On 1/2/06, <b class="gmail_sendername">Jeff Medina</b> <<a href="mailto:analyticphilosophy@gmail.com">analyticphilosophy@gmail.com</a>> wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
You have absolutely no idea whether the computing power "just barely<br>available" to the simulator is anywhere near (as opposed to notably<br>greater than) the amount "just barely adequate for a ground level
<br>simulation of the visible universe". Why are you making such a strange<br>assumption?</blockquote><div><br>
Of course it's a strange assumption, which is why I not only don't make
it but have been arguing against it the whole time. Remember that
the original argument was that a certain experiment may have detected a
slight anisotropy of space or something similar (this seems unlikely in
light of further discussion, but it can be taken as a premise) and that
this implied one's estimate of the likelihood of the simulation
hypothesis should be adjusted upward. Why would it be adjusted upward?
Everyone understands the implied logic: a simulation is short of
computing power and might skimp just enough to let the imperfection
through. I'm arguing that we should _not_ a priori expect the available
computing power to be anywhere near (in either direction), and
therefore the original argument is _not_ valid; the originally
described results do not give us any reason to adjust the estimated
probably of the simulation hypothesis either up or down.<br>
</div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Aside from that, it's also not necessarily unlikely that those values<br>would be near one another. If I were trying to create an interesting
<br>simulation-world, I just might use all of the computing power<br>available to me. Then Russell-in-the-world-I-made would say, "It sure<br>is unlikely that my visible universe's required computing power<br>matches the computing power available to the simulator." And
<br>Sim-Russell would be wrong.<br>
</blockquote></div><br>
Not really, because there'd be a zillion googol Sim-Russells simulated
more efficiently, so the statement would be correct to the tune of a
zillion googol to one.<br>
<br>
- Russell<br>