phi

Some folks at the Parkes radio telescope in Australia have found an interesting radio signal. Some key data isn't public yet, but with my former-SETI-researcher hat on there is one bit of information I find quite intriguing, and some other information that I suspect the mainstream media might latch onto that's really not.

The signal is a narrow band radio signal, which is somewhat interesting because when humans send out intentional interstellar radio signals we use narrow band signals ourselves. But it's really interesting because that makes it easier to tell if the signal is coming from the Earth.

An extraterrestrial signal on a fixed narrow band will appear to change frequency over time because of the Doppler shift *of the earth itself*. You may have seen the science museum demonstrations of Doppler shift using moving sound sources, or noticed changes in tone from vehicles as they pass by. But if the sound source is staying still and you are the one moving, you'll also hear similar shifts. Likewise the earth's rotation creates a shift or "chirp" in the signal as the telescope passes under it. A source that's coming from the earth will not exhibit a "chirp" because source and receiver are moving in tandem. SETI searches all look for this as the primary discriminator for earthbound sources, and it is very easy to detect.

This signal displays a "chirp" and that makes it very interesting. The exact nature of the chirp hasn't been released but I gather that there is a component from the Earth's motion plus another component that's attributable to the motion of the source. That might be useful for attributing the signal to a source whose motion matches, whether that is an astronomical object or merely some manmade space probe or satellite—or even an airplane—in our own solar system.

You may have heard that the signal is coming from the direction of Proxima Centauri b, a nearby Earth-sized planet in its star's habitable zone. This is not interesting at all. Planets like that are a dime a dozen, and there are probably countless others in that same direction.

You may also have heard that the frequency of this signal—980 MHz—is in a reserved band that is not used for radio communication by humans. That is true, but it doesn't mean those frequencies are empty. They're used for radar, and more prosaically by microwave ovens. Just because we're not communicating at these frequencies doesn't mean there isn't a lot of man-made noise there.

Seth Shostak has a great SETI researcher's view on this discovery which I think is spot on, unsurprisingly as he's one of the best, most level-headed folks working in the field today. He gets into the chirp in detail. And he doesn't even mention 980 MHz or Proxima Centauri b.

Today was the longest day of the year here in the Northern Hemisphere. But it wasn't the earliest sunrise, or the latest sunset.

What?

The length of a day -- 24 hours from sunrise to sunrise, on average -- comes mostly from Earth's rotation. But a few minutes of it comes from Earth's orbiting around the sun. It takes Earth 23 hours, 56 minutes, and a bit over 4 seconds to rotate 360 degrees, at the end of which time the stars will be in the same positions in the sky as they were the previous day. But because the sun isn't in the same place in the sky as it was the previous day, Earth has to rotate about 3 minutes and 56 seconds more to bring the sun back into the same apparent position it had been in 24 hours previously.

The catch is that "about". Two things affect this number. First, Earth's orbit isn't perfectly circular. It's closest to the sun in early January and furthest in early July. When it's further, it moves more slowly in its orbit, so it doesn't have to turn as much extra to make a day. When it's closer, it moves more quickly, so it has more to make up. Second, and right now more importantly, at the equinoxes some of the sun's apparent motion is north-south so Earth doesn't have to rotate as far to catch up to the sun's new east-west position. At the solstices, Earth has to turn quite a bit further to make up for same amount of orbital motion, because all of the sun's apparent motion is east-west. This makes the day/night cycle longer than average -- right now, it's about 24 hours 15 seconds from one sunset (or sunrise) to the next. The effect is even more pronounced in December when it's aligned with the eccentricity effect instead of opposed to it.

The longer days mean that here in Boston, though we've missed the earliest sunrise by about a week, we have until June 26 to celebrate the latest sunset of the year.

Last time the three things meme came up I concluded that my most interesting things are more "three unusual things I have done with people who don't happen to read my LJ" than three things I'd done on my own. But on reflection there are some things I think I can put down which won't be quite as much a function of which social circles read this blog. So...

Three things I've done that I'd be surprised if any of my friends (on LJ or otherwise) had:

• Made an astronomical observation with an optical telescope of >5m aperture
  
• Gone on a >1000km bike ride Edit: Finished last in a multi-day bike race
  
• Competed in a motorsport by writing a computer program

Today I learned that Larry Niven can't spell.

Specifically, he can't spell "Klemperer".

Inspired by hammercock, I just did a vanity search and found, among other things, a manual for the first program I ever wrote for money.

I don't know which is more incredible: that the software is still in use, 17 years later, or that it has been running on the same 8086 PC that entire time.