On Wed, Jul 30, 2014 at 8:36 PM, Russell Standish <li...@hpcoders.com.au> wrote:
> On Wed, Jul 30, 2014 at 03:13:01PM +0200, Platonist Guitar Cowboy wrote: > > > > Better than to keep reacting like monkey to his post, which I do, would > be > > imho for the list to focus on precise prerequisites, exercises, reading > > lists to get firmer grasp of AUDA, and make that more universally > > accessible than perhaps overemphasizing Step 7 and MGA. > > > > I don't think step 7 and the MGA have been overemphasised at all - > although step 3 has been done to death it seems. > > The MGA, in particular, is very subtle. I think I have now found a way > of coming to grips with it, and it is a little different to the usual > precis given here. Fundamentally, computational supervenience (which > requires counterfactual correctness, a factor often overlooked in the > summaries) is incompatible with physical supervenience, but only in a > single universe, not a multiverse. Since physical supervenience is > well supported empirically (something JC has banged on a bit about, as > well as Brent), "Well supported empirically" is ok here, but I'd say that majority of scientists take it to be more than that. It's often taken taken for granted without awareness of the potential problems this possibly implies... which is among the reasons the topic remains a hot little tempest in a tea pot. > the only way of rescuing computational supervenience > is to insist that we do, in fact, live in a multiverse. > But that multiverse > entails the universal dovetailer, or at least enough of it to emulate > consciousness, which suffices to get the reversal result of step 7. > > I'm in process of writing this up as a paper, which I'll hopefully > post to this list in the next week or so to be shredded by the > denizens here. > Looking forward to it :-) But you don't need to already cultivate reception response... There was a composer I can't recall atm who would subtitle his early work with "Misunderstood by critics" to maintain the position of "meh, not surprised when I cast my pearls before swine". You could clarify yours with "as shredded by the denizens of Everything list", which is a higher achievement than say Obama's Nobel peace prize, which you could also throw in the title. > > On another note, I've done a calculation of the observer moment > measure given by the universal dovetailer, which shows that the > measure is indeed independent of the chosen reference machine, as we > widely suspected, and hopefully might be comparable with > Solomonoff-Levin's universal prior. More likely, it might be compared > with the Bayesian probability after an infinite number of updates. > Wow, good effort in any case and look forward to that as well! > > Not sure if it belongs in the above paper, as it is necessarily more > technical, but I'll post that to this list in one form or other soon. > > But I'm happy for there to me more discussion of the AUDA too :) The > underpinnings of that theory still seem rather weakly motivated for > me, as opposed to being mandated by computationalism. > Once you take as given the reversal of step 7, it's hard for me to see why/how we can ignore interviewing the relevant machines. Especially in view of the striking correspondences that this seems to point towards up to this early point. I agree on more discussion though to better make up my mind. PGC -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
