Posts by Alan M. MacRobert

1) Message boards : Technical News : Open for Business (Aug 19 2009) (Message 927754)
Posted 21 Aug 2009 by Alan M. MacRobert
Post:
Wonderful, Matt!

However, I believe it's a serious mistake to make "repeats at the same frequency" a screening criterion. Nearly any plausible transmitter will show frequency drifts due to orbital (i.e. accelerated) motion. For instance, if a large beacon transmitter were broadcasting from an 0.1 a.u. orbit around a sunlike star (to collect lots of solar energy, and to keep the transmitter away from the home planet), it would show a frequency drift back and forth ranging over the *entire* 2.5 MHz bandwidth of SETI@home (rough calculation; also depends on the orbital inclination).

For that matter, a lot of what S@H does is *look* for frequency drifts. A drift of 5 Hz/sec crosses the 2.5 MHz bandwidth in 6 days.

So, I really think you should have a version of NTPCkr running that does not filter out persistent signals because they differ in frequency.

I'd bet higher than 10 to 1 odds that if a signal is ever found, it will show frequency drifts (even after correction to the solar system's barycenter to remove Earth's own motion).

Alan MacRobert
Senior Editor
Sky & Telescope

2) Message boards : SETI@home Staff Blog : So where's the science..? (Message 544277)
Posted 11 Apr 2007 by Alan M. MacRobert
Post:
Matt,

I hope you will remember, when planning the science analysis, to include the entire 2.5 MHz spectrum when looking for repeat signals from the same sky location. As I discussed with Dan a few years ago, a broadcast (rather than targeted-at-Earth) transmitter would have to be so powerful that the place to put it would be in close orbit around the local sun -- where it could grow large enough to collect significant solar energy without covering most of the real estate on the home planet or microwave-cooking everything on the planet.

This means that, due to the transmitter's orbital motion, the signal will have a modest rate of frequency drift but a large total frequency range -- which I think happens to be about 2.5 MHz for a transmitter orbiting a sunlike star at 0.1 a.u. (where solar energy is dense but not quite too hot for ordinary structural materials).

In fact I asked Dan if this was why he chose 2.5 MHz for the bandwidth, and I think he said no, it's a lucky coincidence.

Remember any transmitter we're actually going to hear has almost certainly been running for millions of years at least, so the aliens will have had ample time bu now to expand it to astronomical size and power, and this means it's likely to be in a near-star, fast orbit.

Alan MacRobert
Senior Editor
Sky & Telescope
3) Message boards : SETI@home Science : 10th Planet Discovered; Bigger than Pluto (Message 144649)
Posted 30 Jul 2005 by Alan M. MacRobert
Post:

That preprint is about the wrong Kuiper Belt object! Just to keep things confusing, TWO big KBOs were announced today. The one with a satellite is 2003 EL61. The one that's bigger than Pluto is 2003 UB313.

Preprint article that will present further details of the planetoid and its moon at the American Astronomical Society's planetary sciences meeting in Cambridge, UK, on 8 September is available here (pdf file, 113kb).

4) Message boards : SETI@home Science : 10th Planet Discovered; Bigger than Pluto (Message 144586)
Posted 29 Jul 2005 by Alan M. MacRobert
Post:
Breaking news here! A giant Kuiper Belt object has been discovered 97 a.u. from Earth and Sun, definitely larger than Pluto:

http://SkyandTelescope.com/news/article_1560_1.asp
5) Message boards : Number crunching : Allen Telescope Array (Message 117163)
Posted 1 Jun 2005 by Alan M. MacRobert
Post:

Torrey,

The Allen Telescope Array is a completely different (and self-contained) project. We have an article about it at

http://skyandtelescope.com/resources/seti/article_253_1.asp

For descriptions of all current SETI projects worldwide:

http://skyandtelescope.com/resources/seti/article_248_1.asp

Alan MacRobert
Sky & Telescope
6) Message boards : SETI@home Science : An end to it all? (Message 114142)
Posted 23 May 2005 by Alan M. MacRobert
Post:

Last August I asked Dan Werthimer about this. He said that considering interstellar scintillation (which can make a steady signal come and go on a time scale of seconds to hours), and considering the need to get several repeats of a real signal before thinking that it's real, a reasonable goal for SETI@home would be at least nine good scans of most the sky that's accessible to Arecibo. So far, a fair amount of this area has been scanned at least six times.

But analysis of all this data has hardly begun. It will require searching for repeated signal matches across ALL of the 2.5 MHz of spectrum, considering the doppler shifts that are likely for an alien transmitter that's in close orbit around a star (where you would expect a really big, ages-old transmitter to be located -- where it could collect lots of free solar energy forever). This will be a huge job that is not talked about much yet. We're merely listing signals (billions so far), not matching them.

As for your larger question, it's estimated that all the SETI searches to date have analyzed only one part in 10^14 (a hundred-trillionth) of the "search space" that needs to be covered. For more, see

http://SkyandTelescope.com/news/article_1322_1.asp

and

http://SkyandTelescope.com/resources/seti/article_248_4.asp

Alan MacRobert
Sky & Telescope
7) Message boards : Number crunching : How Many Splitters Are Used To Split For SETI Classic? (Message 103696)
Posted 25 Apr 2005 by Alan M. MacRobert
Post:

>> Maybe that is why I have seen, since about June 1, 2003, the flattening of
>> the workunit production graph at Current Progress/ Collect Data/
>> Workunit Production at the SetiClassic Website.
>
> Actually, the reason for that this web page is broken and WAY out of date...

Matt,

Could you tell us what IS the total number of workunits split since the beginning of the project? And the total results received, Classic and BOINC? So that I can post the right numbers on Sky & Telescope's web site. The numbers now on the SETI@home site make it look like for nearly two years there has been gigantic redundancy and waste.

Thanks very much.

Alan MacRobert
Senior Editor
Sky & Telescope
8) Message boards : Number crunching : wasted time (Message 98294)
Posted 13 Apr 2005 by Alan M. MacRobert
Post:
9) Message boards : SETI@home Science : Alien Message In Our DNA? (Message 97896)
Posted 12 Apr 2005 by Alan M. MacRobert
Post:

> Forget waiting for ET to call -- the most likely place to find an alien
> message is in our DNA, according to an expert in Australia.

Actually, I think this idea was first put forth by Timothy Leary, of LSD fame, way back in the 1970s. In a book called "Starseed," if I remember right.

I do think the idea is farfetched, in that if ET visitors had tinkered with our DNA at all, they would have done more with it. As it is, there's every sign that we inherited our DNA completely from natural apes. Artificially inserted code would have been noticed by now unless it's really subtle -- and if you're doing it at all, why be subtle?

Some of the "junk DNA" segments are turning out not to be so junky after all. Partly-overlapping strings of code are turning out to have multiple uses for generating entirely different proteins and structures. Evolution is amazing in the way it finds to do things.

Alan MacRobert
10) Message boards : Number crunching : wasted time (Message 97892)
Posted 12 Apr 2005 by Alan M. MacRobert
Post:

> I am confused, you do not run your computer and now you want BOINC to clean it
> up for you, jeeezzz just run the computer....

SETI@home has always been billed to the public as a screensaver. Install it, forget it. Ordinary people don't want to be computer-process tinkerers any more than they want to be refrigerator tinkerers. They want their refrigerator to work by itself, as advertised.

This doesn't mean they're dumb or contemptible, it just means they've got other things to do on weekends.

We should cut the hardworking Berkeley folks some slack as they fix these problems. But the fact is, for nearly six years S@H has had far more than enough users to analyze all the available data 100% completely -- even with three- or four-times redundancy to catch every (rare) bad result. Beyond that, the excess users are just wasting electricity.

In fact, that's partly why BOINC was created -- to divert the (huge) wasted excess of people into other projects. So, we shouldn't be telling the guy he needs to run SETI@home 24/7.

Alan MacRobert
11) Message boards : Number crunching : Splitter off? WHY? (Message 90373)
Posted 24 Mar 2005 by Alan M. MacRobert
Post:

> I have boxes of tapes (about half a
> year's worth of backlogged data)...

Matt, thanks for the explanation. A question: If there's half a year of backlogged unsplit data, while S@H users are recrunching the same work units more times than is useful, why not split the tapes, clear out the backlog, and put more of us crunchers to good use? I've wondered about this for years. Are the splitters a bottleneck? Dave Anderson once said no -- but then what is?

Thanks again,

Alan MacRobert
12) Message boards : SETI@home Science : What's happened to the seti data from down under ? (Message 60368)
Posted 4 Jan 2005 by Alan M. MacRobert
Post:

Running southern-sky data from Parkes through SETI@home isn't happening yet; many other things are in the pipeline ahead of this (a pretty clogged pipeline apparently), and if funding doesn't show up, it may never happen at all. The last time I heard from Frank Stootman at Parkes, well over a year ago, he said all work on this project had stopped.

Meanwhile, Parkes's own Southern SERENDIP project is still looking for signals; see

http://skyandtelescope.com/resources/seti/article_248_6.asp

Alan MacRobert
13) Message boards : Number crunching : Boinc 4.57 (Message 57970)
Posted 28 Dec 2004 by Alan M. MacRobert
Post:
Overclocking SETI@home should not be done. Before everyone starts flaming, here's why.

SETI@home is sensitive to a single-bit error. That is, an entire work-unit calculation, hours long, can be ruined by a single transistor saying 1 for a billionth of a second when it should say 0.

This is not like running Half Life 2, where one little polygon in a battle scene showing light gray instead of medium gray for 1/24 of a second will go totally unnoticed.

People who say "My overclocked computer works PERFECTLY!" (I bet they're gamers) may just not be seeing trivial errors. But in the long, linear SETI@home calculation, no errors are trivial. They will create bad results that gum up the works (and may get you blacklisted from the data pool).

Think a minute. If your CPU chip was sure NEVER to make an error at the faster speed, don't you think Intel would sell it at that speed?

Alan MacRobert
Senior Editor
Sky & Telescope





 
©2019 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.