Shallow Thoughts

Akkana's Musings on Open Source, Science, and Nature.

Fri, 28 Oct 2005

A Value of "Close" With Which I was Previously Unfamiliar

A very strange article in today's SF Chronicle describes "Mysterious, bright lights in the night sky Wednesday that alarmed or bemused scores of Bay area residents".

It atributes to Andrew Fraknoi, chairman of the Foothill College astronomy department and media hound (it doesn't say that second part), information that "the lights were probably Mars and Venus, two planets that currently appear close together and will probably remain brilliant for another week or two until their orbits begin moving them away from the Earth again."

Aside from the "probably" (I was under the impression that the basic orbits of the major planets were fairly well understood, and that it's fairly rare that a planet suddenly deviates from its regular orbit in a visible way), I found this curious because Venus is currently in the early evening sky -- since its orbit lies inside that of the Earth, it can never appear to move very far from the Sun -- while Mars, a week before opposition, is rising in the early evening and overhead at roughly midnight.

Just to be sure, I checked with XEphem. The angular distance between Mars and Venus is current 146°. They're almost at opposite ends of the sky. This is a definition of "close" with which I was previously unfamiliar.

I don't know if Fraknoi really said this, or if he was simply misquoted by the reporter, David Perlman, the Chronicle's Science Editor. If so, the misquote is quite pervasive -- he repeats several times throughout the article Fraknoi's assurance that the lights (shown in a photograph accompanying the article, indeed close together though we aren't told anything about the lens used to take the photo) must be Venus and Mars.

Other giggle-inducing quotes from the article:

No one except astronomers could offer an explanation.
Well, gosh, you certainly wouldn't want to listen to those egghead astronomers about a question involving lights in the sky.

(Well, okay, in this case you shouldn't listen to them, because the Mars/Venus explanation obviously doesn't fit the observations.)

According to Fraknoi, Mars now far outshines even the brightest of all the stars in the sky, and when skies are clear, the fourth planet from the sun could look even bigger than normal.
Mars at opposition is certainly brighter than any star (except the Sun, of course). It currently shines with a magnitude of about -2.2 (a smaller number means a brighter object; the brightest star, Sirius, is magnitude -1.4. Venus, at the moment, is much brighter than either one at -4.2, as is usual since it's larger, closer, and more reflective than Mars. That might have been worth mentioning.

I can't figure out whether "even bigger than normal" is supposed to refer to size or brightness. Mars is normally a tiny object as viewed from Earth, too small to see much detail except for a few months around opposition every couple of years. Indeed it is much bigger than normal right now (and a lovely sight in a telescope!), as well as brighter; but "even bigger" seems like an odd phrasing for something normally so small.

But since Mars' size isn't visible except in a telescope, Dave thinks "bigger" here was meant to refer to brightness: the misconception that brighter objects look bigger. I shouldn't make fun of this: the great astronomer Tycho Brahe, in the 1500's, was convinced that stars had angular size instead of being point sources. He thought brighter stars appeared bigger, and based his geocentric solar system model on that. That view wasn't disproved until Galileo invented the telescope. It's a common misconception even today, but I'd hate to think the Chronicle's science editor was encouraging it, so I'll stick to the assumption that he really meant size and that the "even" was just an odd journalistic embellishment.

So what were the mysterious lights? I don't know. I didn't see them, and the article doesn't give enough detail to make a good guess. But the photo looks a lot like airplanes or helicopters; at least one of the lights has a couple of smaller lights to either side, usually a dead giveaway for an aircraft.

Update the following day: I wasn't the only one to complain about this article, and the Chron published a paragraph in the Corrections section this morning clarifying that Venus is nowhere near Mars and could not have been related to the lights people reported.

[ 10:57 Oct 28, 2005    More headlines | permalink to this entry ]

Fri, 14 Oct 2005

More On Low Orbits

Wacky Chinese orbital physics are in the BBC again. Today's story tells us how they've corrected yesterday's orbital problems. Quoting from China's "official Xinhua news agency", the BBC tells us:
Xinhua said the craft had deviated from its planned trajectory because of the Earth's gravitational pull.

I can hear them now ... "Darn it! I guess we forgot to take the earth's gravity into account when making our orbital calculations!"

[ 20:40 Oct 14, 2005    More headlines | permalink to this entry ]

Thu, 13 Oct 2005

Darn Those Low Orbits

BBC News Science tells us about the orbital problems of China's manned Shenzhou VI spacecraft.

Gravity has drawn Shenzhou VI too close to earth, the agency said.

Shenzhou VI, which has two astronauts on board, is in a low enough orbit to be affected by the Earth's gravitational pull.

Don't you hate those low orbits that are affected by gravity? Maybe next time they should choose an orbit high enough that it isn't subject to gravity.

[ 19:39 Oct 13, 2005    More headlines | permalink to this entry ]

Wed, 12 Oct 2005

Autumn Winds

The mockingbird is singing. He's been doing that for three weeks now. What's he doing bursting into song all of a sudden in late September and keeping it up for weeks?

All over, animals in the parks are restless. Squirrels are madly digging up nuts from one place, carrying them to another and re-burying them. Chipmunks have appeared, chipping from the bushes as we walk by. I normally don't see chipmunks in the local parks, just ground and tree squirrels. Are they always here, but usually quiet so we don't see them? Or did they migrate in for the season?

An unusual species of large yellow-billed blue bird appeared on the wire above the house. How odd! What's blue, jay sized but has a big bulky yellow bill?

Binoculars provided the answer. A scrub jay with an acorn in its bill! Since then I've seen quite a few yellow-billed Stellar's jays in the local parks as well.

The central area of Alum Rock is filled with a large family of acorn woodpeckers drilling holes in trees, posts, and the walls of the Youth Science Institute building to store their acorns for the winter. The YSI building looks like swiss cheese. A few days after I saw the woodpeckers at work, we went back and the buildings had all sprouted dangling silvery tinsel from all eaves. It seems to be keeping the woodpeckers away. Bad for me (they're cute), good for the YSI.

I saw a couple of nuthatches at Arastradero. A first for me. I don't know if they're migrants, or if they're always there and I've just never noticed before. Arastradero was also thick with white-tailed kites. There are always a few testing the slope currents there, but this time I saw at least four different pairs, maybe more, each with their own territory staked out. Somehow even with that many kites they all managed to stay too far away for me to get a good picture.

The reason for all the time spent at Alum Rock and Arastradero is that we're on the hunt for tarantulas. Every fall, just as the weather starts to get cold, the male tarantulas come out of their burrows and go marching across the trails looking for females. (Maybe the females are marching too. I'm not clear on that.) They're only out for a short time -- maybe a week -- and they're easy to miss. Last year we missed them altogether (but then we lucked out and spotted one later that month while travelling in Arizona).

Anyway, we've had no tarantula luck yet this year. Henry Coe state park had its annual Tarantula Festival already, a week and a half ago. But they always seem to have the festival while the weather's still hot, long before tarantulas show up in any other parks. Maybe Coe tarantulas are a different species which comes out earlier than the others. At any rate, we've seen no sign of them at Alum Rock or Arastradero so far this year.

But back to that singing mockingbird. He doesn't seem to be the same mocker who set up house here this spring and raised three nests of chicks. That one had a very distinctive call note which I haven't heard at all this fall.

But what's he doing singing in autumn? Is he singing as he packs his bags to fly to LA or Mexico? Or confused about the weather? Someone asked that on a local birding list, after noticing thrashers (closely related to mockingbirds) suddenly finding the muse. I reproduce here the edifying and entertaining answer. (Googling, it appears to have been a folk song, though I can't find a home page for the author or anything about the music.)

The Autumnal Recrudescence of the Amatory Urge

When the birds are cacaphonic in the trees and on the verge
Of the fields in mid-October when the cold is like a scourge.
It is not delight in winter that makes feathered voices surge,
But autumnal recrudescence of the amatory urge.

When the frost is on the punkin and when leaf and branch diverge,
Birds with hormones reawakened sing a paean, not a dirge.
What's the reason for their warbling? Why on earth this late-year splurge?
The autumnal recrudescence of the amatory urge.
Written by Susan Stiles, copyright December 1973

[ 22:54 Oct 12, 2005    More nature | permalink to this entry ]

Mon, 10 Oct 2005

How to Search Your Mozilla Cache

Ever want to look for something in your browser cache, but when you go there, it's just a mass of oddly named files and you can't figure out how to find anything?

(Sure, for whole pages you can use the History window, but what if you just want to find an image you saw this morning that isn't there any more?)

Here's a handy trick.

First, change directory to your cache directory (e.g. $HOME/.mozilla/firefox/blahblah/Cache).

Next, list the files of the type you're looking for, in the order in which they were last modified, and save that list to a file. Like this:

% file `ls -1t` | grep JPEG | sed 's/: .*//' > /tmp/foo

In English: ls -t lists in order of modification date, and -1 ensures that the files will be listed one per line. Pass that through grep for the right pattern (do a file * to see what sorts of patterns get spit out), then pass that through sed to get rid of everything but the filename. Save the result to a temporary file.

The temp file now contains the list of cache files of the type you want, ordered with the most recent first. You can now search through them to find what you want. For example, I viewed them with Pho:

pho `cat /tmp/foo`
For images, use whatever image viewer you normally use; if you're looking for text, you can use grep or whatever search you lke. Alternately, you could ls -lt `cat foo` to see what was modified when and cut down your search a bit further, or any other additional paring you need.

Of course, you don't have to use the temp file at all. I could have said simply:

pho `ls -1t` | grep JPEG | sed 's/: .*//'`
Making the temp file is merely for your convenience if you think you might need to do several types of searches before you find what you're looking for.

Tags: , , , , , , ,
[ 21:40 Oct 10, 2005    More tech/web | permalink to this entry ]

Wed, 05 Oct 2005

Blank Survey Messages

I love the messages I keep getting from having attended LinuxWorld a few months ago. They say:
Dear LinuxWorld Attendee:

That's it. That's all they say.

Well, unless you dig deeper. It turns out there's an html part, too, which actually does have content (asking me to participate in a survey). But I would never have seen that if I didn't know how to read the MIME structure of email messages.

You see, the message is sent as Content-Type: multipart/alternative, which means that there's a text part and an html part which are supposed to be equivalent. You can read either part and they should say the same thing. Lots of modern mailers, including Mozilla, mutt, pine, probably Opera, and even Apple Mail can now be configured to give preferences to the text part of messages (so you don't have to squint your way through messages written in yellow text on a pink background, or in tiny 6-point text that's too small for your eyes, and so that you can be assured of safety from web bugs and other snoopware which can be embedded in html mail). Any mailer so configured would show this message as blank, just as I'm seeing it in mutt.

It seems amazing that their web developers would go to the extra trouble of setting up MIME headers for multipart/alternative content, then not bother to put anything in the content. Why not just send as plain html if you don't want to create a text part?

I'd let them know they're sending blank messages, but tracking down a contact address seems like more trouble than it's worth. The mail purports to come from LinuxWorldConference&; the message references URLs from; the actual message (as seen in the headers) came from It's probably generating the bad email messages, but it's hard to say for sure.

Suggestion for web developers: if you write code to send out mass mailings, you might want to check the mail that's actually getting sent out. If your program generates more than one MIME attachment, it's a good idea to check all the attachments and make sure you're sending what you think you are. Look at the actual message structure, don't just glance at the message in the mail program you happen to use. The fact that a message displays in one mailer does not imply that it will display correctly for all users (especially for a survey aimed at Linux users, who use a wide variety of mailers on many platforms and are quite likely to set non-default options). If you don't know MIME, find someone who does to sanity-check your output ... or don't send multiple MIME attachments.

[ 11:30 Oct 05, 2005    More tech | permalink to this entry ]

Tue, 04 Oct 2005

Hacking Mozilla Extension Versions

Mozilla Firefox's model has always been to dumb down the basic app to keep it simple, and require everything else to be implemented as separately-installed extensions.

There's a lot to be said for this model, but aside from security (the need to download extensions of questionable parentage from unfamiliar sites) there's another significant down side: every time you upgrade your browser, all your extensions become disabled, and it may be months before they're updated to support the new Firefox version (if indeed they're ever updated).

When you need extensions for basic functionality, like controlling cookies, or basic sanity, like blocking flash, the intervening months of partial functionality can be painful, especially when there's no reason for it (the plug-in API usually hasn't changed, merely the version string).

It turns out it's very easy to tweak your installed plug-ins to run under your current Firefox version.

  1. Locate your profile directory (e.g. $HOME/firefox/blah.blah for Firefox on Linux).
  2. Edit profiledirectory/extensions/*/install.rdf
  3. Search for maxVersion.
  4. Update it to your current version (as shown in the Tools->Extensions dialog).
  5. Restart the browser.

Disclaimer: Obviously, if the Firefox API really has changed in a way that makes it incompatible with your installed extensions, this won't be enough. Your extensions may fail to work, crash your browser, delete all your files, or cause a massive meteorite to strike the earth causing global extinction. Consider this a temporary solution; do check periodically to see if there's a real extension update available.

More information on extension versioning (may be out of date).

Tags: , , ,
[ 18:47 Oct 04, 2005    More tech/web | permalink to this entry ]