There's a small difficulty with SETI that reminds me of a situation I often encounter in European restaurants. I call it the "synchronicity problem."

Imagine that I've spent two hours savoring the restaurant's house special (usually some species of fish known only to ichthyologists), and want to get the bill and get out. What is the probability that when I stare at the waiter to signal my intentions, he'll be looking at me?

Not high.

This is a SETI problem too. Since we observe a star system, at any particular frequency, for only a few seconds or minutes, can we reasonably expect that during that brief scrutiny the extraterrestrial transmitters are "looking" at us? After all, an alien society will surely be too distant to have picked up our television or radar and thereby know we're here, scanning the skies for signs of life. So why would they be sending messages to us now?

An obvious response is that the extraterrestrials could be beaming a non-stop signal to the entire galaxy. Their strategy for pinging unknown neighbors may simply be "all alien, all the time."

The obvious appeal of a continuous, pan-galactic broadcast has prompted some scientists to invent unconventional schemes for omnidirectional signaling. For example, fifteen years ago G. M. Beskin and A. V. Sannikov, of the Russian Academy of Sciences, figured that advanced societies might co-opt big stars as billboards that would grab the notice of just about anyone in the Milky Way.

Their suggestion was this: Aliens with a yen to yak could set up powerful generators to fire a beam of hard X-rays at so-called super-supergiant stars. These bulky suns, also called hypergiants, are the galaxy's heftiest stellar inhabitants: millions of times brighter than Sol, and considerably larger. Indeed, if the sun were suddenly swapped for the most voluminous of these hypergiants, YV Canis Majoris, the solar system would be stuffed with star all the way to Saturn.

Every astronomically literate society will know of hypergiants, and would undoubtedly burden grad students with their study. Consequently, sooner or later someone would notice if the beam from an alien race lit up the outer layers of a hypergiant, causing it to develop an unnatural X-ray glow. The observer could quickly deduce that advanced aliens were trying to get his attention.

This idea is like the bat signal used by the Gotham City police whenever they need Batman backup. The desperate gendarmes use a powerful searchlight to project a bat logo onto a cloud layer, producing a simple call sign that would be noticed, one presumes, everywhere in the city (including suburban and stately Wayne Manor).

The Bat Signal is a one-bit signaling device, but the X-ray generator lighting up a hypergiant could pulse on and off to send — Morse-code style — a message. Of course, the bandwidth is low because scattered rays from the edges of the star will arrive at the receiver later than those from the center, blurring the pulses. The maximum data rate, even for a star only ten times the diameter of the sun, would be about one bit per minute. That's worse than dial-up.

The limited bandwidth of this scheme is a problem, but there's more bad news. Lighting up a star for signaling purposes takes a lot of power. Beskin and Sannikov estimated that their scheme would require an X-ray generator that was at least 0.01% of the star's light output, which even for a fairly dim hypergiant is 1025 watts. For comparison, that's about a million times more power than required for an all-galaxy radio beacon that could be easily detected by today's SETI experiments! X-raying stars seems like a hard way to get out your message.

Starlight, star bright

Recently, astronomers in Hawaii and France have come up with other ideas for turning stars into beacons — schemes that, once again, advanced aliens might use to get the attention of galactic inhabitants.

John Learned (University of Hawaii) and his colleagues propose something reminiscent of the Russian approach. They suggest that aliens could fire neutrinos into the incandescent innards of a Cepheid variable star, thereby inducing it to slightly change the rate at which it brightens and dims.

Cepheid variables are massive stars (albeit not hypergiants) that pulsate regularly, and are routinely used to calibrate the distances to galaxies. They're useful to astronomers the way lab rats are to medical researchers, and will be noticed and studied by scientists on any world. In addition, they can be seen at intergalactic distances. So causing a Cepheid to change its pulse period is akin to projecting a bat signal, not on some low-lying stratocumulus, but on the moon!

Lamentably, once again the bit rate will be low, and the utility bills high. The former won't exceed one bit per second, and the latter will spin your electric meter at the not-inconsiderable burn rate of 1023 watts. That's somewhat less pricey than the hypergiant scheme, but still nowhere near the much lower cost of an all-galaxy radio beacon.

Finally, there's the scenario proposed by French astronomer Luc Arnold, who suggests that the aliens will build giant, odd-shaped shadow puppets to orbit their own sun, occasionally blocking some of its light. In our search for planets that transit their suns, we might notice one of these artificial occulters. The dimming would be different from that of a round planet if the object is, for example, a giant triangle, or a louvered screen. If we saw one of these strange mini-eclipses, we'd know that someone was at home in that solar system.

But again, this is really slow signaling — the transit would occur for a few hours every year (whatever their year might be), and would only convey a single bit of information: "Aliens are here." And while there's no operating expense (the occulter needn't fire either X-rays or neutrinos. It merely orbits its star), there's the not-inconsiderable expense of building a Christmas tree ornament roughly the size of Earth.

While admirably clever, these schemes to take advantage of stars for signaling strike me as impractical. They're all cursed by high costs and extremely low bandwidth. Radio transmitters or pulsed lasers do enormously better on both counts.

So sure, it's possible that the aliens are tickling stars to get our attention. But looking at it from a cost-benefit point of view, I think these ideas are akin to using smoke signals instead of cell phones. Slow and inefficient, not to mention being a severe annoyance in restaurants.