Archive for July 2006

On Something I Haven’t Read

July 30, 2006

Yet. At Get Religion, Daniel Pulliam discusses an AP article about the book The Language of God: A Scientist Presents Argument For Belief.

The author of the book, Francis S. Collins, is the head of the Human Genome Project. For someone with his stature, the surprise is that he attempts to stake out a middle ground in the Science-Religion controversies extant in the topics of Evolution, Intelligent Design and Biblical Creationism. From the AP article:

He asks scientific skeptics to investigate God with the same open-minded zeal they apply to the natural world, saying that there’s no incompatibility between belief and scientific rigor.

But this does not mean that he accepts either Intelligent Design or Creationism.

He tells fellow evangelicals that opposition to evolution — whether based in the biblical literalism of creationists or “intelligent design” arguments — undermines the credibility of faith. He finds the first line of thought “fundamentally flawed” and says the second builds upon gaps in evidence that scientists are likely to fill in.

I find little to disagree with in these two statements, and where I do detect a difference, he of course speaks with much more authority than I. This sounds like a book I want to read.

Lovin’ Linux

July 30, 2006

It’s been awhile since I last commented on my experiences with Linux. Mostly, that’s a good (very good) things, because my experiences, especially of late, have been great.

I took a new job in June, and much to my surprise I was offered the choice of Windows or Linux on my desktop. That’s the first time I’ve even been given a choice of OSs! I took Linux, and got a P4 2.4 with Fedora Core 5 (that’s a fairly recent Linux mint) installed.

So now, I have Mandriva 2006 at home, FC5 at work, and a little expertise around me. In three days I had a VPN (Virtual Private Network) set up so that I could connect seamlessly to my test environment at work through their firewall, from home. Very cool.

Well, I had noticed that my system partition at home was pretty full, and in fact, I was going to be out of space for new software installs pretty soon. Since I did have my share of problems with the original install, I started looking into partitioning schemes on Linux to find out how to do better, and hoped to rectify one annoying problem that I had had for a few months. (It happens that Mandriva has had problems with its menu system for a few builds. Menu items in it’s equivalent of the start button seem to get lost when some software is removed, and none of the fixes I’ve ever seen seem to rectify the situation. It’s not serious – the software is not lost and can be run, but it is an annoyance.) This week was the perfect opportunity to fiddle with my system at home.

So I (carefully) backed up things of importance and bit the bullet last weekend. Immediately I noticed two things; several minor bugs reappeared, just as they had the first time I installed Mandriva 2006. The cursor was transparent yellow (bad) and moving windows on the desktop left persistent traces on the screen. I had forgotten about that one! This time I knew not to panic – both of these problems would go away with the first round of updates, which was going to happen right after I got the wireless driver going.

You may have heard that wireless connections have been a bear for Linux users, relative to Windows users. That was certainly true in the past when the manufacturers of the cards were loathed to publish their specs. for fear of competition. Without the specs., Linux users would have to wait and/or ‘roll their own’ drivers, which were of obviously lower quality and reliability for a while.

That situation no longer exists. In the years since 801.11g specs. came out (the specs. for the most popular kind of wireless router), several manufacturers have provided Linux drivers, and even better, several independent groups (like at SourceForge) have come out with great drivers for specific chip sets (like MadWiFi, for the Athereos chip set used in my PCI adapter card).

Compared to January, when I had a set-up with an invisible cursor and a balky wireless network card, this was great. I had the wireless drivers built minutes after the install, and I was connected to the internet less than five minutes later.

But I’m not going to tell you that this was problem free. I took my time and installed my usual set of software and got the VPN going, but some pesky details were annoying. In particular I couldn’t get the mPlayer plug-in software working quite right. That’s the Linux version of MicroSofts Media Player, and I couldn’t get it to go. A little research seemed to indicate that I had done things out of order – mPlayer wants codecs installed before the player is installed (and I gotta tell ya, I saw that little detail nowhere in any installation instructions).

Worse, I had misjudged the amount of space I would need for one of my partitions, so I pretty nearly had the same problem I had when I decided to build. So for the second time in a week, it’s back to the installation disks.

I bit the bullet a second time. Did you know that one reason people have so much trouble installing their own operating system is because they don’t do it enough? It’s easy to forget the details. Install a system twice in one week and it becomes…

Easy! The partitioning is now correct, the wireless installation and set up seems almost trivial to me now, VPN is easy, and I have a bunch of software upgrades that I hadn’t gotten around to installing before.

Here are the lessons. I was not happy in January with Mandriva 2006. It seemed a step backwards from a very stable 2005. But early version of Linux software can be pretty rough, and I got obviously buggy early version of both Mandriva 2006 Free and the MadWiFi driver. Both were fixed in fairly short order, and now I really, really like Mandriva 2006. Well, maybe those first updates could have come quicker, but it’s been stable, it recognizes far more hardware by default than previous versions, and has a good feel.
I’m not sure what’s going on with the people working on mPlayer. Going through the users groups showed me pretty quickly that I was not the only one experiencing problems installing that media player. I saw indications that it not only depends on having the codecs installed first, it also needs a completely independent package, RealPlayer, installed prior. In itself, that’s pretty bogus. What’s not acceptable is that this is not indicated anywhere in their installation instructions. What’s infuriating is that they have NO obvious installation instructions on their site. I know mPlayer is a bit OS specific – the install is different in Mandriva from Debian and Fedora, but that has nothing to do with the lack of help on the codecs.

And that problem, severe lack of instructions appears to be endemic to SourceForge too. Love their software, hate their documentation. Almost without fail – no, I’ll amend that to Without Fail – their installation instructions, if they exist at all, are difficult to impossible to find, and never, ever, applicable to the first time installer (I’m not talking about Linux experience here. Every user must install a given package for the first time, ya know.)

With Linux you may have to be willing to break and rebuild your system every so often, or put up with older, tried and true software. But if you’re willing to do that, why are you using Linux to begin with?

Data Is Not Information

July 30, 2006

And information is one heck of a long way from wisdom.

There an awful lot of data in the hockey stick diagram, but not all of it is correct. Here’s just one more data point.

The money quote:

The fact that the earth’s climate changes in cycles from warm to cold to warm, etc. (“Hockey Stick Hokum,” editorial page, July 14) was noted in the late 18th century by Edward Gibbon in his “Decline and Fall of the Roman Empire”: “The reindeer, that useful animal, is of a constitution that supports, and even requires, the most intense cold. He is found on the rock of Spitzberg, within ten degrees of the Pole; he seems to delight in the snows of Lapland and Siberia; but at present he cannot subsist, much less multiply, in any country south of the Baltic.” In the time of Caesar, Mr. Gibbon wrote, the reindeer was native to the forests of Germany and Poland, but in Gibbon’s time the animal was nowhere to be seen in those parts. And between the Age of Caesar and the Age of Gibbon, the Medieval Warming Period and the “Little Ice Age” had taken place.

And In Response To Yesterday’s Post…

July 30, 2006

…this Psychology Today article contends that the care and attention given to children is sometimes (often?) taken to extremes, and has some disturbing consequences.

Contra yesterday’s post, from the NYT noting that early (and improved) childhood health care is demonstrating a virtuous cycle of compounding benefits in later life.

Hat Tip to Glenn Reynolds.

Are We Getting Healthier?

July 29, 2006

This may be a new kind of evolution. Or, at least, it’s a new understanding. According to this Times article, humans as a species are not only taller, bigger and more long-lived than just a century and a half ago, many chronic ailment, such as lung disease and heart disease, are occurring decades later than they used to. That’s not something that can be explained entirely by better medical treatment (better medical treatment of these ailments starts after the disease begins).

The proposed reasons are as unexpected as the changes themselves. Improved medical care is only part of the explanation; studies suggest that the effects seem to have been set in motion by events early in life, even in the womb, that show up in middle and old age.

Ever since baby-boomers discovered the hula-hoop, they’ve been doing things together in large numbers. Now they’re turning 60 in historically very large numbers, and apparently, in about forty more years, a bunch of us will be hitting 100 together. We’ll probably all go to the Rolling Stones 45th farewell tour together that summer.

Lyman Alpha Blobs and Science Reporting

July 29, 2006

You may have caught this post at CNN online.

An enormous amoeba-like structure 200 million light-years wide and made up of galaxies and large bubbles of gas is the largest known object in the universe, scientists say.

It’s taken from a quick piece at Space.com.

But I must say that both, especially the excerpts at CNN, left me far more confused than enlightened. In the space of three different (and short) paragraphs, the author manages to describe “amoeba-like” shapes, filaments and bubbles, and notes sizes of both 200 million and 400 thousand light-years, without much to help the reader understand the significance of either.

Here’s a much clearer write up from Universe Today.

My summary is that very large organized structures in the universe were formed first. That is, the larger something is, the sooner it was formed. Organized structures (and that, indeed, is what were talking about here – identifiable regions where the concentration of gas is about 4 times normal), this size (400 million light-years is about 3% of the size of the visible universe) tell us about forces that acted on matter very soon after the universe came into existence.

The distance between The Milky Way galaxy and the Andromeda galaxy, our closest large neighbor in space, is pretty representative of the separation between galaxies in general. It’s about 2 million light years, give or take. The size of these “blogs” is much more like the extent of clusters of galaxies, which varies widely, but can easily reach the hundreds-of-millions-of-light-years mark. This indicates that the structures are probably implicated in the formation of clusters of galaxies, and indirectly in the formation of galaxies.

The Universe in 5-D

July 21, 2006

String theorists talk alot about 6 “hidden” spacial dimensions on top of the 3 we know about. I don’t know about you, but talking about extra spacial dimensions sort of gives me a headache if I try to imagine them.

The simplest way I know to think about such things (once you get past the idea that there’s just no room in the entire universe for an extra right angle in corners), is to think in terms of analogies. If you imagine a sphere passing through a 2-D sheet of paper, what flat-worlders living on the paper experience as the 3-D sphere and their 2-D universe interact is first a point just when the sphere touches the paper, and then a growing circle as the place of contact passes through the paper, until the circle reaches a maximum and begins to get smaller again, until disappearing as a point.

By analogy, if a 4-D hypersphere were to pass through our 3-D universe, we’d experience it as first a point, then a growing sphere which would reach some maximum size and then start to shrink again until it disappeared. That’s how a finite sized hypersphere would interact with our universe.

But there are other shapes that could potentially interact with our universe. What would a hyper-cube, for instance, look like in our 3-D universe? Maybe like this? Get your red-blue 3-D glasses on before you click on the link. How about a 24 sided object? And I’m not sure what you’d call this thingy but if you’re still wearing your red-blue 3-D glasses, be sure to follow the instructions at the top – for “stereo mode” click the anaglyph choice, and click the ‘edges’ box too.