5) I will have to think about how this works more carefully, since I'm not sure I understand how this would work.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Okay. My ebook on Amazon will help a lot in this effort, it is far more detailed than my FQXi offering.
6) It will be interesting to see who's right. I agree that in some ways technology develops exponentially--if that's what you're suggesting--but I also think that qualitative advances become exponentially harder to achieve in some areas. Projects that require infrastructure or use high energies can be particularly slow. I am afraid at this point we may not even have controlled fusion in 20 years.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
As always, time will tell.
7) You seem to be using distinguishing prediction from foreknowledge in a technical way that wasn't clear to me when I first read your essay. But in any case this is where I think I most strongly disagree with you. A viewing technology that is sometimes wrong can still be tremendously useful. It won't be useful--and may even be harmful--in the specific case where it is wrong. But on balance--if the error rate is low and roughly known--it can be very valuable. Consider diagnostic tests. These routinely produce both false negatives and false positives. But if the error rate is low enough they are still an invaluable tool of medicine. Our actual telescopes inevitably distort images, but nevertheless remain useful to us. I doubt that any machine can run without producing errors. We should certainly seek to minimize observational errors. But we can live with some. In fact, I think we have to.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This is a topic concerning which I sense we are approaching an accord. I very much appreciate your exceptionally clear thoughts and questions.
You raise a vital point above, that if a future-viewing machine were to show an incorrect future, the result could be harmful. This is very true and important. Now, if viewer foreknowledge is to be a category of foreknowledge at all, it must be distinguished by being definite and 100% correct. This is because, a future viewer does not generate any kind of prediction or approximation, it provides a direct image. Now, whatever a future viewer may image is 100% whatever it happens to be (in its own time). So, how could an image of that thing depict something else, even something which only differs from the original by 1%?
Here we come to an issue that will serve to clarify. Viewer foreknowledge of an event could be repeatedly received from all angles, and from very close up. One should be able to count the number of pores on someone's nose in viewer foreknowledge--its resolution could even potentially be extended much more. How could such a kind of technology sometimes be wrong? If viewer foreknowledge were ever wrong, then it would have to be systematically wrong. However, how could it be systematically wrong, if it is just a direct image that can be tremendously zoomed in according to the operator's whim?
You mention diagnostic tests. Yet, I think you now see that what I am proposing is not a test of the future: viewer foreknowledge is full access to the future, as many times as one wishes to see, and with enough detail to satisfy anyone that it is genuine. Yes, it may be a fantasy, but this is the kind of technology I have in mind.
As before, Robert, thanks for your insights and questions. I'd like to know how your thoughts are evolving on these topics when you find the time. All the best!
Aaron