A Short History of Nearly Everything

I've been listening to Bill Bryson's "A Short History of Nearly Everything" for the last couple of weeks, and this book is amazing.  Bill Bryson, most known for various humorous travel books, turned his eyes on the history and progression of science.  It's a journey about what we know about the universe across many disciplines (astronomy, physics, chemistry, biology, geology), and how we came to know that information.  The narrative is very compelling, and often similar in style to James Burke's Connections (so if you loved that, you'll love this.)

He uses big questions to drive the narrative.  The first of which is something that seemed like a simple question, "what is the age of the earth?".  It's somewhat surprising to realize that our current answer of 4.5 Billion years wasn't figured out until the 1950s, and that that discovery was intertwined with the discovery of a massive cover up in the lead production industry on the health effects of lead, and would lead to the banning of the substance for fuel and paint. 

You get to see how the chains of science build upon one another, where a new better answer is made based on what came before, and how over time our methods continue to refine themselves.  The stories on the feuds in the dinosaur hunting communities are incredible.  It also goes to show that individuals shape history much more than they are often given credit for.  This is even more true in the fields of science, where a new discovery or insight often opens up massive new industries or fields of study.  None of modern gene sequencing and DNA analysis would be possible had not a curious researcher decided to take samples from Yellowstone's hot springs and on a lark see if anything was alive in the boiling sulfuric waters.  This is even more amazing given that conventional wisdom at the time assured that no life was possible there.  Decades later we discovered that one of those microbes has a curious ability to crank out DNA copies, thus opening up the modern science of genetics.

I can't say enough good things about this book.  It is a perfect, digestible, approach to science literacy.  Your understanding of the universe will be greatly enhanced in the process, and you'll never quite look at a lump of dirt, a wispy cloud, or the night sky again.

Black Holes don't suck

Last night I learned quite a bit about super massive black holes at the Mid Hudson Astronomy Association month meeting in a great presentation by Dr Barry McKernan.  His overwhelming theme of the evening was that "Black Holes Don't Suck", as he wanted to break that myth.  Everything you've seen in science fiction on black holes is pretty much wrong.

A black hole is just a dark star, so small (relative to mass) that the escape velocity from it's surface is faster than the speed of light.  If you converted the earth into a black hole, it would be the size of a grape.  It would still have the mass of the earth.  The moon would still orbit it just the same.  But you'd be unable to see the thing you were orbiting.  The moon wouldn't fall into the earth, as the celestial mechanics don't change.

A very recent (last decade) discovery is that the hearts of most galaxies contain super massive black holes.  The one at the center of the milky way is 2 Million times the mass of the Sun, though only 10s of times as physically large.  That actually makes it smaller in radius than a lot of the brightest stars you can see at night, like Sirius and Betelgeuse.  Barry started the evening with a movie made of 10 years of observations of the center of the galaxy in which you could see stars moving around in arcs.  Zooming in on the data showed one star in particular whipping around something invisible, like it was a planet going around the sun.  This is part of the data that proved the existence of the super massive black hole in our galaxy.

A subset, less than 1%, of these galactic black holes are consuming dust.  Because dust clouds interact with themselves they effectively slow down over time (converting energy into light and heat).  This creates what are called Active Galactic Nuclei.  They are huge beacons of light at the center of galaxies.  Previously we call these things Quasars, Masars, Magnetars.  In the telescopes of 2 decades ago they look sort of like stars (they are point sources of light), but their light curves are all wrong to a be a star.  Now we can actually see both the "star" and the galaxy they are the center of.  If we get to see the disc top down, they outshine the rest of the galaxy by as much as 1000 times.

I asked the question, "if they outshine the rest of the galaxy, what would they look like if you were in their galaxy".  I saw the speakers eyes light up when I asked the question, so I knew I'd hit on a good one.  The answer, there is so much dust in the discs of galaxies that all that optical, uv, and x-ray light would get absorbed, then re-emitted as infrared.  You'd see a big infrared glow in the direction of galactic center.  You'd also probably see radio jets shooting out like spikes perpendicular to the plane of the galaxy.

All in all this was one of my favorite talks at the group to date.  Very informative, and presenting some great science.  If you have any interest in the stars, and are in the Mid Hudson Valley, you should come check out the group.  Friendly folks, spreading science, what could be better.

Microsoft and Linux

A curious thing happened yesterday, a thing that had been feared for years, Microsoft code started down the road to be included in the Linux kernel.  But, unlike the fears of old, it wasn't slipped in in the middle of the night as a secret time bomb.  It was presented at the front door, going to LKML directly.

Well, maybe it was the side door.  As it wasn't actually a microsoft.com post to LKML, it was actually Greg K-H doing the heavy lifting, as Microsoft is working it's patches in via Novell.  Greg is one of the harshest reviewers out there, so in working through him, these should actually be well up to community standards now.  It also shows some street smarts in not running the gauntlet directly.

I'm still not sure what to make of all of this, as earlier this year Microsoft sued TomTom, and forced crippling of the Linux vfat driver to dodge MS patents.  That being said, I'm also of no illusion that MS speaks with one voice.  Big organizations don't do that, and breaking in lawyers to understand open source principals takes a good few years (I know, I've done it before).

It will be curious to see if these drivers make it in to upstream.  There are plenty of good reasons, and many bad, why they wouldn't.  Far more useful features have managed to not make it mainstream in the past, and nothing draws the lightning like Microsoft.  I look forward to seeing how this will play out.

The Quest for the Wolfram Query Language

When I played with Wolfram Alpha a while ago I was really struck by the fact that it's search box was really useless.  There is a lot of guessing involved to try to get the engine to give you any real information that you didn't already know.  It seemed like it would be far more useful if they just published a Wolfram Query Language that would let you define what you wanted to get out of the system, and how you wanted it to be related.  I'm clearly not the only one with these thoughts.

I was reminded of this lesson by a brief perusal of Wolfram Alpha,
the hype machine's latest gift. Briefly: there is actually a useful
tool inside Wolfram Alpha, which hopefully will be exposed someday.
Unfortunately, this would require Stephen Wolfram to amputate what he
thinks is the beautiful part of the system, and leave what he thinks is
the boring part.

WA is two things: a set of specialized,
hand-built databases and data visualization apps, each of which would
be cool, the set of which almost deserves the hype; and an intelligent
UI, which translates an unstructured natural-language query into a call
to one of these tools. The apps are useful and fine and good. The
natural-language UI is a monstrous encumbrance, which needs to be taken
out back and shot. It won't be.

The full post is well worth the read.

Ubuntu One - Cannonical's storage cloud

I'm quite impressed by how agressively the Cannonical team is getting when it comes to cloud computing.  They're integrating eucalyptus into Ubuntu 9.10, which is open source software that lets you build your own "Amazon-like" cloud.  Eucalyptus even implements the same APIs so that all those hundreds of EC2 applications work with it.

But the Cannonical folks haven't stopped there.  They recently launched Ubuntuone, which is a storage cloud.  Anyone running Ubuntu 9.04 can sign up for an invite (I did last month, and just got mine yesterday).  This provides you with 2 GB of cloud storage for free, or 10 GB for a nominal fee.  The mechanics behind Ubuntuone is an applet that's running which synchronizes $HOME/Ubuntu One directory on changes.  It's not rocket science, but it is seemlessly integrated.

At 2 GB of free space, this isn't for keeping media in sync.  It will do a fair job with text documents, and I've started to put my ebooks and pdfs into it for easy reading wherever I am.  I'm also considering redoing my dot files sharing in this manner, though that will mean symlinking into the Ubuntuone directory, as it doesn't seem like you can share beyond it.

Another interesting feature is a "share with others" on those documents.  That opens this up to be a ghetto version of google docs, at least amongst Ubuntu users.  Again, while this is not rocket science, usability is a huge feature here, and the fact that it is so seemless starts to bring a lot of value to having a whole office on Ubuntu. 

This is where I think Cannonical is making a really brilliant play.  Previously Linux on the Desktop was always about being interoperable with other people's stuff, as it was the edge case, and the value in running all Linux on the desktop was low.  With really useful, Linux only, services like Ubuntu One, there is now an incentive to get everyone there.  The Mac folks have been playing this game for years with all their zeroconf tools that work on a local network, and it definitely helped shore up offices of Mac users.

Kudos to Mark and the Ubuntu folks for thinking past just desktop clones and really starting to push cloud as a concept into Ubuntu across the board.  It makes me excited to be both a Linux and Ubuntu user, and I can't wait to see what they add to my platform of choice next.