Not entirely relevant to the main issues of lesswrong, but possibly still of interest: my talk entitled "von Neumann probes and Dyson spheres: what exploratory engineering can tell us about the Fermi paradox".

Abstract:  The Fermi paradox is the contrast between the high estimate of the likelihood of extraterritorial civilizations, and the lack of visible evidence of them. But what sort of evidence should we expect to see? This is what exploratory engineering can tell us, giving us estimates of what kind of cosmic structures are plausibly constructable by advanced civilizations, and what traces they would leave. Based on our current knowledge, it seems that it would be easy for such a civilization to rapidly occupy vast swathes of the universe in a visible fashion. There are game-theoretic reasons to suppose that they would do so. This leads to a worsening of the Fermi paradox, reducing the likelihood of "advanced but unseen" civilizations, even in other galaxies.

The slides from the talk can be found here (thanks, Luke!).

New Comment
21 comments, sorted by Click to highlight new comments since: Today at 3:42 PM

Video is too slow. Would you share your slides with us?

Why don't you use programs which increase the playback speed? You can still understand most videos if you double the speed. Saves a lot of time!

I did spend 25 minutes watching this video at 2x and I think it was a poor decision compared to reading the slides in 5 or 10 minutes.

Also, it's a big hassle to download videos from youtube, because youtube doesn't want you to do it and plays cat and mouse games. I used youtube's own beta html5 player, which lets me do 2x playback (in mac/safari, but not in mac/chrome wtf). But youtube throttles download speed to normal playback speed, so I had to wait 25 minutes before watching. (Other ways of getting videos do not throttle speed, but back when I downloaded them as files, I usually couldn't get them fast.)

You're right, reading the slides is probably enough for this video but I guess there are some videos which you want to (or have to) watch but are too slow.

I myself use MySpeed. It costs 29$ but you don't have to download the video and can change the playback speed from 0.3 to 3.0. There are probably equivalent programs for free, but my google-fu wasn't strong enough to find them.

I myself use MySpeed. It costs 29$ but you don't have to download the video and can change the playback speed from 0.3 to 3.0. There are probably equivalent programs for free, but my google-fu wasn't strong enough to find them.

You can use mplayer or vlc to stream youtube videos and they have the ability to change playback speed.

Anywhere on LW I can upload PowerPoint files?

The file is gone now. Please re-upload it?

BTW, the LW Wiki has an upload function, but someone needs to change the configurations to allow PowerPoint file type (and perhaps PDF while they're at it).

I'm feeling great about turning on PDF uploads, but allowing PPT uploads includes the ability to upload Office macro malware.

Do we want to allow random people to sign up for accounts then upload potentially malware bearing files?

I'm feeling great about turning on PDF uploads

Please do! This feature could have been (and could) be useful for me at times.

I don't think PPTs are much more dangerous than PDFs since PowerPoint disables macros by default, and PDFs can contain malware as well (that exploit unpatched holes in Acrobat software). And people who want to spread malware can always upload those files elsewhere and then link to them in the wiki or comments.

I put the file back, and contacted Matt (TrikeApps) about the upload function, which I didn't know existed.

Dropbox is good for this.

Alas, this feature hasn't been implemented yet, but if you email me the file I will host it and link it from here.

I never really got the Fermi paradox. There are so many unknowns about "what it takes to reach intelligent life" that I don't think we really can give any estimate - and not seeing any trace of intelligent alien life just should encourage us to consider the "rare earth hypothesis" (that there are some very specific settings regarding Earth, like having a big moon, which are required for intelligent life to evolve, and are very rare in the universe). There are so many things we can forget to consider about Earth (like, the fact it's far from the center of the galaxy, so not exposed frequently to nova) making it special, and so many unknown about the details of evolution of rational life, that it doesn't strike me as impossible that there is in average 1 intelligent life form emerging between 10 billions of years and 15 billions of years after the big bang in a sphere of 5 billions light-years wide, leading to us being alone in our past cone.

Life arose on Earth sometime in the first few hundred million years after the young planet had cooled to the point that it could support water-based organisms on its surface. The early emergence of life on Earth has been taken as evidence that the probability of abiogenesis is high, if starting from young-Earth-like conditions. We revisit this argument quantitatively in a Bayesian statistical framework. By constructing a simple model of the probability of abiogenesis, we calculate a Bayesian estimate of its posterior probability, given the data that life emerged fairly early in Earth's history and that, billions of years later, curious creatures noted this fact and considered its implications. We find that, given only this very limited empirical information, the choice of Bayesian prior for the abiogenesis probability parameter has a dominant influence on the computed posterior probability. Although terrestrial life's early emergence provides evidence that life might be common in the Universe if early-Earth-like conditions are, the evidence is inconclusive and indeed is consistent with an arbitrarily low intrinsic probability of abiogenesis for plausible uninformative priors. Finding a single case of life arising independently of our lineage (on Earth, elsewhere in the Solar System, or on an extrasolar planet) would provide much stronger evidence that abiogenesis is not extremely rare in the Universe.

Interesting, but I was more thinking in the "simple life" => "intelligent life" step than about abiogenesis. As you said, we don't have any strong evidence about the likelihood of abiogenesis, we only have two weak ones : the fact it happened early on Earth, and the Miller's experiment which shows that under right conditions, build blocks of life can arise quickly.

But for the "simple life" => "intelligent life" step, we have even much less evidence. We know it took a long time on Earth, about 4 billions of years. But are we specially lucky (due to the conditions on Earth, and due to random events) ? We just don't know.

There seems to be many evidence pointing to it not being so easy. Consider a parameter like isolation. If your star is too isolated, it won't get enough heavy elements to sustain life. Heavy elements like oxygen, carbon (and worse for iron, calcium, ...) were not created in significant quantities during the big bang. They were created later on in the heaviest (hottest) stars and in supernovae. So your star needs to be close enough to other stars (inside a dense enough galaxy) so it gets those elements. But if there are too many stars nearby, then some will go nova close to your planet before it had time to develop intelligent life, and will wipe all life (or at least, all complex life). So there seems to be a very narrow range of amount of nearby stars for intelligent to be able to develop itself.

The same goes for the mutation rate (dependant on natural radioactivity, strength of magnetic shield, ...). If the mutation rate is too slow, evolution will take longer, and you won't reach intelligent life before your main star goes giant red or nova. But if the mutation rate is too high, complex DNA will be unstable, and complex life won't evolve.

The same goes for so many parameters. For each of them, we don't know much about the acceptable range, nor how likely we are to find it. So... better admit we really don't know about the odds on simple life to evolve into intelligent life, nor about the average time it would take. Not finding any intelligent life is, to me, strong evidence that it takes time and it requires rare conditions.

I think mutation rate would be subject to evolution as well. We can evolve to do more, or less DNA repair (and checks; up to acute radiation poisoning, cell self destruct is good enough).

Also, the effects of radioactivity are presumed linear. Extrapolating from high doses, the doses of up to 10x our natural background must have extremely minor effect on the mutation rate, vast majority of mutations arising due to other causes.

(The radiation releases in nuclear accidents are only a big deal because very huge number of people is exposed, resulting in massive dis-utility even if cancer rate increase is very minor)

Interesting, but

I was just continuing your attack on the common assumptions for the values involved with the Drake equation.

You were quite brilliant, I agree with almost everything you said on the video.

Your public was annoying.

What do the assumptions presented in the video look like with the additional (immediately less likely, I know) assumption that intelligent life is common?

Is it possible that it'd look something like this (the box being the universe and each ellipsoid being the expansion of probes from one civilization)? Basically, the probes from each civilization would expand out and eventually meet other civilization's probes and continually war to maintain territory.

Also, is it more likely that there isn't intelligent life given your assumptions or that we're just among the first intelligent species in the universe and we shouldn't expect to see anything surprising?

New to LessWrong?