been made using AI as a crutch. I hope that it won't be the last.
- yummyfajitas 16y agoThe problem is not that we can't find the answer, but that we don't want to. We have the technology to determine the effectiveness of many policy choices, and we've had it for a while - it's called a randomized trial (basically a great big A/B test). We choose not to use it - it gave us results we didn't like the last time we did. http://www.overcomingbias.com/2010/03/knowing-too-much.html http://www.overcomingbias.com/2010/03/knowing-too-much.html http://www.overcomingbias.com/2010/04/too-much-debate.html http://www.overcomingbias.com/2010/04/too-much-debate.html
- todayiamme 16y agoThan you for teaching me something so important. If you're right, and I know that you are, it explains a lot of things that baffled me as well as why people choose what the choose. Sometimes the best solution isn't the best solution. Thank you.