3 ms·
You can write a model for things you can't measure if you like, but how can you know the accuracy of it or if it can predict future (impossible) measurements. I
by andy_ppp 8y ago
You can write a model for things you can't measure if you like, but how can you know the accuracy of it or if it can predict future (impossible) measurements. It's almost nonsensical to try, unless some alien civilisation comes along with the data I don't see the point.
- minor3rd 8y agoThe title is also clickbaity, since the conclusion is really just that "we can't deduce much of anything... yet".
- nautilus12 8y agoModels like these only work when considering the order of magnitudes as a feasibility assessment. If you can determine a component reasonably has such a low probability that it pulls the entire thing down to infinitesimal probability then that observation is useful. However if the bounds of accuracy are too wide its not useful, this would be more useful with bounds.
- themodelplumber 8y agoYou can measure the logic of using individual components of the model, and the combination of those components, by exposing them to social scrutiny. That's all measurements are: An agreed upon social standard based on shared subjective experiences. Plus, if any part of the model can be abstracted, there's a good chance you'll find leverage points to apply to other disciplines or problems. There's also a socio-political aspect: We are an emotion-driven species. Sharing this kind of stabbing-in-the-dark information may actually drive the advancement of far less speculative technology that could be beneficial to us all.