This article will appear in the next issue of The Institute Letter.

Pluto, the ninth planet in our Solar System was discovered in 1930, the same year the Institute of Advanced Study in Princeton was founded (Pluto was originally classified as the ninth planet in the Solar System. In 2005, the International Astronomical Union decided to call Pluto a dwarf planet). While the Institute hosted more than 6000 members in the following 65 years, not a single new planet was discovered during the same time.

Finally, in 1995, astronomers spotted an object they called 51 Pegasi b. It was the first discovery of a planet in over half a century. Not only that, it was also the first planet around a Sun-like star outside our own Solar System. We now call these planets extrasolar planets, or in short, exoplanets.

As it turns out, 51 Pegasi b is a pretty weird object. It is almost as massive as Jupiter, but it orbits its host star in only 4 days. Jupiter, as a comparison, needs 12 years to go around the Sun once. Because 51 Pegasi b is very close to the star, its equilibrium temperature is very high and it is often referred to as a Hot Jupiter.

Since the first exoplanet was discovered, the technology has improved dramatically and worldwide efforts by astronomers to detect exoplanets now yield a large number of planet detections each year. In 2011, 189 planets were discovered (which is approximately the number of visiting members at the Institute that year). In 2012, 130 new planets were found. As of May 20, 2013 the total number of confirmed exoplanets is 892 in 691 different planetary systems.

Personally, I am very interested in the formation of these systems. We have so much information about every planet in our Solar System, but little is known about all those 892 exoplanets. Digging into this limited dataset and trying to find out how exoplanets obtain their present day orbits is very exciting. Many question pop up even by just looking at the first discovered exoplanet, 51 Pegasi b. Why is it a hundred times closer to its star than Jupiter? Did it form further out? Maybe it was not too different from our own Jupiter in the past? For 51 Pegasi b, we think we know the answer: Yes, it did form further away from the star where conditions such as the temperature are more favorable for planets to form. It then moved inwards in a process called planet migration. For many of the other 891 planets, the story is more complicated, especially when multiple planets are involved. The diversity of planetary systems that have been found is tremendous. We haven’t found a single systems that looks remotely similar to our own Solar System. This makes exoplanetary system so exciting to study!

To do this kind of research one needs a catalogue of all exoplanets. Several such databases exist, but they all share one fundamental flaw: they are not open. These databases are maintained either by a single person or by a small group of scientists. It is impossible to make contributions to the database if one is not part of this inner circle. This bothered me because it is not the most efficient way and it does not encourage collaboration among scientists. I therefore started a new project during my time at the Institute, the Open Exoplanet Catalogue. As the name suggests, this database, in comparison to others, is indeed open. Everyone is welcome to contribute, make corrections or add new data. Think of it being the wikipedia version of an astronomical database.

The same idea has been extremely successful in the software world. With an open source license, programmers provide anyone with the rights to study, modify and distribute the software that they have written. For free. The obvious advantages are affordability and transparency. But maybe more importantly, perpetuity, flexibility and interoperability are vastly improved by making the source code of software publicly available. The success of the open source movement is phenomenal. Every time you start a computer, open a web browser or send an e-mail, it involves programs (mostly the important ones in the background) which are open source.

The success story of open source is largely based on the wide adoption of distributed version control systems (the most popular of those tools is git, used by the Linux kernel and many other major open source projects). These toolkits allow thousands of people to work and collaborate together on a single project. Every change ever made to any file can be traced back to an individual person. This creates a network of trust, based on human relationships. Initially the concept of having thousands of people working on the same project may appear chaotic, risky or plain impossible. However, studies have shown that this kind of large scale collaboration produces software which is better (In the software world better is often measured in units of bugs per line of code) and more secure than using a traditional approach.

Astrophysics lags behind this revolution. While there are some software packages which are open source (and widely used), the idea of applying the same principles to datasets and catalogues is new. Extrasolar planets provide an ideal test case because the dataset is generated by many different groups of observers from all around the world. Observations and discoveries are evolving so quickly that a static catalogue is not an option anymore.

To get people excited about the ideas and philosophy behind the Open Exoplanet Catalogue, I started the Exoplanet Visualization Contest. It is a competition with the goal to come up with stunning and creative ways to visualize exoplanet data. We set no restrictions to the kind of submission. The only requirement was that each submission had to use real data from the Open Exoplanet Catalogue. This led to an extremely diverse set of submissions. For example, we received publication-grade scientific plots, artistic drawings of potentially habitable exo-moons and an interactive web-site. One participant went as far as designing a wearable vest with built-in microcontrollers and displays that show exoplanet data. Thanks to a generous outreach grant from the Royal Astronomical Society in London, we were able to give out prizes to the best submissions. With the help of Scott Tremaine, Dave Spiegel and Dan Fabrycky, two winners were chosen.

The second prize went to Jorge Zuluaga from Antioquia, Colombia. He designed a new way to present exoplanet data such as planetary sizes and equilibrium temperatures.
Those are of particular interest when it comes to determining whether a planet is potential habitable or not. His submission, the so called Comprehensive Exoplanetary Radial Chart is shown above. It illustrates the radii of exoplanets filled with colors representing their approximate equilibrium temperatures.
The chart also shows information on planetary orbit properties, size of host stars and potentially any other variable of interest.

The winner of the contest was Tom Hands, a PhD student from Leicester. He wrote an interactive website ExoVis that visualizes all discovered planetary systems. The project makes use of HTML5, Javascript, jQuery and PHP. One can search for planets, study their orbital parameters and compare them to other systems, all within a web browser.

The Open Exoplanet Catalogue is a very new project. The crucial issue is to reach a large number of regular contributors. Then, the quality of the dataset will outperform all closed competitors in the long run in the same way wikipedia is now much more widely used than the Encyclopaedia Britannica. Looking at the success so far, I am optimistic about the future.

More information about the Open Exoplanet Catalogue, its workflow and data format is available at openexoplanetcatalogue.com.

I would like to thank you for allowing me to apply for your job opening.

First of all, I was very happy after I found the job with the cryptic name among all the other jobs on your state’s HR website. Then, I created an account. I chose a username without any special characters but longer than 8 and shorter than 24 characters. Then I came up with a passport that was readily approved by your security system. It only had to include 8 alphanumeric characters of which 2 but no more than 4 are numbers, at least one punctuation sign and an equal mix of upper and lower case letters. After giving you answers to about ten security questions asking me for the first color of my car and my mother’s maiden name, just in case I can’t remember the secure password that I just created, I was given an account on your system. An account that I will most likely never use again after today.

Then I sat down and wrote my CV, my research statement, my teaching philosophy statement, my publication list and of course the cover letter, in which I emphasize that your institution is so much better than all the others I have applied to. Writing these documents was so easy, thanks to the stringent requirements you listed. In particular, I made sure that the margins are exactly 1 inch wide on a letter sized page, the font is Arial, the font size is 12pt and the line spacing is one and a half. The statements were of course within your page limits. Because every other institution has a different set of requirements and page limits, I ended up rewriting everything from scratch. I then merged all pdf files as you requested and named the file according to my e-mail address followed by a 64 letter hex string. The upload was very uncomplicated after I noticed that there is a file size limit of 1 megabyte and I rewrote my proposal once again.

I finally wrote several e-mails to four senior scientists in my field asking them to write a reference letter for me. They probably can’t wait to drop everything they are working on and write a detailed psychological assessment of me. But of course they had to wait for the 32 letter password I had to send them in able to be allowed to create an account on your website. I imagine that their registration process on your website will be equally enjoyable as mine.

In retrospect, I would like to than you for giving me such a great insight in how your institution works. Because assuming that there is at least some correlation between the academic life at your institution and the experience I had during the process of submitting my application, I’m not interested in your job offer anymore. Maybe you could simple say this next time: “Please send your CV and research statement in pdf format to xxx@xxx.edu by November 10. Please also ask for three reference letters to be sent to the same address.”.

Sincerely,
Hanno Rein

Dear Editor,

Thank you very much for asking me to referee this paper. It is very interesting and it almost certainly deserves publication. The level of detail in this work and the performance gain that has been achieved by the authors is impressive.

However, as much as I like the work by these authors, I am unable to referee the article for you. Together with almost 10000 other scientists, I’ve signed a pledge to not publish, referee or do editorial work for any article that is published by Elsevier (http://thecostofknowledge.com/). I strongly believe that scientific publications should be freely available to everyone. Elsevier is the antithesis to that statement and I don’t want to support their business model. The company makes an annual profit of over one billion dollars and has reported a profit margin of 36%. These exorbitantly high figures are due to disproportionately expensive library subscriptions charged to universities and research institutions which are forced to buy journal subscriptions in large bundles. Elsevier itself is adding almost no value in the peer-review process. The work is done by researchers and referees which are mainly paid with public funds.

Fortunately, scientific publishing is evolving rapidly. For example, I’ve already been able to read this paper a month ago because the authors posted it on the arXiv preprint server which is a free, donation supported service run by a non-profit organization.

For these reasons, I am sorry to say that I cannot assist you in this case. I hope that you will find another person willing to referee this paper. If the authors decide to withdraw their paper and submit it to another journal, I’m more than happy to be the referee.

Kind regards,
Hanno Rein

PS: I’m afraid that the ’30 days full access to Scopus, the world’s largest abstract and citation database of research information and quality internet sources’ that you’ve offered me in your e-mail is not going to change my mind.

This post is password protected. To view it please enter your password below:

Password:

I’m an astrophysicist, working on extra-solar planets and Saturn’s rings. An app on exoplanets was therefore an obvious choice as a Hello World program. In August 2009, I submitted version 1.0, which was a simple list of all discovered planets. The app was static, so I had to resubmit a new version to Apple whenever a new planet was found. I wrote the app just for fun. It was never my intention to earn money with it. There is some advertisement in the app, which helps to pay the fee for the Apple developer program ($100 per year).

In November 2010, after 19 iterations the app arrived at version 4.3 and has matured a lot. It has many OpenGL visualizations, animations and push notifications. Looking at the reviews, people seem to like it. That impression is also supported by the download numbers which have reached 200.000.

I receive many e-mails from people, asking for an Android version. With this post I’m trying to explain my biased view on why there is no Exoplanet app for Android yet.

Last month, I even received an e-mail from Angana Ghosh, the product manager for the Android SDK at Google. She asked me about the possibility of developing for Android. I started programming for the iPhone because I have an iPhone, so there was a hurdle because I didn’t own an Android phone. Google was kind enough to send me a free Droid for development. So that problem was solved.

But there is still no Exoplanet App for Android.

First of all, I am really disappointed by the Android phone. Similar to Woz’s experience, I have the impression that every single app is better on the iPhone. If I start developing an Android version of my app now, it would not be as polished and slick as the iPhone counter part. And because I’m doing this just for fun in my free time, why should I spend time on something that will make thing worse. Suppose I build model trains in my free time. Why would I start building another one if I already knew that it will not look as good as the one that I already have.

Second, the development of iPhone apps is really fun. The entire SDK, including the APIs, the IDE, the documentation, the profiler and debugger are great and well-thought-out. The distribution via Apple’s App store is easy and convenient. The whole discussion about app rejections is largely exaggerated. My app got rejected only once, because it crashed during startup. I’m glad Apple takes a look at it before it get’s pushed to several hundred thousand users.

Now let’s compare this with Android. This is an official video for the Google’s App Inventor. Do I need to say anything else? It’s so shockingly bad, it’s hard to find words that describe it. Of course, there is the real Android SDK. But come on, Google. Do we really want to use apps based on this App Inventor? Well, I don’t want to. And that is also a reason why I’m hesitating to be part of the Android community.

Finally, let’s look at the technical side. From a hardware perspective, there aren’t real differences. There is of course the fragmentation issue. But that also exists on the iPhone. One version of the exoplanet app crashed on older devices which ran iOS 3.1 instead of 3.2. But I’m still able to keep track of all devices on which iOS runs and I fixed this bug quickly. That’s just not possible on Android.

I mentioned earlier that the iOS SDK is so great. Here’s why I think so. Objective C is first of all a great language. It’s just C. Plus something on top of it that makes it easier to handle all those buttons and text field that one necessarily encounters in a graphical user interface. I created a few more apps (Hydro, Gravity Tree) which use the same C code that runs on super computers. It’s very easy to glue that code to an interface with Objective C. Many of the new Objective C features are only wrappers. So, these are still C under the hood, and they are fast!

And then there is UIKit, CoreAnimation and all that stuff that makes an iPhone app look and feel slick. If you use the Exoplanet app, go to the Milky Way screen and click the little (i) on the bottom right. It starts an animation that exchanges an OpenGL view (with points sprites, textures and lines that create the milky way) with a WebKit view (which is used to render the text). That’s done with only four lines of code. And it makes a huge difference compared to simply replacing one view with the other. If you sign up for the Apple developer program, you get access to days worth of videos from previous conferences. By watching those, it becomes clear that this kind of thing is the real difference between iOS and Android. A lot of man power has gone into the development of the user interface classes and their optimization. Maybe you don’t mind the pixel perfect transitions or the responsiveness. But I do and I think they are great.

That’s the reason why there is no Exoplanet App for Android yet. Sorry.

Of course, these numbers don’t necessarily mean anything. There are large errors if a person has a common name or alternative spellings. However, it’s interesting to see that there is indeed a correlation (which is labeled main sequence in the plot).

Update: I’ve made an improved version in which the google searches include the words “exoplanet”, “astronomy” and “astrophysics”. The sum of the number of results is used in the plot. This should avoid most of the issues for people with common names. You can download the scripts to create these plots here and fine tune all parameters.

If you feel that someone is missing from the list let me know and I’ll add him/her. If you’re unhappy with your score, please send me bribes in forms of cakes or iPhones.

PS: Stephen, beware! I’m heading towards you!

Exoplanet HR Diagram (follow link for an svg version)

(original version)

Hanno Rein, John C. B. Papaloizou

Many small moonlets, creating propeller structures, have been found in Saturn’s rings by the Cassini spacecraft. We study the dynamical evolution of such 20-50m sized bodies which are embedded in Saturn’s rings. We estimate the importance of various interaction processes with the ring particles on the moonlet’s eccentricity and semi-major axis analytically. For low ring surface densities, the main effects on the evolution of the eccentricity and the semi-major axis are found to be due to collisions and the gravitational interaction with particles in the vicinity of the moonlet. For large surface densities, the gravitational interaction with self-gravitating wakes becomes important.
We also perform realistic three dimensional, collisional N-body simulations with up to a quarter of a million particles. A new set of pseudo shear periodic boundary conditions is used which reduces the computational costs by an order of magnitude compared to previous studies. Our analytic estimates are confirmed to within a factor of two.

On short timescales the evolution is always dominated by stochastic effects caused by collisions and gravitational interaction with self-gravitating ring particles. These result in a random walk of the moonlet’s semi-major axis. The eccentricity of the moonlet quickly reaches an equilibrium value due to collisional damping. The average change in semi-major axis of the moonlet after 100 orbital periods is 10-100m. This translates to an offset in the azimuthal direction of several hundred kilometres. We expect that such a shift is easily observable.

Preprint

http://arxiv.org/abs/1006.1643

Movies

Simulation EQ40050DT:

• Stochastic orbital migration of small bodies in Saturn’s rings – Hanno Rein and John Papaloizou
• Migration of a moonlet in a ring of solid particles: Theory and application to Saturn’s propellers – Aurelien Crida, John Papaloizou, Hanno Rein, Sebastien Charnoz and Julien Salmon

This is a video of one of the simulations (EQ40050DT, to be precise). Make sure you watch it in HD if possible.

It shows a three dimensional simulation of a small moonlet which is embedded in the rings. Due to collisions and gravitational interactions with ring particles, it will eventually undergo a random walk.