Buhbye Livejournal

So, much like many people today, I'm going to be cleaning out my journal here and moving entirely to Dreamwidth (and probably not actually post any more than I already do, so most people won't notice), courtesy of the shiny new "it's all in Russian we're not going to even pretend this place is for you" Livejournal ToS.

If you want to follow me there, here's me: https://alinsa.dreamwidth.org/

Moving to Dreamwidth...

Just FYI, with livejournal's latest antics (moving the servers to Russia (not that big a deal) and completely disabling https (BIG DEAL)), I'm moving to Dreamwidth (same account name and everything). Feel free to join me there. ;)

Will probably still crosspost for a bit. Not that I post much at all. Er, ever? Ah, well.

This entry was originally posted at http://alinsa.dreamwidth.org/57885.html. Please comment there using OpenID.

Gravity Waves, Just the Fact(oids)

For those who couldn't make it through the epic tome I just posted. Just the factoids from the gravitational wave announcement:

  • The LIGO project has announced what is pretty solid evidence that they have successfully detected gravitational waves. The video of the actual announcement is here.
  • Gravity is, effectively, distortions in the shape of space, caused by the presence of massive objects (the earth, the sun, black holes, galaxies...)
  • Gravitational waves are a 'traveling' version of that distortion, theoretically detectable at great distances from the original source
  • These waves are created by very massive, fast-moving stellar objects, like black holes or neutron stars
  • The actual waves cause a tiny, tiny distortion in space -- on the order of 1/10,000th of the width of a proton. It takes special techniques to even have a chance to detect them.
  • Gravitational waves were predicted by Einstein (in 1916!) as a consequence of the theory of relativity, and were one of the few remaining unproven predictions of that theory.
  • The event that was detected that led to this announcement happened in September. The intervening time has been used to get peer review and rule out other possible sources of the signal (experimental error, etc)
  • There are two (and will soon be more) LIGO installations, geographically separated, to eliminate the possibility that local effects (like earthquakes) would be picked up and mistaken for actual signals
  • The signals from the two locations, along with the predictions for them, and the two signals overlaid with each other, are visible here
  • The event happened in roughly 200ms, with the majority of the energy being released in the final 20ms
  • The frequency of the event is, if converted to audio, within the range of human hearing. You can listen to it here (two versions -- one at the actual frequency, one shifted upwards so humans can hear it better)
  • This is probably the closest any of us will ever be to actually being able to directly listen to the universe.
  • The event detected happened ~1.3 billion years ago, and involved two black holes
  • The black holes were orbiting each other at roughly half the speed of light during the final moments of their merger
  • The black holes were 29 solar masses and 36 solar masses. The resulting black hole was 62 solar masses
  • The missing 3 solar masses were converted into energy in the form of gravitational waves
  • The majority of this energy was released during the last 20ms of the event, making the energy output of the event briefly exceed the energy output of every other star in the rest of the visible universe by a factor of 50.
  • This is not a weak, "oh, that looks kinda like what we expected, maybe this is gravity waves" kind of signal. The signal has a statistical certainty of above 5.1σ, which means it's almost certainly not some sort of random artifact.
  • The statistical certainty doesn't rule out the possibility that we're measuring some other effect not gravitational waves, but given the extremely close match to the predictions of behavior and scale, that is rather unlikely.
  • Gravitational waves are potentially an incredibly valuable astronomy tool, because they are not blocked by the things that block visible light (dust, bright objects, the universe not being transparent yet right after the big bang), so observations made via gravitational wave can not only supplement other types of observation, but can also makes observations that just aren't otherwise possible.
  • You can read the actual paper here
  • A good article with a lot of deeper info is available at physicsworld
  • This list of bulletpoints has a bit of handwaving to save space
  • I am not a scientist, please don't quote me as an authority, though I don't think I made any glaring errors (beyond handwaving) in the above

Gravitational Waves Gravitational Waves Gravitational Waves

I've been rambling about this to tybarbary for the entire morning, talking about this discovery and translating what actually happened and what it actually means (both from a "what happened and what they measured" perspective and from a "what does this mean for the future of science" perspective); Ty kinda went "um, can you post everything you told me to LJ to I can link the hell out of it, since I can't just summarize this coolness in a few tweets?" ... so here you go.

Disclaimer: I am not a scientist. I like to think that I'm well-informed, though. A lot of what I present here is directly from today's announcement and related things, but some is derived from past knowledge and understanding. Also, the theories involved are not set in stone, so you can certainly find counterpoint to some of the things I say elsewhere on the interwebs ("gravity is a distortion of space!" ... "gravity is a quantum particle!"), but generally what I'm presenting is at least one set of accepted theories.

So, anyhow, it was announced this morning (see the video, which is informative if dry), that the LIGO experiment successfully detected gravitational waves (which I will periodically, if incorrectly, refer to as "gravity waves", because I'm lazy). This is a phenomenal achievement, and gives us a view into events in the universe that are spectacular by any measure. But what did they actually measure, what the hell is a gravitational wave, and why would anyone care?

Well, here's the geeky science:

Back in 1915, Einstein proposed the theory of relativity. One part of the theory suggested that gravity, rather than being a force, was actually a distortion of space itself. Much the same way a bowling ball would distort a trampoline if placed in the middle of it, physical matter distorts the underlying shape of space around it. It is the shape of space -- rather than a directly exchanged force -- that makes the planets orbit the sun, or the sun to orbit the milky way.

This was a revolutionary concept; even though it seems simple, this change in perspective opened up entire new areas of scientific study, and has proven to be one of the most testable theories of all time. If you start with "gravity is a distortion of space," there are hundreds of cascading consequences that come from that, hundreds of places where you can go "if space is curved, then X should happen because of it."

The very first experimental evidence that relativity was right (or at least, more right than what we had, which was Newton's laws of gravitation) came shortly after the original idea was published.

Newton's theories suggested that gravity was a direct result of the masses of the two objects involved, and the equations work out so that something without mass -- such as light, which is made up of massless photons -- would not be affected by gravity. Relativity, on the other hand, suggested that they would be affected by gravity, because light has to move through the same warped space as everything else.

In 1919, there was a solar eclipse, which would allow scientists to directly view stars that were visually close to the sun -- something they can't normally do, because the sun is so blindingly bright. Scientists in several places around the globe very carefully watched those stars as they passed near the sun during the eclipse. If Newton was right, the perceived location of those stars should remain the same as they passed near the sun (thus passing through the sun's gravity). If Einstein was right, the position of those stars would appear to shift slightly as the light from them was bent by the distorted space around the sun. Obviously, the latter ended up being what actually happened... otherwise, we wouldn't be having this conversation!

So that's relativity.

One of the ideas that comes from the idea of warped space is that large, quick changes of the amount of mass there is in one area of space should cause a 'ripple' in space, much the same way a pond ripples if you toss a stone into it. The amount of change required to make a measurable wave from any real distance is immense -- something that only quickly moving, super-dense objects like black holes or neutron stars could be capable of creating. Or, more specifically, those super-dense objects orbiting and then merging into each other.

An interesting thing about gravitational waves is that they are a form of energy, radiating away from the objects and events that created them. And, as far as we know, energy can never be created or destroyed (though it can change forms, as in Einstein's famous equation E=mc²). Where does that energy come from?

In the cases we're discussing here, the energy comes from the orbital momentum of the stellar bodies themselves. All else being equal, two dense stars orbiting each other will never get closer, because they have a (roughly) fixed amount of energy (mass and velocity), and those are the only things that determine an orbit. (This is very hand-wavey and ignores things like drag and tidal forces and probably a dozen other things, but for this discussion, it's adequate.)

But we know, observationally, that dense stars orbiting each other do get closer (and faster) over time, moreso than we can explain if we consider all the other factors we can account for, which means those orbits are losing energy. Gravitational waves were suspected to be that lost energy, and the amount being lost closely corresponded to scientific predictions for how much should be lost that way, which was our first indication that gravity waves might actually exist.

If they exist, what do they "look" like and how can we try to measure them?

Gravity waves would manifest as minute distortions in the shape of space -- reality would stretch slightly in one direction, while compressing slightly in another. How slightly? On the order of 1/10,000th of the width of a proton. Protons are subatomic particles so small that we have no way to view them directly, so a fraction of that size is really, really tiny.

To measure changes this small, LIGO used a technique called Laser Interferometry, which basically uses the interference patterns of laser beams to calculate the relationships between them. As the distance the beams travel changes (because of space being distorted), the interference pattern changes, and that can be measured. A sufficiently sensitive system (which is no small feat) can theoretically measure changes this small.

There's a lot of challenges in those measurements, though -- local events can make equipment and buildings shift, minor vibrations caused by any number of things can create false signals, earthquakes (especially tiny ones!) happen... so to help rule out "local" issues, the LIGO folks actually built two detectors, geographically separated, and are building more.

The announcement made today basically boils down to this: In September of last year, the two active LIGO sites both detected a signal, a fraction of a second apart, that had all of the predicted characteristics of an event similar to a black hole merger, and none of the characteristics of a more locally triggered event. Something extremely local (like a large truck driving by) couldn't create the same event in multiple places, and something larger scale (like an earthquake) can be ruled out by the timing of event arrivals at the two sites.

Here's the signals, along with the predictions for what they should have been, and the two signals from the geographically distant locations overlaid on top of each other, just so you can see how well things actually line up:

The reason you haven't heard about this until today is that the team has spent the intervening months having their work peer reviewed, and ruling out other possible causes for such a strongly correlated signal. At this point, the team has very good reason to think the signal they saw (and several smaller ones after) are the real thing, and not just a "oh hey, this thing kinda flickered like we thought it would, maybe this could be gravity waves?"

The timing and magnitude of the signals also tells us a lot about the events that created them. What we know now is that there were two black holes involved, one which was roughly 36 times the mass of the sun ("solar masses"), the other 29 solar masses. The entire signal lasted for about 0.200 seconds (200ms), with the most energetic part of the merger lasting about 20ms. The resulting black hole had a mass of roughly 62 solar masses.

The astute will notice that 29 + 36 gives 65 solar masses, not 62. Where did the rest of the mass go? That missing 3 solar masses is what was radiated away in gravitational waves. It's like our entire sun got completely converted to energy (via E=mc²) three times over, and that energy was released as gravity waves. In 20ms. That's one hell of a lot of energy.

How much energy is that? It's more energy than every other star in the known universe combined emitted during that same period. BY A FACTOR OF FIFTY. Even on astronomic scales, that's an unbelievably large amount of energy. It's so much energy there's nothing we can even compare it to. Supernovas look like distant fireflies, in comparison.

One thing that's really cool about this event is that the frequencies of the waves actually fall into the range of human hearing, if you convert them into audio samples. You can actually hear this event (with a little help from science to convert gravity waves to sound waves). And here that is:

There's two versions there, one is the actual frequency (which is a low rumble), the other is pitch shifted upwards so that you can hear more of the details. And seriously, go listen to it. You are almost directly listening to an unbelievably violent event that happened 1.3 billion years ago. You are listening to the universe. I have no words for this.

So I guess the last thing I want to cover is... okay, we detected gravitational waves, what good does that do us?

The nice thing about these waves is they aren't stopped or muddied by the things that interfere with optical and radio astronomy. They don't go through all the dust in our galaxy and block our view of the other side. They aren't obscured by brighter things in front of them. We should even be able to detect the earliest parts of the big bang via gravity waves, from before when the universe was optically transparent (which is how far back we can see by looking at the cosmic microwave background). It gives us new ways to study the universe around us, much like radio astronomy did with visual astronomy. It lets us see places we've never been able to see before, and improves our view of places we're able to see. It's going to open up entire new fields of astronomy that have never been possible before. That's awesome.

The paper on this discovery is publically available here if you want to read more of the nitty gritty detail.

I also found a decent article that goes into more of some of the details (and almost certainly better written than my posting) at physicsworld.

It is a great day. Science on, Wayne!

I made a thingy!

Just to be a little different, here's a post that doesn't actually involve depression. I know, whoah, right? Anyhow.

So, I made a book! No, I didn't write a book (as that would require actual creativity, of which I have exactly none), but I did do all the work required to get rowyn's completed manuscript actually available -- all the typesetting work for the print edition, and all the little bits and pieces you need to do to get a book available on all the major ebook platform (Kindle, iBooks, Nook, Kobo). I really enjoyed working on this with her, and having a project to work on really does good things for my mental health.

The book, by the way, is A Rational Arrangement, and is really quite good (and I'm not just saying that because I'm biased -- it's getting excellent reviews!). It's a fantasy polyamorous romance book, with a huge pile of worldbuilding and an absolutely delightful cast of characters. You should check it out! The print version, by the way, is very pretty. Yes, I'm biased. The ebook is pretty, too, of course.


And now, to geek out for a moment. So, something I did for this, which I thought was pretty cool, was I ended up with a build process for the book that was not dissimilar to a software build process.

I did the print layout first, using LaTeX -- a typesetting language that's been around since roughly the beginning of time, which can generate output that's second-to-none.

Then on top of that, I had another build step that postprocessed the LaTeX source to produce the epub that most of the ebook sites want. And then that got a post-processing step to turn it into a kindle book. You'd think that last bit would be simple, but it wasn't, because kindlegen is a massively buggy piece of crap that had me doing half its job for it, just so that it would stop dropping entire sections out of the final book. Sigh.

And then, on top of all that, I also postprocessed the LaTeX file into a series of 141 individual book sections, which are currently being posted three times a week at www.rationalarrangement.com. I wrote a few lines of code to automatically schedule those with Wordpress, so Rowyn doesn't have to do anything to post them -- it's almost entirely automatic! Wordpress is also set up to automatically crosspost and announce the new parts to her Livejournal account, to Tumblr, and to Twitter.

One really nice advantage of doing things this way is that nearly everything in all versions of the book is algorithmically generated, which means it's easy to change things. For example, each chapter in the book has a graphical header that looks something like this:

A lot of times, you'd create something like this with Photoshop or a similar tool, which is great, until you want to change it! And as it happened, Rowyn decided on the day she wanted to release the book that she wanted a different typeface for the chapter headings (I don't fault her for this -- it happens!).

I could have spent the morning in Photoshop going "load header, select type, change font, re-center, save file" 37 times, but instead I just had to change one line in a source file, type 'make', and I had a fresh set of output files with the new headers. It took me maybe three minutes to do. I was pretty happy about that.

Having the unified build system also makes it really easy to fix the occasional issues that crop up -- finding those last-minute typos, a character forgetting how many siblings they have, that kind of thing. Rather than have to fix it three or four different times, I just fix it once, and all is well.

Anyhow, I thought that was neat. Go buy the book if you want to check out all my hard work. :)

UPDATE: I had forgotten that I did a video of the build process in action. This is an earlier version (e.g. it doesn't generate the chapter headings), but it captures a lot of it: