Last edit: 2011 May
Introduction
2010 Hardness is a Standardizable Candle for GRBs
2009 Post a Book-Free "Extraordinary Concepts in Physics" Course to the Web
2008 The Transparent Sun has Focal Point in Outer Solar System
2008 A New Type of Energy: Ultralight
2008 Friedmann Cosmology Equation Can be Written as a Sum over all Energy Types
2006 Post a Book-Free Introductory Astronomy Course to the Web
2005 A New Test of Gravity: Lensing of Gravity Itself
2004 Dark Energy Might be Higgs Fields
2004 The Asterisk*: An Online Astronomy Portal
2004 A Unique Way to Compress Astronomy Images
2002 Best Cadence Rates for Sky Monitoring are Variability Time Scales
2001 CONtinuous CAMeras Deployed to Monitor the Night Sky
2000 GRB Pulses Start at the Same Time at Every Energy
2000 GRB Pulses are Scale Invariant over Energy
1999 The Astrophysics Source Code Library
1995 A Modern Curtis - Shapley Debate: The Distance Scale to GRBs
1995 Astronomy Picture of the Day
1994 Most Microlensing Occurs on Background Stars
1994 Microlensing Can Resolve the Surface of Stars
1994 Time dilation Discovery in GRBs
1994 Dim GRBs are Redder than Bright GRBs
1994 Digits of Irrational Numbers
1993 The Brightness Distribution of GRBs is Consistent with a Cosmological Distribution
1993 The View Near a Black Hole
1993 Ultracompact Neutron Stars
1992 Micro Black Holes Might Explain Everything
1989 How to Compute the Detection Probability of Gravitational Lensing
1985 Microlensing is not observable
Over the course of my career, I have had a few ideas of which I am particularly proud. Since it is hard to see what these ideas are from my resume and even my publication list, I have decided to explicitly describe some of the more notable ones here, as immodest as that may be. One idea behind this is to demonstrate to whomever might be interested, in particular aspiring graduate students, that I am not simply "Mr. APOD", but that in fact I am a practicing and publishing astrophysicist. Another testimony to my being an actual scientist is the list of people who have been awarded a Ph D. under my scientific supervision, which can be found here.
Now some the below ideas may not have been all mine, as some were created in collaboration with others, while for others I did not know of previous precedents at the time when I was working on them. I do realize that even given this list, I will likely be primarily remembered for my role in creating, writing, and editing the Astronomy Picture of the Day (APOD).
Oddly, my favorite ideas have not, in my opinion, led to my most cited papers. At the time of this writing, I have been an author on over 65 refereed science papers as part of over 200 science contributions that have been cited over 1500 times. Now not all of these have been my most exciting ideas -- many times papers resulted from ideas created by others or from more straightforward extensions of existing ideas or paradigms. The paragraphs below have turned out to have quite a few interesting stories, many of an autobiographical nature.
Last, I know that my career is not particularly distinguished. I would guess that most professors could produce a list of ideas and stories at least as interesting as what follows.
2010 Hardness is a Standardizable Candle for GRBs
If attributes of gamma ray burst (GRB) spectra can act as a standardizable candle, then the ratio of the number of high to low energy photons, a much easier quantity to measure, might correlate with it, and also be a standardizable candle. And it does. And hardness is particularly easy to compute for GRBs. This idea was developed in coordination with graduate student Amir Shahmoradi, and the published paper on it is: Sharmoradi, A & Nemiroff, R. J. 2010, Monthy Notices of the Royal Astronomical Society, 407, 2075.
2009 Post a Book-Free "Extraordinary Concepts in Physics" Course to the Web
Students study physics partly because the most strange and interesting parts, like black holes and quantum entanglement, intrigue them. Counter-intuitively, however, I could not find a single university that taught an undergraduate course discussing many of these ideas. And I looked. Almost all university physics courses involve balls rolling down inclined planes, or the potential surrounding a charged rod. So I created the course "Physics X: Extraordinary Concepts in Physics" that focused only on these "cool" and "extraordinary" physics concepts. I have made the lectures in this course book-free and freely available over the web here here. I thank the Physics Department Chair at Michigan Tech, Prof. Ravi Pandey, for allowing me to do. I am not sure that many other physics department chairs would allow one of their professors such leeway. My hope, though, is that other physics departments will take this lead -- and even the lectures themselves -- and provide cool courses for students interested in the truly extraordinary concepts in physics.
2008 The Transparent Sun has Focal Point in Outer Solar System
That the Sun's closest gravitational focus, were it transparent, is inside our Solar System, came up in discussion I had back in the graduate school with Prof. Christ Ftaclas, then at the University of Pennsylvania (now at U. Hawaii). At that time, in the 1980s,I was a standard-issue graduate student who enjoyed discussing hypothetical lensing effects. Ftaclas and I picked up on this again later and even presented a poster at the American Astronomical Society about it in 1997: Nemiroff, R. J. & Ftaclas, C. 1997, Our Sun as a Gravitational Lens, Bulletin of the American Astronomical Society, 190, 38.01 Nobody else seemed to know about it, nor, to be honest, really cared.
Later, when the Internet became prevalent, a more powerful literature search found that this idea was already had by others, as early as 1971 (see the reference list of the next paper). This seems to be an example where many people came up with the same idea independently. Still, the older papers did not have a modern model for the density profile of the Sun. I therefore deployed MTU graduate student Bijunath Patla (now a postdoc at Harvard) to looked into the transparent Sun as a gravitational lens more generally. It turns out that our Sun is a very interesting gravitational lens! Our paper about it is here: Patla, B. & Nemiroff, R. J. 2008, Gravitational Lensing Characteristics of the Transparent Sun, Astrophysical Journal, 685, 1297. Patla did a great job with this and it turned into his Ph D. thesis. By the way, the minimum focal length of the Sun, using modern density functions, is about 23.5 AU, just beyond the orbit of Uranus. There are several spacecraft out at least that far already, so the next step really is to equip a future spacecraft with a detector for radiations that go through the Sun, like gravitational waves and some types of neutrinos. Then the Sun would become a 700,000 km diameter focusing telescope.
2008 A New Type of Energy: Ultralight
When teaching an advanced undergraduate course in astrophysics that reviewed cosmology, I had to admit to myself that I not did I not understand why there were two Friedmann equations. Then, after more thought, I had to further admit to myself that I didn't really understand either of them. So I studied them. As I studied them and I came to understand them better, I soon realized that I didn't understand why normal matter, radiation, and (sometimes) a cosmological constant were represented by individual terms, but that other energy forms, like cosmic strings, were absent. I decided that they could have been included -- they were excluded only because an author didn't believe that they have a significant cosmological density. So, for clarity, I put them back in. Hah! The "explicit" Friedmann Equation(s) then appeared to be a sum over the cosmological constant energy (w=-1), domain walls (w=-2/3), cosmic strings (w=-1/3), normal matter (w=0), and radiation (w=1/3). Does anything seem missing to you? Where are the energy type of w=2/3? Or w=1? Or any energy type of w>1/3? I could not find reference to it anywhere. And I looked. I found that fields, including scalar fields, can have any w, and changing w to boot, but I am referring here to "stable" energy types, not fields. So I decided to include the w>1/3 terms "expanded" version of the Friedmann Equation(s). I called them ultralight energy. I didn't expect much ultralight in the present universe because it dilutes so fast as the universe expands. But it may have been important in times near the big bang. Or not.
I think ultralight is cool stuff and I am proud to have thought of it! Unfortunately, my paper on ultralight was rejected because the referee not because it was wrong, but because the referee deemed it "unmeasurable" and "cosmologically insignificant". So I uploaded my manuscript to the physics archive server, arXiv.org, as a preprint and left it at that. Here is the link: Nemiroff, R. J. 2007, The Opposite of Dark Energy, Limits on w=2/3 Ultralight in the Early Universe, astro-ph/0703737". The preprint is rarely cited. Perhaps defiant, I remain proud of it today: Viva ultralight!
2008 Friedmann Cosmology Equation Can be Written as a Sum over all Energy Types
Please see the above entry. With the help of graduate student Patla, I did publish a "teaching paper" about this here: Nemiroff, R. J. & Patla, B. 2008, Adventures in Friedmann Cosmology, A Detailed Expansion of the Cosmological Friedmann Equations, American Journal of Physics, 76, 265. Oddly, although cited only once, this paper is perhaps my "most downloaded paper", as it has been downloaded from ADS now over 2000 times. If you read closely, I do mention ultralight. (Viva ultralight!)
2006 Post a Book-Free Introductory Astronomy Course to the Web
My physics dept. chair, Prof. Pandey, told me that my university, Michigan Tech, would like to develop more online courses, and so encouraged me to put my pending 2006 Introductory Astronomy Course online. I politely declined, of course, because that would take more time teaching astronomy than I had planned, which meant less time for research. But Pandey was persistent, this time stressing that it would help the Physics Dept., and he also promised that I could get a percentage of the enrollment tuition placed into an open fund which could help fund my research. So then I agreed. Soon I even liked the idea.
Now MTU has some classroom studios all set up, and a savvy director to boot, so this actually turned out to be easier than I thought. I soon decided to make the lectures freely available to anyone with a web browser. I was surprised that I could not to find anyone else who put their video astronomy lectures on the web at that time. And I looked. To my delight, the class soon became one of the most downloaded iTunes courses of any university. I reprised the online class in 2008 using only APOD and Wikipedia because it seemed to me that both were now strong enough to carry an introductory class, and this would be cheaper and easier to access for students around the world. I half expected to be kidnapped by the powerful textbook cartels for undermining their business model, but so far they just ignore me (apparently, that's a national pastime). The free 2008 Astro 101 lectures are still online here.
2005 A New Test of Gravity: Lensing of Gravity Itself
No one knows if gravity can lens gravity itself. The earliest root of this idea that I can remember was a question I asked when taking particle physics as a graduate student at the University of Pennsylvania in about 1980. I asked if virtual particles were refracted by the dialectic indicated on the black board, as were real particles. Prof. Segre answered that no, they were not, but he did seem a bit troubled by this at the time and praised the question. The idea next surfaced when I was a postdoc in 1988 and I asked a mentoring professor (Prof. Tsvi Piran at the Hebrew University of Jerusalem) whether virtual particles were deflected by a black hole. As I recall, we were waiting for our bags to arrive in an airport carousel when I finally got up the nerve to ask him. He quickly answered "no." A few minutes later, however, as we dragged our bags through an airport toward a conference, he said it might be more complicated than he originally thought, and said he didn't really know. As before, while learning more and more physics (one typically learns the most physics only after one is already a professor), I have kept an eye on this question. I have never come across physics that gave me a full and satisfying answer. And I looked. So in 2005, after I was tenured, I published the idea. (Note to the reader: please don't publish such speculative ideas unless and until you have tenure -- you may be labelled a unemployable crackpot!) The paper is Nemiroff, R. J. 2005, Astrophysical Journal, 628, 1081.
To my surprise, the referee and editor had only minor quibbles with the manuscript -- I had little trouble getting it published. Some members of the gravitation community, however, seemed to find the idea uninteresting. I don't understand why -- perhaps they are right and the idea is uninteresting, but I haven't yet found anyone to tell me why, or perhaps the answer has more to do with sociology. Regardless, this idea is actually falsifiable. The paper actually takes this effect beyond idle speculation and into a realm where it can be tested and perhaps limited. This test appears to me well different than any yet existing way of testing gravity -- something one might consider exciting. Still, the paper is not highly cited.
2004 Dark Energy Might be Higgs Fields
Since coming to prominence in the late 1990s, I have been continually intrigued by the concept of "dark energy". Because I am a inherently a vain person, I thought that I myself could help humanity to better understand it. Unfortunately, I was handicapped by my knowledge of particle and string physics not being on a research level. Then again, however, if one bites off only a small piece of a puzzle, and analyzes only that in some detail, one still might make a contribution.
So when reading up on dark energy, I noticed two seemingly juxtaposed ideas. The first was that dark energy was some sort of field, say a simple scalar field, that had an unknown origin. The other was that the Higgs fields that create inertial mass might be visible in the universe, but seems not to be. So, being a simple person, I decided to see if the Higgs field could be the very field that creates dark energy. Even before this, I had the whim that perhaps a field that actually creates gravity was the dark energy, but this seemed to vague to pursue.
I intrigued graduate student Patla, who loved to discuss theoretical ideas, to look into this with me. We wrote down the key Higgs equations and found that it was mathematically possible that they could evolve into dark energy, but that "tooth fairies" were needed to force the equations to do unexpected things. Now one might expect that any mathematical framework that involves a "tooth fairy" would be discredited and unpublishable, but that would be wrong. It turns out there are many "tooth fairy"-heavy physics papers all over the literature. I am not making that up. Which fairies you tolerate appears to be partly a matter or taste. Unfortunately, the tooth fairies we invoked did not fit with the tastes of the referee, who rejected the paper, although not maliciously. If I had more interest and initiative, I might have persevered, argued longer, and possibly gotten it published. But I did not. Therefore, this too ended up as an arXiv preprint only manuscript, found here: Nemiroff, R. J. & Patla, B. 2004, Decaying Higgs Fields and Cosmological Dark Energy, astro-ph/0409649. The manuscript is only lightly cited. One day I do hope to return to this idea.
2004 The Asterisk*: An Online Astronomy Portal
After the Night Sky Live (NSL) sky monitoring network (described below) went online with its associated web pages, I thought it would be useful if people viewing the network images could communicate. I envisioned that people would exchange software updates, science results and software, future sky camera hardware ideas, and consult each other about unusual things being seen in the images. I was helped greatly in this endeavor by Lior Shamir, a graduate student who is excellent with computers. After a short while I realized that few people really cared about discussing the images, rather they just wanted to see the most recent ones near their location.
Disappointed, rather than fold the online bulletin board, I reformatted it. I renamed it the Asterisk*, primarily so that people could comment on APODs. The name was meant to convey that each of these comments could be considered to be an asterisk (*) to those APODs. Also, the "aster" part of asterisk means "star". My hope was to leverage APOD's popularity and drive people to know about and contribute to discussions of the Night Sky Live project and images as well.
As usual, I was wrong again. Even with this wide on-ramp, few people seemed to care about commenting on any aspect of the Night Sky Live project. The APOD discussion board section, however, seemed to be going quite well, and some knowledgeable people began posting there and even answering questions about the posted APOD image and even astronomy.
Therefore, a few years later, as the NSL project was concluding, I had closed all NSL discussion and fully moved the Asterisk into a bulletin board that would bolster APOD in as many aspects as possible. In 2010, I renamed the bulletin board Starship Asterisk, trying to encourage new people to post by envisioning the online board as a space ship. My most recent thought is to keep all of this but also open up more sections, like ASCL, dedicated to more professional astronomers. One hope is that this may bolster amateur sky enthusiasts. The main page for the Asterisk is
here.
2004 A Unique Way to Compress Astronomy Images
A graduate student of mine who specialized in computer science, Lior Shamir, once seemed insulted by the idea that astronomers might figure out a better way to compress images than computer scientists. Image compression was a mature field and competitive field, he related, driven by increased revenue to big corporations that could compress and uncompress information efficiently. Still, it seemed to me that astronomy images, in particular ones that had dark backgrounds and bright stars, might be unique enough to utilize their own compression algorithm(s). The background for this was our own sky monitoring project, Night Sky Live, which was generating hundreds of images every night that were being sent over to our main computer at MTU, and using lots of bandwidth in the process. After convincing him of the uniqueness of the approach, Shamir and I brainstormed and came up with an astronomy specific method to maximize image compression, in particular a photometry specific method.
A key part of the best idea we came up with was to do a preliminary pass on the digitized images and find the stars. Assuming these locations were only a small fraction of the image, the star image pixels could be recorded exactly as measured, whereas the dark background could be highly compressed. Since in many astronomy images stars make up only a small fraction of the total image area, this technique could yield very large compression rates while leaving the ability to detect changes in stellar brighness untouched. Shamir and I were unable to find anything like this published anywhere. And we looked. So we coded this up, tested it, implemented it in practice as part of our Night Sky Live sky monitoring project, and wrote up a paper. The paper is here: Shamir, L. & Nemiroff, R. J. 2005, PHOTZIP: A Lossy FITS Image Compression Algorithm That Protects User-defined Levels of Photometric Integrity, Astronomical Journal, 129, 539.
The knowledgeable but anonymous referee did inform us of a precedent paper in a conference proceedings many years ago (Press 1992), which is cited, but significant differences remained. This paper appears to be quite useful for the astronomy photometry community, but is rarely cited -- I'm not sure why. The routine, remains available through the Astrophysics Source Code Library (ASCL), described below.
2002 Best Cadence Rates for Sky Monitoring are Variability Time Scales
In the early days of sky monitoring, in particular when the Night Sky Live project that started in 2000, I decided to present a conference paper on the best cadence rates to detect things like GRB optical afterglows and microlensing. This seemed simple enough. I soon realized that I had little idea how this should work -- my preconceptions were wrong. Still, I kludged something together that seemed OK at the time (I mean, how hard could this be?), and presented the conference paper anyway. I don't think that conference proceedings were ever published. I later determined that was good, because I came to again realize that I still didn't know what I was talking about. Determining optimal cadence rates seemed trickier than I thought, at least for me.
Well after the conference, I tried again to revisit the idea with a new determination to understand myself what must be obvious to others. But I kept getting it wrong. I could find nothing written on the subject. And I looked. Finally, slowly, one glaring conceptual error after the next, I realized how to do it right. It was right because it just made so much sense. So I wrote up a paper. I was very proud of my work, and I worked very hard on this paper, trying to make it as general as possible, and even included examples from major sky monitoring projects that might soon exist. Surely my work would be met by warm hugs and gratitude. And who better to administer the first warm hugs and gratitude than my scientific mentor Paczynski, who identified himself as the referee of the paper.
So Paczynski rejected the manuscript. This was particularly unnerving since previously it was hard to find places where we had a sustained disagreement. Paczynski said that sky monitoring was too dependent on details of specific site details to make the generalizations that I made in my manuscript. Surprised, I tried to argue but he didn't want to hear it. He suggested that the only way the paper could get published is with a new referee. The next referee, Prof. Andrew Gould, who also identified himself, was also an acquaintance. He was also somewhat skeptical, but after a few pitched (but polite) exchanges, I was able to convince him that this was good stuff, or at least publishable stuff.
Yes, In my biased opinion, this is good stuff. There are times when I think something over and over and over and eventually it becomes so simple I have a hard time explaining it to anyone because it has all become just so obvious. This was one of those cases.
Additionally, over the years I have come to realize at this is also important stuff partly because it is at the base of a classic conundrum common to many aspects of life and science: how often do you check for something? How often does a warthog check for nearby tigers? How often do you look out the window to see if your ride is here? Etc. Determining cadence rates is fundamental, and astronomical applications are not only similar but keys to telescope efficiency. The published paper is here: Nemiroff, R. J. 2003, Tile or Stare? Cadence and Sky-monitoring Observing Strategies That Maximize the Number of Discovered Transients, Astronomical Journal, 125, 2740. I don't remember who the editor was that accepted the paper even over Paczynski's objections, but (s)he deserves warm hugs and gratitude. To my delight, this paper does get cited a bit, and it is downloaded quite a bit more. Still, my endless vanity being what it is, I thought it would be cited more.
Epilogue: When I was an associate professor, a well-respected full professor here at MTU (Prof. Kostinski, atmospheric sciences) stopped by my office and said that he thought I might be ready to apply for full professor status, but first he wanted to see a good paper that I had written since becoming an associate professor. I showed him this paper. A few days later he returned to my office, said he thought this was good stuff, and encouraged me to apply for full professorship. I was indeed promoted. That was quite gratifying!
2001 CONtinuous CAMeras Deployed to Monitor the Night Sky
To actually monitor the entire night sky required real hardware. Almost all astronomy hardware at this time consisted of small field of view, high magnification devices. I wanted to do just the opposite -- I wanted to deploy wide field of view devices that monitored and recorded as much of the sky as possible, as often as possible. This, I felt was the best way to monitor variability on the night sky. And this variability could include the slight variability of stars to the great variability of novas, supernovas, and the optical transients to GRBs.
For this I was fortunate to pique the interest of (then) graduate student Wellesley Pereira and MTU colleague Prof. J. Bruce Rafert. Rafert and his own graduate student were putting much effort into developing the MTU 16" telescope into a classical research telescope. If you know me, though, you might realize that is a bit too normal for my tastes. Although perhaps Rafert thought my sky monitoring project as a bit strange, he continually provided much expertise, particulalry when it came to actual sky observations, because, well, I was nearly clueless. On request, Rafert would come to my office and lecture me on how real photometric measurements were actually made and computationally reduced.
At first Pereira and I designed and built a device designed to stare at Earth's northern spin axis -- near Polaris -- and keep recording the surrounding and changing sky in that direction. With this strategy, the camera mount would only need to spin and not move about and find locations all over the sky. Fortunately, my NSF CAREER grant was worded vaguely enough that it could be adapted to pay for the CCD. We set up our wide-field CONtinuous CAMera 1 (CONCAM1) on the edge of the roof of MTU's small college observatory. As planned, we took many pictures of the same part of the night sky. Still, there was too little data to do any real science at that point. Also, it also appeared that it would be too much money to order the parts for, and build, the multiple CONCAM1s needed to monitor the entire sky all at once. It was when contemplating the corner I had painted us into when I had what I thought was a better idea -- fisheye CONCAMs. Fisheye CONCAMs would see only the brighter stars, but they could monitor the entire sky on a much more limited budget.
To the best of my knowledge, nothing like this had ever been done. Moreover, it seemed counter to any observational astronomy approach of which I was aware. Still, as usual, I did an extensive search for previous similar projects, and during this search, I did find a precedent. In the US southwest, there was deployed one US Army fisheye digital camera that was being used to monitor clouds. I therefore emailed a person listed on that web page. From the web pages and the emailed response, from what I could tell, although this camera did record some stars, their setup was not equipped for astronomy and in particular, appeared to do no astronomical photometry. And their setup seemed to cost well over $100K. After some thought and investigation, I thought I could design and deploy a real astronomical camera for much less.
I soon found myself searching hardware sites for components and doing back of the envelope estimates trying to get the output from any commercial fisheye lens to fall onto the observing area of any commercial CCD. At first they seemed all to be mutually exclusive -- all fisheye lenses I could find only made image circles well wider than listed CCD dimensions. Also, the focal lengths of fisheye lenses all tended to be too short for practical use. Then I lucked onto a web site that advertised a defocusing lens, and I realized that putting one of these in the optical path could increase the focal length just enough to enable a practical device. And the costs were not excessive, particularly if I used a common commercial fisheye lens and other common commercially available parts. I ordered the parts and deployed Pereira to the lab to make sure this really worked. It did. So then we gave all the parts and a water proof case to the Physics Department's industrious machinist, Dave Cook, and asked him to make it all fit together. My mantra is that I wanted light to go in the top, power to go in the bottom and data to go out the bottom of the box. I also wanted no moving parts. Cook did a masterful job, and we had our first "observatory in a briefcase" done, which we then called CONCAM1 (again).
I was quite proud of our new CONCAM1. It so happened that Pereira and I were soon on the road to a conference in Santa Barbara, California, so we decided not only to show off our CONCAM1 at this conference, but to go to the headquarters of the Santa Barbara Instrument Group (SBIG), and demonstrate how their ST7 CCD was being used in our novel device. SBIG president RIchard Holmes was quite accommodating, and he seemed to admire our novel concept and design. Sadly, he was not so impressed that he offered any substantial discount on future SBIG CCDs, though. In retrospect, I believe he saw this as an opportunity to sell even more of his SBIG CCDs. Holmes did come in useful on some occasions, though, when trying to troubleshoot our future CONCAM devices.
The following year had me spending significant time trying to control SBIG CCDs and data flow with Linux, as this gave us full control of shutter, dark frame, and image characteristics. Also, Pereira, Rafert, Cook, and I kept upgrading our CONCAMs to larger CCDs and fisheye lenses with greater throughput. The very next step was a device we called CONCAM2, which was ready to deploy somewhere in the world other than the cloudy winter skies near Houghton, Michigan. Thus started our next great adventure: trying to find a perch.
Social connections are important in any field, and in this case it was only through an acquaintance I had made in my previous life as a postdoc researching GRBs at NASA's Goddard Space Flight Center that made the key difference. This acquaintance, Dr. Scott Barthelmy, knew of a building that at Kitt Peak National Observatory that NASA could well make better use of. In a previous incarnation, the building housed something called the Explosive Transient Camera that used a bank of CCDs too look for transients on the night sky. So here was a case of another precedent that I had missed. Since this project had concluded, Barthelmy was looking for new projects that might make good use of their Kitt Peak Building. Putting our CONCAM on the roof of this building seemed like a natural. The site had dark skies, high bandwidth, power, and was in the US. After Barthelmy realized that I was incapable of bolting two screws together, he was the one who actually installed our CONCAM2 on the roof. That night, in April 2000, we took the first fisheye images of our all sky monitoring project.
As usual, I didn't understand the immediate result. The camera appeared to be recording bright stars, but oddly there were faint bands of light running across the images that looked a little like really long clouds that were mostly transparent. I reasoned there was something wrong with our instrument. Later, I found out we were seeing ripples in the atmosphere called gravity waves. Then something unexpected happened yet again that would change the course of the project, yet again.
At the Kitt Peak canteen, someone I didn't know came up to me, said that he heard that we were taking wide angle night sky images, and asked to see one. Apparently he wasn't interested in the stars in the field, he was interested to see the CLOUDS in the field. I guess he wanted to know how good his observations of the previous night would be. So CONCAM2 made an immediate splash not as a star monitor or a transience monitor, but as a cloud monitor. Interesting. The CONCAM project now had a better selling point.
Now hyped as a cloud monitor, we were soon able to get CONCAM2s -- and then CONCAM3s -- up on almost every major observatory in the free world. We charged $20K. a device. Locations included Mauna Kea (Hawaii), Wise Observatory (Israel), Canary Islands, Cerro Pachon (Chile), Siding Spring (Australia), South Africa. At its peak, the (now renamed) Night Sky Live (NSL) project had eleven (11) CONCAMs running simultaneously all over the world, reporting images back to NightSkyLive.net at MTU continuously. Notable collaborators included Prof. Noah Brosh at Tel Aviv University, and Dolores Perez-Ramirez who was a postdoc with us for a while. Pereira and Shamir also leveraged the project and data for their Ph D. theses.
What I didn't expect is that in order to keep the NSL project running efficiently, I would slowly have to shift from being a scientist to being a technical consultant. Fortunately, after Pereira graduated, graduate student Lior Shamir arrived and did most of the programming the ran the NSL network. Even so, and even with no moving parts, these CONCAMs kept running into a series of (usually minor) problems and the host observatories were counting on us to solve them. Just keeping the network up and running was becoming a full time job.
Sadly, this was a job I might have kept for at least a little while longer, but could not. Although many of the observatories housing CONCAMs indeed liked having a real time all-sky cloud monitor, more and more of them felt that they could create and deploy their own CONCAM-like device themselves, for less money than we were charging, and have more internal accountability to boot. Even so, I got enough observatory collaborators together to cobble together an ambitious grant proposal to the US National Science Foundation (NSF) for a new round of even better sky monitors, dubbed CONCAM4s. A proposal that was was turned down. Without new money, the NSL project was doomed. So we kept recording as much data as we could, handed off the existing CONCAMs to their host observatories, and gradually shut down the project.
Although the Night Sky Live project did not survive, I find solace in that it created or bolstered precedents that are still important in astronomy today. For example, many major observatories now have their own fisheye cloud monitors, which make their own major telescopes more efficient. Also, the drive to continually monitor the entire night sky has only gotten stronger. So much so that, in retrospect, perhaps the fisheye approach to CONCAMs might not have been ambitious enough.
In terms of productivity, the project did produce several papers and several firsts, in my opinion. For example, Tte NSL project was the first project to monitor most of the night sky, most of the time. A conference paper announced this here:
Nemiroff, R. J. et al. 2003, Expanding Fisheye Webcam Network Now Capable of Monitoring Most of the Night Sky, Bulletin of the American Astronomical Society, 202, 3.03 . The NSL project was the first to trigger on a (possible) rapid optical transient, lasting only minutes, in real time. This possible transient was seen by two different CONCAMs in different parts of the world at the same time, and resulted in this paper:
Shamir, L. & Nemiroff, R. J. 2006, OT 060420: A Seemingly Optical Transient Recorded by All Sky Cameras, Publications of the Astronomical Society of the Pacific, 118, 1180. Years of CONCAM data analyzed later to create the best limits on naked eye optical transients yet made. That paper, where the statistics are trickier than they look, is here:
Shamir, L. & Nemiroff, R. J. 2009, Frequency Limits on Naked-Eye Optical Transients Lasting from Minutes to Years, Astronomical Journal, 138, 956 .
2000 GRB Pulses Start at the Same Time at Every Energy
Some ideas appear so obvious that it is hard to believe that they aren't common knowledge. Conversely, if one looks at data in ways that aren't common, sometimes obvious things may just pop out. This is one of those cases.
Partly to check if gravitational lensing commonly occurs in gamma ray bursts (GRBs), my colleagues and I continually printed out prompt emission light curves. Since lensing is supposed to be achromatic -- the same at every color -- I developed a simple program that would just print out a time-aligned version of all four energy bands of recorded gamma-ray emission from the BATSE instrument on board NASA Great Observatory Compton. Given my luck, of course, I was not able to find any evidence of gravitational lensing. (Actually, I was able to find comedic solace in that this appeared to be a case where the entire universe WAS conspiring against me.)
Along the way, however, I stared at quite a few GRB light curve plots, all stacked in energy so that they their start time were all co-aligned. Hundreds of plots. Eventually, without running a single computer program or doing any analyses other than holding up the straight bottom of a stapler, it seemed quite evident to me that when a pulse in a GRB begins, it begins at all plotted energies at the same time. Now since GRB pulses tend to overlap each other, most cases were not so clear, but whenever a pulse seemed to begin in isolation, this seemed to hold up.
I asked people about this, assuming this was common knowledge, but nobody seemed to know about it. I could find nothing written about this. And I looked. So I found some relatively clear cases of this, branded this as the "Pulse Start Conjecture" and published it as part of a greater paper here:
Nemiroff, R. J. 2000, The Pulse Scale Conjecture and the Case of BATSE 2193, Astrophysical Journal, 544, 805.
The title of the paper refers to another GRB characteristic, potentially even more interesting, that is described below.
2000 GRB Pulses are Scale Invariant over Energy
Another facet of GRBs, as indicated above, appeared to be more subtle and hence surely more controversial. The same printouts of GRB pulse light curves as measured in different energies, when held up to a light and tilted just right, seemed consistent with the idea that GRB pulses really had the same shape at all energies. It other words it seemed, upon inspection, that the only real differences between GRB pulse light curves at any two energies was scale factors on the X and Y axes. Therefore, were a light curve at one energy somehow printed on rubber graph paper, it could be stretched in X and Y to fit right on top of the same GRB pulse measured at another energy. This seemed to me to be saying something about GRB pulse physics, although exactly what I didn't know. As usual, I could find nothing written about this. And I looked. So I first presented it at a meeting, and then wrote it up. The paper was the same paper that I announced the Pulse Start Conjecture (discussed above) and can be found here:
Nemiroff, R. J. 2000, The Pulse Scale Conjecture and the Case of BATSE 2193, Astrophysical Journal, 544, 805. The paper gets a lot of reads but only a moderate amount of citations, although the pace appears to be picking up.
1999 Continuous Monitoring of the Entire Night Sky is Now Possible
I expect this entry to be a controversial one. This idea starts with a daydream: the Starship Enterpise from Star Trek (tm) is on a mission and needs to know the history of a common star on their main view screen. A bridge officer says that data on that star is only available back to ... some date. The daydream ended then because it occurred to me that this date had not yet occurred. At the time of this daydream, to the best of my knowledge, there was only data on specifically interesting stars, and single archived plates and prints of much of the sky taken about 50 years ago. There was, at that time, to the best of my knowledge, no sky monitoring program that would regularly record the brightness of a random nearby star. So I thought about what would be needed to start one.
This search did turn up some history. For one, a conversation with Prof. Paczynski indicated that he had already been thinking along similar lines, and referred me to a conference proceedings he had written the previous year that I had not seen.
As an aside, this is not the first time that I presented Prof. Pacynzski with an idea I thought was original but that he had thought up previously. This time I remember relaying to him another daydream that seems to recur at the beginnings of some of our discussions. In this daydream, I tell him, I am slashing through the dense forests of some faraway place in the world, hot on the trail of some valuable treasure. Yes, it's a bit lit Indiana Jones. Anyway, following my tattered map, I feel that I am finally getting close to the treasure when I come across a clearing. In the clearing sits Prof. Paczynski. He is already searching for the same treasure, but he has a better map and actual hot food. "Oh there you are," Prof. Paczynski tells me. "I have been expecting you. It is good to see you. Have a seat -- there is much work to be done!"
Actually realizations that Paczynski (or someone) had previously similar ideas were not totally disheartening. Although my pride and vanity were dinged, I was happy to see that this idea was actually a good one. Sometimes, OK many times; OK most times speculative ideas I follow end up crashing spectacularly (I keep a folder of them, by the way). Sometimes the idea just does not work out at all, sometimes one of my premises is wrong (or way wrong), sometimes I rediscover what should have been obvious, sometimes my math is faulty, and sometimes I missed a key word that would have shown me that this subject has been thought about for years already.
Anyway, the search also turned up an existing sky survey that even Prof. Pacynski didn't know about. In turned out that the GROSCE project (PI: H. S. Park, a wonderful and thoughtful woman working at LANL) had been monitoring the sky as part of the GRB afterglow project since the mid 1990s. When not chasing GRB afterglows, the GROSCE wide angle cameras would tile the sky and record data. To the best of my understanding, this really was the first sustained sky monitoring project. The GROSCE data was then put on tapes and placed in a filing cabinet. I actually got to see many of the tapes piled up in the filing cabinet in LLNL. Park and I discussed ways of "liberating the data" so that scientific analyses could be done. I therefore joined Park's project and thought about trying to make their data more generally available. We continually rediscovered, however, that this would be an expensive and time consuming undertaking.
Nevertheless, I decided to write a paper on this and try to actually quantify how difficult it would be to monitor the sky down to almost any limiting magnitude. Paczynski's conference paper in Japan, which I finally located, turned out to be mostly qualitative and discussed what discoveries such a survey might make. I myself had taken for granted that many discoveries would be made because the universe was so variable and because so little time monitoring of the sky had ever been done. Therefore, what I wanted to explore was how many telescopes would be needed, how much data must be recorded, etc. to actually get a useful sky monitoring project to work. I could find nothing written on the subject. And I looked. I therefore worked with a Prof. Rafert who was in my department and had a better understanding of telescopes and practical observational astronomy than I had. The paper is here:
Nemiroff, R. J. & Rafert, J. B. 1999, Toward a Continuous Record of the Sky, Publications of the Astronomical Society of the Pacific, 111, 886. I believe that Prof. Code was the referee, who made some good suggestions.
In retrospect, however, it now seems to me that this paper was not ambitious enough. Our analysis only went down to a visual limiting magnitude of 20. Sky monitoring has now become big business, with many efforts now attempting records of the sky down past visual magnitude 25. I did not think the community would put THAT much money into this type of effort. I was wrong. Oddly, even though this paper was published openly and even though I circulated the paper at an Aspen Conference on Sky Monitoring a few years later, the current sky monitoring community ignores it. I am not sure why. For example, I don't think the PanSTARRS or LSST community has ever cited this paper. To my delight, though, to the year he died, Prof. Pacyznski cited this paper almost every chance he got.
1999 The Astrophysics Source Code Library
Computers have quickly changed science, but some aspects of science publishing are slower to change. It used to be, before computers, that when you published a paper, you gave enough information to make your results reproducible. Now so much science is done with computers that creating a separate computer routine to reproduce the results in a paper is prohibitive. Therefore, to reproduce those results, the computer code(s) should be made available, in my opinion. People could then not only run the code, but inspect the source text to see what approximations were being used and how algorithms were being implemented.
Unfortunately, very few papers make available the computer codes that enabled them. Oddly, at this time, no real venue existed for presenting these codes to the public. And I looked. So I decided to create the Astrophysics Source Code Library (ASCL). I recruited a friend and fellow astrophysicist Prof. John Wallin to help, but as time went on he acted more as an advisor than an editor or contributor. The announcement of this library is here: Nemiroff, R. J. & Wallin, J. F. 1999, The Astrophysics Source Code Library: http://www.ascl.net/, Bulletin of the American Astronomical Society, 194, 44.08. Later, Dr. Philip Helbig volunteered to help out, but soon he went on to bigger and better projects as well. A conference poster on ASCL involving him is listed here:
Helbig, P. & Nemiroff, R. J. 2000, Current Status of the Astrophysics Source Code Library, Bulletin of the American Astronomical Society, 197, 116.05.
Initially, we opened a web page http://ascl.net/ for this, and solicited source codes for a while. I figured this would be popular. I figured that people who received grant money to develop codes would want to gain as much exposure for these codes as possible. I figured that astronomers would want to show the world their coding skills and allow their codes to be used by others. I figured wrong.
Getting people to post their codes in ASCL was like pulling teeth. Nobody submitted anything. We had to go out there and get them ourselves. Codes, it turned out, tended to be surrounded by their own insular communities, many of which have the attitude that everyone important knows about their code(s) already. After helping to gather about 40 codes, I drifted onto other projects, and looked for someone to step forward as a new ASCL editor. In 2010, in an effort to make the Asterisk into a better general astronomy portal, I moved the ASCL codes there and used the help of an intelligent volunteer (Alice Allen), to help me build the library. In 2011, Prof. Peter Teuben and (former student, now Professor) Shamir have stepped forward to help this new Asterisk version of ASCL.
1995 A Modern Curtis - Shapley Debate: The Distance Scale to GRBs
In 1994, I became concerned that being right in science is somewhat irrelevant. What is more important is to have good connections. When investigating this, I decided to check into the famous Curtis-Shapley debate on the Nature of the Universe. It seemed to me, at the time, that Shapley was wrong but hired on as the head of Harvard Astronomy, whereas Curtis was right but remained obscure. I wanted to dig deeper in this to see if being right had helped Curtis at all. And why did Harvard hire the guy who was so wrong?
The background of this was the current raging debate about GRBs. Some people, like me, thought that GRBs occurred far away, at cosmological distances. Others thought that GRBs occurred in our own Milky Way Galaxy. It seemed to me clear that our side was right -- GRBs had to be far, far away. But it also seemed to me that that I would soon be cast out of astronomy because I was not prestigious enough and my work was sub-par. The other side, most seemingly secure at the University of Chicago, would be wrong wrong wrong and more wrong but either currently employed or soon to be employed by prestigious institutions in prestigious jobs. Now that just didn't seem fair. I therefore wondered what really happened the last time something like this came up -- whatever became of Heber Curtis?
Therefore in September 1994, I began checking into this in some detail. One seemingly extraneous piece of information I came across was that the Curtis - Shapley debate occurred on April 16, 1920. I wondered how long ago that was. To my surprise, would be exactly 75 years the next April. I actually had to do the arithmetic several times because I was sure it couldn't be such a significant anniversary. But it turned out that the coming April was indeed the 75th anniversary. Could a new and similar debate be set up? Could a modern version of the Curtis - Shapley debate be staged?
I quickly interested my officemate (Dr.) Jerry Bonnell, and as we started checking into more and more details, things fell into place nicely, at least at first. We found that the 1920 debate actually occurred on April 26, but that it happened right in Washington, DC, the nearest city to Greenbelt, MD, where Jerry and I worked (most usually) on GRB research. After a false start, we even traced the debate to the Natural History Museum and found that the auditorium still stood and was available, if need be, for a new debate.
I drew up a candidate "dream team" of people who could be involved in this modern GRB anniversary of the classic 1920 "Nature of the Universe" debate between Shapley and Curtis. Of course I was not prestigious enough to present the cosmological point of view, so I invited my mentor Paczynski. It was clear that the Galactic point of view should be championed by Lamb. Rees liked to volunteer, at that time, his neutrality, so he would make an excellent moderator. I recall that I had to be careful to invite these people in a very specific order so that others would be more likely to agree.
Paczynski related to me that he invented the GRBs-are-cosmological idea himself because John Bahcall asked him to speak about "something" at an IAS lunch and this crazy idea seemed likely to promote debate. Actually, the idea that GRBs might be cosmological preceded Paczynski (by van den Bergh, for example), but he didn't know that at the time. Oddly, in the run-up to the debate, it appeared that my group (Norris, Bonnell and myself) were keeping better tabs on the latest GRB results, so that sometimes I felt the need to brief him on recently relevant papers. Conversely, as described below, I would not have been studying GRBs at all if it weren't for Prof. Paczynski.
Actually, Paczynski initially declined participating in the debate, as he would be taking a sabbatical in Japan. Realizing the debate could not come together without him, I then offered him $10K to attend, arguing this was a unique opportunity. Prof. Pacynski then accepted, but then declined the money. I am not making this up. Given Prof. Paczysnki's attendance, the rest of the program quickly fell into place. Prof. Lamb agreed to argue the Galactic point of view for GRB origins, while Prof. Rees moderated. Professor Trimble and Dr. Fishman agreed to give background introductory talk.
It has been called to my attention that some people thought that the idea to hold this debate originated with Prof. Paczynski. To be clear, It did not. This topic actually came up in the planning of an honorary symposium the year after Prof. Pacyznski passed away. In response to his request, I sent Prof. Draine a series of emails I sent out in 1994 detailing how this debate got started, and he seemed satisfied that this settled the matter. Still, Prof. Paczynski agreeing to participate made the event come together -- I don't think it would have occurred without him.
The editors of the Astronomical Society of the Pacific agreed to publish the proceedings in their journal, with Jerry and I acting as the defacto referees. These contributions are here:
A series of web pages about the 1995 debate can presently be found
here.
In 1996 Bonnell and I then organized the
Distance Scale of the Universe debate between Tammann and van den Bergh, while in 1998 Bonnell and I organized the
Nature of the Universe debate between Peebles and Turner.
1995 Astronomy Picture of the Day
This whole essay was written, in part, to generally communicate that I had and still have a career in astronomy beyond the Astronomy Picture of the Day (APOD). Yes, I created APOD in 1995 along with Jerry Bonnell. The idea was ours and if others had a similar idea, we were unaware of it. And we looked. Up until at least the middle of 2011, Bonnell and I have chosen all of the images and written the text for (just about) every APOD. One way to learn more about how APOD started and is currently produced is to read an interview I did with the journal Communicating Astronomy with the Public here:
here.
1994 Most Microlensing Occurs on Background Stars
Certainly the luck of timing played a factor in this idea. My Ph D. thesis was completed in1987 and, as described below, was about microlensing. Microlensing was then actually discovered in the early 1990s. Being pre-programmed on the subject, it was relatively easy for me to make relevant comments on how future microlensing searches might be improved. This idea turned into one of these comments. It seemed to me that taking field observations, it would be the quite likely to see microlensing for stars actually below the magnitude limit of the survey, as compared to the stars being surveyed. The reason is that lensing would boost the brightnesses of these stars over the survey limit. A driver for this is that, typically, stars just below the survey limit actually outnumbered survey stars. I literature search turned up nothing on this idea. So I wrote this up and it was published quickly. The published paper is here:
Nemiroff, R. J. 1994, Magnfication Bias in Galactic Microlensing Searches, Astrophysical Journal, 435, 682. Although unbeknownst to me at the time, it appears that several others had this idea nearly concurrently, and papers by them are usually much better cited.
1994 Microlensing Can Resolve the Surface of Stars
This idea resulted in a published theoretical prediction that clearly came true. The luck of timing also played a role in this idea. The first graduate student I ever mentored, Thulsi Wickramasinghe, now a professor at the College of New Jersey, and I would frequently have fun discussing the dramatic microlensing observational results that were being reported during the mid-1990s. One discussion centered on my Ph D. thesis idea that microlensing light curves were so generic that one could tell embarrassingly little about the lensing event, including the mass of the lensing star and the size of the source star. One potential case that would add more information to the generic microlensing light curves, however, was if the center of the lensing star crossed the disk of the source star. Then not only would be light curve look different, but more information would be available, as then the ratio of the (angular) Einstein ring size to the (angular) source star size became important. A search turned up nobody who had this idea previously. So we wrote this up fairly quickly, and the reference to this paper is here:
Nemiroff, R. J. & Wickramasinghe, W. A. D. T. 1994, Finite Source Sizes and the Information Content of MACHO-Type Lens Search Light Curves, Astrophysical Journal, 424, L21. It turned out that this idea was also thought independently by others at about the same time. For reasons I don't understand, though, this paper is much better cited than the previously mentioned paper. To see the accuracy of our predicted star-crossing light curve, one can compare, for example, Figure 1 in our above paper with Figure 1 of this observational paper.
1994 Time dilation Discovery in GRBs
Searching for time dilation in GRBs was actually my colleague (Dr.) Jay Norris' idea, and he also did most of the work. In 1993, when I was a postdoc just starting to work at NASA's GSFC under Norris' supervision, Norris asked me what the time dilation signal would be for GRBs, and whether it would be a function of the cosmological model. After double checking this myself, I confirmed what he likely already knew -- that at a redshift z, the expected time dilation for a GRB (or anything) would be just (1+z). I also helped Norris to realize that it was possible to make a bright GRB artificially look like a dim GRB, just add the right type and amount of noise. This allowed Norris and our group to create two seemingly identical set of GRBs -- one really dim and the other bright but artificially dimmed -- to see if they had the same average duration. Given much debate with other groups, we found a strong signal.
Oddly, other groups kept showing up claiming that this measurement was inaccurate and that their was no time dilation signal. One reason for this was that other groups thought that GRBs occurred in our own Galaxy and hence there was no reason to expect a cosmologically-created time dilation effect. Another reason was that there were many detailed steps in doing this right, and Norris knew better than anyone how to avoid common missteps. Usually, after looking at their methods, Norris (et al.) was able to figure out what they were doing wrong, and sometimes, they would acquiesce. After a while collaborators (Jerry) Bonnell and I started thinking of Norris as a western gunslinger who kept being challenged by newcomers to see if Norris was the indeed the "fastest gun in the west", which in this case meant "was really measuring time dilation." New claims therefore became tiring, not because we thought we were wrong, but because we not only didn't want to shoot the newcomers, we didn't even want to spent the time needed to explain to the newcomers why they got shot. Still, Norris showed tremendous patience.
The Norris et al. group published several papers on this, but here is perhaps the most prominent:
Norris, J. P., Nemiroff, R. J., Scargle, J. D., Kouveliotou, C., Fishman, G. J., Meegan, C. A., Paciesas, W. S., & Bonnell, J. T. 1994, Detection of a Signature Consistent with Cosmological Time Dilation in Gamma Ray Bursts, Astrophysical Journal, 424, 540. Note that when the galactic / cosmological debate became resolved (in favor of GRBs being cosmological), this result became much less controversial. Still, the novelty of the claim, as well as the controversy, caused this paper to become very well cited.
1994 Dim GRBs are Redder than Bright GRBs
During the "GRB Distance War" of the mid-1990s, my GRB collaborators and I were looking for other ways to tell if GRBs were at cosmological distances. Our core group (Jay Norris, Jerry Bonnell, and myself) was perhaps some of the more vocal proponents of cosmology, as shown by conference proceedings and publications of the day. Still, were a signal consistent with cosmology not there, we would surely acquiesce since first and foremost we were really interested in truth.
Now one way that near and far GRBs might be different would be if one group had a different average color than the other. In other words, redshift should also work on spectra. Paczynski was quick to point out to me that this did NOT necessarily mean that more distant GRBs would be measured as being more red than nearby GRBs -- the answer really depended on K-corrections, spectra, and detection criteria. Dim GRBs could turn out to be more blue, for example.
To see what the data showed, I led a collaboration that compared the color (termed "hardness" in the X- and gamma-ray bands) of groups of GRBs that had bright peak flux and dim peak flux. We found that the dim group was more red to a statistically significant degree. This result was also met with controversy, but since it was easier to check than time dilation -- and more ambiguous to interpret -- this color effect was less controversial. The paper is here:
Nemiroff, R. J., Norris, J. P., Bonnell, J. T., Wickramasinghe, W. A. D. T., Kouveliotou, C., Paciesas, W. S., Fishman, G. J. & Meegan, C. A., 1999, Gross Spectral Differences Between Bright and Dim Gamma-Ray Bursts, Astrophysical Journal, 435, 133L
1994 Digits of Irrational Numbers
Once a summer, our research group's boss, Jay Norris, would go out west for a month and continue to build his retirement home. When he came back he would have stories of clearing brush, making roads, and ordering logs, many times complete with pictures. During that month away, however, it seemed that my officemate Jerry Bonnell and I were "home alone" with a whole lot of NASA computer power. So we would embark on speculative projects that we might not have other times of the year.
In the summer of 1994, one subject that Bonnell and I argued about was random numbers. One train of logic we explored was whether the digits of pi (3.14159... ) could be some sort of Encyclopedia Universica. Say you went to your local library and digitized all of the books, starting from the first page of the book in the front left of the ground floor and ending with the last page of the book at the back right of the top floor. Say for each letter you substituted a numeric code, such as 01 for A, 02 for B, 00 for a blank, or even used ASCII codes. Then you could run all of these codes together, for all of the books, and form one really long line of numbers. Next,one might place a decimal point in front of the first number and make a really long fraction. At this point, if one had a really good ruler, one could draw a line on a page where the length of the line was exactly this fraction of the length of the whole page. Then you would have encoded all of the local library in a single line, and to recover that information all you would have to do would be to measure the exact length of this line and the exact length of the page. And ignore quantum mechanics.
This number, however, could be made much shorter. The reason is that there is much redundancy in this fraction, what with only 26 or so numbers used, and with all the repetition. For example, instead of coding the six digits for the word "the" all of the time, perhaps the word "the" could be indicated by only two digits, hence greatly shortening the fraction. Taking this example to extremes, any number system that has inefficiencies or redundancies can be shortened yet further by re-encoding them using a shorthand for the redundancy. This turns out to be one of the bases for modern digital compression routines. The important point here, though, is that a maximally compressed library should look just like a string of random numbers to someone who does not know the key to deciphering them. The difference, therefore, between a string of digits that contained complete knowledge, and a string of digits that contained essentially zero knowledge, was impossible to tell without a key.
The next rung on this latter of unusual logic was to wonder out loud if there were, in fact, any strings of digits out there that could, in theory, contain all of the knowledge of the universe. The one that came to mind first was pi. Perhaps in the digits of pi was encoded all of the knowledge in the universe. If so, how would we know? So Bonnell and I set out to look. We set out to find many digits of pi and see if there was any pattern, any key that would tell us how to unlock the digits and hence unlock all of the knowledge in the universe. I am not making this up.
The web was quite young back then, but crude searches found that two brothers at Columbia U., the famous Chudnovsky brothers, were busy computing many miliions of digits of pi. Unfortunately, we felt intimidated about even trying to contact them. We understood from reading some articles about them that they were also curious whether the digits of pi were appeared truly random, and hence understood that they would not be sending us these digits so that we could check it ourselves.
So perhaps we could compute pi ourselves. This proved more difficult that I had anticipated, and the arctangent routines that presented themselves seemed slow and complicated to me. But what about the natural number "e"? Perhaps the digits of this e encompassed a Encyclopedia Universica? After some preliminary investigation, it appeared that it was much easier to compute "e" than pi. And so we started to compute the digits of e.
As luck would have it, I received a new computer on my desk just then. It was a DEC Alpha, and although it looked like a standard issue PS, it was rumored to have almost supercomputer speeds for integer operations. So I started running FORTRAN routines that computed e. I created version after version, timing them to see which methods were the fastest. Soon thousands of digits would come out after just a few seconds. I would send these digits over to Bonnell and he would search them for randomness. They always seemed to be completely random.
As the routines to compute e became faster and faster, more and more digits would be computed. Soon it became standard to let the computer run over night and see what happened the next morning. A leader for our computer support team eventually asked what was going on, and I told him. Instead of castigating me, he helped tune the internal hardware and software to run even faster, enabling a run a 10 million digits. Which we made, setting a (then) world record. This record is recorded on this modern Wikipedia page on e
here. Apparently, our record surpassed the existing record held by Stephen Wozniak on the Apple II.
Besides e, Bonnell and I decided to also try to compute the digits of other irrational numbers, such as the square root of two and the square root of the other single digit integers. To compute sqrt(2), I first tried Newton's method, but found that it tied up my computer for too long as it took increasingly long to compute the next set of increasingly numerous digits. I therefore searched for a method that was easier to start and stop. What seemed to work best for me was, again, a home-grown code where I "guessed" the next digit, computed how accurate this guess was, and then corrected this guess given its accuracy. This way I could keep all of the previous numbers and compute just one new digit at a time, stopping any time I liked. And it seemed just as fast as Newton's method.
Actually, I didn't so much develop this algorithm as remember it from when I was in junior high school. Then, when my family would go to the local department store (specifically Wanamakers), I would go off on my own and frequently try out the newest calculators they were selling. At first, these calculators did not have the square root function, so I liked to try them out by guessing the square roots of integers, and then multiplying my guess by itself to see how close I was. Eventually, from seeing how close the last guess was, I became pretty good at guessing the next digit in a square root sequence.
Who knew this strange bit of experience would become relevant 20 years later? I was so inherently familiar with this method that it was easy to code, although as with the e algorithms, I was able to keep making them faster and faster as time went on. Soon I was email the files to Bonnell, who would check their accuracy by rapidly multiplying the number by itself to see if an integer came out. Bonnell also searched for non-randomness.
The name of the file we created with the digits of sqrt(2) and e was just called "a.out". These files became very very large for the time, as the number of digits ballooned into the millions. Coworkers kept asking Bonnell what was a.out because it was so large it was clogging up parts of the computer system. For one thing, trying to load the file into the web browser Mosaic, the only browser then available, tended to freeze of crash the browser.
A few times a year, at Paczynski's invitation, I attended the "Tuesday lunch" of astrophysicists at the Institute of Advanced Study in Princeton, NJ. Sometimes I would volunteer the latest results of our groups GRB analysis. One Tuesday, although it was off topic, I mentioned our digits results. To my surprise, the famous physicist Freeman Dyson came up to me afterwards to ask me about it. He was most interested in the routine I was using. I described it but he seemed most interested to find out how fast it ran in terms of how long it took to compute N digits -- did it take only some number times N, or did it take some number times N squared. I thought for a bit and then estimated that the routine only ran as fast at N squared. I trumpeted the fact that the routine could be stopped at any digit, an advantage, it seemed, over Newton's method. Still, Dyson was disappointed and no longer wanted to continue the conversation.
As fast as I could compute them, Bonnell kept testing these irrational digits for randomness. He never round any. I guessed testing the digits was a good thing to do regardless of the result, since if a correlation had been found, that would have been even more interesting! My own experience is that the result of any experiment I am involved in will be the one that causes the least gain in knowledge and therefore the least publishable result. Alternatively, however, were the correlation had led to a key to a Encyclopedia Universica, that would have been even better right up until people used this knowledge to end the world.
Given that the digits appeared truly random, which they did, the next thing Bonnell and I did was write a paper on the randomness of the digits. Now even though this was out of our field, it seemed to me that since nobody else had a million digits of e or sqrt(2) (and we looked), and the digits of pi did not appear anywhere on the web, this approach was not only unique but useful. To our surprise, several other commonly available random number generators showed cleared correlations to Bonnell's programs, which made these digits appear good. We said in the manuscript that we were making these digits freely available and they show no randomness of which we are aware.
To our chagrin, the manuscript was rejected. I don't really recall the criticism of the referee, but the memory I have now is that what we were suggesting was just too strange. I don't think we argued much with the referee, we just let the manuscript drop. In modern times we would submit the rejected manuscript to arXiv and it would still be available, but in those times arXiv was not very prominent, especially in computational things, and so we didn't submit it there either. I now fear that this manuscript has been lost. Oh well.
What we did do is to archive many of these digits on the web. For this, we created the Digit Warehouse in 1994, which survives to this day
here. Oddly, one of the largest files, the 10 million digits of e, was taken down because at the time it was freezing and crashing many browsers. I saved these digits in four chunks, and can even remember their file names. Unfortunately, I have now lost track of these files. Still, first million digits of e and the square roots of all single digit integers are there, and can be used to stump people who claim they can do any math problem in their head (try this one: what's the millionth digit of sqrt(2)?).
We also submitted these digits to Project Gutenberg, an online project that makes freely available many familiar texts, many of them quite famous. Therefore, many of these digit files are freely available there as well.
A few years ago, Bonnell and I found out that we were the authors of a new book titled "The Golden Mean". This came as a surprise as neither of us could remember writing such a book. And we have written, so far, two books together. Both of which we remember writing. Anyway,
being the curious sort, I decided to order the book from amazon.com, where it was listed. I paid the listed price, gave my address, and waited for the book to arrive. No book. Soon I got an email saying something like the was "unavailable".
But what was going on? To the best Bonnell and I can presently understand, Project Gutenberg made available much of their collection to Amazon.com, possibly for free. Amazon would then format these books and sell them. Apparently, Bonnell took our file for the sqrt(5), and created the golden ratio, sometimes known as the golden mean, simply by adding one and dividing the whole thing by two. He then made this file available to Project Gutenberg as well. Then, apparently, Project Gutenberg made the digits available to amazon.com. So I should have expected a book with lots and lots of numbers. The web page says 20,000, in fact. Currently, though, the "book" is available for free on the Kindle.
1993 The Brightness Distribution of GRBs is Consistent with a Cosmological Distribution
This is an idea that originated with Prof. Paczynski. Then graduate student (now Professor) Thulsi Wickramasinge and I explored into the newly arriving BATSE GRB data to see if cosmology and deep space volume effects could explain the lack of dim GRBs. After some analyses, it appeared that it could. We used peak flux as our standard candle. Paczynski could see the effect qualitatively in the data before we started, and was not surprised at the result. This was one of the first clear observational indications that GRBs occurred at cosmological distances. The paper is here:
Wickramasinghe, W. A. D. T., Nemiroff, R. J., Norris, J. P., Kouveliotou, C., Fishman, G. J., Meegan, C. A., Wilson. R. B., & Paciesas, W. S. 1993, The Consistency of Standard Cosmology and the BATSE number versus brightness Relation, Astrophysical Journal, 411, L55. Although asked explicitly by me, Prof. Paczynski declined my request to explicitly include his name on this paper because he worried that all of this work would all be attributed to him, which he did not feel was fair. This paper was very well cited right up to the year that actual GRB redshifts showed that GRBs were, in fact, cosmological.
1993 The View Near a Black Hole
Explaining the origin of this idea leads to a few more interesting stories. While a postdoc in the early 1990s at the Naval Research Lab in Washington, DC, I was asked by Dr. Kent S. Wood, my supervisor, if I would add to a computer code his group was developing about novas and supernovas. At that time, they were considering photons as always flying on straight lines, and wanted to increase the codes' accuracy by considering photons to be on trajectories curved by gravity (as viewed from infinity). A a former mentor of mine, Prof. Christ Ftaclas, had written a photon-bending code previously, which I was given, and so my task would be even easier were I do adapt his code.
So I looked at his code. I could not understand almost any of it. And it wasn't the code that was bad, it was me. Now, true, the code's parameters were not the most descriptive, but try as I might, I could not understand why the code did most of the things that it did.
I therefore decided it might be easier to write my own code from scratch. So I tried. It was hard, but actually kind of fun. I had little idea, though, whether my code was working at all. In theory, I could compare my code to the Ftaclas code to see if the results were the same, but in practice I couldn't really even get the Ftaclas code to run. Besides returning all zeros. And I was too proud to admit that.
I therefore decided to test my code by writing out data that could be plotted as images, and seeing if the images made sense. As that was progressing, I realized that there was this strange looking monitor in the local computer room called the "Gould 9000". I am not sure that it was named correctly, as it does not seem to be related to the supercomputer of the same name available in the 1980s. Whatever its real name, it had the (then) amazing ability to display 1000 x 1000 graphics in 16 colors. And it was almost never being used. So after asking around, I was given permission to use it. So I learned how to use it.
I therefore started testing my high gravity photon deflecting algorithms by plotting them on the "Gould 9000". One good method I found to test the code was to create an artificial surface near the event horizon and project a map of the Earth onto it. Then, the code would show a distorted map of the Earth after lensing, and I could see if it made sense. At first, the images would make no sense, and I would to back and fix the code. Then the images would make no sense and I would go back and find no obvious error in the code. Increasingly, I realized that the code was right and my intuition was wrong. That is one way I know that a code is working. I then had to "practice" calibrating my new intuition by asking the code to do different cases. Of course, as soon as my intuition was calibrated, I was able to implicitly assert that I always expected the lensed images to be this way.
Unexpectedly, people would come by and comment on the images of the distorted Earth that were being put up. People who were only acquaintances were showing unusual interest in my work. Strange! Soon I was taking requests -- "can you show Germany?" Things like that. I began to believe that I was onto something here.
The next part of this story relates to my troublesome curious nature. Sometimes, I like to explore spaces, for instance opening unlabelled doors to see what is inside. So in the mainframe room in my NRL building, there was this cabinet that was usually closed. More than one, in fact, but when I looked inside this one, there was a strange looking device that looked like a classical movie camera that appeared to have some sort of computer interface. I asked about it and was told that it attached to the "Gould 9000" and could be programmed to take images automatically through a computer interface. What was better, the project for which the device was ordered had run its course, and so the movie camera was not presently in active use. I was therefore given permission to use it.
I therefore starting making movies of the images. Not digital movies -- actual 16-mm film movies. I had them developed at a commercial photo lab, and I bought a 16-mm movie projector to see the bits and pieces as they were produced. It was a bit of work but my curiosity to see how the next bit looked made it a lot of fun. Now I had never before seen anything like these movie clips (and I would have remembered), so I decided to make a separate project out of this and write a paper on what it looks like to go to such a compact object. Also, I included the sky distortions so that I could do the black hole case as well. Actually, the black hole case was easier than the case that included the surface -- the neutron star case -- because all I needed to track was the sky distortions. From the finished 16-mm film, I made VCR tapes (I paid for all of this myself). I then discussed the results in detail with Dr. Wood and Professors Ftaclas and Paczynski. In particular I remember Paczynski agreeing to meet me at his house so that we could watch the tape on his home VCR. Each asked lots of questions, but ultimately they all appeared to me to be reasonably satisfied with the integrity of the results and even somewhat impressed. I submitted the paper to a teaching journal, who sent it to three referees. One hated it, one thought it OK, and one liked it so much he wanted to become a collaborator. The paper on this was published here:
Nemiroff, R. J. 1993, Visual Distortions Near a Neutron Star and Black Hole, American Journal of Physics, 61, 619. This paper was cited very little at first but has been cited quite a bit more recently.
Back to the movie: to my chagrin, wherever I debuted the movie, there was usually someone who objected, figuring that I must have gotten it wrong. The first case of this was at a meeting of the American Astronomical Society meeting in 1991. There I had the movie playing on a loop near my poster presentation. Dr. Steve Maran, the AAS press officer, asked me to do a press conference on the movie. So I did. A prominent professor attended the press conference and kept heckling me, saying things like "You don't know that!" and "That can't be right!" I have decided not to mention this person's name. I tried to answer this professor's concerns as best I could, but he was several years older and significantly more prestigious than me, and I think I looked like I was out of my league. Conversely, I don't think this had a large effect since the press conference was not very well attended anyway, possibly because I did not represent a prestigious enough institution. I did hear that CNN ran a few seconds of the video, though. In the many years since this video was created, however, this prominent professor since apologized -- repeatedly in fact -- and now seems to go out of his way to acknowledge and support my future work.
I also presented the video in VHS form (not PAL) at a conference on gravitational lensing in Germany. There, a separate and even more prominent professor watched the video for a few second and proclaimed out loud that it seemed quite wrong. Fortunately, Prof. Paczynski was RIGHT THERE and immediately retorted that the video appeared scientifically accurate to him. The two then continued on their way while discussing it.
In the next few years, I re-ran the codes creating digital outputs which I posted to a web page I created for the videos at NASA's GSFC in 1994. This page still stands today. (It might be one of the oldest continually available pages on the web, altough I have made a few minor changes to it over the years.) That page is
here. The page was once designated an "Internet Goldmine" by Computers in Physics section of the American Institute of Physics, back when they did that sort of thing. Also, upon request, I received an endorsement for the scientific accuracy of the video from Prof. Kip Thorne. I did email Prof. Stephen Hawking to solicited an endorsement of the science in the video, but never received a reply. A version of the movie played in a greater educational video about black holes for a number of years, around the year 2000, at the American Museum of Natural History in New York City. The experience I gained in creating the web page highlighting the video later helped out when creating APOD with Jerry Bonnell.
Epilogue: I never ended up contributing the photon-bending subroutine I was originally enlisted to write to the nova or supernova code project. Sorry!
1993 Ultracompact Neutron Stars
When investigation light bending effects in high gravity environments such as black holes (as described above), I needed to test my codes in cases where the gravity was strong, in particular near the event horizon of the black hole. Some of the most interesting theoretical cases occurred when the object was inside its own photon sphere, meaning the sphere where gravity bends photon trajectories into circular orbits. I investigated these case by artificially painting the surface of the Earth on these near black-hole compact objects. Naturally, I assumed that neutron stars could just be this dense and compact, but when I looked into finding real cases, to my surprise, I found few reference to such objects. It seemed such compactness was not allowed for standard equations of state, and even for some General Relativistic energy conditions. Still, published scientific papers postulate cases that break GR energy conditions almost every day, so why can't I? So I decided to dub these very compact objects "ultracompact neutron stars". My postdoctoral supervisor, Kent S. Wood, seemed intrigued, and suggested I publish it. After recruiting (then) fellow NRL postdoc (now Prof.) Peter A. Becker , we worked out some details and possible physical behavior, and we published this paper:
Nemiroff, R. J., Becker, P. A. & Wood, K. S. 1993, Properties of Ultracompact Neutron Stars, Astrophysical Journal, 406, 590. The paper, while fun to write, was essentially ignored.
1992 Micro Black Holes Might Explain Everything
In a paper for the now defunct journal "Comments on Astrophysics", I wrote a defense of the possibility that GRBs were cosmological. In this paper, however, I gave three toy models for GRBs. One, in particular, was so bold it is quite humorous in hindsight, and most definitely wrong. Still, it remains one of my favorite ideas, for its crazy attempt to solve numerous observational problems in astronomy with a single invention. The invention was very low mass black holes, on the order of the size of the Earth or so. These black holes would solve three long standing problems in astronomy with one fell swoop. They would a) make up dark matter in its entirely, b) dive through neutron stars creating GRBs, and c) fall into the Sun and lower the central temperature so as to reduce the Sun's neutrino output, therefore also solving the solar neutrino problem. To recap: bold, funny, wrong.
1989 How to Compute the Detection Probability of Gravitational Lensing
This was my first "A level" paper, in my opinion. The idea behind it was to generalized the probability of stellar lensing technique I used in m Ph D. thesis, to more types of gravitational lenses and, in theory, all types of observational criteria. I mapped out the details that the limits one uses to detect gravitational lenses themselves create volumes in space within which the lenses must fall to be so detectable. Not only did this seem to me to be a good way to compute the probability of lensing more accurately and generally, but these detection volumes looked cool, were easy to use computationally. I knew they worked because I used them myself quite frequently to estimate lensing probability.
To my misfortune, this paper followed by a few years another paper published by three very famous astrophysicists which was then considered the last word on the subject of probability in gravitational lensing. Perhaps egotistically, I considered my technique more general, although it certainly could reduce to what the other paper had published. The result was that my paper was cited only occasionally, but the paper by the famous people was cited much more commonly. Oh well. My paper is here:
Nemiroff, R. J. 1989, On the Probability of Detection of a Single Gravitational Lens, Astrophysical Journal, 341, 579. The paper was moderately well cited for a while but now seems to have been forgotten about.
1985 Microlensing is not observable
This entry also involves some stories that are perhaps more controversial than any of the others, since it involves the history of the idea of microlensing. It has several facets and I expect some people will not believe. Anyway, here goes.
As a graduate student, in 1985, I took up an interest in a very new area of astrophysics dubbed gravitational lensing. At that time, I got into frequent discussions and even polite disagreements with Professor Christ Ftaclas, who at the time was on the faculty of the University of Pennsylvania. I remember getting many things wrong, including thinking for a while that a point lens would make a disk out of a background point object. I remember finally having the obvious revelation that a gravitational lens really created a ring out a small source far in the background, not a disk. Of course I then pretending that I always knew it was that way all along.
Anyway, I decided to demonstrate to Ftaclas that lensing of stars in in our Milky Way Galaxy was detectable. So I did an estimation of the probability of the effect and proudly showed it to him. Ftaclas correctly showed me that my technique was flawed and that I was mistaken -- my estimation did not show that the stars in our Galaxy would show detectable gravitational lensing. Not to be deterred, I felt that if I could make my calculation more accurate, the effect would finally show itself to be probable. So I came up with a better way to estimate the probability of lensing, and showed it again to Ftaclas. Finally, I would make him see. Once again, however, he showed me flaws in my estimate. Back at the drawing board, this time I really went much further and computed the probability in a fully three-dimensional sense, creating what I called a "Lensing Ellipsoid" where the lens must fall to be detected, and then integrating over the stellar density in the volume. My previous estimates had all been two dimensional. This time, Ftaclas seemed impressed, but with a change of a single decimal point in one of my numbers, showed me that my estimate was again low.
However, I decided to try to publish the "Lensing Ellipsoid" method, even if the end result was that stellar lensing was unlikely to be seen. I therefore wrote out the text longhand and sent it in to the journal Astrophysics and Space Science. One reason I sent it to this journal was because this journal did not have page charges. And since I did not have a grant, I could not pay the page charges. To my delight, editor Prof. Kopal accepted the paper even without sending it out to a referee, asking that I type it up and resubmit it. So I did, and the published version appeared here: Nemiroff, R. J. 1986, Random Gravitational Lensing, Astrophysics & Space Science, 123, 381.
So here comes the controversial part. If you read this paper, you will not see Paczynski cited. When I wrote that paper, I did not know of Prof. Paczynski and his recent work on microlensing. It turned out that Paczynski also turned his thoughts to stars lensing stars, which he dubbed microlensing, except that he better understood technology and therefore found a case that was actually likely to be found. One might say that I chanced across the idea of microlensing at nearly the same time as Paczynski, but did not have the astronomical background nor technological savvy to suggest observational campaigns that could actually find cases of microlensing, as he did.
That, however, is not the end of the story. During 1985, Prof. Wiita, another professor at Penn, showed me that preprint from Paczynski. I think people at Penn had considered my interest in gravitational lensing somewhat nutty before, but after seeing the microlensing preprint from Paczynski, they suddenly felt otherwise. I always wondered if Ftaclas felt bad for trying to convince me that microlensing was practically unobservable, or for not suggesting the cases that worked as done by Paczynski, but I have never asked.
What did happen is that, unfortunately, the good Prof. Ftaclas did not get tenure at Penn and soon made plans to leave. He and Wiita, who actually knew Paczynski personally, hatched a plan to introduce me to Paczynski and for me to ask Paczynski to be my scientific supervisor for my Ph D. So one day Ftaclas and I went over to Princeton. Ftaclas introduced me saying that I had been thinking about lensing. At first Paczynski seemed skeptical of even speaking with me, but when my turn came to speak, I described my Lensing Ellipsoid concept and how its shape was dependant on actual detection thresholds. I quickly used it to estimate some lensing probabilities. Paczynski conceded that I seemed to understand how lensing probabilities were done. When I tried to trumpet the novelty of my approach, however, Paczynski retorted sharply that everyone has their own way of thinking of things, but it all amounted to the same. I felt rebuffed, but there must have been something intriguing about me or my approach because Paczynski then agreed to keep talking with me about lensing issues. To my chagrin, Paczynski then tried to sell me one of his own projects. In particular he wanted someone to think about what happens when many lenses were involved, a situation called "high optical depth". To help, he gave me a code he wrote on lensing. I did end up using this code, but at a later time, and not for that.
What actually happened after that was that about once a month I would drive over from Penn to Princeton and review for Paczynski the latest ideas that I had about lensing. He was the only one I knew who even understood them. We would sometimes debate them so intensely that if anyone else was in the room, they surely must have felt left out of the verbal rapid-fire ping-pong match. Most typically, Paczynski would shoot down my ideas one after the other -- but oddly now encouraged me to continue. I remember, after a few of these sessions, he even passed up an opportunity to tell me other ideas he was working on because, he said, I didn't seem to need them. One day, he surprised me and said "That's a good one. I don't think that one has ever been done. That could be important."
The idea he bolstered an idea about using microlensing to differentially magnify the broad line regions of quasars, hence possibly indicating the dynamics of the broad line region. This paper eventually appeared here:
Nemiroff, R. J. 1988, Astrophysical Journal, 335, 593. Actually, I offered to include Paczynski's name on this paper but he declined. After that, it seemed he was signed on for the long haul as my defacto thesis advisor. Other successes eventually followed. I wrote up several of them including Lensing Ellipsoids at the front (I was defiant to the end) into a Ph D. thesis that he signed. As indicated by the above stories, I kept in contact with him at least a few times a year until his unfortunate death from brain cancer in 2008.
Nemiroff, R. J. 1995, The 75th Anniversary Astronomical Debate on the Distance Scale to Gamma Ray Bursts: an Introduction, PASP, 107, 1131;
Trimble, V. 1995, The 1920 Shapley-Curtis Discussion: Background, Issues, and Aftermath, PASP, 107, 1133;
Fishman, G. J. 1995, Gamma-Ray Bursts: an Overview, PASP, 107, 1145;
Paczynski, B. 1995, How Far Away are Gamma-Ray Bursts?, PASP, 107, 1167;
Lamb, D. Q. 1995, The Distance Scale to Gamma-Ray Bursts, PASP, 107, 1152;
Rees, M.J. 1995, Concluding Remarks, PASP, 107, 1176.
Bonus Material Not Listed Above
The Main Sequence Blues
(I wrote this when I was a graduate student at UPenn in 1984. It appeared in Astronomy Magazine.)
Once upon a time there was a five-solar-mass star named Brutus. This five-solar-mass star was the biggest, meanest five- solar-mass star in the whole spiral arm. Now the latest fad in Brutus's stellar neighborhood was to get an energy high by burning hydrogen in your core (where no one could see). At night, when few people were looking, Brutus and his buddies would line up on the Hertzsprung-Russell diagram and "do" hydrogen. They called themselves the Main Sequence.
Now Brutus could party up a storm: he could "do" more hydrogen than his smaller friends, and could frequently be seen even in the wee hours of the morning all energized up, spitting up photons like crazy.
Soon Brutus burned all his core hydrogen, and the law demanded he leave his friends on the Main Sequence and go off to galactic prison. Brutus was first assigned to the Red Giant Branch. Some thought this was just a phase he was going through, and that he would be back soon, but for Brutus there was no return.
"Hey man, do HELIUM, man," his big new friends would say. "You'll reach a new high. Do it, man, do HELIUM." Since all his 'Giant buddies were into helium so, eventually, was Brutus, becoming more energetic than ever before. But before Brutus expected it he burned out, he ran out of helium to burn in his core. Brutus became depressed and contracted within himself.
"Hey man, do CARBON, man," some of the cooler stars would say. "It'll blow your mind, it'll dump your depression, it'll ignite your innards. Don't be a dwarf star, man, DO it!"
So Brutus did carbon, and it was even more fantastic than they said, it was the best of any element yet. The photons pulsed through Brutus's photosphere like never before in his entire life. But after almost no time his carbon burned out, and none of the other stars would sell him theirs, not even his down and out white dwarf friends. The only thing Brutus could do was burn carbon around his core, which he did, as he did helium and hydrogen as well, but it just didn't turn him on like it used to. Brutus needed something else.
"Should I do SILICON, man, should I?" he asked his friends.
"That degenerate stuff? Are you crazy? I wouldn't touch that stuff. No way. Uh-uh, "they all said. "Just cool out, OK?"
"You can't do it, you're not big enough," his smart friends would say. "Its 'da body weight, you don't have enough, so stay off the silicon, man. Stay off the silicon, you hear?"
But Brutus wouldn't listen. When nobody was looking, he tried to do silicon.
One day one of Brutus's old flames came by the old stellar neighborhood asking about him. "Whatever happened to that big hunk of gas?" she asked.
"Oh look, THERE he is, " some star replied, indicating a super bright flare in the sky that slowly faded to nothing. Even stars that knew Brutus were shocked.
"Too bad," she said, "he never did know his limitations."
"Nonsense," Brutus's friends disagreed, "it was fate."
On one thing they did agree: the answer was blowing in the wind.
(I also wrote this as a graduate student in 1987, and it also appeared in Astronomy Magazine. It was later reprinted by an online publication titled "Dark Matter", which accounts for the "DM" abbreviation below. It is more than a little bit silly.)
Mornings just wouldn't be the same without him. He really does light up our life. But did we ever thank him, even once? No, never. We should consider ourselves lucky that he doesn't get overheated. Dark Matter Magazine thus decided to thank that wonderful star of ours, Sol, for all those billions of years of free sunlight; tax free sunlight. We also thanked him for that really great gravitational field of his that keeps us from cavorting around the galaxy. To our surprise he granted us an exclusive interview. We found out that being a star isn't always sunshine.
Dark Matter: Mr. Sun, how did you become a star?
Sol, our Sun: Please, call me Sol. I answered an ad. "Galaxy Now Forming, Stars Needed." The ad made it look real easy. Just sit around, make hydrogen into helium. Have everyone look up to you. Have a career where you can really shine. And I always thought I had what it took to be a star, you know. I was young, I had great big clouds of hydrogen.
The actual triggering event was the tragic death of a star right in our own neighborhood. This shocked me and a lot of other large young gas regions into several new stages of development. It totally changed our future. Ironically, many of us eventually decided we wanted to become stars ourselves.
DM: Your future must be very bright.
Sol: Well it's not so easy being a star. First of all, you have to be made of the right stuff. And not only spunk and perseverance, also 75% hydrogen and 25% helium. But I have all that. It's the work schedule that's really impossible. 24 hours a day, 7 days a week, you really can't take a day off without people getting annoyed.
DM: Are there a special class you had to be part of?
Sol: Yes, luminosity class. I, of course, wanted to be in the first class, but it turned out that was only for the brightest super-stars: the industry giants. It turns out I was put at the bottom. Class five. The other classes all made fun of us. "Dwarf stars" they would call us.
DM: How did you do in your spectral studies?
Sol: Not very well. I wasn't in the hottest class there either. I thought I had "A" potential, but after a while I was afraid they would give me an "F". But it even got worse: "F" wasn't low enough for me. I got a "G". Oh, I felt bad for a while, I went through a real solar minimum. But soon I found out lots of stars get "G's"; or even lower. Turns out "G's" out-number "A's" 30 to 1.
DM: Are you happy with your home galaxy: the Milky Way?
Sol: Well, when I signed on, I thought we were going places. I thought we would take on the universe, as a team. Adventure, action, 100 billion stars, all together in one galactic union. Now that, I thought, was power. But all we ever do is go in circles. I've been around the center 60 times or so, it's no big deal. Spiral arms - one millennium your in them - the next your out - they're fickle. The Andromeda galaxy is really no better; it's the same story there. Now the Virgo Cluster - that's where all the local action turned out to be.
I mean the Milky Way is OK and everything. It's a little dusty, I guess, but at least it's stable. We haven't been disrupted by any other galaxies or anything. We even have our little groupies: the Magellanic Clouds, galaxies like that. It's home.
DM: When did you decide to have planets?
Sol: Very early in life, although I'm not exactly sure when - that part of my life was very nebulous. I didn't really plan to have planets because I never had a steady companion. They just sort of spun off of my care-free early life style.
DM: Do you have trouble keeping the planets in line?
Sol: I'll say. I thought it would be easy - I'm hundreds of times bigger than they are. Have some planets, they said, they won't perturb you much. But they're running rings around me. Mercury won't keep to a Newtonian orbit - the darn things' precessing all over the place. Uranus fell over. Jupiter's getting all spotty. Saturn has ring-around-the-collar. And Earth: please turn down your radio; the emission is much too loud, the other planets are complaining.
DM: Do you have any complaints?
Sol: Well, there is one: privacy. I know that when you become a star you give up a lot of personal privacy. But what happened was ridiculous. There are people looking at me at all times of the day. And they're not just looking, they're using telescopes. Big telescopes. They look at my back, my front; no place is sacred. Privacy is really a problem.
And sometimes I break out in those unsightly sunspots. You know, those dark magnetic depressions. You would hope they would look the other way, save me some embarrassment. No way. They take pictures. It's incredible. I break out in sunspots and next thing I know I'm on the cover of Astronomy Magazine or something.
And you know what really gets me. Sometimes I can't help myself. Sometimes I accidentally let go of a little gas. It's natural, it happens to everybody. "Solar Flares" they call it. They make movies of it. They tell their friends. Soon everyone is watching. I'm so embarrassed.
DM: This bothers you?
Sol: Oh yeah. Sometimes I get little mad. I think of implementing a little photon tax. Not much, just a penny or two per photon, or something like that. Just to let you know I'm here. Get a little respect once in a while. I could go on strike, you know. A few months without sunshine and you people would ante-up.
DM: How do you get along with the other stars in the solar neighborhood?
Sol: Pretty well. I don't interact with them as much as I used to. Sometimes we joke around, throw some snowballs at each other. But I'll tell you though, they're not happy with the privacy situation either. They claim that when you guys aren't looking at me during the day, you're looking at them at night. And with even bigger telescopes. Don't you ever stop?
DM: What's in the future for you?
Sol: Well, I'm not really so unhappy with my present job. It's a steady gig. Maybe in a few billion years I'll apply for Red Giant status. I don't have the helium yet. That's the key to being a good Red Giant - helium. But I'm saving up what I have. I'll get there. After that I'll probably just retire. Fade away slowly. Nothing like Brutus - he was crazy.
DM: We would like to thank you for all those years of warmth and sunlight. I think I speak for the whole planet when I say we're really grateful. Is there anything you wish to request of the people of earth?
Sol: Yes. Batteries. This fusion stuff is not going to last forever. Please send me 200 billion batteries. Didn't you people read the instruction manual? Size "Double D". And not those cheapies you find on sale.
Also a neutrino generator. One of mine seems to be on the blink.
Of Bits, FLOPs, and Free Parameters
by Robert J. Nemiroff (with apologies to Dr. Seuss)
(I wrote this in 1990 (or 1991) shortly after I arrived as a postdoc at the US Naval Research Laboratory. I had just attended a cosmology conference. I just couldn't get this idea out of my head, and soon decided to write up my own unsolicited conference summary. I kept trying to work on other things, but I would find myself daydreaming up new and wittier prose for this review. Soon I just gave in, pushed other work aside, and wrote it out. I recall having a strange sense that I was doing both the right thing and the wrong thing at the same time. But I just had to do it -- the prose had to come out. Even after my first complete draft, I obsessively nitpick edited it perhaps 100 times before submitting it, trying to get the science just right. My understanding was that there was considerable disagreement between the conference proceeding editors as to whether to include this unsolicited review, and it went back and forth for a while. I guessed that at least one editor thought it would be quite an insult to the prestigious person they had invited to write the real conference review to include this whimsical addition. Still, after a while, they did accepted it. I may have Prof. Virginia Trimble to thank for that, but I am not sure.)
New data had come to Skyville and all the funded residence rushed off to see it at the town meeting. The meeting hall buzzed with news of the new data and the technological inventions which made it possible. "They've got a Floating Frozen Microwave Mapper," someone overheard. "Did you hear about the Tree-Pronged Bit Swapper?" whispers repeated. Skyville was bustling with excitement.
Finally the meeting was called to order and the observers got up to speak about the new data. "10^12 bits" the observers said, showing some of them to a hushed audience. The audience was impressed: this was much more data than last year, and more impressive. Some of the data showed fascinating features like great groups and winding walls of galaxies, while other data showed featureless photon fields and because of this were just as interesting. It seemed more of the sky was explored during the last year than in all the history of Skyville combined.
Next it was the computationalists turn to speak about the improved computer calculations completed over the last year. "10^16 FLOPs" they said, letting the audience see some in action with a million mass movie. Again the audience was impressed, and again it was more than ever in the entire history of Skyville.
Next the theorists got up to speak on the current state of Skyville's understanding of the sky. "3 free parameters" the theorists said. An unsettling murmur spread over the audience. Skyville was not impressed. This was also more than last year, and more free parameters usually meant less understanding. Skyville yearned to understand the sky. Their hope now was that this new data would help increase understanding.
As the meeting continued Skyville learned more and more about the Sky they loved so much. But a complete understanding without free parameters eluded everyone. Try as they might, no one in Skyville could postulate a comprehensive theory to explain the data. It was as exciting time, but because of this it was also a time with an undercurrent of frustration.
Skyvillians made many comments to each other during the meeting as they strived to understand the sky. Here's what some of them said:
"Hoyle, Hubble, toil and trouble . . ."
"Frankly, my dear, I don't believe CDM."
"Read my MIPS."
They also sang of old and changing times:
"High H_o, high H_o, Hubble's constant we soon will know . . ."
"Omega_B, you're not half the value you used to be . . ."
"All in all it's just another indication of Walls . . ."
As the meeting drew to a close the theorists spoke about the likely understanding of the Sky in the next year, in light of the new data. "4 free parameters" they said sorrily. It seemed all of Skyville groaned collectively. And what was worse, many of the free parameters were either newly added, not very popular, or not well understood. Many of the theorists were dismayed, but others took this as a challenge. Some thought that it was this very lack of understanding that drove Skyville to its present frenzy.
Next the computationalists spoke about the future of Skyville's ability to compute the Sky. They were preceded by rousing rumors of even more improved computational ability. Surely this would help. "A Max VAX Universe Tracker," one computationalist was heard to say. "A Rapid Reading Correlation Catcher," whispers repeated. The computationalists called for order and spoke confidently about their expectations in the next year: "10^17 FLOPs," they forecasted. Skyville's excitement increased.
But the most impressive rumors were of new observational technology which would allow even newer, even more vast data sets. "A Quantum-Quick Quasar Quaffer," repeated one Skyvillian to another. "A Tinsel Tiled Galaxy Grabber," "A New SNU Neutrino Snarfer," "A Flying Refocused Photon Finder," rumored others. A hush descended as the observers finally spoke about what they expected in the next year: "10^13 bits," they promised proudly. Skyville roared in approval as the meeting ended. With this pronouncement, exciting times in Skyville were sure to continue.
(I wrote this in 1989, but I don't remember why. I was, of course, a big fan of the original Star Trek TV series, having watched every episode numerous times. I remember when I was a teenager I would come home from school and, many times, watch a rerun of Star Trek. My dad would come by and ask "Haven't you seen this episode before?" to which I would reply "So?". I realize that many people have spoofed Star Trek before, but it was a lot of fun taking my turn! )
Kirk: Engineering. Scotty?
Scotty: Aye Captain.
Kirk: Fix the warp engines, Scotty.
Scotty: But the warp engines are working fine captain.
Kirk: Scotty look, whenever we need the warp engines, they go blinky. So start fixing them now.
Scotty: But captain, they're in perfect working order . . .
Kirk: NOW Scotty. Fix 'em. Kirk out.
Captain's log, Stardate -34,212.6. The common cold. Already this mission 3 crew members have come down with it. Bones tells me he's doing everything he can but I can see the intense strain building on his face. How many people can he treat at once? How many tissues do we have left? How long can the crew remain silent while fellow officers sneeze and blow their noses?
Kirk: Sick Bay. Bones?
McCoy: McCoy here.
Kirk: About that third crew member . . .
McCoy: He's dead, Jim.
Kirk: He's what? All he had was a cold!
McCoy: I'm sorry, Jim. I did everything I could.
Kirk: Dead, from a cold? Bones, what ever happened to those twins we sent you in Episode 12 with the back problems?
McCoy: Their dead, Jim.
Kirk: And Ensign Cadaver from Episode 4 with the stubbed toe?
McCoy: He's dead, Jim.
Kirk: Bones, do you have a medical degree?
McCoy: No, Jim.
Kirk: Security, escort "Dr." McCoy to the brig.
Spock: Captain, if I may intrude for a moment, I estimate that the chances a healthy person dying of a common cold are 17 million 256 thousand and 12 to 1. Kirk: You really think so, Spock? Well were finally gonna test this little logical guessing game you like to play. We'll see. Computer on.
Computer: Working.
Kirk: What are the chances of a healthy person dying from a common cold.
Computer: 17,256,011.
Kirk: Not 17,256,012 ?
Computer: Negative.
Spock: Yes it is, Captain. I believe the computer is malfunctioning.
Computer: 11, Captain. I believe Spock is malfunctioning.
Spock: 12, Captain.
Computer: 11.
Spock: 12, 12, 12, 12, 12.
Computer: 11, 11 . . .
Bailey: What the &*%! does it matter!
Kirk: Thank you Mr. Bailey. (Note to the reader: Bailey was the one at the helm in the "Corbomite Maneuver" Episode who became upset when Sulu read off the exact time until the ship exploded.) We've had enough of your emotional outbursts. You're relieved. Mr Spock, thank you for your fascinating commentary. Mr. Chekhov, take the helm. Chekhov: But Captain, Sulu already has the helm.
Kirk: Not there, Chekhov, to his right. Where Bailey went emotional on us.
Chekhov: But Captain, there is not really so much for me to do there. The helm doesn't really need two people sir.
Kirk: NOW Chekhov! (Chekhov takes the helm.)
Computer: Aaaachooo!
Uhuru: Captain. We have an emergency distress signal coming in over all frequencies.
Kirk: Thank you, Lieutenant, put the signal on ship-wide intercom. Spock, get a fix on the signal's origin. Chekhov, Sulu - push some of those buttons and look busy.
Spock: The signal was originating from Starbase 4, Captain. They have just been destroyed by a Klingon Battle Cruiser.
Kirk: That music - you hear that? That dum - dum -dum. Who did that? (Nobody answers). Sulu, set course for Starbase 4.
Sulu: We're there Captain.
Kirk: Good work, Sulu!
Sulu: We've been there all along Captain. If you remember Captain, we're suppose to guard the starbase against enemy attack.
Kirk: Thank you, Mr. Sulu, I don't need constant reminders of our purpose. Uhuru, let's here a replay of that distress message.
Uhuru: But Captain, it says . . .
Kirk: NOW, Lieutenant.
Distress Message: Enterprise, Enterprise, come in Enterprise. You're supposed to be guarding us, you morons! Wake-up out there! Klingon Battle-Cruiser this sector firing at will - what's the matter with you people! Major damage - all sectors. OK, enough of this - get me the Klingons. Klingons, Klingons, come in Klingons. We hereby fully surrender, denounced our Federation membership, and request permission to tell you all our Federation secrets. Klingons, Klingons, come in Klingons . . . (a loud explosion is heard, the bridge rocks with people being thrown everywhere. One might consider this peculiar considering they were listening to a recording. Then silence).
Kirk: Spock, analysis.
Spock: I believe Chekhov has a point, Captain. You don't really need two people at the helm.
Computer: Aaaaachooo!
Kirk: No Spock, I mean . . . Damn it! Just a minute. (Punches intercom button.) Brig. Bones? The computer has a virus or something. Look into it.
McCoy: But Jim, I'm a doctor not a computer programmer. Or I was a doctor. Ok, look, I'm not really a doctor, but I preformed a brain transplant on Mr. Spock once - not bad, huh? Wait a minute! Brain transplant - computer virus. By golly, Jim, I'm becoming a miracle worker, someday I'll be able to cure the common cold!
Kirk: Wonderful. Security, take the computer to the brig so Dr. McCoy can fix it. Spock, how's the starbase holding up with our protection.
Spock: It's completely destroyed sir. I believe it was the Klingons. The same ones that appear to be firing at us now.
Kirk: Sulu! Damn-it! Tell me these things! (Sulu points to Chekhov who has fallen asleep and silently insinuates that he didn't want to wake him.) CHEKHOV! Wake up! Keep your eyes on the screen!
Chekhov: What? Huh? With Yeoman Rand? Where? Oh yes, Captain, sorry. Klingons. Sulu was watching them, Captain. Oh yes, your right Captain, that's really my job. But with there so little to do and . . . well, Ok, I guess you know that argument. Klingons? Oh yes, Klingons. Oops, there he is. He's firing again Captain. Uh oh, that last hit was bad. We lost photon torpedoes and phasers. Oops! Drat, now the shields are gone too sir. Sorry, my fault, I should have been paying more attention.
Uhuru: Damage reports coming in now, Captain. All decks, sir. Heavy damage in the upper and middle decks, sir.
Spock: That excludes engineering. Maybe the warp engines are undamaged, Captain! Kirk: Yes! Thank you, Mr. Spock! Engineering! Scotty! Prepare to warp us out of here. Scotty: Aye Captain, I'd be happy to oblige you, but when fixing the warp engines, which, by the way, weren't really broken, we accidentally caused the magneto-anti-dylithium neutrafier to graze the electro-ferra- . . .
Kirk: What?
Scotty: The warp engines captain, they don't work anymore. I suggest we buy new ones sir. These are always broken.
Kirk: Fine. Impulse engines Scotty, on the double. Get us out of here.
Scotty: Well Captain, when you said to fix to warp engines even though they weren't really broken, it occurred to me that the impulse engines weren't broken either. So we decided to fix them as well.
Kirk: Now Scotty, full impulse engines.
Scotty: Don't you like where we're at, Captain?
Kirk: NOW, Scotty.
Scotty: Aye, we're re-assembling them now Captain. They should be ready by next episode.
Kirk: Spock, analysis.
Spock: Death, Captain. Soon we will all die hideous painful deaths at the hands of Klingons, Captain.
Bailey (from intercom): Oh my god, now you've done it! We're all gonna die!
Kirk: Bailey, shut up. You're relieved. Scotty, take your biggest and strongest men, and send them out to push. We must get out of here.
Scotty: Aye, Captain. That's all well and good, Captain. But the laws of physics say that unless there is something to anchor your feet against, Captain, pushing is just no good. It's just action and reaction sir.
Kirk: Action and reaction? What the . . . Scotty! That's it! You just earned your pay for this week! Reaction! The warp engines can finally do us some good after all these years of malfunctions! Scotty, you've done it!
Scotty: You sound like you need some rest, Captain.
Kirk: Scotty, have the warp engines dismantled: we will eject them toward the Klingon ship.
Scotty: Aye Captain! Yes! That will push us away from the Klingons and give them a bogus warp drive at the same time! It'll take them generations to understand why that thing breaks down only they need it most!
Spock: Fascinating, Captain, but I calculate that we have only 9 minutes and 12.5 seconds stop gloating and blow us to smithereens, Captain.
Kirk: 9 minutes, 12.5 seconds? Computer on.
McCoy (from intercom): It's dead, Jim.
Uhuru: Captain, I have the Klingon Commander on line 3 Captain. He wants us to surrender before he blows us to smithereens.
Kirk: Uhuru, put him on the screen. (The ugly mug of a Klingon appears on the screen.) Hey Klingons, are you crazy? Don't you know we have Corbomite now? Get lost.
Klingon Captain: Corbomite? Really, Captain Kirk. Very nice. Yes, we saw that episode. Is it as good as our new Super-Improved- Corbomite?
Kirk: Damn. Super-Improved-Corbomite! Now we're *&^$#. Spock, analysis.
Spock: We're really *&^$# now, Captain.
Klingon Captain: Yes, quite. May I remind you that it is Klingon tradition to have the enemy surrender completely and unconditionally before we blow them to smithereens. Would you be kind enough to comply?
Spock: That is illogical, Captain. He's going to blow us to smithereens anyway. Why should we surrender first?
Kirk: You have preformed admirably, Captain. Let me be the first to congratulate you on a spectacular victory. Myself and the crew - we are very much impressed. In fact, we'd like to give you a present. It's our warp drive, sir. Think of it like the ear of a bull won in bull fight. You deserve it, Captain. We won't be needing it after you blow our ship to smithereens causing each of us hideous painful death, so we want you to have it. Klingon Captain (Wiping away tears): You really think so? We can have it? Oh how the other Klingon Captains would be jealous! You would really give us your warp drive? Kirk: Certainly. And oh yes, the ship's computer, too. It, ah . . . it . . . helps run the warp drive. Yeah, that's it. It helps run the warp drive. Why don't you install it, zip around the galaxy a few times, and see how you like it. In the meantime, we here on the Enterprise can write up a full surrender speech and present it to you when you come back - just before you blow us to smithereens.
Bailey (from intercom): Oh my God, we're all gonna die!
Kirk: Security. Get Mr. Bailey in his quarters. Stuff him in the warp drive or something, just shut him up.
Klingon Captain: It's a deal!
Kirk: Transporter room. Kirk here. Prepare to beam the warp drive and the ship's computer over to the Klingon ship. Beam on my signal. . . Ok, now.
Transporter Room Ensign: Are you sure you want to do this Captain? Weren't we supposed to push . . .
Kirk: YES, Ensign. I WOULD appreciate it if you and everybody would stop questioning my orders.
Transporter Room Ensign: But Captain, I . . .
Kirk: NOW, Ensign. (The sound of the transporter is heard. A few minutes later the Klingon ship's lights go out, and it lies dormant, powerless and tilted in space.) Kirk: Mr. Spock - please tell us again what is the chance that a healthy person will die from a common cold?
Spock: Yes, Captain, you are indeed correct. I was in error. It is 12,981,011 as the computer originally predicted. But you must admit that this supposed error helped in keeping our galaxy free from Klingon devastation. And it also helped us to find out more of the, shall we say "unusual" past of our beloved Dr. McCoy. So you might think of it as something of a gambit, Captain.
McCoy (from Brig intercom): Hey, Captain, guess what? My mail- order diploma just came in this morning's mail! I'm a REAL doctor now, Jim! No matter what that pointy-eared half-Vulcan says! I'm a real doctor now! Yippie! Can I return to the Sick Bay and talk Nurse Chapel out of making baby eyes at Spock?
Kirk: Why not? And Mr. Sulu, let's get out of here. Warp 5. Straight ahead.
Sulu: But Captain, we just gave the warp . . .
Kirk: Sulu, please DON'T question my orders.
Sulu: But the planet is straight ah . . .
Kirk: NOW, Sulu.
Sulu: Aye, Captain.
Every smiles good-naturedly as the closing music is played. The Enterprise then slams right into the planet killing everyone.
Bailey (from the great beyond): Oh no, now we're all dead! I knew this would happen. NOW maybe they'll listen to me when I complain!