Science is what we have learned about how to keep from fooling ourselves.
― Richard Feynman
Last week was another trip to the Multimat conference, a biannual meeting of scientists who solve the equations of multiple material flows under highly energetic and compressible conditions. I’ve always greeted the meeting with great enthusiasm and returned to work bristling with new ideas and inspiration. In many ways this community is the tip of the intellectual spear for modeling and simulation capability. It also marks an anniversary of sorts, four years of blogging. My very first post here was about the San Francisco edition of the meeting that coincided with my 50th birthday (https://williamjrider.wordpress.com/2013/09/13/thoughts-about-multimat2013/). Two years ago we had a wonderful meeting in Wurzburg Germany
(https://williamjrider.wordpress.com/2015/09/11/multimat2015-a-biannual-festival-on-computing-compressible-multiple-materials/). Every meeting was wonderful and this was no exception, except in one very important and perhaps essential regard; the meeting seemed devoid of the usual exciting intellectual inspiration. What happened to the big ideas that flowed so easily in every previous meeting? Is it my imagination, or have the exciting new ideas dried up?
Do stuff. be clenched, curious. Not waiting for inspiration’s shove or society’s kiss on your forehead. Pay attention. It’s all about paying attention. attention is vitality. It connects you with others. It makes you eager. stay eager.
― Susan Sontag
This assessment might seem rather harsh, but upon reflecting on the previous meetings, it holds up under considerable scrutiny. Each previous meeting was full of moments where you are watching someone else’s talk and thinking, “I wish I’d thought of that, this is brilliant”. This is exactly what makes conferences so wonderful and important to attend; you get to cherry pick people’s best ideas accumulated at great effort all at once. In the moment these ideas seem like Athena springing fully formed from Zeus’ brow! Your colleagues get to look like the geniuses they are and present their most creative and ingenious thoughts in an intellectual banquet (https://williamjrider.wordpress.com/2014/06/13/why-do-scientists-need-to-attend-conferences/, https://williamjrider.wordpress.com/2014/10/27/a-holistic-view-of-attending-conferences/). The reason for attending conferences isn’t to give talks; it is to learn new things taught by the smartest people you know. It is to meet and let ideas breed openly over laughter, food and drinks. You give talks as an act of repayment for the knowledge you are granted by
being in the audience. Giving talks is pretty low on the list of reasons, but not in the mind of our overlords, which starts to get at the problems I’ll discuss below. Given the track record of this meeting my expectations were sky-high, and the lack of inspiring ideas left me slightly despondent.
A few more thoughts about the meeting are worth pointing out before getting to the dialog about fresh ideas, their importance and postulates for their recent absence. The meeting is attended by a collection of computational scientists (mathematics, physics, engineering,…) dominated by the nuclear “club”. This means American, French and British with a smattering of Russians and Chinese – who couldn’t come this year for undisclosed reasons. These scientists for the most part work at their nation’s respective nuclear weapons’ labs. Occasional others attend like Israelis (an unacknowledged member of the club) along with a handful of Czechs, Italians, and Germans. As such the meeting serves as a proverbial checkup on the intellectual health of this important area of science at the West’s nuclear weapons Labs. This year’s checkup should give everyone pause, the state of health is declining. There is a real lack of creative energy surrounding the heart of our most important codes. Many important codes are built around a powerful hydro-solver that produces accurate, physically relevant solutions to the multi-material “hydrodynamic” equations. Previous meetings have seen a healthy resurgence of new ideas, but that upswing seems to have come to a staggering halt. These labs have also provided a deep well of inspired research that has benefited the broader scientific community including weather, climate, astrophysics and a broad swath of engineering use of computation.
In my opinion the reasons for this halt in creative energy are simple and straightforward. The foolhardy worldwide push for exascale computers is sucking the air out of the room. It is gobbling up all the resources and attention leaving nothing for new ideas. This completely obsessive and unwise focus on the hardware is attempting to continue – the already dead – Moore’s law. This push is draining the community of vitality, resources and focus. The reasons for the push are worth considering because they help define the increasingly hostile nature of the modern world toward science. The computers being build for the future are abysmal to use and the efforts to move our codes to them are sucking all the energy from the Labs. Nothing is left of creative work; nothing is left for new ideas. Simply put, the continued use of old ideas is hard enough if you add these generally unusable computers to the mix. The reason is simple; the new computers completely suck. They are true monstrosities (in the classic definition of the word) and complete pieces of shit as scientific hardware. They are exactly the computers we don’t want to use. The price of forcing them down our throats is the destruction of research that isn’t associated with simply making these awful computers work. Worse yet, the return on the massive investment of effort will be vanishingly small in terms of added modeling and simulation capability.
As noted this whole direction is a foolish attempt to breathe life into the already rigid corpse of Moore’s law. Now dead at every scale of computing and already a decade deceased at the level of computer chips – note the death of Moore’s law and the ascendency of cell phones is strongly correlated, and that probably is not a coincidence. The truth of our real performance on computers is far more dire and damning of this entire effort. We have been getting an ever-lower proportion of the potential performance on our computers for 25 years. Each computer has a peak performance measured on silly hardware friendly benchmarks that no one gives a flying fuck about (the dense linear algebra LU decomposition, Linpac). This silly and useless benchmark is how we crown the fastest computer! Our actual code performance on these machines is truly atrocious and gets worse every year. The dirty little secret is that its been getting ever worse every year. It was god-awful 20 years ago, and it has just gotten worse. Zero is a pretty good approximation to the proportion of the performance we get – generally much less than one percent. We mindfully ignore the situation just like one would ignore a cancer threatening to devour our lives. The attitude is generally, “look away, nothing to see here”. The exascale program is that cancer metastasized.
Part of the discussion about exascale needs to acknowledge the nature of choices and priorities in research. In isolation, the exascale program is an unambiguous good; it would be genuinely awesome to have – usable – exascale computers (https://williamjrider.wordpress.com/2014/09/19/what-would-we-actually-do-with-an-exascale-computer/). This good needs to be weighed in terms of its cost and the impact of alternatives. It needs to be viewed through the lens of reality too. If one looks at the raw cost, the opportunity cost and collateral damage, and under this examination we can see that the exascale program is a massively negative force in science (https://williamjrider.wordpress.com/2016/06/27/we-have-already-lost-to-the-chinese-in-supercomputing-good-thing-it-doesnt-matter/, https://williamjrider.wordpress.com/2016/05/04/hpc-is-just-a-tool-modeling-simulation-is-what-is-important/, https://williamjrider.wordpress.com/2016/10/19/why-china-is-kicking-our-ass-in-hpc/, ). In isolation without considering anything else, it is a clear positive. In the context of lost opportunities and effective use of available resources, the program is an unmitigated disaster. We will all be poorer for it as it lays waste to potential breakthroughs we will be denied in its wake. In today’s world we talk about things in isolation, free of nuance and trade spaces that would make for a more robust and responsible conversation. Our leaders are irresponsible in the extreme for taking down this path with no real discussion, or any debate taking place. The message in the trenches is “do what you’re paid to do and quit asking questions”.
The really dirty secret is that chasing exascale as a route to scientific discovery is simply bullshit of the highest and most expensive order. We would be far better served by simply figuring out how to use the hardware we already have. Figuring out how to efficiently use hardware we have had for decades would be a difficult and worthy endeavor. The punch line is that we could get orders of magnitude in improved performance out of the hardware we’ve been using for decades. By simply figuring out how to get our codes working more efficiently on the computers already existing would meet most scientific goals without eviscerating the rest of computational science in the process. Instead we chase goals that are utterly meaningless. In the process we are destroying the research that has true and lasting value. The areas being ignored in the push for exascale have the capacity to provide far more scientific capability than even the most successful exascale program could possibly deliver. This brings me back to the meeting in Santa Fe and the lack of energy and exciting ideas. In the past the meeting has been a great survey of the active work from a creative and immensely talented group of people. As such this meeting is the proverbial canary in the coalmine. The ideas are dying right in front of our eyes.
This outcome is conflated with the general lack of intellectual vigor in any public discourse. The same lack of intellectual vigor has put this foolish exascale program in place. Ideas are viewed as counter-productive today in virtually every public square. Alarmingly, science is now suffering from the same ill. Experts and the intellectual elite are viewed unfavorably and their views held in suspicion. Their work is not supported, nor is projects and programs dependent on deep thinking, ideas or intellectual labor. The fingerprints of this systematic dumbing down of our work have reached computational science, and reaping a harvest of poisoned fruit. Another sign of the problem is the lack of engagement of our top scientists in driving new directions in research. Today, managers who do not have any active research define new directions. Every year our manager’s work gets further from any technical content. We have the blind leading the sighted and telling them to trust them, they can see where we are going. This problem highlights the core of the issue; the only thing that matters today is money. What we spend the money on, and the value of that work to advance science is essentially meaningless.
Effectively we are seeing the crisis that has infested our broader public sphere moving into science. The lack of intellectual thought and vitality pushing our public discourse to the lowest common denominator is now attacking science. Rather than integrate the best in scientific judgment into our decisions on research direction, it is ignored. The experts are simply told to get in line with the right answer or be silent. In addition, the programs defined through this process then feed back to the scientific community savaging the expertise further. The fact that this science is intimately connected to national and international security should provide a sharper point on the topic. We are caught in a vicious cycle and we are seeing the evidence in the hollowing out of good work at this conference. If one is looking for a poster child for bad research directions, the exascale programs are a good place to look. I’m sure other areas of science are suffering through similar ills. This global effort is genuinely poorly thought through and lacks any sort of intellectual curiosity.
Moving our focus back to exascale provides a useful case study of what is going wrong. We see that programs are defined by “getting funding” rather than what needs to be done or what should be done. Arguments for funding need to be as simple as possible, and faster computers are naïve enough for unrefined people to buy into. It sounds good and technically unsophisticated people buy it hook line and sinker. Computers are big loud and have lots of flashing lights to impress managers, politicians and business people who know no better. Our scientists have been cowered into compliance and simply act happy to get money for doing something. A paycheck beats the alternative, and we should feel happy that we have that. The level of inspiration in the overall approach has basically fallen off a cliff, and new ideas are shunned because they just make things complicated. We are left with the least common denominator as the driving force. We have no stomach for nuance or subtlety.
Priority is placed on our existing codes working on the new super expensive computers. The up front cost of these computers is the tip of the proverbial cost iceberg. The explicit cost of the computers is their purchase price, their massive electrical bill and the cost of using these monstrosities. The computers are not the computers we want to use, they are the ones we are forced to use. As such the cost of developing codes on these computers is extreme. These new computers are immensely unproductive environments. They are a huge tax on everyone’s efforts. This sucks the creative air from the room and leads to a reduction in the ability to do anything else. Since all the things being suffocated by exascale are more useful for modeling and simulation, the ability to actually improve our computational modeling is hurt. The only things that benefit from the exascale program are trivial and already exist as well-defined modeling efforts.
Increasingly everything is run through disconnected projects that are myopic by construction. The ability to do truly unique and groundbreaking science is completely savaged by this approach to management. Breakthroughs are rarely “eureka” moments where someone simply invents something completely new. Instead, most good research is not made through connections to other good research. Conferences are great incubators for these connections. Well-defined and proven ideas are imported and redefined to make contributions to a new area. This requires people to work across discipline boundaries, and learn about new things in depth. People need to engage deeply with one another, which is similarly undermined today by project management and information security focus. The key thing is exposure to new and related areas of endeavor and basic learning. The breakthroughs come episodically and do not lend themselves to the sort of project management in vogue today.
It isn’t like I came back with nothing. There were a couple of new things that really fall into the category of following up. In one case there was a continuation of a discussion of verification of shock tube problems with someone from Los Alamos. The discussion started in Las Vegas at the ASME VVUQ meeting, and continued in Santa Fe. In a nutshell, we were trying to get cleaner verification results by dividing the problem into specific regions associated with a particular solution feature and the expectation of different rates of convergence for each. We found something unexpected in the process that doesn’t seem to follow theoretical expectations. It’s worth some significant follow-up. A mysterious result is always something worth getting to the bottom of. The second bit of new intellectual blood came in direct response to my talk. I will also freely admit that my contribution wasn’t the best. I haven’t had any better luck with a good free energy at work to energize my work. The same exascale demon is sucking my intellectual lifeblood out. I simply reported on a here-to-fore unreported structural failing of solvers. In summary, we find systematic, but small violations of the second law of thermodynamics in rarefactions for modern and classical methods. This shouldn’t happen and violations of the second law lead to unphysical solutions. All of this stems from identifying a brutal problem (https://williamjrider.wordpress.com/2017/06/09/brutal-problems-make-for-swift-progress/ ) that every general-purpose code fails at – what I call “Satan’s shock tube” with 12 order of magnitude jumps in density and pressure approximating propagation of waves into a vacuum.
We cannot live only for ourselves. A thousand fibers connect us with our fellow men; and among those fibers, as sympathetic threads, our actions run as causes, and they come back to us as effects.
― Herman Melville
Before closing I can say a thing or two about the meeting. None of the issues dulled the brilliance of the venue in Santa Fe, “the City Different”. While I was disappointed about not enjoying the meeting in some exotic European venue, Santa Fe is a fabulous place for a meeting. It is both old (by American standards), yet wonderfully cosmopolitan. There is genuine beauty in the area, and our hotel was nearly perfect. Santa Fe boasts exceptional weather in the fall, and the week didn’t disappoint. It has a vibrant art community including the impressive and psychedelic Meow Wolf. It was the Drury Plaza hotel placed in a remodeled (and supposedly haunted) old hospital. Two short blocks from the plaza, the hotel is enchanting and comfortable. We all shared two meals each day catered by the hotel’s exceptional restaurant. Having meals at the conference and together with the participants is optimal and makes for a much-improved meeting compared to going out to restaurants.
We had a marvelous reception on the hotel’s rooftop bar enjoying a typical and gorgeous New Mexico early autumn sunset with flowing drinks, old friends and incredibly stimulating conversation. American laws virtually prohibit government funds paying for alcohol, thus the drinks were courtesy of the British and French governments. One more idiotic prohibition on productivity and common sense that only undermines our collective efforts especially creatively and collaboratively. These laws have only gotten more prescriptive and limiting. We no longer can pay for meals for interview lunches and dinners, much less business meetings. None of this is reflective of best practice for any business. The power of breaking bread and enjoying a drink to lubricate human interactions is well known. We only hurt our productivity and capacity to produce valuable work by the restrictions. We are utterly delusional about the wisdom of these policies. All of this only serves to highlight the shortcomings in the creative energy evident from the rather low level of vibrancy exhibited by the lack of exciting new ideas.
Never underestimate the power of human stupidity.
– Robert A. Heinlein
rse. Most of the activity for working scientists is at the boundaries of our knowledge working to push back our current limits on what is known. The scientific method is there to provide structure and order to the expansion of knowledge. We have well chosen and understood ways to test proposed knowledge. A method of using and testing our theoretical knowledge in science is computational simulation. Within computational work the use of verification, validation with uncertainty quantification is basically the scientific method in action (
If the uncertainty is irreducible and unavoidable, the problem with not assessing uncertainty and taking an implied value of ZERO for uncertainty becomes truly dangerous (
may prove deadly in rather commonly encountered situations. As systems become more complex and energetic, chaotic character becomes more acute and common. This chaotic character leads to solutions that have natural variability. Understanding this natural variability is essential to understanding the system. Building this knowledge is the first step in moving to a capability to control and engineer it, and perhaps if wise, reduce it. If one does not possess the understanding of what the variability is, such variability cannot be addressed via systematic engineering or accommodation.
systematically is an ever-growing limit for science. We have a major scientific gap open in front of us and we are failing to acknowledge and attack it with our scientific tools. It is simply ignored almost by fiat. Changing our perspective would make a huge difference in experimental and theoretical science, and remove our collective heads from the sand about this matter.
willful uncertainty ignorance. Probably the most common uncertainty to be willfully ignorant of is numerical error. The key numerical error is discretization error that arises from the need to make a continuous problem, discrete and computable. The basic premise of computing is that more discrete degrees of freedom should produce a more accurate answer. Through examining the rate that this happens, the magnitude of the error can be estimated. Other estimates can be had though making some assumptions about the solution and relating the error the nature of the solution (like the magnitude of estimated derivatives). Other generally smaller numerical errors arise from solving systems of equations to a specified tolerance, parallel consistency error and round-off error. In most circumstances these are much smaller than discretization error, but are still non-zero.
The last area of uncertainty is the modeling uncertainty. In the vast majority of cases this will be the largest source of uncertainty, but of course there will be exceptions. It has three major components, the choice of the overall discrete model, the choice of models or equations themselves, and the coefficients defining the specific model. The first two areas are usually the largest part of the uncertainty, and unfortunately the most commonly ignored in assessments. The last area is the most commonly addressed because it is amenable to automatic evaluation. Even in this case the work is generally incomplete and lacks full disclosure of the uncertainty.
repeated using values drawn to efficiently sample the probability space of the calculation and produce the uncertainty. This sampling is done for a very highly dimensional space, and carries significant errors. More often than not the degree of error associated with the under sampling is not included in the results. It most certainly should be.
Every September my wife and I attend the local TeDx event here in Albuquerque. It is a marvelous way to spend the day, and leaves a lasting impression on us. We immerse ourselves in inspiring, fresh ideas surrounded by like-minded people. It is empowering and wonderful to see the local community of progressive people together at once listening, interacting and absorbing a selection of some of the best ideas in our community. This year’s event was great and as always several talks stood out particularly including Jannell MacAulay (Lt.
Col USAF) talking about applying mindfulness to work and life, or Olivia Gatwood inspiring poetry about the seeming mundane aspects of life that speaks to far deeper issues in society. The smallest details are illustrative of the biggest concerns. Both of these talks made me want to think deeply about applying these lessons in some fashion to myself and improving my life consequentially.
We have transitioned from an animal fighting for survival during brief violent lives, to beings capable of higher thought and aspiration during unnaturally long and productive lives. We can think and invent new things instead of simply fighting to feed us and reproduce a new generation of humans to struggle in an identical manner. We also can produce work whose only value is beauty and wonder. TeD provides a beacon for human’s best characteristics along with a hopeful forward-looking community committed to positive common values. It is a powerful message that I’d like to take with me every day. I’d like to live out this promise with my actions, but the reality of work and life comes up short.
TeD talks are often the focus of criticism for their approach and general marketing nature strongly associated with the performance art nature. These critiques are valid and worth considering including the often-superficial nature of how difficult topics are covered. In many ways where research papers can be criticized increasingly as merely being the marketing of the actual work, TeD talks are simply the 30-second mass market advertisement of big ideas for big problems. Still the talks provide a deeply inspiring pitch for big ideas that one can follow up on and provide the entry to something much better. I find the talk is a perfect opening to learning or thinking more about a topic, or merely being exposed to something new.
not identify a single thing recommended in Pink’s book that made it to the workplace. It seemed to me that the book simply inspired the management to a set of ideals that could not be realized. The managers aren’t really in charge; they are simply managing the corporate compliance instead of managing in a way that maximizes the performance of its people. The Lab isn’t about progress any more; it is about everything, but progress. Compliance and subservience has become the raison d’etre.
is progressive in terms of the business world. The problem is that the status quo and central organizing principle today is anti-progressive. Progress is something everyone is afraid of, and the future appears to be terrifying and worth putting off for as long as possible. We see genuinely horrible lurch toward an embrace of the past along with all its anger, bigotry, violence and fear. Fear is the driving force for avoiding anything that looks progressive.
Still I can offer a set of TeD talks that have both inspired me and impacted my life for the better. They have either encouraged me to learn more, or make a change, or simply change perspective. I’ll start with a recent one where David Baron gave us an incredibly inspiring call to see the total eclipse in its totality (
Durkee finding a wonderful community center with a lawn and watched it with 50 people from all over the local area plus a couple from Berlin! The totality of the eclipse lasted only two minutes. It was part of a 22-hour day of driving over 800 miles, and it was totally and completely worth every second! Seeing the totality was one of the greatest experiences I can remember. My life was better for it, and my life was better for watching that TeD talk.
Another recent talk really provoked me to think about my priorities. It is a deep consideration of what your priorities are in terms of your health. Are you better off going to the gym or going to party, or the bar? Conventional wisdom says the gym will extend your life the most, but perhaps not. Susan Pinker provides a compelling case that social connection is the key to longer life (
struggle is for good reasons, and knowing the reasons provides insight to solutions. Perel powerfully explains the problem and speaks to working toward solutions.
other reason that I usually don’t. I will close by honoring the inspirational gift of Olivia Gatwood’s talk on poetry about seeking beauty and meaning in the mundane. I’ll write a narrative of a moment in my life that touched me deeply.
movie “Fight Club” again. This is my 300th blog post here. Its been an amazing experience thanks for reading.
McDonalds for my first job. I was a hard worker, and a kick ass grill man, opener, closer, and whatever else I did. I became a manager and ultimately the #2 man at a store. Still I was 100% replaceable and in no way essential, the store worked just fine without me. I was interchangeable with another hard working person. It isn’t really the best feeling; you’d like to be a person whose imprint on the World means something. This is an aspiration worth having, and when your work is truly creative, you add value in a way that no one else can replicate.
of an incubator for aspiring scientists. You were encouraged to think of the big picture, and the long term while learning and growing. The Lab was a warm and welcoming place where people were generous with knowledge, expertise and time. It was still hard work and incredibly demanding, but all in the spirit of service and work with value. I repaid the generosity through learning and growing as a professional. It was an amazing place to work, an incredible place to be, an environment to be treasured, and made me who I am today.
e scientific culture there were relabeled as “butthead cowboys,” troublemakers, and failures. The culture that was generous, long term in thought, viewing the big picture and focused on National service was haphazardly dismantled. Empowerment was ripped away from the scientists and replaced with control. Caution replaced boldness, management removed generosity, all in the name of formality of operations that removes anything unforeseen in outcomes. The modern world wants assured performance. Today Los Alamos is mere shadow of itself, stumbling forward toward the abyss of mediocrity. Witnessing this happen was one of the greatest tragedies of my life.
importance. Everything is process today and anything bad can be managed out of existence. No one looks at the downside to this, and the downside is sinister to the quality of the workplace.
Instead of encouraging and empowering our people to take risks while tolerating and learning from failure, we do the opposite. We steer people away from doing risky work, punish failure and discourage lesson learning. It is as if we had suddenly become believers in the “free lunch”. True achievement is extremely difficult, and true achievement is powered by the ability to try to do risky almost impossible things. If failure is not used as an opportunity to learn, people will become disempowered and avoid the risks. This in turn will kill achievement before it can even be thought of. The entire system would seem to be designed to disempower people, and lower their potential for achievement.
the knowledge necessary to mentor others. This was a key aspect of my early career experience at Los Alamos. At that time the Lab was teeming with experts who were generous with their time and knowledge. All you had to do was reach out and ask, and people helped you. The experts were eager to share their experience and knowledge with others in a spirit of collective generosity. Today we are managed to completely avoid this with managed time and managed focus. We are trained to not be generous because that generosity would rob our “customers” of our effort and time. The flywheel of the experts of today helping to create the experts of tomorrow is being undone. People are trained to neither ask, nor provide expertise freely.
It is where we find ourselves today. We also know that the state of affairs can be significantly better. How can we get there from here? The first step would be some sort of collective decision that the current system isn’t working. From my perspective, the malaise and lack effectiveness of our current system is so pervasive and evident that action to correct it is overdue. On the other hand, the current system serves the purposes of those in control quite well, and they are not predisposed to be agents of change. As such, the impetus for change is almost invariably external. It is usually extremely painful because the status quo does not want to be rooted out unless it is forced to. The circumstances need to demand performance that current system cannot produce, and as systems degrade this becomes ever more likely.
and not lose all the good things in the process. Bad things, bad outcomes and bad behavior happen, and perhaps need to happen to have all the good (in other words “shit happens”). Today we are gripped with a belief that negative outcomes can be managed away. In the process of managing away bad outcomes, we destroy the foundation of everything good. To put it differently we need to value the good and accept the bad as a necessary condition for enabling good outcomes. If one looks at failure as the engine of learning, we begin to realize that the bad is the foundation of the good. If we do not allow the bad things to happen, let people fuck things up, we can’t have really good things either. One requires the other and our attempts to control bad outcomes, removes a lot of good or even great outcomes at the same time.
The reasons for not estimating uncertainties are legion. Sometimes it is just too hard (or people are lazy). Sometimes the way of examining a problem is constructed to ignore the uncertainty by construction (a common route to ignore experimental variability and numerical error). In other cases the uncertainty is large and it is far more comfortable to be delusional about its size. Smaller uncertainty is comforting and implies a level of mastery that exudes confidence. Large uncertainty is worrying and implies a lack of control. For this reason getting away with choosing a zero uncertainty is a source of false confidence and unfounded comfort, but a deeply common human trait.
If we can manage to overcome the multitude of human failings underpinning the choice of the default zero uncertainty, we are still left with the task of doing something better. To be clear, the major impediment is recognizing that the zero estimate of uncertainty is not acceptable (most “customers” like the zero estimate because it seems better even though its assuredly not!). Most of the time we have a complete absence of information to base uncertainty estimates upon. In some cases we can avoid zero uncertainty estimates by being more disciplined and industrious, in other cases we can think about the estimation from the beginning of the study and build the estimation into the work. In many cases we only have expert judgment to rely upon for estimation. In this case we need to employ a very simple and well-defined technique to providing an estimate.
speaking, there will be a worst case to consider or something more severe than the scenario at hand. Such large uncertainties are likely to be quite uncomfortable to those engaging in the work. This should be uncomfortable if we are doing things right. The goal of this exercise is not to minimize uncertainties, but get things right. If such bounding uncertainties are unavailable, one does not have the right to do high consequence decision-making with results. This is the unpleasant aspect of the process; this needs to be the delivery of the worst case. To be more concrete in the need for this part of the bounding exercise, if you don’t know how bad the uncertainty is you have no business using the results for anything serious. As stated before the bounding process needs to be evidence based, the assignment of lower and upper bounds for uncertainty should have a specific and defensible basis.
To some extent this is a rather easy lift intellectually. Cultural difficulty is another thing altogether. The indefensible optimism associated with the default zero uncertainty is extremely appealing. It provides the user with a feeling that the results are good. People tend to feel that there is a single correct answer. The smaller the uncertainty is the better they feel about the answer. Large uncertainty is associated with lack of knowledge and associated with low achievement. The precision usually communicated with the default, standard approach is highly seductive. It takes a great deal of courage to take on the full depth of uncertainty along with the honest admission of how much is not known. It is far easier to simply do nothing and assert far greater knowledge while providing no evidence for the assertion.
consider this experiment to be a completely determined event with no uncertainty at all. This is the knee jerk response of people is the consideration of this single event as being utterly and completely deterministic with no variation at all. If the experiment were repeated with every attempt to make it as perfect as possible, it would turn out slightly differently. This comes from the myriad of details associated with the experiment that determine the outcome. Generally the more complex and energetic the phenomenon of being examined is, the greater the variation (unless there are powerful forces attracting a very specific solution). There is always a variation, the only question is how large it is; it is never, ever identically zero. The choice to view the experiment as perfectly repeatable is usually an unconscious choice that has no credible basis. It is an incorrect and unjustified assumption that is usually made without a second thought. As such the choice is unquestionably bad for science or engineering. In many cases this unconscious choice is dangerous, and represents nothing more than wishful thinking.
n. Are things worse where I am, or better than the average? For most of my adult life, I’ve had far better conditions than average, and been able to find great meaning in my work. Is the steady erosion of the quality of the work environment a consequence of issues local to my institution or organization? Or is it part of the massive systemic dysfunction our society is experiencing?
government, led by an incompetent narcissistic conman without a perceptible moral compass. Racial tensions, and a variety of white supremacist/right wing ultra-Nationalists are walking the streets. Left wing and anarchist groups are waking up as well. Open warfare may soon be upon us making us long for the days where sporadic terrorist attacks were our biggest worry. A shit storm is actually a severe understatement; this is a fucking waking nightmare. I hope this is wrong and I could simply find a better place to work at and feel value in my labors. I wish the problem was simple and local with a simple job change fixing things.
Work is an important part of life for a variety of reasons. It is how we spend a substantial portion of our time, and much of our efforts go into it. In work we contribute to society and assist in the collective efforts of mankind. As I noted earlier, I’ve been fortunate for most of my life, but things have changed. Part of the issue is a relative change in the degree of self-determination in work. The degree of self-determination has decreased over time. An aspect of this is the natural growth in scope of work as a person matures. As a person grows in work and is promoted, the scope of the work increases, and the degree of freedom in work decreases. Again this is only a part of the problem as the system is working to strangle the self-determination out of people. This is control, fear of failure and generic lack of trust in people. In this environment work isn’t satisfying because the system is falling apart, and the easiest way to resist this is controlling the little guy. My work becomes more of a job and a route to a paycheck every day. Earning a living and supporting your family is a noble achievement these days, and aspiring to more simply a waking dream contracting in the rear view mirror of life.
t can unleash people’s full potential through allowing them to fail spectacularly and then fully support the next step forward. Today, cowardice and mistrust dominate and even marginal failure results in punishment. It is corroding the foundation of achievement. It makes work simply a job and life more survival than living.
comfortable and automatic. In many cases culture is the permanent habits of our social constructs, and often defines practices that impede progress. Accepted cultural practices are usually done without thinking and applied almost mindlessly. If these practices are wrong, they are difficult to dislodge or improve upon.
or fundamental human needs, but most are constructs to help regulate the structures that our collective actions are organized about. The fundamental values, moral code and behaviors of people are heavily defined by culture. If the culture is positive, the effect is resonant and amplifies the actions of people toward much greater achievements. If the culture is negative, the effect can undo and overwhelm much of the best that people are capable of. Invariably cultures are a mixture of positive and negative. Cultures persist for extremely long times and outlive those who set the cultural tone for groups. Cultures are set or can change slowly unless the group is subjected to an existential crisis. When a crisis is successfully navigated the culture that arose in its resolution is enshrined, and tends to persist without change until a new crisis is engaged.
We see all sorts of examples of the persistence of culture. The United States is still defined by the North-South divide that fractured during the Civil War. The same friction and hate that defined that war 150 years ago dominate our politics today. The culture of slavery persists in systematic racism and oppression. The white and black divide remains unhealed even though none of the people who enslaved or who were enslaved are still alive with many generations having passed. The United States is still defined by the Anglo-Saxon Protestant beliefs of the founding fathers. Their culture is dominant even after being overwhelmed in numbers of people and centuries of history. The dominant culture was formed in the crucible of history by the originating crisis for the Nation, the Revolutionary war. Companies and Laboratories are shaped by their original cultures and these habits and practices persist long after their originators have left, retired or died.
(earthquakes, floods, hurricanes, famines, …). These events can stress people and existing cultures providing the sorts of crises that shape the future to be more resilient to future disasters. Human events such as wars, trade, and general political events provide both the impact of culture in causing or navigating events, as well as producing crises that shape cultural responses and evolution. We can continue down this line of thinking to ever-smaller cultures such as organizations and businesses are influenced by crises induced by the larger systems (natural or political). This web of culture continues to smaller and smaller scale all the way to communities (towns, regions, schools, families) each having a culture shaped heavily by other cultures or events. In every case a crisis is almost invariably necessary to induce change, cultures are resistant to change unless something painful provides direct evidence of the incapacity of existing culture to succeed.
culture of the broader scientific community. This culture exists within the broader network of cultures in society with give-and-take between them. In the past science has provided deep challenges to prevailing culture, and induced changes societal culture. Today the changes in main societal culture are challenging science. One key aspect of today’s culture wars is lack of support for expertise. One of the key rifts in society is mistrust of the elite and educated. The broader society is attacking and undermining educational institutions across the board. Scientific laboratories are similar in makeup and similarly under assault. Much of this broader assault is related to a general lack of trust. Some of this is a reaction to the surplus of trust granted science in the wake of its massive contributions to the resolution of World War 2 and the Cold War. These successes are waning in memory and science is now contracting for a distinguished role societally.
ns suffers a bit relative to the other Labs, as does the rigor of computed results. Los Alamos was the birthplace of all three labs and computational work, but always puts computation in a subservient role compared to experiments. This leads to a mighty struggle between validation and calibration. Often calibration wins out so that computed results are sufficiently close to experiment. Sandia excels at process and rigor in the conduct of calculations, but struggles at other aspects (at least in a relative sense). The whole verification and validation approach to simulation quality comes from Sandia reflecting the rigor. At the same time institutional support and emphasis are weaker leading to long-term effects.
All this texture is useful to think about because it manifests itself in every place computational science is done today. The scientific culture of any institution is reflected in its emphasis, and approach to the conduct of science. The culture produces a natural set of priorities that define investments and acceptable quality. We can speak volumes about how computational work should be done, but the specific acuity to the message is related to preconceived notions about the aspects. For example, some places are more prone to focus on computing hardware as an investment. In terms of the competition for resources, the purchase of hardware is a priority, and a typical route for enhancement. This becomes important when trying to move into new “hot” areas. If the opportunity falls in line with the culture, investments flow and if it is out of line the institution will miss it.
omputational science is a relatively new area of endeavor. It is at most 70 years old as practiced at Los Alamos; it is a new area of focus in most places. Sometime it is practiced at an institution and added to the repertoire as a new innovative way of doing work. In all these cases the computational work adopts the basic culture of the institution it exists within. It then differentiates based on the local conditions usually dominated by whatever the first acknowledged success is. One of the key aspects of a culture is origin stories or mythological achievements. Origins are almost invariably fraught situations with elements of crisis. These stories pervade the culture and define what success looks like and how investments in the future are focused.
Where I work at Sandia, the origin story is dominated by early success with massively parallel computers. The greatest success was the delivery of a computer, Red Storm. As a result the culture is obsessed with computer hardware. The path to glory and success runs through hardware; a focus on hardware is culturally accepted and natural for the organization. It is a strong predisposition. At Lawrence Livermore the early stages of the Laboratory were full of danger and uncertainty. Early in the history of the Lab there was a huge breakthrough in weapons design. It used computational modeling, and the lead person in the work went on to huge professional success (Lab Director). This early success became a blueprint for others and an expected myth to be repeated. A computational study and focus was always expected and accepted by the Lab. At Los Alamos all roads culturally lead to the Manhattan Project. The success in that endeavor has defined the Laboratory ever since. The manner of operation and approach to science adopted then is blueprint for success at that Laboratory. The multitude of crises starting with the end of the Cold War, spying, fires, and scandal have all weakened the prevailing culture, and undermined the future.
and knowledge. This always manifests itself with a lack of questioning in the execution of work. Both of these issues are profoundly difficult to deal with and potentially fatal to meaningful impact of modeling and simulation. These issues are seen quite frequently. Environments with weak peer review contribute to allowing confidence with credibility to persist. The biggest part of the problem is a lack of pragmatic acceptance of modeling and simulation’s intrinsic limitations. Instead we have inflated promises and expectations delivered by over confidence and personality rather than hard nosed technical work.
When confidence and credibility are both in evidence, modeling and simulation is empowered to be impactful. It will be used appropriately with deference to what is and is not possible and known. When modeling and simulation is executed with excellence and professionalism along with hard-nosed assessment of uncertainties, using comprehensive verification and validation, the confidence is well grounded in evidence. If someone questions a simulations result, answers can be provided with well-vetted evidence. This produces confidence in the results because questions are engaged actively. In addition the limitations of the credibility are well established, and confidently be explained. Ultimately, credibility is a deeply evidence-based exercise. Properly executed and delivered, the degree of credibility depends on honest assessment and complete articulation of the basis and limits of the modeling.
One of the major sins of over-confidence is flawed or unexamined assumptions. This can be articulated as “unknown knowns” in the famously incomplete taxonomy forwarded by Donald Rumsfeld in his infamous quote. He didn’t state this part of the issue even though it was the fatal flaw in the logic of the Iraqi war in the aftermath of 9/11. There were basic assumptions about Hussein’s regime in Iraq that were utterly false, and these skewed the intelligence assessment leading to war. They only looked at information that supported the conclusions they had already drawn or wanted to be true. The same faulty assumptions are always present in modeling. Far too many simulation professionals ignore the foundational and unfounded assumptions in their work. In many cases assumptions are employed without thought or question. They are assumptions that the community has made for as long as anyone can remember and simply cannot be questioned. This can include anything from the equations solved, to the various modeling paradigms applied as a matter of course. Usually these are unquestioned and completely unexamined for validity in most credibility assessments.
stay the course and make standard assumptions. In many cases the models have been significantly calibrated to match existing data, and new experiments or significantly more accurate measurements are needed to overturn or expose modeling limitations. Moreover the standard assumptions are usually unquestioned by peers. Questions are often met with ridicule. A deeply questioning assessment requires bravery and fortitude usually completely lacking from working scientists and utterly unsupported by our institutions.
personality. This proof by authority is incredibly common and troubling to dislodge. In many cases personal relationships to consumers of simulations are used to provide confidence. People are entrusted with the credibility and learn how to give their customer what they want. Credibility by personality is cheap and requires so much less work plus it doesn’t raise any pesky doubts. This circumstance creates an equilibrium that is often immune to scientific examination. It is easier to bullshit the consumers of modeling and simulation results than level with them about the true quality of the work.
One of the biggest threats to credibility is the generation of the lack of confidence honesty has. Engaging deeply and honestly in assessment of credibility is excellent at undermining confidence. Almost invariably the accumulation of evidence regarding credibility endows the recipients of this knowledge with doubt. These doubts are healthy and often the most confident people are utterly ignorant of the shortcomings. The accumulation of evidence regarding the credibility should have a benefit for the confidence in how simulation is used. This is a problem when those selling simulation oversell what it can do. The promise of simulation has been touted widely as transformative. The problem with modeling and simulation is its tangency to reality. The credibility of simulations is grounded by reality, but the uncertainty comes from both modeling, but also the measured and sensed uncertainty with our knowledge of reality.
n oversold and any assessment will provide bad news. In today’s World we see a lot of bad news rejected, or repackaged (spun) to sound like good news. We are in the midst of a broader crisis of credibility with respect to information (i.e. fake news), so the issues with modeling and simulation shouldn’t be too surprising. We would all be well served by a different perspective and approach to this. The starting point is a re-centering of expectations, but so much money has been spent using grossly inflated claims.
limitations. The difference that simulation makes is the ability to remove the limitations of analytical model solution. Far more elaborate and accurate modeling decisions are available, but carry other difficulties due to the approximate nature of numerical solutions. The tug-of-war intellectually is the balance between modeling flexibility, nonlinearity and generality with effects of numerical solution. The bottom line is the necessity of assessing the uncertainties that arise from these realities. Nothing releases the modeling from its fundamental connection to validity grounded in real world observations. One of the key things to recognize is that models are limited and approximate in and of themselves. Models are wrong, and under a sufficiently resolved examination will be invalid. For this reason an infinitely powerful computer will ultimately be useless because the model will become invalid at some resolution. Ultimately progress in modeling and simulation is based on improving the model. This fact is ignored by computational science today and will result wasting valuable time, effort and money chasing quality that is impossible to achieve.
This is a nice way of saying this is usually a sign that the quality is actually complete bullshit! We can move a long way toward better practice by simply recalibrating our expectations about what we can and can’t predict. We should be in a state where greater knowledge about the quality, errors and uncertainty in modeling and simulation work improves our confidence.
One of the issues of increasing gravity in this entire enterprise is the consumption of results using modeling and simulation by people unqualified to judge the quality of the work. The whole enterprise is judged to be extremely technical and complex. This inhibits those using the results from asking key questions regarding the quality of the work. With the people producing modeling and simulation results largely driven by money rather than technical excellence, we have the recipe for disaster. Increasingly, false confidence accompanies results and snows the naïve consumers into accepting the work. Often the consumers of computational results don’t know what questions to ask. We are left with quality being determined more by flashy graphics and claims about massive computer use than any evidence of prediction. This whole cycle perpetuates an attitude that starts to allow viewing reality as more of a video game and less like a valid scientific enterprise. Over inflated claims of capability are met with money to provide more flashy graphics and quality without evidence. We are left with a field that has vastly over-promised and provided the recipe for disaster.
A great question is an achievement in itself although rarely viewed as such. More often than not little of the process of work goes into asking the right question. Often the questions we ask are highly dependent upon foundational assumptions that are never questioned. While assumptions about existing knowledge are essential, finding the weak or invalid assumptions is often the key to progress. These assumptions are wonderful for simplifying work, but also inhibit progress. Challenging assumptions is one of the most valuable things to do. Heretical ideas are fundamental to progress; all orthodoxy began as heresy. If the existing assumptions hold up under the fire of intense scrutiny they gain greater credibility and value. If they fall, new horizons are opened up to active exploration.
questions have led to the best research, and most satisfying professional work I’ve done. I would love to recapture this spirit of work again, with good questioning work feeling almost quaint in today’s highly over-managed climate. One simple question occurred in my study of efficient methods for solving the equations of incompressible flow. I was using a pressure projection scheme, which involves solving a Poisson equation at least once, if not more than once a time step. The most efficient way to do this involved using the multigrid method because of its algorithmic scaling being linear. The Poisson equation involves solving a large sparse system of linear equations, and the solution of linear equations scales with powers of the number of equations. Multigrid methods have the best scaling thought to be possible (I’d love to see this assumption challenged and sublinear methods discovered, I think they might well be possible).
As a result as problems become larger, the proportion of the solution time spent solving linear equations grows ever larger (the same thing is happening now to multigrid because of the cost of communication on modern computers). I posed the question of whether I could get the best of both methods, the efficiency with the robustness? Others were working on the same class of problems, and all of us found the solution. Combine the two methods together, effectively using a multigrid method to precondition the conjugate gradient method. It worked like a charm; it was both simple and stunningly effective. This approach has become so standard now that people don’t even think about it, its just the status quo.
The second great question I’ll point to involved the study of modeling turbulent flows with what has become known as implicit large eddy simulation. Starting in the early 1990’s there was a stunning proposition that certain numerical methods seem to automatically (auto-magically) model aspects of turbulent flows. While working at Los Alamos and learning all about a broad class of nonlinearly stable methods, the claim that they could model turbulence caught my eye (I digested it, but fled in terror from turbulence!). Fast forward a few years and combine this observation with a new found interest in modeling turbulence, and a question begins to form. In learning about turbulence I digested a huge amount of theory regarding the physics, and our approaches to modeling it. I found large eddy simulation to be extremely interesting although aspects of the modeling were distressing. The models that worked well were performed poorly on the structural details of turbulence, and the models that matched the structure of turbulence were generally unstable. Numerical methods for solving large eddy simulation were generally based on principles vastly different than those I worked on, which were useful for solving Los Alamos’ problems.
I should spend some time on some bad questions as examples of what shouldn’t be pursued. One prime example is offered as a seemingly wonderful question, the existence of solutions to the incompressible Navier-Stokes equations. The impetus for this question is the bigger question of can we explain, predict or understand fluid turbulence? This problem is touted as a fundamental building block in this noble endeavor. The problem is the almost axiomatic belief that turbulence is contained within this model. The key term is incompressible, which renders the equations unphysical on several key accounts: it gives the system infinite speed of propagation, and divorces the equations from thermodynamics. Both features sever the ties of the equations from the physical universe. The arguing point is whether these two aspects disqualify it from addressing turbulence. I believe the answer is yes.
Another poorly crafted question revolves around the current efforts for exascale class computers for scientific computing. There is little doubt that an exascale computer would be useful for scientific computing. A better question is what is the most beneficial way to push scientific computing forward? How can we make scientific computing more impactful in the real world? Can the revolution of mobile computing be brought to science? How can we make computing (really modeling and simulation) more effective in impacting scientific progress? Our current direction is an example of crafting an obvious question, with an obvious answer, but failing to ask a more cutting and discerning question. The consequence of our unquestioning approach to science will be wasted money and stunted progress.
rust in people and science. Trust is a powerful concept and its departure from science has been disruptive and expensive. Today’s scientists are every bit as talented and capable as those of past generations, but society has withdrawn its faith in science. Science was once seen as a noble endeavor that embodied the best in humanity, but generally not so today. Progress in the state of human knowledge produced vast benefits for everyone and created the foundation for a better future. There was a sense of an endless frontier constantly pushing out and providing wonder and potential for everyone. This view was a bit naïve and overlooked the maxim that human endeavors in science are neither good or bad, producing outcomes dependent upon the manner of their use. For a variety of reasons, some embedded within the scientific community, the view of society changed and the empowering trust was withdrawn. It has been replaced with suspicion and stultifying oversight.
The largest portion and most important part of this process is the analysis that allows us to answer the question. Often the question needs to be broken down into a series of simpler questions some of which are amenable to easier solution. This process is hierarchical and cyclical. Sometimes the process forces us to step back and requires us to ask an even better or more proper question. In sense this is the process working in full with the better and more proper question being an act of creation and understanding. The analysis requires deep work and often study, research and educating oneself. A new question will force one to take the knowledge one has and combine it with new techniques producing enhanced capabilities. This process is on the job education, and fuels personal growth and personal growth fuels excellence. When you are answering a completely new question, you are doing research and helping to push the frontiers of science forward. When you are answering an old question, you are learning and you might answer the question in a new way yielding new understanding. At worst, you are growing as a person and professional.
Once this creation is available, new questions can be posed and solved. These creations allow new questions to be asked answered. This is the way of progress where technology and knowledge builds the bridge something better. If we support excellence and a process like this, we will progress. Without support for this process, we simply stagnate and whither away. The choice is simple either embrace excellence by loosening control, or chain people to mediocrity.
