Reed's Ruminations: A Blog by Dan Reed

On WSPRs, Geeks, and Luftballons

A few weeks ago, a large, high altitude balloon carrying a large payload of Chinese surveillance equipment was shot down by a U.S. Air Force F-22 Raptor using an AIM-9X Sidewinder missile.  The F-22 was scrambled from Langley Air Force Base in Virginia, the downed balloon landed offshore in South Carolina, and an intensive intelligence salvage operation began immediately.  This was a large, likely multiple ton, balloon with a substantial array of solar cells, navigational and directional capability, and electronic packages. Because my wife and I own a beach house in North Carolina, near the South Carolina border, I received local warnings to report and deliver any floating debris to the authorities.

In response, Secretary of State, Antony Blinken, canceled his planned trip to Beijing, and Sino-American geopolitical relations, already strained, sank even further, with public accusations and recriminations flying fast and furiously. The resulting consternation ignited political posturing by pundits across the political spectrum, which, in turn, led to strident calls for a public statement from the U.S. President.

How could this happen?  Why wasn’t the balloon detected earlier?  Why was it allowed to drift across the continental United States before being shot down? Members of the U.S. Congress wanted answers, and they got them via classified national security briefings. Meanwhile, NORAD increased its aerial and near-space surveillance, given heightened national awareness and concern about other objects in the sky. 

International law is clear that satellite overflights of a country (i.e., from space) are allowed.  Conversely, countries have the complete right to control the airspace over their territory, though international law is a bit uncertain about the rules at very high altitudes. As a practical matter, might makes right; if you can shoot it down over your territory, you have the right to do so. (Remember the history of the U-2 and SR-71 spy planes and note that the aerial photograph of the balloon above was taken from a U-2.) Relatedly, radio transmissions are controlled by International Telecommunications Union (ITU) treaty, as part of the United Nations.

Putting aside the very real and important national security questions behind the Chinese balloon and the calculations and mis-calculations that led to its path over the United States, the public hysteria and political ennui brought to mind the West German band Nena’s song, Neunundneunzig Luftballons (99 Air Balloons), which became an international hit in 1983 as an anti-war protest. It’s a poignant and melancholy Cold War story of a group of air balloons accidently triggering a 99-year global war, leaving only devastation in its wake.  (You can find it on YouTube, and the German lyrics are just as haunting today as they were forty years ago.)

            Original German                                          English Translation
            Neunundneunzig Jahre Krieg                        Ninety-nine Years of War
            Ließen keinen Platz für Sieger                       Left no room for winners
            Kriegsminister gibt's nicht mehr                    There is no longer a Minister of War
            Und auch keine Düsenflieger                        And also no jet planes
            Heute zieh' ich meine Runden                      Today I'm doing my rounds
            Seh' die Welt in Trümmern liegen                 See the world in ruins
            Hab' 'n Luftballon gefunden                          Found a balloon
            Denk' an dich und lass' ihn fliegen               Think of yourself and let him fly

An English version of the song, entitled 99 Red Balloons, was released shortly thereafter, with somewhat different lyrics, changed to scan with the original melody, and with a more American-centric story. However, many felt it lacked the punch and ennui of the German original, and it failed to achieve the same chart success. My always limited German language skills have atrophied with time, but I am inclined to agree.

But Wait, There Are More!

In the wake of the balloon shootdown, questions were asked; fingers were pointed; answers were demanded. Were there other, heretofore undetected, balloons floating over the United States and Canada? It turned out the answer was yes, and UFOs, literally – unidentified, flying objects – were back in the news. Soon, an unidentified balloon was reported to be drifting over Alaska and heading into the Canadian Yukon, followed by another over Lake Huron, the latter described as “small, metallic balloon with a tethered payload below it.”

After diplomatic consultations between the U.S. and Canada, both were shot also down by the U.S. Air Force. (Technical aside:  One should be impressed at NORAD’s surveillance capability, which can identify extremely small objects in the large airspace over North America.)

Were aliens in flying saucers involved?  Was North America under threat?  Was there a new Cold War “balloon gap,” analogous to the infamous “missile gap?” Inquiring minds wanted to know.

In a “you can’t make this stuff up” farce, a comedic mashup of fiction and reality – Dr. Strangelove (“You can’t fight in here.  This is the War Room!”), Neunundneunzig Luftballons, and national security fears – the news was soon filled with speculation that the United States may have used a $100M advanced fighter jet and $500K missile to shoot down a $250 amateur science project, a pico-balloon carrying a ~10 gram, extremely low power amateur “ham” radio transmitter. (See Hobby Club’s Missing Balloon Feared Shot Down By USAF.) 

The group potentially responsible for a balloon traversing America’s airspace is – wait for it – the Northern Illinois Bottlecap Balloon Brigade (NIBBB), a group of pico-balloon amateur radio  and science enthusiasts whose club was named in homage to the 2009 animated movie Up. If you are thinking these people hardly look like a threat, you would be right; the NIBBB notes that the club’s youngest member is only eleven years old.

One of their balloons, with amateur radio call sign K9YO-15 (balloons use the FCC call sign of the owner and a distinguishing number), was launched from Libertyville, Illinois on October 10^th, 2022 and had circumnavigated the Earth multiple times before the club lost contact with it.  The balloon’s last reported position was at 38,910 feet off the coast of Alaska on February 10^th, 2023.  The NOAA HYSPLIT wind model projected the balloon to be over the Yukon on February 11, 2023.

On that same day, NORAD downed an unidentified airborne object in the Yukon. The Pentagon issued a press release, and the Canadians searched briefly for the downed fragments but found nothing. 

However, given the small size of the balloon and the remote environment, it is unlikely that any debris will ever be found, leaving it uncertain if K9YO-15 was indeed the object shot down.  The NIBBB responded with equanimity as well, noting:

As has been widely reported, no part of the object shot down by the US Air Force jet over the Yukon territory has been recovered. Until that happens and that object is confirmed to be an identifiable pico-balloon, any assertions or claims that our balloon was involved in that incident are not supported by facts.

Even as the klieg lights of sudden and unexpected fame can be harsh and unforgiving, the NIBBB website shares the group’s passion for pico-balloons and amateur radio, including detailed construction details and a bill of materials.

That’s quite the story, with lots of jargon.  Let’s deconstruct this just a bit, starting with pico-balloons.

That’s My Science Project

What’s a pico-balloon, you say?  As the name suggests, it is a very small balloon, such as the Mylar kind filled with helium one might purchase from a hobby store or a florist shop. As anyone who has brought one to a birthday or graduation party knows, they can be purchased for a few tens of dollars, and are at most a few feet in diameter.

Most such balloons are soon discarded, though a few escape, intentionally or unintentionally, and float into the sky, never to be seen again. How high can they go?  The answer depends on the characteristics of the balloon material, gravity, and the Archimedes principle.  Scientifically, a balloon rises until it reaches neutral buoyancy (i.e., the balloon’s average density equals that of the surrounding air and the buoyant force balances gravity).

Simple Mylar party balloons are unlikely to rise to altitudes beyond 6-9 kilometers (20,000-30,000 feet), limited by either helium leakage or material stresses. In contrast, hardier balloons made of polyethylene or latex are neutrally buoyant up to fifteen kilometers (~50,000 feet). Bear in mind that these pico-balloons are much, much smaller than weather balloons ,and are often less than one meter in diameter. As such, their payload is also quite small, only 10+ grams (0.02 pounds for those of us in non-metric countries), and they are easily affected by weather, even frost.  For this reason, experienced pico-balloon aficionados prefer balloons that can rise to higher altitudes above most weather.

Under Federal Aviation Administration (FAA) 14 CFR Part 101, such small balloons can be launched by anyone, anywhere in the United States, without the need for FAA approval or reporting.  The only requirements are that the balloon’s payload weighs less than six pounds, and the balloon is launched in an unrestricted area. Given the low weight of pico-balloons, the jet stream can easily carry a balloon around the world, and there are many tracked and reported cases of such circumnavigation.

Why might anyone want to do this?

The truth is that people do such things because they are possible and they are cool. After all, humans did not build the Gossamer Albatross to usher in a new era of human powered aviation. They did it because building a human powered airplane that could cross the English Channel was the union of technological prowess and human inspiration; it was the epitome of cool. (See Science: It’s About the Wide-Eyed Wonder.)

Amateur radio pico-balloon enthusiasts are no different. Yes, the technical answer is that amateur “ham” radio enthusiasts often use pico-balloons to study the atmosphere and radio propagation characteristics, but the joy is in the construction and the operation. In such cases, the payload consists of a GPS tracker for location, a WSPR or APRS radio transmitter, an inexpensive micro-controller such as an Arduino to manage the transmitter, and a few solar cells to power everything. A complete package, including the balloon, can be assembled for at most a few hundred dollars.

What’s WSPR, you say?  It’s an acronym for Weak Signal Propagation Reporter (WSPR). The protocol was designed by Nobel Laureate and avid ham radio operator, Joe Taylor, and is intended to test propagation paths on the medium frequency (MF) and high frequency (HF) amateur radio bands using very low (and I do mean very low) power transmissions. A typical WSPR message contains the licensed amateur radio operator’s call sign, the Maidenhead grid locator of their rough location, and the transmitter power in dBm.  The entire message fits in 50 bits. Transmitting such a 50-bit WSPR message takes 110 seconds at 1.46 baud.

Okay, let’s now reconstruct the phrase “pico-balloon carrying an extremely low power amateur ham radio transmitter.”  There are several design options of varying price and sophistication, but here are the elements of a globe-circling amateur balloon, all within the expertise of amateurs and even (with supervision) high school science enthusiasts

• One small balloon (~$50 from Yokohama Balloon or ~$150 for a slightly more expensive research version from Scientific Balloons)
• Helium canister for balloon charging (~$75)
• Lightweight (~10 gram) payload, custom designed or purchased (e.g., ~$90 ZachTek WSPR controller or ~$130 QRPLabs LightAPRS-W tracker) h containing
  • GPS tracker
  • Microcontroller (e.g., Atmel 8-bit AVR microcontroller)
  • WSPR and/or ARPS transmitter
  • Solar cells for electrical power
• Dangling 0.005-inch wire dipole antenna of resonant length for the desired frequency (~33 feet for the 20m radio band)
• Open source software
  • Microcontroller configuration (call sign, Maidenhead locator, …)
  • Flight path prediction (e.g., NOAA HYSPLIT model or com)
  • Pico-balloon tracking collaborations (e.g., https://amateur.sondehub.org/ or http://lu7aa.org/wsprset.asp)

I would have been thrilled beyond measure to have had the opportunity to participate in such an activity as a student. Such projects are precisely the kind of experiential science education that can excite students, filling them with the realization that science is not a desiccated recitation of facts, but rather is an endeavor filled with passion and excitement. We need many more such experiences if we are to attract the millions of additional STEM workers this country needs to remain globally competitive. (See For Science and Society, The Future Begins with Better Dreams.)

My Geek Bona Fides

During the COVID-19 pandemic, my days (and often my nights) were consumed with managing the evolving crisis and ensuring the academic enterprise of the University of Utah continued with as few disruptions as possible.  Given the stress and uncertainty, I regularly counseled others to spend some time away from work doing something fulfilling and emotionally satisfying, lest they be overwhelmed by uncertain and rapidly changing events. (See The Simple Things Matter, Most of All Now.)  Knowing I should heed my own advice, I sought solace and distraction in my own happy place – science and technology. 

My first stop was my digital astrophotography hobby, which though always rewarding, proved uncertain, given the vagaries of Utah weather.  More practically, the late nights were incompatible with my day job due to sleep deprivation. After that, I turned to writing (See My Balm in Gilead) and, finally, to my tried and true friend, computing technology.

First, I developed an end-to-end suite of software for low-power, wireless sensors, using an inexpensive ESP32 microcontroller, LoRaWAN low power wireless transceivers, environmental sensors, and open source software tools.  All of this served as a citizen science complement to the larger SAGE project of which I am a part. (See Cyberinfrastructure: Lots of Room at the Edge.)

With the sensor suite complete, I studied for and passed the three FCC-mandated examinations – technician, general, and amateur extra – to become a fully licensed amateur radio operator.  The study materials gave me an opportunity to refresh my memories of electricity and magnetism (E&M) from undergraduate physics and also to delve more deeply into antenna and radio transceiver design.  The geek in me wished the American Radio Relay League (ARRL) examination materials had been a bit more mathematical, as many of the concepts are best understood via their roots in Maxwell’s equations.

All of which brings me back to pico-balloons and the “because we can” excitement.

First, armed with an ICOM-7300 radio and a trusty PC, I fired up WSJT-X and transmitted WSPR messages (radio callsign KK7EUJ, maidenhead grid locator DP40cp) at one milliwatt on the 20M HF band (14.0956 MHz) using my house’s gutter as an end-fed-random-wire (EFRW) antenna. Reflecting the effectiveness of WSPR encoding in a high noise, low power environment, these WSPR messages have been received by amateur stations across the entire continental United States, as shown at the right.

Second, I have written extensive code for the ESP32 and its smaller, 8 and 16 bit cousins, the ESP8266 and Arduino, the latter based on the Atmel 8-bit AVR microcontroller. Just as the proliferation of powerful, open source software has enabled construction of complex toolkits, so too have open source libraries of hardware components and powerful, open source design tools (e.g., KiCAD) facilitated design of inexpensive, custom hardware boards. As a complement to the wireless sensor library, I have also specified the design of custom printed circuit boards (PCBs), with GPS, sensors, and ESP32, which have been inexpensively fabricated in varying quantities by one of many fabrication firms.

Put another way, I realized I had followed the same path as my pico-balloon compatriots, using open source software and easily designed open source hardware to build new and interesting projects, ones that in the past would have been only within the reach of nation-states.

Coda

Sino-American relations remain deeply strained, and not just because of the surveillance balloon. The global battle for 21^st century geopolitical hegemony is underway, waged via economic competition, trade sanctions, diplomacy and political rhetoric, and military posturing. There seems little doubt that we are entering a new political era, one fraught with uncertainty.

In the midst of all this, I was on Capitol Hill just days after the downing of the Chinese balloon, discussing the importance of greater U.S. research investments and developing the domestic STEM workforce. I have been working with the National Science Board (NSB) on the intersection of dual use technologies, basic scientific research, economic competitiveness, and national security. In addition to a variety of meetings on Capitol Hill, I have invited the White House National Security Advisor, Jake Sullivan, to join the NSB in discussions.

Meanwhile, no alien UFOs have been confirmed, unless the definition of alien includes a loose affiliation of geeks (young and old), all fascinated by the combination of electronics, software development, atmospheric sciences, mathematics and coding theory, ionospheric electromagnetic propagation, and small balloons, and all wishing to share their scientific passion with others.  Could they be?  They might be. The signs and omens are all there. Yes, they are.  I am proud to say those are my people.

I am a geek, albeit an aging (and balding) left-handed, myopic one, some of those adjectives necessarily being implied by the geek nom de guerre.  I will try not to trigger an international incident by launching any pico-balloons of my own, but I will not rest in encouraging the passion of discovery in young minds.

Nena WSPRed rightly about the Luftballons:

        Man, who would have thought
        That it will come to this
        Because of ninety-nine balloons

Cyberinfrastructure: Lots of Room at the Edge

Twenty years ago, while director of the National Center for Supercomputing Applications (NCSA) at the University of Illinois, I gave a series of presentations on the future of cyberinfrastructure.  Looking through my presentation archives, I found this diagram, which emphasized building up (i.e., larger and more capable computing systems and scientific instruments) and building out (i.e., ubiquitous access to computing systems and instruments of all sizes, including distributed sensor/actuator networks).

At the time, we did not understand that fungal infections from chytridiomycosis were in part responsible for widespread death and extinction among amphibian species. I suggested with some seriousness that we instrument selected biological environments with MEMS smart dust and track flora and fauna movements in situ. Or, as I put it at the time, “Why not give every frog an IP address?” Of course, understanding ecosystem evolution eventually became the raison d'etre for the for the National Ecological Observatory Network (NEON) – more on that later. I also had several conversations with then National Science Foundation (NSF) Director, Rita Colwell, about the prospects for HPC-enabled in silico computational modeling of biological processes, something only now becoming possible. (See In Silico Whole Cell Modeling.)

With the perspective afforded by twenty years, how have we fared in realizing this integrated vision?  Building up, we have done well in most contexts, but we now face the twin realities of escalating costs and technology challenges at very large scales, both for computing platforms and for scientific instruments. It is unclear how many multibillion-dollar instruments or computing platforms we have either the political or economic will to construct. As for building out, it is a mixed story of broader access to midrange commodity Linux clusters and new instruments, but we are far from having created a ubiquitous scientific infosphere.

Building Up

In 2002, the first terascale computing systems had just become operational, including the NSF TeraGrid, which connected NCSA, SDSC, Argonne National Laboratory, and Caltech via a then extraordinarily fast 40 Gb/s transcontinental network. (See NCSA@30: The Revolution Continues.) At the time, we were still dreaming of petascale data storage systems, and research discussions about the path to petascale computing were well underway. The 1996 and 1999 Caltech Petaflops workshops at Bodega Bay and Santa Barbara were seminal in shaping the future that because petascale computing.  As this photograph shows, we have all aged a bit in the intervening twenty five years, and sadly, a few of our colleagues are no longer with us.

Today, terascale and petascale computing systems via GPU accelerators are widely available, spanning desktops to laboratory and university cluster environments. In May 2022, the latest incarnation of the TOP500 list of (some of) the world’s fastest computers was revealed at the ISC High Performance conference in Hamburg, Germany. There was much hype and chest thumping as the Oak Ridge National Laboratory (ORNL) Frontier system broke the exascale barrier based on the high-performance LINPACK (HPL) benchmark for dense matrix factorization.

I say “some of” the world’s fastest computers because submission to the TOP500 list is entirely voluntary, and some organizations chose not to submit entries, for a variety of reasons, some political and economic, some technical, and some related to national security. In particular, China has at least two exascale systems (OceanLight and Tianhe-3), but has not submitted either of them to the TOP500 list, likely to avoid further inflaming tensions with the United States.

Meanwhile, the very definition of exascale is itself now subject to debate, both because the dense linear algebra of the HPL benchmark is widely acknowledged as no longer a good predictor of multidisciplinary application performance and because mixed precision arithmetic is now common in many applications, particularly those with AI-accelerated components. When convolved with system and chip energy constraints, semiconductor fabrication challenges, and ecosystem shifts, a sense of quo vadis now pervades the field. (In a recent essay and associated article, my colleagues Jack Dongarra, Dennis Gannon, and I offered a few thoughts on strategic directions.)

In the world of computing, the locus of innovation has focused on the small and many (think smartphones and the Internet of Things) and the few and large (think commercial cloud infrastructure and leading edge HPC systems); scientific instruments are dominated by a similar dichotomy, though the large and few receives most of the attention and visibility.  As for large-scale scientific instruments, one need look no further than the aging Hubble Space Telescope, the recently upgraded Large Hadron Collider (LHC), the International Thermonuclear Experimental Reactor (ITER) being built by a global consortium, the soon to be operational Vera Rubin Observatory (nee Large Synoptic Survey Telescope) and James Webb Space Telescope (JWST), or the three proposed and under construction Extremely Large Telescopes. Biology has joined the physical sciences in large-scale instrumentation, spanning the Human Genome Project to the National Ecological Observatory Network (NEON).

Like leading edge supercomputers, these big instruments are the shiny objects featured in press releases and popular science stories, and well they should be, for they produce scientific results achievable via no other mechanism. Each involves hundreds, thousands, and sometimes tens of thousands of scientists and engineers; each requires billions of dollars, euros, or yuan to design, build, and operate; and produces torrents of data that demand petascale data archives and global content distribution networks. Thanks to these new instruments, the associated explosion of scientific data has been hugely democratizing, allowing many researchers to focus on the key intellectual questions rather than the mechanics of data acquisition. 

However, this brave new world and the shift from data paucity to plethora have also raised new questions about data preservation and triage, shared and individual responsibilities, FAIR access, and the costs of data retention, as mandated by government sponsors.  (See Research Data Sustainability and Access and My Big Scientific Data Are Lonely.) As the late Nobel Laureate Herbert Simon once noted,

What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.         

In my humble opinion, we have not yet fully internalized the social, political, economic, and technical implications of the shift from data scarcity to data ubiquity.  What data we keep and for how long, how we manage access, and how we distribute the very real costs of data management and preservation have not yet been fully resolved.  Serious science policy issues remain unresolved.

On Being the Right Size

Although our intellectual curiosity is rightfully unbounded, our financial resources are not. What are the right sizes and relative numbers for our computing platforms and our scientific instruments?  Is it all about faster and bigger, which in almost any realistic budget scenario, also means fewer?  In the limiting case, as illustrated by ITER and the LHC, fewer may well mean only one per planet, financially sustained only via a global scientific partnership.

Nor does the fewer and bigger choice only affect the high end, it squeezes investment in the long tail of campus and laboratory infrastructure as well. These are important science policy questions, especially when one realizes that the day-to-day, “bread and butter” work of science is conducted with many instruments of much smaller size, educating students in scientific processes.  (See On Being the Right Size: Science Scaling and Power Laws.)

Make no mistake; form and scale absolutely follow function, as the British geneticist J. B. S. Haldane first wrote in a now famous essay entitled On Being the Right Size. (You can read the original essay as originally published in Harper’s Magazine, skipping the political pontification at the end.)  In it, Haldane offered cogent arguments about surface area to volume ratios, structures, respiration and energy, noting:

The most obvious differences between different animals are differences of size … it is easy to show that a hare could not be as large as a hippopotamus, or a whale as small as a herring. For every type of animal, there is a most convenient size, and a large change in size inevitably carries with it a change of form.

Computing systems and scientific instruments are no different. A simple refractor telescope design cannot be expanded unchanged to create a 30-meter optical observatory; the materials properties do not scale linearly, and the control systems differ markedly. A child’s horseshoe magnet cannot be scaled to create a 100-tesla pulsed research magnet. Nor is an exascale computer just a big deskside cluster, despite some comparable elements. The lack of self-similarity at scale means new design approaches are needed in different operating regimes, each with associated design costs.

Building Out

Returning to my twenty-year-old diagram, we have made progress in building out, with midrange computing clusters now more common and new classes of midrange instruments (e.g., cryo-EM) changing the face of science. Cyberinfrastructure training materials are now widespread, and we even host student cluster construction competitions; building a Linux cluster is no longer an arcane art. Indeed, one can now build an inexpensive Raspberry Pi cluster with almost all the same software tools found on leading edge supercomputers.

Despite this progress, our intellectual fascination with a small number of large instruments and computing platforms often blinds us to the power of distributed discovery and innovation with large numbers of instruments at smaller scales. Despite this, we have made enormous progress in sensor miniaturization, perhaps even more than Feynman envisioned in his seminal, “There’s Plenty of Room at the Bottom” nanotechnology lecture. (See Come to the Supercomputing Big Data Edge.) In this sense, we have a long way to go to realize the vision of a ubiquitous infosphere. Let me illustrate with a specific example.

SAGE: A Software Defined Sensor Network

Over the past two years, I have greatly enjoyed playing a small part in the NSF-funded SAGE project, which is seeking to build a flexible and extensible hardware and software cyberinfrastructure for AI at the edge.  Led by Pete Beckman, SAGE builds on many ideas from Charlie Catlett’s urban science and environmental science Array of Things (AoT) project, which deployed sensors on utility poles across the City of Chicago and surrounding areas.

AoT engaged community residents in defining acceptable data use policies, explored data driven social science by direct measurement, and stimulated citizen science and K-12 hands-on education. For a review of the AoT project and the motivating lessons for SAGE, I recommend this IEEE Xplore article, which several of us wrote recently.

SAGE is based on the notion of software-defined, configurable sensors that include edge computing, standard sensor interfaces, a set of baseline sensors, and extension interfaces for environment-specific sensors, along with backend data storage and analysis facilities.

A standard “wild SAGE” node (named for its hardened packaging and fault tolerant design to run unattended in harsh conditions, or “in the wild”)  includes the following:

• One or two NVIDIA Xavier NX GPUs with Wi-Fi, Ethernet, and a full Linux stack
• Power over Ethernet (POE) sky and ground facing cameras
• An optical rain sensor and a microphone
• Relative humidity, barometric pressure, and temperature sensors in a Stevenson shield (that’s the box on the left in the photograph, which also contains a Raspberry Pi to manage the Bosch BME 680 sensor)
• POE and USB connectors for additional, domain-specific sensors

Why deploy AI at the sensor edge?  First, the volume of streaming data may preclude storage, either due to bandwidth limitations or storage capacity.  For these same reasons, the LHC’s detectors rely on real-time triggers detect and record only a small subset of particle collisions. Second, urban and environmental monitoring bring strong security and privacy concerns, particularly involving video. By extracting events and features entirely at the edge, the raw video need never leave the sensor, obviating many privacy concerns (e.g., the timing and frequency of pedestrian interactions may suffice, without saving any identifying video). Third, AI at the edge means the sensor can adapt dynamically to changing conditions (e.g., increasing measurement frequency based on a detected event or even triggering alternative measurements at other locations). Fourth, latency constraints may demand a response time possible only when decisions are made at or near the edge.

In SAGE, the key to this flexibility is an end-to-end architecture – Waggle AI@Edge – that allows domain scientists to focus on their science, rather than building generic infrastructure.  One need only connect new sensors, develop domain-specific AI plugins using widely available AI tools (e.g., OpenCV, PyTorch, and TensorFlowLite) to process sensor data in situ on SAGE nodes, and use tools such as Jupyter notebooks to analyze and correlate data.

To demonstrate this integration capability, the SAGE project has been working with NEON to integrate inexpensive SAGE nodes for augmented environmental monitoring and with California and Oregon wildfire researchers to detect and respond to the growing prevalence of wildfires. There are now SAGE plugins for vehicular traffic detection, bird song classification, cloud cover, smoke, and other flora and fauna events. 

At Utah, we have deployed one SAGE node at the Natural History Museum of Utah (NHMU), which allows us to track environmental conditions and local flora and fauna.  Additional SAGE nodes are being deployed in downtown Salt Lake City to monitor PM 2.5 particulate air pollution and to study advanced wireless technologies in collaboration with the NSF POWDER project.

Much More Room at the Bottom

At ~$15,000 each, SAGE sensors are inexpensive compared to multibillion dollar mega-instruments.  Yet even this price is prohibitive for many research contexts, limiting their deployment in science and community education projects where resources are even more bounded and precious.  Fortunately, there is much more room at the bottom.  In this spirit, members of the SAGE education team are working on lower cost SAGE versions based on the Jetson Nano and the Raspberry Pi.

However, imagine a cyberinfrastructure world containing even lower cost sensors -- tens of thousands of them at ~$50 each! Leveraging a wealth of inexpensive and open source hardware and software, it is now possible to create and deploy a powerful sensor infrastructure for just a few hundred dollars.

Let’s start with server infrastructure. A Raspberry Pi with a complete Linux software stack, including Docker and Kubernetes virtualization, a high-capacity SD card for system software and data storage, Wi-Fi, Ethernet, and LoRaWAN long-distance communication transceiver can be assembled for approximately $100.

Such a system can readily serve as a base station for hundreds of microcontroller nodes, each supporting a wide variety of ~$5 sensors. As an example, one of my former colleagues at Iowa fabricated the sensor node shown at right based on my specifications. This node includes

• Powerful ESP32 Tensilica microcontroller with Wi-Fi and Bluetooth LE
• Bosch BME 280 pressure, temperature, and humidity sensor
• LoRaWAN low bandwidth, long distance communication
• I²C, analog, and digital Grove, solderless sensor connectors
• Battery and/or solar power interface for long-term episodic operation

Such sensor nodes can be readily designed using open source hardware tools such as KiCad and fabricated in quantity for less than $100 each, or similar ones can be purchased commercially (e.g., Heltec Wi-Fi LoRa32 V2 or Adifruit Huzzah32 Feather with Feather Wing addon sensors).  Equally importantly, they and the Raspberry Pi base stations can be integrated directly with environments such as SAGE. Just as commodity clusters opened computing to a wide range of practicing scientists, inexpensive sensors can be the missing and catalyzing element to realize the vision of an immersive sensor/actuator infrastructure.

Over the past two years, as a research proof-of-concept, I have developed a generic software infrastructure for Raspberry Pis and ESP32 sensors, while Charlie Catlett has been exploring complementary hardware and software models based on inexpensive home automation tools.  We are now drafting a “how to” book, targeting K-12 STEM education and citizen science.

Cyberinfrastructure Futures: The Missing Elements

We have come far in building out a 21^st century cyberinfrastructure, but there is more to do if we are to fully exploit the opportunities now afforded by scale – both small numbers of large computing systems and instruments and large numbers of small computing systems and instruments. As we imagine a future world of software defined sensors, I believe we must address the following research challenges, and I encourage all of us think about how we might mount national research programs to explore each of them.

An Extensible Cyberinfrastructure Baseline. At present, we lack a generic sensor toolkit that is broadly applicable to diverse measurement environments.  The SAGE project is an early attempt at designing and deploying such a generic infrastructure, drawing on lessons from the earlier Array of Things (AoT) project, but much more experimentation and targeted development is required to separate systemic needs from domain-specific science.  Even more importantly, the lifetime of these longitudinal sensor data surveys may last for years or even decades, placing an even greater premium on organizational and infrastructure stability.

Intentional Programming at Scale.   Task or node level specification is central to most of our programming models, yet so many of the questions that matter with distributed sensors are systemic ones involving identification of either data outliers (e.g., vibrations on this bridge exceed norms) or broad trends (e.g., rainfall along the severe storm front is approaching critical levels).  Equally importantly, such data should trigger certain global actions (e.g., if viral loads in sewage rise, increase global sampling rates). These global, “if this happens, then do this” intentional specifications, with some tolerance for sampling errors and uncertainty, are not typical of our programming models, but are ideal for specifying system behaviors and responses. We need new research on systemic, intentional programming models and specification systems (e.g., based on AI qualitative reasoning and zero shot planners).

Run Forever Heterogeneity. Those of us in computing are intimately acquainted with forklift hardware replacement, when old hardware gives way to new systems.  Depending on facility capacity, there may be some period of system overlap, but inevitably the forklifts remove the old hardware, leaving only a fully integrated, new system in their wake. Alas, the real world of distributed sensors bears much more resemblance to that of embedded real-time systems, where replacing sensors can be both costly and impractical.  Sometimes this is because sensors are deployed in critical infrastructure with hard real-time constraints and uptime quality of service expectations. In other cases, the cost of replacement may simply be prohibitive. 

For example, replacing the Chicago AoT nodes required access to City of Chicago utility bucket trucks, something possible only rarely and as a low priority relative to operational utility needs.  Similarly, replacing sensors in remote field sites may require extensive and difficult travel.  One consequence of such difficulties is extensive sensor heterogeneity, with new sensors and technology operating in concert with older, less capable sensors.  This has profound implications for both programming models and software configuration and deployment, as well as the types and resolution of data being captured and kinds of in situ analysis possible. 

Bags of Interdisciplinary Data. In addition to disciplinary access, much of the value of longitudinal sensor data accrues from cross-disciplinary data fusion, asking questions never envisioned by any of the original collectors of the data. The latter necessarily requires integration across diverse data schemas, none of which were designed with cross-disciplinary questions in mind.  We need to embrace data schema heterogeneity as an inevitability rather than as an impediment, encouraging development of new techniques that support “come as you are” data analysis.  None of the search firm giants have a master schema for web pages; rather, they index data and build search engines capable of responding to a wide range of queries.  We need to draw insights and lessons from there experiences.

A Data Economy. Finally, government mandates for data preservation and access are creating new challenges for researchers and research institutions alike.  What should be saved and for how long?  Can such answers be determined solely by the constituent disciplines or should there be a combination of discipline-specific and university specifications?  When can and should data be discarded?  Are those choices based solely on usage frequency, uniqueness, or cost of replication relative to cost of retention?  Research librarians long ago adopted deaccession policies, triaging holdings and negotiating interlibrary loan policies for rare or unique artifacts?  We need a coherent data economy that balances data utility against costs, driven by scientific needs.

Coda

We have come far in developing a coherent national and international cyberinfrastructure for research and innovation.  However, there are still unmet needs and unresolved infrastructure and research questions surrounding distributed instruments and large numbers of inexpensive sensors.  It is time to reason together and build the missing pieces.

Geek Gifts for the Holidays

The holiday season is here, with lights and festive celebrations of all kinds. Normally, I am loathe to offer holiday product endorsements, perhaps because, like you, I’ve been jaded by too many marketing campaigns and late night infomercials. (See The Ghosts of Holiday Shopping: Past, Present and Future.) Other than filing my taxes on time, acting before midnight tonight has never changed my life.  However, in a spirit of beneficence, I do make one exception to aid two connected, but suffering groups. I speak of course of geeks and their gift giving friends and family.

Let’s be honest; it’s hard to buy for geeks. Few would know that a signed copy of The Art of Computer Programming is a rare and precious thing, perhaps exceeded in value only by a Knuth reward check.  A copy of a Big Bang Theory DVD or a word of the day calendar just isn’t the same, though it helps if it is Klingon. 

So what’s a friend or family member to do when your geek smiles wanly and says thank you – for the socks? Fear not; even though the holidays are here, it’s not too late to trade those socks for a gift that will bring a delighted grin to your geek’s face, whatever their age.

Turing Tumble: Logic Is Child’s Play

Maybe you have a young geek, one interested in all things science related, and computing in particular. They may be too young for “the talk,” where you explain the dangers of unprotected ports, the subtleties of functors and futures, or the central dogma of the Church-Turing thesis, but they are still old enough to avoid specious logic, develop algorithms, and create working code.  (See Just the Facts, Ma’am: Reasoning is Not Dead, Jim.)

Turing Tumble ($69.95, ages 8+) is a game where players can build mechanical computers powered by marbles to solve logic puzzles.  In the spirit of LEGO-powered Turing machines, Turing Tumble exposes the logic of computers in a tangible way.  Developed by University of Minnesota professor and backed by a Kickstarter campaign, Turing Tumble is the real deal, combining marbles, levers, gears, and bits to teach logic and algorithms in an accessible way.  Just don’t let the adults play with it, or your child geek may never get it back.  And yes, I’m playing with mine over the holidays.

AWS DeepRacer: Start Your Engines

If you have an adult geek, particularly one who writes code for a living, this is the gift for you. AWS DeepRacer is a bit pricey at $399, but it’s the only gift your adult geek will want or need. AWS DeepRacer is a 1/18^th scale model car and Amazon Web Services (AWS) reinforcement learning toolkit to train your car for competition in either a real or a virtual racing league.

The DeepRacer can be also be driven manually via a joystick, and with an Intel Atom processor, HDMI port, and Ubuntu Linux, it can double as a serviceable, albeit somewhat slow, desktop computer. (Pro tip: Your GPU-accelerated Jetson Nano ($89 from Nvidia) is a great offline reinforcement learning training tool.)

If your geek is throwing around words and phrases like TensorFlow and generative adversarial network, while railing about deep fakes and social responsibility, relax.  Even if you think Sagemaker is the nom de plume of the family matriarch and her secret stuffing recipe, your geek knows that a reward function is more than a second helping of gravy and stuffing.  I will be racing mine in the living room tomorrow.

PDP-11 Emulator: When Nostalgia Reigns

If your geek’s native language is FORTRAN and they speak wistfully of the days when toggling absolute binary into the computer front panel was a rite of passage, you have an aging geek. They are likely to grow pensive when someone mentions the Great Editor War, modality, and meta keys. Viva Emacs; viva vi; those were heady and difficult days, when friends were forced to choose sides, and flames filled the nascent Internet. (See You Might Be a Computing Old Timer If …)

The PIDP-11 ($250) is a Raspberry Pi emulation of the iconic PDP-11/70, complete with hardware front panel lights and switches. If you want to impress your geek friends, comment on the orthogonality and beauty of the PDP-11 instruction set and smile knowingly. Even better, tell them the simh software supports all the old PDP-11 operating systems your geek knew and loved – RT-11, RSTS/E (my favorite), and UNIX V5, V6, and V7.

With the PIDP-11, you can toggle in code, or boot your operating system of choice, play with all the old system software tools, and even write or run old code. TECO lives!  (Warning: Some soldering is required to assemble the front panel, or you can skip the kit and do everything in software, albeit without that heady joy of blinking lights and toggle switches.)  Mine is on order.

Save the Detritus

Now it is possible you live in one of those increasingly rare places where the gray fog of nothingness lurks in the distance and same day or overnight package delivery is not possible. However, you still have the raw materials in your possession needed to satiate the emotional and intellectual yearnings of a nascent Time Lord, Dungeons and Dragons role player, Skywalker acolyte, or wannabe Gandalf.

Wired’s 2011 review of the All Time Best Toy Ideas ($0) exposes the truth: the detritus of your gift wrapping, along with some yard debris, holds the key to happiness.  Stick (Glamdring), box (TARDIS), string, cardboard tube (lightsaber), and dirt, these are the enduring toys for geeks and the rest of you, though the purist might insist on the addition of rock, hole, and mud puddle.

I’d explain more, but a launch window has been posted for my LEGO Saturn V.  I need to finish the LEGO launch gantry before the Mobile Launcher Platform can roll out of the Vehicle Assembly Building to pad 39A. If you were a geek you would understand why that is important.

Idea Bandwagons

Does it ever seem that we are imagination challenged, bound to imitate or follow a popular idea or approach, whether successful or not? (At this point, I could insert a quasi-obligatory reference to lemmings plunging into the sea. For the record, lemmings do not commit mass suicide; that is a persistent urban myth.)

Once an idea gains traction, it often seems everyone adopts a "me too" approach. There is, of course, intellectual safety in following a crowd. After all, if "everyone" is doing it, then ipso facto it must be a good idea. Alas, not necessarily. In Gartner hype cycle parlance, this mimicry can happen early, during the "peak of inflated expectations," or later, riding the "slope of enlightenment," if the idea survives the dreaded "trough of disillusionment."

My AI friends still talk ruefully about the long and difficult AI winter that followed early, unrealistic expectations (e.g., human language translation in the 1960s). Funding became scarce and the community shrank. Today, of course, we are basking in the AI spring, when many early hopes and predictions are being realized in dramatic fashion, powered by new machine learning ideas, large volumes of data, and powerful computing systems. Despite AI's enormous success, we have not yet realized the dream of strong AI, though I retain fond hopes. And yes, I am an optimist; I do not share the fears expressed by some about strong AI. However, I digress – back to viral memes.

Uber, the ride sharing service, has spawned seemingly endless variants, all described in pitch decks as Uber for X – a new instance of the sharing economy. As Wired noted last year, Uber for X has been the headline of more than 400 news articles. If you are thinking that the Uber for homework opportunity (there are several startups) might not be the next, well Uber, you are almost certainly right. There is, as former Federal Reserve chair, Alan Greenspan, called it, some irrational exuberance.

Following the success of AirBnB, a new category of startups, somewhat whimsically called MaaS (Mom as a Service), has exploded. (Let me just state right now, I apologize to moms everywhere. You were right; we should have cleaned up our rooms.) These MaaS companies provide services such as cooking, cleaning, laundry, and food; among the best-known food examples are Grubhub (prepared food delivery) and Blue Apron (cooking ingredients delivery). As one who is culinary-challenged, I hope this segment fares well.

Because humans are involved, this faddish behavior is largely domain invariant. As such, it extends to research and scholarship as well. In the best cases, a triggering discovery stimulates the cognizant research community to shift directions and build a rich and robust intellectual infrastructure around a novel idea. The discovery of the CRISPR/Cas9 gene editing technique is a wonderful example; research groups worldwide quickly pivoted to investigate the technique and its applications. In other cases, a technology crosses a threshold (e.g., price or performance) and that quantitative change enables qualitatively new approaches to old problems. Field programmable gate arrays (FPGAs), long a niche technology, are now the subject of intense research as AI accelerators.

Whether in startups or research, advantage often accrues to first movers, those who pioneer a new idea and explore its applications. Doing so requires imagination – the ability to step out of the incremental, "me to" mindset and take transformative intellectual (and often financial) risks. That is not easy, both because creative thinking is challenging and because failure is often punished. Now there's an opportunity – IaaS (Imagination as a Service). Let me get to work on that research proposal and pitch deck.

Effervescent Clouds: Nimble Computing Infrastructure

N.B. A version of this essay appeared as a guest blog post for the EU Cloud Lightning project

What's a cloud? The word means many things to many people. In true Airplane! movie parody style, one is tempted to say, "It's a visible mass of condensed water vapor floating in the atmosphere. But, that's not important right now."

Some might say that our cloud confusion is indicative of a fog, a low hanging cloud. Indeed, in public perception, the cloud is a vaguely understood and amorphous capability that lurks behind our smartphones, enabling electronic communications, web searches, social networks, e-commerce, and 24x7 news and streaming media. It is a mysterious "something, somewhere" that is only noticed when it is inaccessible.

For those of us in computing, the cloud is more than a buzzword or meme, it is a set of technologies that are transforming the provisioning and delivery of computing cycles, storage and services, with profound implications for government services, research discovery and innovation, and businesses large and small. Given the public confusion and sometimes overhyped marketing, it seems worthwhile to dispel the fog and illuminate a few compelling features of clouds.

In a very real sense, public clouds – those hosted by cloud providers for on-demand access – have much in common with the timesharing systems of the 1960s and 1970s, albeit at unprecedented scale. They shift the capital costs of provisioning, along with the expertise required for operation, from users to providers, allowing users to focus on their core expertise (e.g., research, business, government) and pay for only what computing they need, when they need it.

In turn, cloud elasticity allows servers to be provisioned and de-provisioned dynamically in response to changing interest and demand. One of the few things worse than having your brilliant idea ignored is having it fail publicly from too much attention. Many an organization has first been thrilled to see their product or service be reported in the press and go viral, only to be horrified when on-premise servers collapse under the load of success; this is known as the slashdot effect, named after one of the popular "news for nerds" web sites. Without cloud elasticity, the global phenomenon of Pokémon Go would not be possible.

Although not required, many clouds also virtualize the underlying computing, allowing multiple services and software stacks to be co-resident on the same physical hardware, further reducing costs. With Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS) and containerization, developers can choose the level of abstraction, from low level to high level, needed to support existing and new applications.

By allowing innovators to focus on their ideas rather than computing infrastructure, the pay-as-you-go cloud model has made it easier, faster and cheaper to bring ideas to life, unleashing a second dot.com boom of cloud-based services, unprecedented citizen access to government data, and new approaches to scientific discovery via big data.

The latter has sometimes been called the fourth paradigm, where new sensors, data analytics and deep learning have shifted the research approach from "what experiment should I conduct" to "what insights can I extract from available data." It has also stimulated interest in convergence architectures for high-performance computing (HPC) and data analytics that combine the best elements of cloud and HPC capabilities. (See my essay, Exascale Computing and Big Data: Time to Reunite.)

Implicit in this client-cloud model is ubiquitous, reliable and inexpensive wired and wireless broadband services. Without broadband access, clouds are simply isolated sets of computers and data, unconnected to the world of mobile devices and the nascent Internet of Things. To further ensure reliability, most public cloud providers geo-distribute their pools of servers. Even private clouds (i.e., those operated by governments and companies for their own use) often geo-distribute their servers at multiple data centers. Should disaster strike at one site, services can be shifted automatically to another site, perhaps even in another country or continent.

Finally, the unprecedented scale of cloud deployments, with commercial data centers each costing over $1B, has stimulated new approaches to energy efficiency and cooling, component failures and reliability, customized and open hardware, software defined networks (SDNs) and software defined storage (SDS), and system provisioning and management. In short, we are in a time of radical change in the computing ecosystem, all against a backdrop of Moore's law limitations.

What's a cloud? It's an elastic, on-demand, often geo-distributed, frequently virtualized, flexible hardware, software and services infrastructure that powers the 21^st century knowledge economy. And that is really important right now.

Related articles

Exascale Computing and Big Data: Time to Reunite

Raspberry Pi? Why Yes, I’ll Have Pi Cluster

The Raspberry Pi, a credit card sized, $35 computer, is a runaway success, exceeding even the wildest expectations of anyone in the hobbyist or educational communities. The latest version contains a 900 MHz quadcore ARM processor, 1 GB of DRAM, multiple USB ports, a 10/100 Ethernet port, an HDMI interface and 40 GPIO pins (for external device interfaces). Permanent storage is via a Micro SD card slot.

Assuming one has either an HDMI-capable digital television or desktop computer display, the only additional requirements are a USB keyboard/mouse and a Micro SD card of at least 8 GB capacity. Chances are most of us have all of these; if not, another $30 suffices for all but the display. Thus configured, the Raspberry Pi can run several variants of Linux and, as well as the forthcoming embedded version of Windows10.

All of this takes me back to a time in the early 1980s, when I walked to graduate school every day (in the snow, uphill both ways), Pac-Man was a state of the art video game, and the Rubik's Cube was bringing recreational group theory to the masses (though they didn't know it). A DEC VAX 11/780 with 2-8 MB of DRAM running the Berkeley Software Distribution (BSD) of UNIX was the de facto standard research environment for academic computer science.

At a time when batch computing and timeshared university mainframes defined academic research, these BSD VAX systems, along with the National Science Foundation (NSF) Coordinated Experimental Research (CER) program and the CSNET academic network, were transformative. It is no exaggeration to say that modern, experimental computer science can be traced directly to this environment. All of which is why Bill Joy once called the 1 MIPS of the VAX 11/780 the "electron volt of computing."

The Raspberry Pi is a more capable system in almost every way than that departmental VAX – faster, with a larger memory and bigger secondary storage system – as well having a richer base of open source software and access to the global Internet. It is a very tangible and inexpensive manifestation of Moore's Law. However, to quote from the movie Airplane!, that's not important right now. (I am serious. And don't call me Shirley.)

What is important is the transformative effect of the Raspberry Pi and related, inexpensive computing devices on both education and innovation. (The Kickstarter-funded Adapteva Parallela multicore processor is another.) Raspberry Pi's now serve as robot controllers, research sensors, web servers, and software development platforms; they are the darling of hobbyists and the maker movement.

However, the creation nearest and dearest to my heart has been the widespread development of Raspberry Pi Beowulf clusters for parallel computing education. These systems use inexpensive Ethernet switches to create systems similar to the Caltech Cosmic Cube, a seminal design for message-based parallel computing created by Chuck Seitz and Geoffrey Fox in the early 1980s.

Although not fast by today's parallel computing standards, these Raspberry Pi clusters allow students to construct complete parallel systems, ranging from small, development clusters (4-8 nodes) to larger (40-120+ nodes) research systems. More importantly, they expose all aspects of hardware and software configuration, bringing experiential parallel computing to a broad new class of students. (The construction recipe for ORNL's Tiny Titan is one of many such examples.) Even more importantly, the clusters let students run real-world parallel applications.

I would have dearly loved to have one of these clusters for parallel computing research back in the early 1980s. At that time, Richard Fujimoto and I were writing Ph.D. theses at UC-Berkeley and Purdue, respectively, about various aspects of message passing machines. We then collaborated on a book about what we then called multicomputer networks. We have come a long, long way since then.

Yes, I will have some Raspberry Pi. In fact, I'll have a cluster.

Exascale Computing and Big Data: Time to Reunite

N.B. I also write for the Communications of the ACM (CACM). The following essay recently appeared on the CACM blog. In addition, you can find the video of an interview and the full text of the CACM article, co-authored with Jack Dongarra.

In other contexts, I have written at length about the cultural and technical divergence of the data analytics (aka machine learning and big data) and high-performance computing (aka big iron) communities. I have euphemistically called them "twins separated at both." (See HPC, Big Data and the Peloponnesian War and Scientific Clouds: Blowin' in the Wind.) Like all twins, they share technical DNA and innate behaviors, despite their superficial differences. After all, in a time long, long ago, they were once united by their use of BSD UNIX and SUN workstations for software development.

Since then, both communities have successfully built scalable infrastructures using high-performance, low cost x86 hardware and a rich suite of (mostly) open source software tools. Both have addressed ecosystem deficiencies by developing special-purpose software libraries and tools (e.g., SLURM and Zookeeper for resource management and MPI and Hadoop for parallelism), and both have optimized hardware for their problem domains (e.g., Open Compute hardware building block standardization, FPGAs for search and machine learning, and GPU accelerators for computational science).

Like many of you, I have seen this evolution firsthand, as a card-carrying geek in both the HPC and cloud computing worlds. One of the reasons I went to Microsoft was to bring HPC ideas and applications to the nascent world of cloud computing. While at Microsoft, I led a research team to explore energy-efficient cloud hardware designs and new programming models, and I launched a public-private partnership between Microsoft and the National Science Foundation on cloud applications. Now that I am back in academia, I am seeking to bring cloud computing ideas back to HPC.

In that spirit, Jack Dongarra and I recently co-authored an article for the Communications of the ACM on the twin ecosystems of HPC and big data and the challenges facing both. Entitled, Exascale Computing and Big Data, the article examines the commonalities and differences, and discusses many of the unresolved issues associated with resilience, programmability, scalability, and post-Dennard hardware futures. Most importantly, the article makes an impassioned plea for hardware and software integration and cultural convergence.

The possibilities for this convergence are legion. The algorithms underlying deep machine learning would benefit from the parallelization and data movement minimization techniques commonly used in HPC applications and libraries. Similarly, the approaches to failure tolerance and systemic resilience common in cloud software have broad applicability to high-performance computing. Both domains face growing energy constraints on the maximum size of feasible systems, necessitating shared focus on domain-specific architectural optimizations that maximize operations per joule.

Perhaps most important of all, there is increasing overlap of application domains. New generations of scientific instruments and sensors are producing unprecedented volumes of observational data, and intelligent, in situ algorithms are increasingly required to reduce raw data and identify important phenomena in real time. To see this, one need look no further than applications of machine learning to astronomy, which now include automated object identification. Conversely, client plus cloud services are increasingly model-based, with rich physics, imaging processing and context that depend on parallel algorithms to meet real-time needs; augmented reality applications are one such exemplar.

The explosive growth of Docker and containerized software management speaks to the need for lightweight, flexible software configuration management for increasingly complex and rich software environments. My hope is that we can develop a unified hardware/software ecosystem that leverages the technical and social strengths of each community. Each would benefit from the experiences and insights of the other. It is past time for the twins to have a family reunion.

Through A Google Glass, Darkly

N.B. I also write for the Communications of the ACM (CACM). The following essay recently appeared on the CACM blog.

Whether de regueur or merely an occupational hazard, a love of technological gadgets is commonplace in computing. We know the endorphin-fueled euphoria of an early adopter, one willing to tolerate the failings of immature and possibly transient technologies.

Oh, how I loved my RIM 870, which made me a two-thumbed typist and allowed me to send email on the DataTAC network, long before the word BlackBerry conjured any vision other than breakfast jam! I wore a Microsoft SenseCam, learning that email, coffee and meetings defined my life.

My desk and closet are filled with "might have been" and "never were" technological artifacts. From a subcompact Poqet PC through an Apple Newton PDA to Ricochet network cards to a collection of bulky, CGA-based head mounted displays (HMDs), I have played with my share of bright, shiny toys. Today, Google Glass shines oh, so bright, with technological promise and cultural uncertainty.

All of which reminds me of the first time I pulled out an early Motorola cellphone in an airport terminal, after an airline canceled one of my flights. This was in the early days of AMPS service, when mobile telephones were large and bulky and roaming service had yet to appear. As I called my office to rebook travel, a group of businesspeople gathered around me in wonder at this amazing and heretofore unknown technology. Years later, I was reflecting on the societal transformation wrought by mobile telephony as I listened to a middle manager obliviously pace an airport concourse while systematically firing the unsuspecting members of some company team via cellphone. In ictu oculi, sic transit gloria mundi!

I have been wearing Google Glass (hereafter, Glass) intermittently as both a technical assessment of utility and as a social study in human dynamics and expectations. What does the future hold? Prediction is very difficult, especially about the future. Attribute the quote to whomever you choose – Yogi Berra, Niels Bohr, or Mark Twain – its veracity remains.

Technical Experiences

My first impressions of Glass were generally positive, but I expected that; I have decades of Pavlovian conditioning to embrace technology prototypes with joyful abandon. Software configuration took most of an afternoon, as I connected my social networks to Glass and configured news feeds to balance timeliness and frequency, lest breaking news of the underwater basket weaving championship interrupt my reflections on the dismal state of research funding. Not surprisingly, these networks were heavily Google-centric, though I found the inability to send images to an arbitrary email address somewhat limiting. Ok, Glass, take a picture …

Because I am sufficiently myopic that escaping a broom closet is visually challenging, I wondered if it would be practical to wear Glass in front of my normal glasses, rather than fitting it to the custom Glass frames. Pleasantly, the answer is yes, if one carefully adjusts the prism. The screen is relatively bright and unobtrusive, and the head tilt adjustment angle simplifies screen activation and menu selection.

Is Glass a substitute for a smartphone? No, but it is not intended as such. Rather, it is a social and technical experiment in digital immersion and social commerce, from instant context sharing to walkabout video conferencing. In all of those things, it succeeds.

Social Challenges

Far more interesting than the technical features of Glass have been the social reactions it has engendered. When I am in a private setting, I always inform people that I will not take photos or record video using Glass without permission. I also invite people to wear Glass and experience it for a few minutes to allay fears. In public, I use Glass with great care, with as much deliberation and clarity I would use with a camera or smartphone.

The sometimes visceral reactions to the public use of Glass are rooted in deep social concerns about privacy and cultural norms, themselves shaped by the broader worries about government surveillance, data breaches, and identity theft. Glass also occupies something of an uncanny valley of privacy, arousing primal fears that the sanctity of one's sensorium is being violated. The negative reactions have also been fueled by the annoying and sometimes foolish behavior of some Glass users. (See Google Glass "dos and don'ts," or how to avoid being a Glasshole.)

Each new technical change brings social and legal debates regarding acceptable use, debates made more pressing and contentious by the increasing rate of change. In the U.S., we are still defining the bounds of the first amendment on the right to take photographs in a public place, despite over 150 years of photographic technology experience and an abundance of case law.

One of the creators of mobile telephony once remarked to me that he was still surprised to see individuals speaking into thin air about intensely personal topics. That dystopian science fiction writer, Phillip K. Dick, favored the ancient line, "For now we see through a glass, darkly ..." I could wax on at length in a metaphorical and epistemological exegesis of this prescient juxtaposition of technology, privacy and the nature of change, but a heads up news feed on Glass has captured my attention.

Looking Forward

It is not surprising that we are struggling to delineate the bounds of privacy and normative behavior in an age of ubiquitous digital devices. Some acceptable and normative issues of wearable technology will be decided by law, some by social consensus, and still others by concurrent technological evolution and generational acceptance. In the meantime, good sense should be applied liberally.

Despite all the uncertainties, I remain an optimist. Ok, Glass, send a message to the future …

Simplifying Communication

N.B. I also write for the Communications of the ACM (CACM). The following essay recently appeared on the CACM blog.

How often have you picked up a scholarly journal in a discipline far removed from your expertise, only to be stymied and mystified by the disciplinary jargon? It can be humbling and intimidating when one fails to understand the meaning of all the words in an article's title or the abstract. When coupled with the contextual knowledge often implicitly assumed by the authors, the gulf of understanding yawns wide and deep.

This epistemological and linguistic chasm separates and isolates even within the broad tent of our own discipline, which spans everything from the fundamental theory of computability to the professional practice of informatics. If you have any doubt, open the ACM Digital Library and scan a few articles in a specialty far removed from your own.

In a world where discoveries increasingly lie at the boundaries of traditional research disciplines, simplifying communication and encouraging multidisciplinary dialog and partnership have never been more necessary. In almost every case, computing is an essential element of disciplinary and multidisciplinary research. Thus, it is time for us to embrace writing as a collaborative enabler, rather than a research burden.

Channeling Strunk and White

All too often, we academics write in a strange argot of disciplinary esoterica that can obscure the very ideas we seek to communicate. If you have ever encountered an article like the following, you know what I mean.

"Spatiotemporal domicile proximity to retroverting domestic ruminants," I. B. Smart, I. A. Postdoc & O. Authors, International Journal of Bovine Mobility, Vol. 123, No. 11, pp. 2143-2147, 2013

However linguistically facile and intellectually adept, the authors and putative ruminant experts failed to say what they really meant ("wait for the cows to come home") and why that might matter.

In a similar spirit, the late Richard Hamming once famously noted, "The purpose of computing is insight, not numbers." The academic publishing cognate is best summarized as, "The purpose of writing is communication, not obscuration." There is also an important corollary, "Write to communicate, not to impress or intimidate." Yes, subtlety and nuance are important, but they are mere handmaidens to clarity and lucidity.

Even when we avoid these linguistic traps, another, equally deadly one waits to ensnare – turgid and passive prose that invites only slumber. As anyone knows who has either served as a journal editor or reviewed a seemingly endless stack of conference paper submissions, passive, wordy and meandering prose makes identifying the key ideas and assessing their importance even more difficult.

Technical papers are not page turning spy novels, nor should they be, but they can still be interesting, clear and engaging as they convey the essential facts. As a writer, one's job is to make the reading easy; you want your papers to be read and appreciated.

The Message is the Message

It is always dangerous to write an essay about writing, lest one be lampooned for the very deficiencies one seeks to highlight. Such is life. My goal is to focus attention on an important issue.

While continuing to pursue core research in our own discipline of computing, I believe we must also communicate effectively with our peers in the arts and humanities, science and engineering, medicine and public policy. We cannot all be polymaths, but as writers, we can do more to lower the disciplinary drawbridges and invite readers into our intellectual castles.

One Life to Live

A few years ago, a foundation ran a billboard campaign with an image of Kermit the Frog and the caption, "Eats Flies, Dates a Pig, Hollywood Star. Live Your Dreams." Beyond being a huge fan of the Muppet Show and its sardonic humor, especially as offered by Statler and Waldorf, I have always loved this billboard because it captures an essential element of human life, the desire to make a difference, to live life with the passion of commitment. It is a feeling Thoreau captured in Walden, to the enduring delight of secondary school English teachers across North America:

I went to the woods because I wished to live deliberately, to front only the essential facts of life, and see if I could not learn what it had to teach, and not, when I came to die, discover that I had not lived.

I was reflecting on this recently as the world lost two innovators in computing. About the first, Steve Jobs, much has been written, but far less about another pioneer of computing, Dennis Ritchie.

Remembering Ritchie

UNIX and C transformed our notions of operating systems and programming, seeding a cornucopia of software tools and technologies. I still have my dog-eared copy of the 1978 Bell Systems Technical Journal (BSTJ), which was devoted to the "UNIX Timesharing System."

I was a graduate student then, and it brought my first exposure to UNIX on a VAX 11/780 at Purdue. We ran VMS part of the time and UNIX the rest, largely because VMS had a decent FORTRAN compiler. The numerical analysts among us used FORTRAN, but for the rest, it was UNIX and C. It was a heady time, when new tools and approaches were transforming academic computing research from a largely theoretical discipline to an experimental science. I created a compiler for John Backus' FP language using lex and yacc on that VAX, and I wrote a Ph.D. dissertation using troff.

As da Vinci supposedly remarked, "Simplicity is the ultimate sophistication." UNIX was elegant in its simplicity and functionality. It and the C language inspired multiple generations of hackers (and I use that word in its original, pure sense) to create so many things. It is not exaggeration to say that one can date operating systems as pre-UNIX and post-UNIX.

Richie's passing was not noted with great fanfare, but he made a difference. All of us in computing honor that.

Meditations on Dreams

In between the endless work deadlines, the email deluge and the frenzied travel, take a deep breath and ask yourself what matters and what you are doing. The answers may surprise you, particularly as mortality seems more than a tale told by the wise and aged to the young and uncertain. Those are the essential facts Thoreau described.

Throughout my life, I have been able to live many of my dreams. Far too many people are not so fortunate. (For the record, eating flies, dating a pig and being a Hollywood star have not been among them.) I have a few dreams left, both in computing and in life, and I suspect you do too. In my case, there is a book or two trying to escape via my fingers. (See Eudora, You Got the Love?)

Seize the opportunity. Live your dreams. Make a difference. It is all that any of us can hope to do.