Future Babble: A Review and Commentary

 

  • McClelland & Stewart (October 12, 2010)
  • ISBN-10: 0771035195

Book Review and Commentary by Mike Darwin

The success of cryonics, both in absolute and relative terms, arguably depends upon the accuracy and precision with which we (cryonicists) can predict the future. Our ability as seers is important in the absolute sense, because failure to accurately anticipate the requisite social, economic and scientific developments necessary for the success of cryonics would mean that we are wasting our time, energy and money – and perhaps should  concentrate those assets on other strategies for survival (or more simply, stop tilting at windmills and enjoy our life in the here and now). Our predictive ability is also important to cryonics’ success relatively, since failure to accurately foresee the short- to intermediate-term future of cryonics is very likely to erode our credibility with both the general public and the professional and scientific communities and result in failure to anticipate lethal problems that might otherwise have been avoided.

If you doubt that this is so, there is a simple on-line “game” that you can “play” that was developed by cryonicist and computer programmer Brook Norton.  It is called The Cryonics Calculator: Derivation of Cryonics Probabilities, and it allows you to enter the risk of various possible failure modes for your hypothetical (or real) cryonics organization and then see what happens to the probability you that you will remain cryopreserved long enough to be revived: http://www.cryonicscalculator.com/index.php?option=com_content&view=article&id=2&Itemid=3.The results might be described as the reverse of compound interest: small risks for any short period of time become lethal risks over long periods of time. In plugging scenarios into the The Cryonics Calculator, I was also reminded of the liability of complex systems with hundreds or thousands of critical components to failure, even if the per component reliability is 99%. Spacecraft, as any Shuttle engineer will tell you, are a good example of this phenomenon.

So, how do we do in predicting the future? That question isn’t hard to answer in the case of most cryonicists, because there is a fairly large base of written material available to peruse in making an assessment. The answer is that we do horribly. Really horribly.

Of course, cryonicists are by no means the only people interested in predicting the future. To some extent, everybody wants to know what tomorrow holds. Economists, politicians, investors, corporations, in fact just about every human institution and enterprise, has a strong incentive to accurately predict what lies ahead.  Indeed, many people make their livings doing just that; stock market analysts, commodities advisers, government intelligence analysts, and even the neighborhood fortune teller are all  paid to peer into the future and tell us what lies in store. In answer to the question of how well these more conventional (and vastly more respected) seers perform, Canadian journalist Dan Gardner wrote the book Future Babble: Why Expert Predictions Fail and Why We Believe Them Anyway. Gardner’s conclusion, informed heavily by the research of Philip Tetlock, Professor of Psychology at the University of Pennsylvania,  is that the experts, be they economists, petroleum experts, futurists, or political pundits are about as accurate in forecasting the future as as a group of “dart-throwing monkeys.”

In fact, on average, you’d be better off making decisions about what is to come based on a simple coin toss, or deciding that “things will stay about the same.” The first question that comes to mind is, “why are the experts (and indeed humans in general) so bad at predicting the future?” Gardner explores the answers to this question in clear, easy to understand terms in text that is as concise as it is fast paced. At the most basic level, predicting the future suffers from the problems of complexity and chaos that are inherent in the real world. Want to know when “peak oil” production will occur? How hard that can be to figure out? There is clearly a finite amount of oil on the planet, it would seem we know how much is left, and it is certainly easy enough to plug in various numbers for the rate at which oil is being consumed. What’s so difficult about that?

As it turns out, even such a seemingly simple problem is enormously complex. Knowing where and how much untapped oil exists is more difficult than it seems. Technological advances cannot only make formerly unreachable oil accessible, it can also make long abandoned oil fields formerly considered “exhausted” highly productive.  And, as prices rise, previously economically nonviable sources of oil, such as oil sands, become cost effective to recover. While there is no question that oil will eventually run out, there is a huge difference between that happening in the 1980s, versus it not having happened 20 years later. Accuracy isn’t enough; precision is critically important as well.

If complexity weren’t a bad enough problem, to it can be added the problem of chaos, as in chaos theory. Modern chaos theory originated with the work of mathematician and meteorologist Edward N. Lorenz, who noticed that even infinitesimal changes to the numbers used in maths models of weather prediction resulted in radically altered outcomes.  It was Lorenz who discredited linear statistical models in meteorology and who famously asked, “Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas?” The answer is, yes, it can, and thus was born the term “the butterfly effect.”  Chaos powerfully limits both accuracy and precision in predicting the behavior of complex systems, of which the everyday world is certainly one.

A central point that Gardner considers is Tetlock’s study (and resulting book) Expert Political Judgment: How Good Is It? How Can We Know? (2005) which describes his 20-year long prospective study in which 284 experts in many fields, from university professors to journalists, and with many ideological orientations, from ultra-left Marxists to libertarian free-marketeers, were asked to make 28,000 predictions about the future. Tetlock found their performance dismal: they were only slightly more accurate than chance. His study was complex, but his conclusion was brutally simple: the experts were not only worse than run of the mill statistical models, they could barely eke out a tie with the proverbial dart-throwing chimps. And there was no difference in ideological bias; capitalists and Marxists performed equally poorly.

None of this should be too surprising. Lots of other authors have explored this phenomena in detail, most notably Tetlock himself  (i.e., Expert Political Judgement), and Nassim Taleb, in his superb book Fooled by Randomness (and the later in The Black Swan). The useful things about Gardener’s book are that it presents these ideas in a highly readable and accessible format, and that it explores the underlying psychology and biology of why we humans are such “seer-suckers.” We just can’t help coming back for more – usually from the same “discredited” experts who misled us only a few years, months or even weeks before.

Implications for Cryonics

Recently, in preparation for another piece of writing, I hauled out my copy of science fiction author Robert Heinlein’s 1980 book, Expanded Universe. Included in the book are his essays “1950 Where To?” and “The Third Millennium Opens.” The former are his predictions about the year 2000 made in 1950, and the latter are his predictions about the year 2001, made from the vantage point of 1980. In reading these, it is impossible to conclude anything other than that Heinlein was terrible, in fact ridiculously terrible at predicting the future.  “Where to?” is 7 pages long, whereas his attempt to justify and waffle on the failed predictions he makes there runs to (a pathetic) 29 pages!  Heinlein was neither stupid nor ignorant; he had access to some of the best  scientific, technical and military minds of his day (as did future forecasters Herman Kahn and Robert Prehoda) and yet he failed utterly to see what lay even 20  years ahead of him, as did virtually all of the other technological seers before him.

What does this mean for cryonics? At first glance the news would seem to be all bad. It is pretty clear that we can’t predict the future, even the very near term future (5-10 years), either in terms of technological advances or man-made or natural catastrophes.  The future remains as it has always been; not just to be seen “through a glass darkly,” but not to be seen at all. However, there is some more hopeful news summarized in Gardner’s book (and present in considerably greater detail in Tetlock’s superb book Expert Political Judgment), which I believe has real and useful application to cryonics. Not all seers in Tetlock’s study were equally bad. Some were truly  terrible, and those were invariably the experts who informed their decision making on the basis of an ideological agenda. It did not matter if the experts were Marxists or Capitalists; to the extent their decision making was ideologically based, it was invariably less accurate. The best decision makers relied on multiple sources of data, entered the problem solving process with minimal biases, and had little or no ego investment in their conclusions. In other words, they were willing to revise their thinking, admit errors and reevaluate their conclusions as necessary. That’s a fairly uncommon trait in humans, even amongst scientists.

The Directors, Officers and in particular the Chief Executive Officers of cryonics organizations are the ones on whom the proximate responsibility rests for shepherding the organization’s members and patients into the future.  In the past, no attention has been given to how these people should be selected. In large measure this has been because the pool of candidates has been vanishingly small, and all too often almost anyone willing to serve had to be accepted, for lack of any alternative. Hopefully, the future will offer more choice, and if and when it does, it would behoove us to carefully examine the background and the corpus of writing of those whom we choose to lead us. We should look for the accuracy and precision of their past decision making, as well for the extent to which they are “calibrated” in their decision making. If a person says (on average) that he is  ~80% confident his predictions will come true, and in fact, ~80% of them do prove correct, then he is 100% calibrated. This is important, because knowing how much confidence to place in your judgment is often crucial. Overconfidence can be a killer, as can endless waffling and the inability to act.

Beyond the leader as seer there are, of course, many duties and qualities required. These are beyond the scope of consideration here. However, it seems a good place to start that we not empower people to decide our futures who are demonstrably terrible at predicting it. Not just ‘flip of the coin bad,’ but truly terribly bad. Such people, it turns out, are fairly easy to spot by examining the corpus of their past work and decision making. This is quite different than looking at a “markers,” such as economic success. A used car salesman, a stock broker, or a huckster of commemorative coins may be tremendously financially successful. The question that should be asked in such cases is, “At whose expense?”

____________________________________________________________

Afterword

A few months ago, I was scanning (digitizing) some back issues of Cryonics magazine from 1988, and I happened to notice I had written (with assistance from Steve Harris, M.D.) an article predicting the future of medicine 20 years hence, entitled The Future of Medicine, Cryonics, January, 1988 pp. 31-40: http://www.alcor.org/cryonics/cryonics8802.txt and in Cryonics, February 1988, pp 10-20: http://www.alcor.org/cryonics/cryonics8803.txt. I had forgotten I’d even written the article! You can read it and see how well (or poorly) I did.

That article led me to more comprehensively review my writings over the years. The results were interesting. For those of you who write, publicly or privately, I can promise you that rereading your writings in the decades to come will be a fascinating undertaking. Socrates famously said, “The unexamined life is not worth living.” Well, maybe, but I think that just perhaps, the unexamined life may be a lot more fun.

 

 

This entry was posted in Cryonics History, Cryonics Philosophy, Culturomics, Economics, Philosophy, Uncategorized. Bookmark the permalink.

8 Responses to Future Babble: A Review and Commentary

  1. unperson says:

    Such predictions typically come from those at the top of the social hierarchy, or are published by such. These predictions are disseminated by radio, tv, magazines, which are all relatively powerful entities, and are thus treated by people are being high up in the social hierarchy. This gives them their weight, however useless the value of their predictions.

    I think the more important and informative question is why do men not realize the inherently hierarchical structure of their pack animal species?

    We formally recognize how dominance hierarchies mold and influence the behavior of other animal species. Why do we not discuss how the dominance hierarchy decides for us what things are worthy of discussion?

    I discuss such things, but I am not at the top, and the top does not disseminate my opinions. Thus I am discredited, at least by those who value only opinions disseminated by those at the top, which is to say 99.99 percent of all humans.

    Drs Somit and Petersen discovered this as well when they covered these same topics in their book DOMINANCE, DEMOCRACY AND DARWINISM. One of them was quoted as being disappointed that their book was ignored. But as a logical extension of their theories, it should have been ignored, and it was.

    I recently viewed a video from about 30 years ago of some New Guinea tribes that met white men for the first time. Up to that time they have lived only a simple forest, stone age existence. It was interesting to see that the near non-existence of a social hierarchy. I think that larger the society, the stronger the social hierarchy.You can see this also in the San/Bushmen societies of Africa–the near lack of social hierarchy. In fact in their small tribes, the people actively fight against the ascendance of people to the top. One westerner social scientist saw a concrete example of this when, to thank them for helping him in his studies of them, he bought them a large steer for a farewell feast. The tribespeople disparaged the worth of the steer, when in fact it was a huge bonanza. But in their hierarchy hatred, they resisted the rise of some to the top of the hierarchy, hence the insults about his steer.

    In modern society we have accepted, unknowingly, societal complexity and its positive rewards (technology via specialization), and also received the negative rewards of an out of control societal elite due to a runaway hierarchy.

    • admin says:

      It’s interesting and ironic to read your comments, because they are “top down” focused from a “bottom” perspective. By this I mean that you are doing exactly what you accuse the people at the top of the “social hierarchy” of doing, namely considering the problem of prediction only from the “top.” Many (and maybe even most) predictions, including most that relieve people of their money and cause harm, are not made by experts who use the apparatus of the media or government. Rather, they are made by experts who operate in everyday life. I’ve no shortage of examples here, because I’ve experienced some of them personally, and watched others suffer, as well. You need to buy life insurance, tires that will last a long while, or need advice about computers or software? Chances are pretty good you asked a “local” expert – someone you know who seems knowledgeable, or who is certified, trained, or otherwise credentialed in such a way as to give the impression they are “expert” in the area you need advice about. All too often these people are WRONG. This goes on every day…and the surprising thing is that you (and I) are most likely to return to them for advice in the future when the stakes are highest and the problem is the most complicated.

      A few weeks ago I attended a lecture at The Old Operating Theatre in London. It is the oldest operating “room” now in existence and was created at St. Thomas’ Hospital in 1822. A large group of students from outside London was present (14-16 year olds) and, as the lecturer knew I was present, I was sometimes asked for commentary. One of the issues discussed was trephination, and that led the lecturer to inquire about Congresswoman Gabriel Giffords, who had been shot in the head some weeks previously. I noted that despite its compelling logic, and despite excellent performance in early, small trials, decompressive craniectomy, wherein a large portion of the skull is removed to accommodate cerebral swelling, had recently been demonstrated not to be effective in improving either survival or neurological outcome in a large, prospective, randomized clinical trial with the clever acronym of TOPHAT. After the lecture, a group of students wanted to talk with me about how it was possible that so many useless, disgusting and clearly harmful treatments had predominated in medicine and surgery for so long, for centuries, in fact? How was this POSSIBLE, they wanted to know?

      I told them that one good way to understand this was to look at politics. Politics is usually seen as authoritarian, and organized from the top down. But in fact, in both the UK and the US, politicians mostly get their start from the bottom up. People just don’t start paying attention to them until they get to near the top tier. Most politicians start out small and begin running for office with no large scale support, beyond a party endorsement, which is pretty easy to get… The important point about their being local is that the people who elect them will have a lot of opportunity to MEASURE their effectiveness and to MEASURE the effectiveness of their specific decisions. BUT THEY DON’T. If a politician, great or small, advocates that a tax be levied or removed, that a law be enacted, or that any other measure be taken, there is no systematic and rigorous SCIENTIFIC evaluation of the outcome. In other words, did it work? How well did it work? If not, why not, and what were the consequences? Since this is never done in politics, the issues are decided on the basis of ideologies, opinions, bullying, appeal to emotions, and clever rhetoric uninformed by facts.

      A consequence of this is that decision making goes back and forth like the tide in the Thames: conservatives, labour, conservatives, labour, until the system becomes so dysfunctional that another, new kind of nutter is minted and empowered. At no point are the DECISIONS examined by scientific means and their practitioners held accountable. And, the funny thing is, no one thinks this is at all unusual. It’s just the way politics is, and people even say, “Well that’s politics!” with exasperation and resignation. That’s how it was in medicine and still is, to a surprising extent today. Semmelweiss, Hunter, Koch, Pasteur – these men all helped to provide a more scientific foundation for medicine. However the advent of systematic scientific medicine, so called Evidence Based Medicine, is only about 15 years old!

      One reason you might benefit from reading FUTURE BABBLE is that Gardner answers an even more fundamental question than the one about why we are hierarchical, and that is why are we such suckers for people who obviously can’t predict anything???? [Actually, I suspect the answer to that questions overlaps with the answer to the question of why we are so hierarchical.] The answer to that question is that it is in large measure a function of the way our brains are configured as a result of being evolved to function in an environment that is radically different from the one we currently inhabit. We are configured to see patterns where none exist and we have a strong survival-linked incentive to believe we can be sure, or at least less uncertain, about the future. Those are very likely biological realities that will be difficult to work around.

      Finally, I largely agree with your observations about hunter-gatherer societies and authoritarianism. A book you might really enjoy reading that examines this in detail is CRAZY HORSE AND CUSTER: The Parallel Lives of Two American Warriors by Stephen Ambrose. Ambrose makes a pretty convincing case that the primary reason for the conquest of the Amerindians by the Americans was their inability to understand, let alone embrace, hierarchical social structure. — Mike Darwin

  2. gwern says:

    > That article led me to more comprehensively review my writings over the years. The results were interesting. For those of you who write, publicly or privately, I can promise you that rereading your writings in the decades to come will be a fascinating undertaking. Socrates famously said, “The unexamined life is not worth living.” Well, maybe, but I think that just perhaps, the unexamined life may be a lot more fun.

    But you can’t check your predictions unless you write them down, indeed…

    I’ve spent a great deal of time over the last year recording my own predictions: http://predictionbook.com/users/gwern At last count, I had around 900 registered and 202 closed/judged.

    I’ve found it a valuable exercise in teaching me the truth of Roy Amara’s saying: “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.”

  3. Mark Plus says:

    Speaking of medical progress, Skeptical Inquirer published an assessment of the war on cancer which concludes that modern medicine doesn’t have much to show for all the money thrown at the problem. By contrast, medicine has made substantial progress against cardiovascular diseases:

    The War on Cancer A Progress Report for Skeptics
    http://www.csicop.org/si/show/war_on_cancer_a_progress_report_for_skeptics/

  4. Mark Plus says:

    Related to the prediction issues, cryonics faces a “Hayekian” sort of problem in finding and coordinating a vast amount of dispersed and tacit knowledge which needs to come together to solve our technical challenges. So far we’ve used a technocratic, central-planning approach which has resulted in (1) no progress, and (2) wastes of fortunes on follies like the Saulsoleum in Texas.

    Clearly we need to try something radically different: When you have a really hard problem, ask everyone to help you try to solve it.* I’d like to see cryonics organizations develop an open innovation model, perhaps based on the X Prizes, to provide the incentive for other people to spend their own money and employ their own minds on well posed technical goals related to cryonics and applicable to our immediate needs. According to the experience of the X Prizes, a $10 million prize leads the contestants to spend on the order of $100 million of their own money in developing new technologies to solve the engineering challenge!

    Moreover, we may already have good contracts within the X Prize foundation to help us. One well known cryonicist sits on the X Prize Foundation’s board of trustees, and possibly more if another famous inventor there follows through with his interest in cryonics by signing up (based on confidential information I’ve received about the second fellow):

    http://www.xprize.org/about/board-of-trustees

    Incentive prizes have the potential to offer the cryonics movement the benefit of far more resources than we could acquire ourselves. We have free money just sitting there on the table. Why don’t we try to pick it up and improve our chances for survival?

    *The Japanese government should consider an incentive prize to help it deal with its nuclear disaster.

    • admin says:

      Mark, your comments are accurate, but your focus on a “prize” as the solution to them doesn’t seem a good idea to me. Maybe because I’m too close to the technical problems? I don’t know. I guess my first question is, “what is your sample size of “prizes” that makes you think this approach works?” I think the Rutan X-Prize was interesting, and it may even prove to be worthwhile. But it is not a demonstrated meaningful step in advancing spaceflight – and it won’t be until there is proven follow-on.

      How would you structure an X-prize for cryonics? I’ve heard people suggest that the feat be something like reviving a small mammal from vitrification. That sounds good, unless you understand what’s involved. An X-Prize for the longest lived mouse, or for achieving “space” flight privately, are in a completely different class than recovering a rat from -130 deg C. Why? Because the fundamental theoretical and technical barriers to spaceflight were solved over 40 years ago. Similarly, the lifespan of mammals has been manipulated with some success since the 1930s. Ask for incremental advances in these two areas and you are likely to get them.

      There is also the hard reality that virtually all of the expertise in solid organ cryopreservation now resides in a building in Fontana, CA. That is a barking mad state of affairs! I’ve seen the technology required to achieve the modest level of success in this area to date, and it is enormously complex, “arty” and daunting. Just mastering the various models (such as small animal organ transplantation) is a black art. This is true of both spaceflight and gerontology, as well. The difference is that by comparison to solid organ cryopreservation, these very hard to master and “art-intensive” technologies are well distributed. Because rocketry is so useful for warfare, nation-states have kept the expertise viable – and the advent of communications and weather satellites has helped, as well. Almost every good gerontology lab in terms of rodent liespan work can trace its pedigree back to Roy Walford at UCLA. It is a high art to do lifespan experiments because so much can go wrong. The US, and other governments have spent a lot of money on aging and they will likely continue to do so. That’s why these other labs exist.

      Leaving cryonics out of it, organ cryopreservation has become completely disenfranchised and the way the remaining outpost is structured is 100% not conducive to that changing. One of the great and powerful things about academia is its relative openness, and its strong incentive for teaching, training and mentoring. I’ve met some of the finest minds in critical care medicine, and I can tell you that they are all very much aware they are going to retire someday. They inhabit a culture where it is incumbent upon them to teach others and to mentor others to continue their work; and even to help found centers of study elsewhere in the world. A culture of secrecy and presumed immortality is not conducive to the dissemination of ideas. If you look at history of any scientific advance or revolution, it becomes apparent that different intellectual loci are essential to progress. Each institution inevitably develops its own culture and that culture will invariably limit its creativity. There are no exceptions to this.

      Reading FUTURE BABBLE made me reflect on the singular lack of prophetic ability of my good and long-time friend Greg Fahy. Greg has been reassuring us that vitrification is “just around the corner” for at least 20 years now! It’s not Greg’s fault that it hasn’t happened yet; but he does have to take responsibility for being a lousy prophet. And more importantly, we have to stop believing him. I mean, what’s up with us that we keep buying into the prophecies of a guy who has a 100% track record of failure as technological forecaster? And we are not talking about an n of 1 here, either!

      As far as a cryonics X-Prize goes, were the knowledge and skill base there, the ideal goal, IMHO, would be to demonstrate consistent recovery of integrated EEG activity in a mammalian brain (excluding rats and mice) following stable cryopreservation (i.e., cooling to and storage at -100 deg C for at least 7-days). That’s the ideal. Next best is Kenneth Hayworth’s Brain Preservation Technology Prize. The problems with scaling up brain fixation and room temperature vitrification (polymer embedding) are very different than the problems of scaling up cryopreservation from tissue culture to whole organs. I’m not sure I agree with Ken’s estimate of “absolute certainty” that it is possible to adequately fix and RT vitrify a human brain. But, it IS more of a technical problem than a theoretical one when compared to scaling up organ cryopreservation from tiny volumes of tissue to whole organs.

      Ironically, where organ vitrification stands (or should I say rests?) right now is also due largely to a series of technical problems: cryoprotectant toxicity, cryoprotectant equilibration, viscoelastic injury, and fracturing. That’s four really serious and daunting technical problems. Nevertheless, it is my hunch that were the research more disseminated and being worked on by a broader cross section of minds (and institutional cultures), organ cryopreservation would be a done deal – if not now, then within the decade. But alas, that isn’t the case.

      Fixation is looking better and better ;-0! I wish Ken luck and godspeed. — Mike Darwin

      • admin says:

        This morning I realized that I did not adequately state the problems with the M-Prize for the longest lived mouse. There was a subtle flaw in that prize’s goal, and that was that it asked for (and got) and extension of the AGING PROCESS. Slowing aging is not halting or reversing aging: it is not rejuvenation. Rather, it is dragging out the extant dying process a little longer and in a little better health, if you are lucky. This is what I call the “red herring of the calorie restriction approach” to life extension. Sure, we’ll take what we can get in terms of life extension, but calorie restriction is literally a dead end. It slows age related pathologies and essentially converts animals into rodent centenarians or supercentenarians. Have you looked at pictures of human centenarians or supercentenarians (Google supercentenarians under images and take a look)? Argh! Horrible!They are OLD and the look awful. They are basically “crisps” of human beings – the fried and withered bits left over that constitute the least amount of human tissue that can remain alive. That is not what we want.

        If the M-Prize had been to rejuvenate a mouse or simply to stop aging for a fixed and appreciable period of time, say 4 months, I think it unlikely that there would have been many takers, and more unlikely still that there would be any winners.

        For such “incentivizing” prizes to work, the theoretical problems generally have to be largely or completely overcome, and the requisite expertise has to be widespread or reasonably easily achievable. So, for example, an X-Prize to teleport a 1 gram integrated circuit (or other complex structure) over a distance of, say, 15 feet, or an X-Prize to demonstrate a working “molecular assembler” aren’t very likely to be successful. — Mike Darwin

  5. Abelard Lindsey says:

    I don’t know where to say it. So, I will say it here.

    I am sick and tired of hearing people describing the crash of ’08 as a “black swan” event.

    The U.S. economy had two bubbles, the dot-com bubble of the late 90′s and the housing bubble during the oughts. Both bubbles were obviously fake growth, as are all bubbles, and a crash was inevitable. There was nothing “black swan” about the crash of ’08. It was entirely predictable and inevitable.

Leave a Reply to gwern Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>