SETSIs (was Re: seti@home WILL NOT WORK)

Robert J. Bradbury (
Thu, 8 Jul 1999 01:47 PDT

> Brian Atkins <> wrote:

> > "Robert J. Bradbury" wrote:
> that much more justifiable), but at least RC5 is guaranteed
> to produce a result (if it gets enough computers/time), while
> SETI@HOME never will.

> Wrong. You actually have no way of knowing whether the SETI
> effort will succeed or not; this is just your not-so-humble
> opinion.

Of course, all of this is opinion. I can't even *prove* that I'm not an alien nano-clone. :-)

> As for RC5, THAT is a complete waste of time. We
> already all agree on what currently makes for a "secure"
> level of encryption.

At least until they get an operational quantum computer (which is probably 20 years out if it is even doable).

Question: Does anyone know if there is an encryption methodology that will work if QC cracks the factoring problem?

> Wasting your CPU time on cracking yet
> another stupid RSA message is the height of boredom.

Well, I'll agree that the results aren't going to be either surprising or disappointing.

> Well the fact of the matter is: you can't really say for sure
> what a SI will want to do with its time. You can blather on
> all day, but it's all a guess.

I can't "say for sure" that we won't get blown to bits by a near-earth crossing asteroid tomorrow morning. All I can do is argue that it is very unprobable and construct my life around that premise. The people doing SETI@home, IMHO, are doing a lot of work with a poor apreciation for the very low probability of success. Which to my mind is sad because with a little work we could move forward on multiple fronts instead of only one.

> In fact you can't even be
> completely sure that there are even any SIs in this universe.

True, but I can sift a body of evidence and create a reasonable argument that the number of SIs hugely outweighs the number of nearby civilizations at our level.

Some of this evidence would include:
(a) 90% of the mass in the galaxy is "missing". (b) The gravitational microlensing researchers have done simulations

      based on their *observed* frequency of gravitational microlensing
      events and postulate there are *400 billion* objects with
      masses averaging 0.3 Solar masses they *can't see* orbiting
      our galaxy.  That is about 2x the number of stars that we can see.
  (c) Astronomers looking at other galaxies don't see the number
      of Red giant stars they expect.  In some cases there seems
      to be evidence of an IR "glow" (which they attribute to dust).
  (d) Kardashev and a couple of other Russian astronomers have gone
      through the IRAS data (the best infrared survey data we have
      so far) and have discovered some interesting anomalies in
      the directional distribution of "cold" objects around our galaxy.
  (e) The stars in the galactic halo in globular clusters are very
      low in metal (meaning they are very old, perhaps older than
      the best estimates of the age of the universe -- or that they
      may be the leftovers from SI metal extracting mining operations).
  (f) M-brain architectures (layered supercomputers surrounding stars)
      seem to be optimal computing architectures for many problems
      and have heat emission signatures that are very difficult to
      detect using current technologies.

So taking Ockham's Razor, which of the following would you choose: (1) That the longest lived "entities" in the galaxy will be

      technological civilizations that have evolved to the limits
      allowed by *known* physics (which as far as I can tell are
      something like M-brains).
  (2) That the galactic halo consists of strange "concentrations"
      of brown dwarfs orbiting small black holes (one proposed
      explanation using "known" astronomical objects without any
      theory as to how such things came to be formed).
  (3) That the galactic halo is composed of mysterious "new" particles
      (WIMPs) that explain at least (a) and (b) above?

Since SETI has been unsuccsessful so far, people have argued for lowering the frequencies for some of the Drake Eqn parameters. Some prefer to mess with f_l (evolving life), some with f_i (intelligent life), or f_c (communicative life) but then you always end up arguing that we shouldn't exist and you are bucking the trends in current findings in astronomy & biology.

As we ourselves provide evidence that most of the parameters in the Drake equation lead to the development of intelligent, communicative "technological civilizations" (TCs) and doing *any* SETI work assumes there are at least 2 TCs, then, one is really concerned with the final "L" parameter of the equation: "the mean Lifetime TC's are in a "communicative state".

Given the FERMI Paradox ("I see no aliens") and the lack of SETI success lots of handwaving has to be done to explain why TC's pass out of a communicative state.

So lets examine this with the "UFE" (Universal Fatality Equation): Pa = the probability a civilization is knocked out by an Accident

(asteroid, comet, blazar, supernova, etc.) and stays knocked out Pd = the probability that a civilization De-evolves or decays away

       (i.e. once you have evolved survival and preservation instincts
        and intelligence, somehow you "lose" them; this might include
        losing an interest in space.)
  Pn = the probability of *No* nanotechnology (even after thousands of
       years of R&D)
  Ps = the probability of permanent Stagnation (either environmentally-
       or self-imposed) at some "level of growth" beneath the limits
       imposed by physics (your material/energy limits)

Now, if Pa + Pd + Pn + Ps = 1 for *all* technological civilizations that have ever evolved (in billions of years of galactic history), then:

No technological civilization in our galaxy has ever evolved to the limits imposed by known physics -- i.e. No SIs.

This is a the inversion of (1) above. There are some interesting interactions in this equation, because as you reduce Pa or Pd, you probably increase Pn which in turn may reduce Ps. Of course one can argue that an increased Pn causing a singularity of the grey goo type increases Pa, but you have to argue it for all civilizations.

[I'm open to suggestions on how to improve the UFE.]

Unless you argue a very low level of TCs (which makes SETI@home kind of silly), or for the universal application of the UFE, then I don't think you can escape the evolution of SIs. If even a single SI evolved more than ~100,000 years ago, then you have to accept that our galaxy should theoretically be at Kardashev Type III Civilization level (i.e. extensively developed). Sagan and other showed many years ago that even using "current-level" technologies, interstellar travel is feasible. That means that whole-galaxy colonization is possible and would occur in short time periods (millions of years). Nanoprobe colonization managed by SIs makes it happen much faster.

Why hasn't everything been colonized? If you take (a-e) above, it would appear that most of it *has* been colonized. Whether SIs do in fact colonize (or whether many or all SIs evolve locally) and whether they fight over resources remain interesting questions.

Why do we still see stars? My best explanations to date is that the central part of the galaxy isn't a very safe place to be (black holes, blazars, supernovas, etc.) and its simply not worth the time and energy trying to keep the garden weeded. The other possibility is along the lines that Greg has proposed which kind of evolves from John Ball's "Zoo hypothesis", which is that perhaps the centers of galaxies (with all of their natural "hazards") are used for cultural/meme breeding grounds. Or perhaps after billions of years of existence and consuming a majority of the available matter/energy SIs have just lost interest in it. Or it may be the least expensive way to breed an increasing supply of materials that can be used to store their prodigious memories.

So lets fit the observationional data to the theories --

- They are here watching everything we say and do (because nanotechnology seems within the laws of physics *and* nanotechnology allows this level of "spying" undetected by us. Since they haven't "contacted" us it isn't because they aren't able, it is because they have reasons for not doing so. or
- They aren't here and are simply content to watch us from afar. (Because they certainly have the observing capacity to continuously monitor the development of our planet.) or
- They are completely disinterested in us -- we are beneath the nematodes.

Now, it would appear that the development time for a technological civilization from the pre-radio stage to the post-singularity stage is ~100 years on a planet similar to ours. In the post-singularity stage, you don't communicate with "broadcast" radio-waves -- you communicate with tightly focused high power lasers aimed directly at the person you want to talk with (presumably another SI) because you then have the technology to find them and directed optical communication provides the highest bandwidth at the lowest energy cost. It is worth noting that this idea is slowly catching on in the SETI community and there have been several conferences devoted to Optical SETI.

So, for the SETI@home project to work, as has been suggested, by "evesdropping" on similar level civilizations, they would have to be within ~100 light years of us. If they are beyond that their signals would have already passed us by. If you plug the number of Earth-like planet supporting stars in 100 cubic light years into the Drake equation (rather than the number of stars in the galaxy) then your probability of success gets very small. It is also true that the calculations on the power levels required (for a continuous omnidirectional signal) require a level of civilization that we won't reach for some 10's of years. So the detection "non-intentional" signals from a civilization like ours is limited to something like 10-20 light-years, further lowering your Drake Equation probability. Once you get to the level of civilization where you can send a high-power signal omnidirectionally (or have sufficiently good telescopes that you can see planets where someone might be listening for a low power directional signal) you are very near the singularity boundary. For us, NASA may get the "Terrestrial Planet Finder" spacecraft up around 2010 *iff* they stay on the current proposed schedule. So if we use our own civilization as a benchmark, we might have the power and/or know where to send a signal by perhaps 2020. That looks to be to be pretty close to singularity time.

I will acknowledge that using our civilization as a predictor for development paths of other civilizations is a really poor statistical exercise, but its the best we can do at this point.

So, when I go through all this stuff carefully, I reach the conclusion that SETI@home will not work and people have just been hoodwinked. Now, this is being done by people who don't know about nanotech, don't really understand computer (and therefore SI) architectures, and who would be shocked by concepts such as "uploading". I'm not against them, I plan to educate them and perhaps if I'm lucky change a few minds.

However, this is the Extropian forum, so I feel pretty comfortable telling the people here (who are aware of the afore mentioned concepts), that you are being really silly if you think this will work.

> So you are saying there is nothing currently available better
> than SETI@Home to spend your time on? Or did I misunderstand.
No, the 2MASS data is available. They have released part of it in a very compressed form. However, you would have to write the application yourself to compute the black body radiation curves from the data (not very difficult I believe).

The question is whether you have "thinking" time to spend on it or just "computer" time. If you have just computer time there isn't any game other than SETI@home. But to my mind you should put the SETI@home computers in suspend mode and donate the value of the electricity saved to an organization that can pay someone to write a competing application to do an analysis of the 2MASS data. That is because the 2MASS data *will* produce interesting things, though due to the frequencies they use in ground observations, I would say the probability for SIs, unless they are nearby, is low. Though not as low as SETI@home because there are potentially many more long-lived SIs than nearby life-supporting planets. in their radio-communications stage.

After they start releasing the Sloan Digital Sky Survey data in a couple of years the Sloan data and the 2MASS data combined will give you I think 8 data points to compute the temperature of each object. Then there will be plenty of work for all the computers we can lay our hands on. We will still however be limited because to get the frequencies where the cold SIs are the brightest (and therefore most detectable at a distance) requires a space-based survey. NASA and ESA are working on such spacecraft but I don't think they go up until after 2005 and even then I suspect the data will be difficult for us to get for a while.]

> > software to process the 2MASS data and then we would
> > have a crack at finding those darn SuperIntelligences.
> Why that's a great idea! Instead of spending your time dissing
> everything, you could actually take some of the advances the
> SETI people have made in distributed computing and use it
> for something else!

First you have to get the SETI people to see the possibility, and since they have been at this for nearly 35 years in the face of virtually no results, it is going to take some *strong* arguments to shake their "faith".

I'm only at the beginning of this road and didn't realize that the 2MASS data was available until about a week ago. I strongly agree that the distributed computing model is a great way to go for astronomy research as well as promoting interest in space. What I'm dissing on is the fact that the SETI people gloss over some of the details (like the transmitter power requirements) that make people think the probability for aliens that we can communicate with is >> 1 instead of << 1. You never hear about these things unless you delve into the literature in some detail.

> Let me know when you
> have that software for the 2MASS stuff ready. Ok? Ok.

It will be September at least before I have a chance to talk to people. What I'm going to try to do (given this discussion) is see if I can convince the SETI@home people to talk to the 2MASS people.

> > 6th cousin, twice removed from the
> > S.F. Author "Raymond Douglas Bradbury"
> What, is that a name drop or what?

I'm going to have to add more smileys for you, thats for sure.... People often acuse me of going way out on a limb, so I like to provide them with a genetic explantion for it. :-)

Ah noooo, its my genes doing it to me again... Conscious evolution will never work...