Risk Assessment Form (3: Fill & Download for Free

First, we must disambiguate: the term artificial intelligence can mean two very distinct things in practice:Computerized decision-making. This is the traditional meaning of artificial intelligence, where a mechanical system makes policy decisions directly to save human time and effort. Examples include donor kidney matching algorithms, customer-facing virtual assistants in government offices, and even handwritten address recognition by the postal department.Computer-aided decision-making. This is the use of computer simulations, numerical forecasts and predictive analytics to inform decisions and recommendations made by humans. Examples include fire risk modelling, predictive policing, and decision support systems (such as IBM Watson).This disambiguation is needed because:The distinction is blurry…Both computerized and computer-aided decision-making often use the same techniques (logistic regression, k-means, algorithmic mechanism design, etc.) to arrive at the same decisions.This makes them effectively the same things under the hood, and thus have a common source for problems — bad choices of variables to model, quality and size of dataset to generalize, faulty assumptions, etc.… but the long-term risk profile in policy making is very different.Fully computerized decision-making is expected to be able to performa) securely, b) without supervision or with minimal supervision, c) armed solely with task-specific information, and d) with high accuracy even in unforeseen circumstances.Computer-aided decision making is not subject to such high performance expectations. Because it is a research tool, it is expected to be slower, more error-prone, requiring human intervention to adapt it to novel situations, in some cases with transparent methods open for anyone to inspect.Thus, not only are the kinds of risks and impacts different: the pressures that shape each’s improvements are divergent, and will result in individual safety approaches for each new application.I’m not going to address the many benefits both kinds of approaches offer — it is transparently obvious that computer approaches (whether autonomous or assistive) have a substantial role to play in government policy, whether it be in predictive efforts or in time savings and efficiency.What this answer will do, instead, is talk about unintended effects of this adoption in policymaking, as well as sensible risk mitigation strategies in policy. This will clarify the boundaries computer assistance or automation should have.Where is the Current Risk Right Now?Despite popular media depictions to the contrary, many of the dangers of AI use in policy currently come from computer-aided decision-making rather than computerized decision-making.This has several reasons:Adoption of computerized decision-making has been slower in government than in the private sector[1]owing to a shortage of expertise and a change-resistant culture.Where adopted, it has primarily been used in relatively narrow domains such as document drafting, virtual assistance, license plate reading, document translation and request routing[2] with low potential for both negative impact and size of negative impact.Computer-aided decision-making, in contrast, is both used more widely and more systematically by organizations — for example, estimates of how many police forces in the UK are using some form of algorithmic profiling are between 14% to 33%. [3]The negative impacts of computer-aided decision-making are significantly higher since they areused in high-impact areas across society[4], such as determining whom to lend to, whether individuals are likely to recidivate or should receive bail, and risk of child exploitation in familiesmore prone to systemic bias, rather than individual bias, on account of data quality and data engineering issuesconsidered more “reliable” because they are evaluated by a human layer despite the comparatively higher likelihood of mistakes at the human layer, meaning less security and regulation overall.This is not to say that computerized decision-making does not have concrete safety problems or potential for abuse or significant harmful applications in government, such as military use — they do, and there have been widespread efforts to address these technically and procedurally. It is almost certainly where the risk will be in the future, in things like predictive coding where small errors can be amplified enormously and detrimentally for end users.However, their current impact in public policy specifically is limited, so, unless it becomes more prominently used or involves higher-cost systems, focusing on it is not productive. The rest of this answer will focus only on computer-aided decision making.What are the Biggest Issues Surrounding Computer-Aided Decision Making?By far and away, the most frequently cited problem of computer-aided decision making is unfair systemic discrimination — that is, an algorithm violating one of the three criteria below[5][5][5][5]:1. Anti-classification: The model is fair if it does not use protected characteristics or proxies from which protected characteristics can be inferred.2. Classification or outcome error parity: The model is fair if protected groups receive equal proportion of positive outcomes, or equal proportion of errors.3. Calibration: An algorithm is well-calibrated if the risk scores it gives to people reflect the actual outcomes in real life for the people given those scores. Equal calibration definitions of fairness say that an algorithm should be equally calibrated between protected groups. For example, among those given a particular risk score, the percentage which then results in the predicted outcome should be the same between protected groups (e.g. men and women).Algorithmic bias is widespread and predominant[6][6][6][6][7][7][7][7]— I highlight just two instances of their impact below, taken from the UK government’s landscape survey on the subject, and a third from the Wikipedia article on artificial intelligence use in government:A 2015 study showed that many of the algorithms used by insurance companies in the US to create quotes for car insurance were relying on credit scores more heavily than driving records. This meant that in Florida, an individual with a clean driving record but poor credit score could end up paying $1,552 more for car insurance than the same driver with a drink driving conviction but an excellent credit score. …… Big Brother Watch have flagged concerns over potential biases in the facial recognition systems currently being trialled by some UK forces. In 2018 they identified high rates of misidentification (where an individual was inaccurately identified as a possible person of interest—i.e. a false positive), averaging around 95% across all trials, although this is beginning to change …… One example is the use of risk assessments in criminal sentencing in the United States and parole hearings. Judges were presented with an algorithmically generated score intended to reflect the risk that a prisoner will repeat a crime. For the time period starting in 1920 and ending in 1970, the nationality of a criminals's father was a consideration in those risk assessment scores.Today, these scores are shared with judges in Arizona, Colorado, Delaware, Kentucky, Louisiana, Oklahoma, Virginia, Washington, and Wisconsin. An independent investigation by ProPublica found that the scores were inaccurate 80% of the time, and disproportionately skewed to suggest blacks to be at risk of relapse, 77% more often than whites.[8][8][8][8]Where does algorithmic bias come from? One is invited to read reviews on the subject — suffice it to say human bias can occur before data curation, during feature selection, and after post-processing. In essence, fixing algorithmic bias end-to-end is a procedural problem that both reflects existing biases in collection and amplifies it during curation. Approaches to resolving algorithmic bias are still nascent: one hopes to see more work done on this front in the next few decades.The next most commonly cited issue is that computer-aided decision-making in policy often needs to be able to operate under stringent data protection standards such as HIPAA or FERPA in the US.This raises a number of challenges for improving computer-aided decision making:Not only must data be sufficiently anonymised before sharing or inclusion in a predictive or classification model, the model should not be able to inadvertently reveal the identity of anyone. This is surprisingly hard — public datasets can still be used to identify people even after de-anonymisation.A number of techniques have emerged in tackling the problem of sufficient anonymization such as k-anonymity and differential privacy. These are both fascinating topics providing rich guarantees on privacy.Somewhat sadly, usage of k-anonymity and differential privacy can have some negative effects on the accuracy of certain prediction models[9][10]. The task of being able to improve prediction accuracy in the wake of anonymization efforts is case-specific and must be tackled accordingly.Not all computer-aided decision making techniques in policy need to impact human groups, by the way — for example, forest management has a history of using computer-aided decision making to predict woodstack availability and model ecosystem growth! This happy group is immune to the issues I have discussed above, so one needs only look to forestry management for an example of a happy marriage between computer aid and public policy.The diligent reader is recommended to consult Law and AI, an excellent blog on the state of issues considering artificial intelligence across all sectors, for more information on all of these issues.What Should We Do?There is almost no consensus on this subject. We are still very much in the dawn of a new era regulating the use of computer-aided or computerized decision making.Below, I offer some recommendations based on the complaints and suggestions I've sampled from reviews on the subject for different risk mitigation strategies.Government agencies need to make the details of their algorithms public.One of the most significant issues preventing research into improvement is that there is wide variability in just how much different agencies are willing to share about their techniques or data sources — we still don’t know just how widespread some systemic deficiencies are for this reason.Open data is essential to every meaningful investigation. Without a culture of open data and information sharing amongst government bodies and the public, we cannot expect improvements in algorithmic decision-making or assistance in decision-making from a regulatory standpoint.Some regulation and/or process certification is going to be necessary to result in reduced systemic bias.This is a popular and common view per opinion polls, though some suggestions in this regard — such as an FDA-style approval process or organization — are too heavyweight (in my opinion) for something as low-barrier-to-entry as software development. In researching this question, I’ve found recommendations for crowdsourced regulations or citizen input to ensure that claims of bias are voiced and handled responsibly — this kind of grassroots participation is going to be completely necessary for government-led usage of computerized/computer-aided decision-making.Beyond ensuring a tight feedback loop on bias management, some process guidelines for ensuring data quality and minimizing bias amplification should exist. Recent legal efforts, such as the Algorithmic Accountability Act, miss the mark in several ways, but engineering organizations such as the IEEE are currently working on standards that seek to minimize algorithmic bias[11]. These kind of standards can easily be adopted into government agencies and government subsidiaries such as in the area of public health, where regulation already exists and compliance is high.Operators should regularly audit themselves for bias.[12]It isn’t going to be enough to just have regulation — it’s going to be necessary for organizations to approach themselves in a way that holistically checks for bias in the decision-making process. How this’ll work or look like is still very much up in the air — incentivising or encouraging companies to publish bias impact statements routinely is one good way to do it.Computer-aided decision making needs to be easily adjustable and modifiable, so governments need to adopt modern software techniques.Technical debt is a huge problem in many organizations, and computer-aided decision making systems are no exception. Scully et. al. wrote a remarkable paper outlining the many anti-patterns that crop up in machine learning pipeline design[13]that should be required reading for every data engineer or DevOps practitioner — often, these anti-patterns can cause immense bottlenecks in publishing improvements over time beyond a certain point of entrenchment.For public policy programs, being able to respond quickly, efficiently and effectively is going to be key to handling impacts. Implementing modern software practices — breaking data and culture silos, adopting an ownership model for production services, reducing time to deploy and ship — are all going to be vital to accomplish that.This quadrifecta — regulation and compliance, active citizens, motivated agencies, easily adaptable technology — are, to my mind, the key ingredients to minimizing AI risks in public policy. Tight transparent feedback loops with quick implementation times and a corrective focus on structural causes are theoretically the safest kind of systems to operate and develop, and that attitude should be reflected in public policymaking.tl;dr I think AI should play a large role in public policy because of its many benefits, and the risks (mostly limited to algorithmic bias right now) should be mitigated via structural change in how policymakers manage the algorithm development lifecycle to include best practices, rapid iteration, an active citizen feedback loop, and a culture of self-analysis.Footnotes[1] 5 challenges for government adoption of AI[2] https://ash.harvard.edu/files/ash/files/artificial_intelligence_for_citizen_services.pdf[3] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/819055/Landscape_Summary_-_Bias_in_Algorithmic_Decision-Making.pdf[4] Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy - Kindle edition by Cathy O'Neil. Politics & Social Sciences Kindle eBooks @ Amazon.com.[5] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/819055/Landscape_Summary_-_Bias_in_Algorithmic_Decision-Making.pdf[5] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/819055/Landscape_Summary_-_Bias_in_Algorithmic_Decision-Making.pdf[5] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/819055/Landscape_Summary_-_Bias_in_Algorithmic_Decision-Making.pdf[5] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/819055/Landscape_Summary_-_Bias_in_Algorithmic_Decision-Making.pdf[6] weapons of math destruction[6] weapons of math destruction[6] weapons of math destruction[6] weapons of math destruction[7] How to be Human in the Age of the Machine: Amazon.co.uk: Hannah Fry: 9780857525246: Books[7] How to be Human in the Age of the Machine: Amazon.co.uk: Hannah Fry: 9780857525246: Books[7] How to be Human in the Age of the Machine: Amazon.co.uk: Hannah Fry: 9780857525246: Books[7] How to be Human in the Age of the Machine: Amazon.co.uk: Hannah Fry: 9780857525246: Books[8] Algorithmic bias | Wikiwand[8] Algorithmic bias | Wikiwand[8] Algorithmic bias | Wikiwand[8] Algorithmic bias | Wikiwand[9] https://thesai.org/Downloads/Volume5No11/Paper_26-A_Comparison_of_the_Effects_of_K-Anonymity_on_Machine_Learning_Algorithms.pdf[10] Privacy and data anonymization from a data scientist’s point of view[11] P7003 - Algorithmic Bias Considerations[12] Algorithmic bias detection and mitigation: Best practices and policies to reduce consumer harms[13] https://papers.nips.cc/paper/5656-hidden-technical-debt-in-machine-learning-systems.pdf

The question 'Does aluminum really cause the disease of Alzheimer's?' could be more accurately re-stated as 'Could exposure to aluminum cause Alzheimer's disease and could a causal link even be proven?' simply because aluminum is such a pervasive element in modern life, it's practically impossible to pinpoint frequency, duration and dosage of exposure at the individual level, let alone establish a cause-and-effect linkage between this one element, aluminum, on the one hand, and a complex, obviously multi-factorial disease such as Alzheimer's (AD) on the other hand. Simply, conclusive data's lacking. Rather epidemiological support of link between cumulative aluminum exposure and correlative risk of developing AD is confusing and inconclusive.This answerOutlines some basic facts about aluminum as it pertains to degree and variety of biological exposure.Summarizes conclusions from some recent meta-analyses, and systematic and umbrella reviews on the link between Aluminum and AD.Aluminum’s Pervasive in Human Foods, Daily Use Products and EnvironmentThird only to oxygen and silicon in its prevalence, aluminum is estimated the most abundant metal in the Earth's crust (1). Although there's as yet no evidence it is metabolized or even metabolizable by living things, its exponential industrial use from mid-20th century onwards has likewise exponentially increased human exposure to it. Though since the early 1970s, the pervasive soda can is a poster child of such use, it's far from the only one since aluminum has now become ubiquitous in not just human food and drinks but also in construction and the aircraft industry. After all, industrial aluminum use is pervasive, being used in everything from water treatment to generate drinking water to cosmetics, food, medical use and vaccines (see sequentially below from 1, 2),'The largest markets for aluminium metal and its alloys are in transportation, building and construction, packaging and in electrical equipment. Transportation uses are one of the fastest growing areas for aluminium use. Aluminium powders are used in pigments and paints, fuel additives, explosives and propellants. Aluminium oxides are used as food additives and in the manufacture of, for example, abrasives, refractories, ceramics, electrical insulators, catalysts, paper, spark plugs, light bulbs, artificial gems, alloys, glass and heat resistant fibres. Aluminium hydroxide is used widely in pharmaceutical and personal care products. Food related uses of aluminium compounds include preservatives, fillers, colouring agents, anti-caking agents, emulsifiers and baking powders; soybased infant formula can contain aluminium. Natural aluminium minerals especially bentonite and zeolite are used in water purification, sugar refining, brewing and paper industries.'Source of ingested aluminum is thus either natural, or through food and drug additives and daily use products, which constitute both consumer as well as occupational exposure (in the form of work in aluminum production and user industries).Natural: from its presence in foods grown in aluminum-containing soils. This can vary widely since aluminum compounds are more soluble in low pH soil, which is often the consequence of acid rain. This in turn increases aluminum content in plants animals and surface water (1, 3). Drinking water is another source since Flocculation - Wikipedia, a commonly used water treatment process, uses aluminum salts (1, 2), though the concentration is estimated low, <0.2mg/liter (4).Food and drug additives: With regard to aluminum in foods, starting sometime in the late 19th century and progressively more so since the mid-20th century, large-scale industrial food production the world over has enabled the abrupt and dramatic switch from a largely unprocessed to processed diet, the so-called 'Western' diet. Doing so has only increased aluminum bioavailability, especially human oral exposure. Such additives are found in dairy (milk, processed cheese, yogurt), staples (cereals, flours, grains), sweets (sugar, jams, jellies, baking sodas, powdered or crystalline dessert products (1, 2, 4). Use in food thus ranges from anticaking agents to buffers, emulsifying agents, firming agents, leavening agents, neutralizing agents and texturizers (2, see below from 4).While diet-based aluminum consumption is estimated to be ~10mg/day, over-the-counter drugs such as analgesics and antacids can increase this by several grams per day (5). Aluminum hydroxide for example is a common antacid ingredient that helps neutralize stomach acid while aluminum in antacids helps increase bioavailability of the active ingredient which is typically poorly soluble in the stomach's acidic environment (4).Daily use products: Cosmetics (perspirants, sunscreens, lotions, pigments), cookware, packaging are aluminum-containing daily use products. Aluminum's heat conductivity explains its pervasive presence in cookware. Leaching from cookware and packaging is estimated to add 2 to 4mg of aluminum per day in food, representing ~20% of daily aluminum intake (6, 7, 8).Estimated daily exposure between countries varies as much as 4-to 8-fold (see below from 4). This makes the task of estimating cumulative exposure in epidemiological studies attempting to discern a link between aluminum and AD or any other disease all the more challenging.Aluminum and Alzheimer's Disease (AD): Conclusions from Meta-analyses, & Systematic and Umbrella ReviewsAlzheimer's disease is classified as either the less frequent familial (1 to 5%) or the far more prevalent late-onset AD (LOAD), which is presumed the outcome of complex genetic, epigenetic and environmental interactions. Since hereditary factors fail to explain most AD cases, environmental factors have become prime research focus.Aluminum emerged as a candidate in the 1960s when a 1965 study observed neurofibrillary tangle (NFT)-like degeneration after directly injecting aluminum into rabbit brains (9), i.e., lesions similar but not identical to those considered a hallmark of AD. A 1973 study followed-up with the report of higher levels of Aluminum in post-mortem AD brain samples (10).Numerous mechanistic studies in the succeeding decades have proven inconclusive. After all, brain tissue degeneration in AD may simply make it better suited to accumulate metals such as aluminum. How to prove cause and effect? In addition, AD brain increase in aluminum levels isn't always accompanied by aluminum level increase in CSF (cerebrospinal fluid) with some studies suggesting it does and others not (11).Thus, establishing a conclusive link between increased human bioavailability of Aluminum and Alzheimer's disease remains elusive. For example, aluminum use in cosmetics such as antiperspirants became a focus of attention starting in the 1980s and yet, after decades of cumulative study, the FDA concluded (12),'The agency does not find the current evidence sufficient to conclude that aluminum from antiperspirant use results in Alzheimer’s disease.'Epidemiological studies trying to establish link between aluminum exposure through food and risk for AD are extremely complicated since it's present in such a wide variety of foods. Since AD's assumed to require years if not decades to develop, such studies would have to monitor aluminum exposure not just long-term but also at great depth, examining large study populations so that subset numbers remain large even after stratification, all amounting to a prohibitively expensive proposition.OTOH, epidemiological studies that tried to establish link, if any, between aluminum exposure through drinking water or occupational exposure and risk for AD have more promise since there's less ambiguity about the degree of daily and cumulative exposure. One 2016 meta-analysis of 8 such studies (4 drinking water, 4 occupational) on a total population of 10567 individuals found a significant association between aluminum exposure and risk for AD (13). Specifically, this study established chronic exposure to aluminum increased AD risk by 71%, where chronic exposure was defined as >100µg/liter aluminum in drinking water or its equivalent occupational exposure.A 2016 umbrella review of systematic reviews and meta-analyses (14), also concluded suggestive link between aluminum and AD. Other factors that also showed up as suggestive included factors as disparate as education, herepesviridae infection, low frequency electromagnetic fields and NSAIDs. OTOH, factors they concluded were highly suggestive included cancer, depression at any age, physical activity (high level being protective). However, the authors concluded cautiously (see below from 14, emphasis mine),'Several risk factors present substantial evidence for association with dementia and should be assessed as potential targets for interventions, but these associations may not necessarily be causal.'Thus, as of 2017, there is no consensus on whether and how aluminum exposure, specifically its bioavailability, influences AD risk. This is because numerous meta-analyses and systematic reviews that examined the totality of aluminum exposure, not just through drinking water or occupational exposure but also through cosmetics, over-the-counter drugs, processed food, vaccines, found the evidence to be inconclusive. For example, following a massive 2007 systematic review (1) that concluded there was little unambiguous evidence that aluminum exposure increased AD risk, a 2014 systematic review examined in great detail a total of 469 peer-reviewed studies, delving into not just exposure sources but also routes, amounts and potential toxicity to different organ systems. Evaluating the data by comparing to existing standards and guidelines for aluminum, it too concluded that (see below from 15, emphasis mine),'The results of the present review demonstrate that health risks posed by exposure to inorganic Al depend on its physical and chemical forms and that the response varies with route of administration, magnitude, duration and frequency of exposure. These results support previous conclusions that there is little evidence that exposure to metallic Al, the Al oxides or its salts increases risk for AD, genetic damage or cancer'Bibliography1. Krewski, Daniel, et al. "Human health risk assessment for aluminium, aluminium oxide, and aluminium hydroxide." Journal of Toxicology and Environmental Health, Part B 10.S1 (2007): 1-269. https://www.researchgate.net/profile/Robert_Yokel/publication/5764224_Human_Health_Risk_Assessment_for_Aluminium_Aluminium_Oxide_and_Aluminium_Hydroxide/links/0fcfd50187fc6a9eb4000000.pdf2. Yokel, Robert A. "Aluminum in food–the nature and contribution of food additives." (2012): 203. http://uknowledge.uky.edu/cgi/viewcontent.cgi?article=1030&context=ps_facpub3. http://www.who.int/ipcs/publications/jecfa/reports/trs940.pdf4. Vignal, C., P. Desreumaux, and M. Body-Malapel. "Gut: An underestimated target organ for Aluminum." Morphologie 100.329 (2016): 75-84. http://www.spritzer.com.my/wp-content/uploads/2016/09/Vignal-Morphologie-Gut-Al-2016.pdf5. Reinke, Claudia M., Jörg Breitkreutz, and Hans Leuenberger. "Aluminium in over-the-counter drugs." Drug Safety 26.14 (2003): 1011-1025.6. Jorhem, Lars, and Georg Haegglund. "Aluminium in foodstuffs and diets in Sweden." Zeitschrift für Lebensmitteluntersuchung und-Forschung A 194.1 (1992): 38-42.7. Wang, L., D. Z. Su, and Y. F. Wang. "Studies on the aluminium content in Chinese foods and the maximum permitted levels of aluminum in wheat flour products." Biomedical and environmental sciences: BES 7.1 (1994): 91-99.8. Al Juhaiman, Layla A. "Estimating Aluminum leaching from Aluminum cook wares in different meat extracts and milk." Journal of Saudi Chemical Society 14.1 (2010): 131-137. http://ac.els-cdn.com/S1319610309000751/1-s2.0-S1319610309000751-main.pdf?_tid=45465cbe-999a-11e7-968e-00000aab0f02&acdnat=1505427532_7874942c940894faf91bf7c71b0a5f059. Klatzo, Igor, Henryk Wiśniewski, and Eugene Streicher. "Experimental production of neurofibrillary degeneration: I. Light microscopic observations." Journal of Neuropathology & Experimental Neurology 24.2 (1965): 187-199.10. Crapper, D. R., S. S. Krishnan, and A. J. Dalton. "Brain aluminum distribution in Alzheimer's disease and experimental neurofibrillary degeneration." Science 180.4085 (1973): 511-513.11. Kapaki, Elisabeth N., et al. "Cerebrospinal fluid aluminum levels in Alzheimer's disease." Biological psychiatry 33.8 (1993): 679-681.12. FR Doc 03-1414013. Wang, Zengjin, et al. "Chronic exposure to aluminum and risk of Alzheimer’s disease: A meta-analysis." Neuroscience letters 610 (2016): 200-206. http://ac.els-cdn.com/S0304394015302512/1-s2.0-S0304394015302512-main.pdf?_tid=a04de4d8-99a9-11e7-90f5-00000aab0f02&acdnat=1505434127_91565c1d9cca3832c1bc29264a97ff4214. Bellou, Vanesa, et al. "Systematic evaluation of the associations between environmental risk factors and dementia: An umbrella review of systematic reviews and meta-analyses." Alzheimer's & Dementia 13.4 (2017): 406-418.15. Willhite, Calvin C., et al. "Systematic review of potential health risks posed by pharmaceutical, occupational and consumer exposures to metallic and nanoscale aluminum, aluminum oxides, aluminum hydroxide and its soluble salts." Critical reviews in toxicology 44.sup4 (2014): 1-80. https://www.researchgate.net/profile/Thomas_Wisniewski/publication/265790897_Systematic_review_of_potential_health_risks_posed_by_pharmaceutical_occupational_and_consumer_exposures_to_metallic_and_nanoscale_aluminum_aluminum_oxides_aluminum_hydroxide_and_its_soluble_salts/links/54ca2caf0cf298fd2627a86d.pdf

I would argue that the Moon is by far the best place to send humans in the near future for many reasons. But not for Helium 3. I’ll get to that in a minute.Inside look at one of the ideas for the ESA moon village, using 3D printing on the Moon for the radiation shielding. Image credit Foster + Partners / ESA. Their new director, Professor Johann-Dietrich Woerner is keen on taking us back to the Moon first, and has an exciting vision for a lunar village on the Moon as a multinational venture involving astronauts, Russian cosmonauts, maybe even Chinese taikonauts, and private space as well.It is great for science. I think we are bound to have science bases there similarly to the ones in Antarctica - and the ESA lunar village is a great way to get started on it.It is even, surprisingly, of astrobiological interest. With a near certainty it has meteorites in the permanently shaded lunar craters that have organics preserved for billions of years from Mars, early Earth and even Venus. It may even have intact fragments of ammonites and earlier Earth lifeforms. After a simulated impact on the Moon, fossil diatoms are still recognizable, and indeed the smallest ones are intact, complete fossils. There must be a lot of material from the Chicxulub impact on the Moon, which may also contain fragments of larger creatures such as the ubiquitous ammonites of Cretaceous seas. Perhaps the Moon will be one of the best places for fossil hunters in our solar system, outside of Earth.Artist's impression of Cretaceous period ammonites, courtesy of Encarta. The Chicxulub impact made these creatures extinct. It hit shallow tropical seas and the ejecta could have sent fragments of Cretaceous period sea creatures such as ammonites all the way to the Moon. Fragments in the cold polar regions may even have the organics preserved.The Moon must have meteorites from Mars too, for us to pick up, also from early Venus, from before its atmosphere became as thick as it is now. Early Venus might have had oceans and might have been as habitable as early Earth and Mars. However, most of that record is probably erased even if we get to explore the surface of Venus. It was resurfaced by volcanic processes around 300 hundred million years ago. It doesn't have continental drift, and the leading explanation for its young cratering record is that Venus may have had superplumes so large they resurfaced the entire planet. Its atmosphere is so thick that no meteorites get from Venus to Earth right now, and anyway, a modern meteorite or a sample return would tell us nothing about early Venus. That leaves any Venus meteorites on the Moon as our best, and maybe only way to find out about early Venus, including any biology from those times. That's especially so if they have the organics preserved. For more on this see Search for life from Mars, Venus, or the Earth - on the Moon in Meteorites! below.The Moon also must have collected organics from the comets and asteroids of the early solar system that bombarded Earth, so it can help give us an inventory of the organics that lead to kick starting life on Earth. For more about all this, see Charles Cockell's paper: Astrobiology—What Can We Do on the Moon?For more on this see the chapter Search for past life from Earth on the Moon in my Touch Mars? book.It is a great place for certain kinds of telescopes - infrared and very long wave radio. Also the optical telescopes we build there can be joined together to get long baseline optical interferometry because it is stable geologically with only minor tremors and small earthquakes. Also larger versions of the Arecibo telescope could eventually be built in the craters there and liquid mirror telescopes.The Moon itself is also far more interesting than we realized with Apollo. That includes the ice at the luanr poles - which is of great scientific interest, whether or not it is useful as a resource, and lunar caves probably over 100 km long and some may be kilometers wide. It is also known to be still active with argon vented, and some small areas that have been resurfaced recently, with the processes involved not clear yet - there are many mysteries to sort out. For more on this see the Moon science surprises chapter of my Case for Moon First book.And - it is differentiated with wide variations in mineral content - it is not at all the bland uniform “all the same” body that it is sometimes made out to be. There were many processes on the early Moon that concentrated and separated out minerals although many of them are different from the Earth and there are still processes going on depositing meteoritic iron and materials from the solar wind - and probably also ice from comets at the poles. The processes includeFractional crystallization - as a melt cools down, some minerals crystallize out at a higher temperature than others so form first. They then settle or float, so remove the chemical components that make them up from the mix, so changing its formula, leading to new crystals to form in a sequence.Gravitational settling, lower mass material floats to the top.Volcanic outgassing can concentrate materials such as iron, sulfur, chlorine, zinc, cadmium, gold, silver and lead.The processes that lead to volatiles condensing at the poles - which it seems can also concentrate silver tooProcesses unique to the Moon (perhaps electrostatic dust levitation may concentrate materials)?Volatiles brought in as part of the solar windAsteroid and micrometeorite impacts bring materials from asteroids to the lunar surface such as iron and possibly platinum group metals etc.For more on this see the Metals section of my book.ASSETS AND RESOURCESThen - it actually does turn out to be a great place for in situ resource use, and to set up human habitats. Also for export of materials to the Earth and LEO. I’d go so far as to say that if you can’t set up a human settlement on the Moon at the lunar poles, or in the lunar caves, you probably can’t do it anywhere off planet.Some of the main assets areSolar power 24/ 7 at the lunar poles.Dust is easily managed (no dust storms)Meters deep regolith good for radiation shielding from the solar storms and cosmic radiationThe regolith has nanphase iron in it which makes it possible to melt a surface layer using microwaves - useful for constructing a landing pad - and for a layer of glass to keep the dust away from the habitatIt is easy to land and take off. You can get to lunar orbit with only about half of your spaceship mass as rocket fuel. Also the landing is reversible until the last minute. Apollo 11 could have aborted back to orbit right up to just before the moment of touchdown (not possible on Mars or the Earth). Also it happens on a slow enough timescale to permit manually piloted landing as Neil Armstrong and Buzz Aldrin demonstrated (not possible on Mars)The hard vacuum is itself a resource - it may eventually be the prime place for manufacturing computer chips in the Earth- Moon system.Easy to make solar panels in situ using mainly in situ resources.It seems to have a fair bit of ice at the poles. Though the evidence is rather contradictory, the LCROSS impact data does suggest the presence of ice in pure crystal form. If so it may have larger deposits of pure ice. These are of scientific interest of course, but assuming there is a fair bit, can probably also be used for human settlersThe lunar caves may be the largest in the solar system that we know of - we don’t know how big they are but the Grail evidence suggests over 100 km long for the longer ones and tens to hundreds of meters wide but they may easily even be many kilometers in diameter and still be stable in the low lunar gravity.It has many useful metals - including calcium which in vacuum conditions is as good as copper, also aluminium.It has iron - pure iron, not the oxides - mixed in with the regolith and also probably as ore bodies from meteorite impacts. This is very variable in its platinum group and metal content. But amongst its thousands of craters it probably has at least a few that were made by particularly platinum rich meteorites. And - if we are lucky - well there are signs of magnetic anomalies from the impactor that formed the south pole Aitken basin. It may well be platinum rich - that’s Paul Spudis’ business case for the Moon - he thinks it may be a rich source of platinum. I don’t know for sure - but it seems likely that amongst all those craters, at least a few happen to have been made by platinum rich iron meteorites and have tons of ore just below the surfaceFor more on this see the The Moon is resource rich section of my book.It is also a great place to find out about human impacts on the solar system as we explore it. For instance the problem of trash. What do we do with the many tons of trash that build up around a settlement every year? What about footprints on the Moon, does the entire area around the base get covered in them? What about fuel for the rockets landing and taking off? Does the contamination of the lunar surface cause problems for science studies and if so, how do we minimize these effects? What about other organics and contamination and waste products from human bases, how does this impact on science and indeed on other human activities there? It is something we can study in much simpler situations than further afield. The Moon is huge and it doesn’t have a connected environoment, apart from ballistic motion of gases and electrostatic levitation of dust. What you do in one part of the Moon will for the most part only affect the nearest few square kilometers. So we can learn about these issues without the risk that the entire Moon gets contaminated in ways that we find a nuisance before we understand what we are doing in practice.For more on this see the Trash on the Moon - testing ground for planetary protection measures for a human base section of my bookLIFEBOATS ON THE MOONIt’s also the place in our solar system that I think has most potential for industrial exports to LEO or to Earth. I don’t know if it is possible even for the Moon to make this economic as a business. But if you can’t do it on the Moon you can’t do it anywhere probably. And even fi it is not economic as a business it could still be a very useful sideline for reducign costs. E.g. maybe your main objective is to make a lunar railway, but your byproducts include the platinum group elements and gold which you don’t need in situ and export to Earth to offset some of the costs of your habitats on the Moon.That’s partly because it is so easy of access from Earth. You can get there in a couple of days. This makes it far safer . The ISS has lifeboat spacecraft attached to it at any time that can take you back to Earth within hours. It is practical to add lifeboats to a lunar habitat to get you back to Earth within two days kept constantly stocked with food, fueled and ready to go. We can’t at present do this with voyages further away.The main problem there is life support. You can't test life support intended for a zero g environment on the ground, not properly. The ISS has had numerous life support issues which were only fixed due to resupply from Earth. See this list of some of them. None were immediately dangerous, and some were relatively minor but some of them would have been fatal on a timescale of months.If issues like that arose on a spacecraft like the ISS as far away as Mars many of those issues would have lead to the entire crew dying as they could never have got their spaceship back to Earth in time. The same would be true of other issues that arise over long timescales only, e.g. damage to equipment or to hull integrity from a micrometeorite - food gone off, harmful microbioal films build up, fire fire, or release of harmful chemicals, damaging vital equipment for life support, or essential provisions.The retired Canadian astronaut Chris Hadfield, former commander of the ISS, interviewed by New Scientist, put it like this in their article "We should live on the moon before a trip to Mars""I think ultimately we’ll be living on the moon for a generation before we get to Mars. If the world and the moon were threatened and the only way to preserve our species was to launch from Earth, we could go to Mars with yesterday’s technology, but we would probably kill just about everybody on the way.""It’s as if you and I were in Paris, paddling around in the Seine in little canoes saying, 'We’ve got boats, we’ve got paddles, let’s go to Australia!' Australia? We can barely cross the English Channel. We’re sort of in that boat in space exploration right now. A journey to Mars is conceivable but it’s still a lot further away than most people think."For more about this see the Lifeboats on the Moon section of my Touch Mars? book.TRANSPORT TO / FROM EARTHMost authors talk about how the ice at the poles of the Moon will make it easy to transport materials back to Earth, and especially ice itself as an export to LEO. I think this is at least a possibility but not yet proven.However there is one way we can achieve easy transport to / from Earth that is already worked out. I am a big fan of the Hoyt cislunar tether system. This doesn’t get anything like the attention it deserves. The compelte system which synchronizes a tehter so that its tip is stationary momentarily on the lunar surface for materials to be loaded and unloaded once per cycle would be rather an engineering marvel. But even if they don’t achieve that, to be able to hop from the lunar surface up a few meters, catch onto the tether and end up boosted with no extra fuel all the way to LEO and then get caught by another tether for a gentle re-entry - that’s a big game changer. The ingenious thing is that it is powered by the potential difference between the lunar surface and the Earth which is further down in the gravitational well. It works not unlike a syphon which lets you syphon water over the edge of a bath.It is cleverly designed and the details are important - but the basic idea is simple. Just two tethers and material flowing both ways, from Earth to the Moon and from the Moon back to Earth. So long as more material is moved from the Moon to Earth then that generates an excess of power much like the way a difference in water levels powers a waterwheel and that then can be used to keep both tethers spinning. Both tethers alternately are raised and lowered in orbit depending on whether they receive or send their loads to the other tether.The thing is - it is made of ordinary materials. You can build it with kevlar. It doesn’t need carbon nanotubes. Also its total mass is low. That means that if you are sending payloads to the Moon once a week for a year, you would already save on the total payload if you set up this tether system first (given that once it is set up, you no longer need any fuel at all to send payloads to the Moon).At least - that’s for the tether system itself. It assumes some infrastructure on the Moon of course to load and unload payloads at that end and you need to be mining the Moon - but it can just be dummy payloads of luanr regolith if needed, maybe used for radiation shielding in LEO. At any rate - whether they do it in the first few years or a decade or two later, it is not a distant future thing like the space elevator.For more about this see the Exporting materials from the Moon section of my book.WHY HELIUM 3 IS NOT LIKELY TO BE A MAJOR ASSET ON THE MOONThe Moon is a source for helium 3, deposited in the regolith by the solar wind, and some say that helium 3 will be of value for fusion power in the future because it is not radioactive and doesn't produce radioactive waste products. If so, small amounts of helium 3 from the Moon could be worth a lot on Earth and be a useful commodity to export. Apollo 17's Harrison Schmidt is a keen advocate of helium 3 mining on at a reasonable rate at a reasonable rate the Moon.However, we don't yet have fusion power plants at all, and one able to use helium 3 is a tougher challenge. Frank Close wrote an article in 2007 describing this idea as "moonshine" saying it wouldn't work anyway. Frank Close says that in a deuterium - helium 3 tokamak, at normal temperatures for a tokamak, the deuterium helium 3 reaction proceeds so slowly that the deuterium would instead fuse with itself producing tritium and then fuse with the tritium (the original article is here, but it's behind a paywall). For a critical discussion see also the Space Review article The helium-3 incantationSee also Mining the Moon by Mark Williams Pontin. If you can use much higher temperatures, six times the temperature at the centre of the sun by some calculations, the helium 3 will fuse at a reasonable rate, but these are temperatures way beyond what is practical in a tokamak at present. The reason such high temperatures are needed for a tokamak is because the plasma is in thermal equilibrium and has a maxwellian distribution which means that to achieve a few particles at very high temperatures you have to heat up a lot of particles to lower temperatures to fill up the maxwellian distribution so that just a few will react. This is potentially feasible for the lower temperatures of DT but not feasible for the higher temperatures of 3He 3He.However if you use electrostatic confinement, a bit like a spherical cathode ray tube with the fusion happening at the center where the negatively charged "virtual cathode" is, then the particles are all at the same high energy and the result is much more feasible with lower power requirements. This is the approach of Gerald Kulcinsky who achieves helium 3 fusion in a reactor 10 cm in diameter. However though it does produce power, it produces only one milliwatt of power for each kW of power input so is a long way from break even at present.Gerald Kulcinski who has developed a small demonstration electrostatic 3He 3He reactor 10 cm in diameter. It is far from break-even at present, producing 1 milliwatt of power output for each kilowatt of input. See A fascinating hour with Gerald KulcinskiPerhaps this line of development will come to something. Perhaps one way or another we will achieve helium 3 fusion as the enthusiasts for helium 3 mining on the Moon hope. However it is early days yet, and we can't yet depend on this based on a future technology that doesn't exist yet.However even if we do achieve helium 3 fusion, it might not be such a game changer for the lunar economy as you might think. Crawford says (page 25)" that to supply all of our energy from Helium 3 would mean mining 5,000 square kilometers a year on the Moon, which seems ambitious (and would mean the whole Moon would only last 200 years). So, even if we develop Helium 3 based fusion, and it turns out to be a valuable export, it's probably not going to be a major part of the energy mix.Even more telling, he also calculates that covering a given area of the Moon with solar panels would generate as much energy in 7 years as you'd get from extracting all the Helium 3 from that region to a depth of three meters.Also - there are many other ideas being developed for nuclear fusion, such as laser fusion, and the Polywell which has the same advantage that no significant radiation is produced when it uses fusion of boron and hydrogen. I think it is far too soon to know whether or not the helium 3 on the Moon will be an asset in the future when we achieve nuclear fusion power. For a summary, see ESA: Helium-3 mining on the lunar surface.This doesn't mean that there is no point in helium 3 mining however. As Crawford suggests (page 26)", Helium 3 is useful for other things, not just for fusion power. It's used for cryogenics, neutron detection, and MRI scanners, amongst other applications, so some Helium 3 from the Moon could be a valuable export right away, even if it doesn't scale up to the huge quantities you'd need for Helium 3 based power generation on Earth. You'd get it automatically as a byproduct while extracting the more abundant volatiles from the solar wind in the regolith, so it might well be a useful side-line to help support lunar manufacturing economically as part of the mix along with everything else.(this is from the Helium 3 section of my book)CASE FOR MOON FIRST AND THE VALUE OF ASTROBIOLOGY ON MARSFor more about the case for the Moon and the many benefits of sending humans there see my Case For Moon FirstI wrote that book originally because I care deeply about the science value of Mars for astrobiology. It concerns me a lot that NASA are contemplating sending humans there before they can assess what effect our microbes could have on any native Mars life - especially in the case of a crash of a human occupied ship on the Mars surface. They agree that so long as there are habitats there for our microbes to survive in, that the process of introducing Earth microbes to Mars would become irreversible as soon as humans land on the planet - never mind crash there. For more on this see NASA's plan for safe zones - based on finding Mars life easily in my Touch Mars?A crash could spread the debris with microbes over hundreds of kilometers of the Mars surface, by analogy with the Columbia space shuttle crash on re-entry to Earth. For more on this see the Elon Musk's fun but dangerous trip to Mars section of my Touch Mars? bookYou might wonder, who cares, the life there is probably only microbial? Or at most perhaps lichens? Well - it could easily be microbes from an earlier era of evolution, for instance RNA only microbes. After all that is one hypothesis for the structures in ALH84001, and whether or not those structures are early living cells, it is a possibility for early life on Mars. Then, it is possible that Mars life has hardly evolved since then. Or it could be highly evolved. Nobody seems to know if Martian conditions would favour vastly more rapid evolution than on Earth, or evolution that ran to a standstill at a very early stage of evolution and has barely changed since then, or something roughly parallel to evolution on Earth, so that it has reached about the same stage as us.Anyway - if it is an early form of life like that, it could be amazingly precious for science, fill in a vast gap in our understanding of how life evolved - and yet be very vulnerable.For more on this see the What if Mars has really tiny cells - like the structures in the Mars meteorite ALH84001? section of Touch Mars?And it need not have been exposed to Earth life. People talk a lot about panspermia - transfer of lifeon meteorites from Mars to Earth or Earth to Mars - but it has never been establisehed in either direction yet. The obstacles are formidable even though life is also far hardier than previously thought. Especially from Earth to Mars, the shock of ejection at a speed fast enough to LEAVE the atmosphere at the Earth escape velocity is huge. Also at such speed as that - it has to travel through the atmosphere at such speed that it is already a fireball in the first few hundred meters as it leaves the surface of the Earth - otherwise it is far too slow to leave Earth’s gravitational field. Any photosynthetic life on the surface would be destroyed by the fireball. It then has the vacuum of space, cold, cosmic radiation, solar storms and then has to find a habitat on Mars when it gets there, from the interior of a rock (remember the exterior was roasted already as it left Earth’s atmosphere) and if in modern Mars in an environoment with almost no running water even when during its occasional periods of somewhat thicker atmosphere.It may never have happened in that direction for billions of years and it is also possible that life from Earth never got to Mars (depends how robust Early Earth life was).For more on ths see the What about Zubrin's meteorites argument? section of Touch Mars?HOW THE MOON TURNS OUT TO BE BETTER THAN MARS FOR HUMAN SETTLEMENTSAnyway - so that was my original motivation for writing this book. I thought - if we can find that the Moon is good enough to at least delay the colonization of Mars attempt - it gives us some breathing space and time to assess the impact of humans on Mars before they actually get there.But - as I worked on the book I was amazed to find that not only is the Moon a good second fiddle to Mars - it is actually a far better place for an attempt at human settlement. In one comparision after another.Meanwhile, many of the ideas for Mars colonization can be used on the Moon - though not all. If you look at the Moon with “Mars colonization” preconceptions you will keep thinking about how it is not as good as Mars.That’s because solutions devised for Mars don’t necessarily all of them work on the Moon - especially e.g. using the CO2 to make fuel from hydrogen. But then - maybe you don’t need to do that on the Moon. It takes less fuel to get back to Earth anyway - and you can use the Hoyt cislunar tether sytem which doesn’t work on Mars - or use water mined at the lunar poles. You also have abundant solar power 24/7 at least at the lunar poles. And the pure iron on the lunar surface and the hard vacuum are major assets compared to Mars, also the higher levels of solar power are never blocked out by Mars dust storms. Then you have the easy way you can clear areas of its surface from dust by glazing it. None of that will work on the Mars, because of its atmosphere and the winds blowing the dust around.So those are all disadvantages of the Mars atmosphere. While CO2 is not at all needed for greenhouses - for space habitats if it is a perfect closed system then human breathing + composted or burnt plant wastes produce all the CO2 you need forthe next crop and ifyou have to import any food then you get a CO2 excess that has to be scrubbed. And the Mars close to 24 hour day is much less of an asset than you’d think if you realize it means it gets bitterly cold at night, so cold that the CO2 freezes out as dry ice at night even in the tropics for many days of the year.Meanwhile things like the “suitport” spacesuits to keep out dust are useful in both places.For more on this Mars or Moon spectacles and the old woman young woman illusionTERRAFORMING AS A PLANET CENTRIC APPROACH THAT WOULD TAKE THOUSANDS OF YEARSAs for the idea of terraforming Mars - the Kim Stanley Robinson books are science fiction with a lot of fudging of figures to fit the story into a timescale of a few generations. In reality the optimistic projection of the Mars society is for a thousand years to get to the point where you have a pure carbon dioxide atmosphere with trees and humans using air breathers - based on very optimistic projections of amounts of dry ice on the surface. It could be tens to hundreds of thousands of years after that to reach a breathable atmosphere (you have to extract so much carbon from the atmosphere that you create a meters thick layer of organics over the entire surface of the planet, in the low light levels of Mars) and that’s with vast use of megatechnology. When you think also of what you could do with that level of megatechnology in decades rather than centuries, for other space activities - well -it becomes more of a fantasy future than a likely near future reality. Also with much to go wrong and the prospect of making the planet less rather than more habitable for humans in the worst cases.Why use so much mega technology to create an atmosphere tens of kilometers thick when you only need the bottom couple of meters, and to cover a planet with seas filling up bone dry desert sand with water to depths of hundreds of meters (perhaps by bombarding it with comets) when you just need a few inches depth of water for aquaponics or growing plants in normal soil, or perhaps meters depth of soil for trees? Why commit to making an entire planet into a human habitat when you only need a few square meters to square kilometers to start with? And when it comes to the area available for colonization - well it would take a long time to exhaust even the areas available on the Moon and in lunar caves. But if we do eventually have billions and then trillions in space - you get much more surface area for far less resources invested if you use the materials from the asteroid belt. Enough to build habitats sufficient to have a surface area total of a thousand times that of the land area of Earth just from the asteroid belt.If we do ever engage in terraforming, it seems like something we’d do at a later stage, not one of the first things we attempt in space. Learn our lessons in smaller scale habitats. After all there was plenty to go wrong even with Biosphere II never mind trying a similar project for an entire planet in one go.And we can also try paraterraforming, starting with the Moon - covering large areas of the surface with habitats - or indeed, building large habitats in the lunar caves. It’s actually much easier to live in the lunar caves than many realize - the caves have a stable temperature, shieled from cosmic radiation, may be vast inside, already “built” as a structure to live in. The main disadvantage is the 14 day lunar night. However humans don’t need an acre or even a thousand square foot of agriculture to grow enough food for one person, but only 30 square metres. Also nowadays we can use LED lights optimized for growing plants at only 100 watts pwer square meter. With these sorts of figures it is feasible to deal with the lunar night using only battery storage and other energy storage solutions and solar power. For that matter, the Russians in their early BIOS-3 experiments in the 1970s found out that there are many plants that can crop fine with 14 days of darkness every month so long as the roots are cooled to just a few degrees above zero during the long lunar night.For more on this, see my An astronaut gardener on the Moon - summits of sunlight and vast lunar caves in low gravity in MOON FIRST Why Humans on Mars Right Now Are Bad for Science .And for lunar paraterraforming and even terraforming, see Terraforming or paraterraformingBUT WHAT ABOUT ZUBRIN?Zubrin makes what seem to be knock down arguments to his supporters - that we need to go to Mars right now and that there is no need to protect Earth or Mars in the process. However, convincing as he may sound, it’s important to know that when it comes to planetary protection then almost nobody is in agreement with him;. He is not the decision maker.For the background on planetary protection going back to Lederberg in 1957, and how Zubrin’s arguments seem within that context, see Planetary protection - researches by Sagan and Lederberg onwards - and Zubrin's argumentsMy main concern is with NASA, and only indirectly his effect on them. He can’t get them to ignore planetary protection altogether as he is not supported in that by the planetary protection department or COSPAR - the international group who discuss planetary protection or anyone. He can get them to put the Mars mission on the top of the agenda but I think with Trump his influence on US space policy is waning somewhat.My hope is that as we start to explore the Moon and space geeks get excited and fired up by the new things we are finding out there - and faced with the practicality of actually being able to go into space and visit the Moon and do cool things there right nowm that the Mars fever will calm down.SpaceX is the only commercial company in support of Mars first - Blue Origins and everyone else is in the New Space business is focused on LEO and the Moon.Under Obama, the US was the only country focused on Mars first and is now changing itts stance towards Moon first. The rest of the world including Europe, China and Russia have always been keen on returning to the Moon first.Also just about all astronauts agree that we should go to the Moon first. Even Buzz Aldrin, one of the keenest advocates for humans to Mars, thinks we should go to the Moon first - he was misquoted by Obama as his “been there done that” was meant as a joke as he made clear in his autobiography published soon after.It just makes more sense, apart from planetary protection issues. Iam not saying that they agree on the need to find out more about Mars before landing humans there. Their focus on going to the Moon first is mainly for safety and practical reasons and many of them are thinking in terms of a follow on mission to Mars a few years later. But it can give us a breathing space to both find out more and raise public awareneess of the need for planetary protection and get it more discussed so that we make a wise informmed decisions for Mars. It will also help take the pressure off from those who are desparate for humans to go further away than LEO and currently see Mars as the best way to do that.It will be a few years before we have humans on the Moon too probably, at least for Russia, the US and EU because of our commitment to the ISS. It is very expensive to maintain and there’s no suggestion at the moment of taking on a second big project for humans in space at the same time. Though I’ll be sad to see the end of thre ISS when it de-orbits probably some time in late 2020s, the funding will then go into these Moon missions and the ESA village will be international and also involve private / public partnerships and collaborations. It makes it far more feasible than the ISS was for development and expansion plus there is so much we’d be learning about the Moon.China or some other country could surprise us with astronauts to the Moon beforoe then. But there is no sign of that at present and indeed China are keen to join ESA in the lunar village concept.Meanwhile we have many robotic missions to look forward to .This year we have China and India both sending government missions and four teams for the lunar X Prize by end of March and astrobiotic, who were the leadingcontender for the X Prize until they pulled out probably some time this year or next. We will soon be at the point where we have multiple robotic missions to the Moon every year - which we can do because it is cheaper and easier to get to the Moon than to Mars - and that will make a big difference with streaming HD video as our robots explore the caves, surface and lunar poles and then find the best location for a human base and start prepairing for it. ?And lots of science discoveries streaming back.This is not going to detract from Mars robotic exploration but is additional to it.That’s how I see things unfolding. I am much more hopeful about a planetary protection friendly future exploration of space than I was, say, two years ago. And I think people will be surprised at how many minor but niggling issues there are on the Moon such as the trash, rocket exhaust and contamination of the site around the base for human investigations. So leading them to take more care as we explore further afield rather than just landing on Mars and hitting all those issues for the first time there,THE MOON AS AN ASTEROID CATCHERWhatever you have in Near Earth Asteroids you also have on the Moon. Asteroids have impacted on the moon for billions of years so it is like an asteroid collecting station - smaller material mixed in the regolith, metallic ore bodies buried in craters probably, iron core of the impactor that made the Aitken basin splashed out over the rim of it possibly by the magnetic evidence, and there is ice at the lunar poles.The main advantage of the Moon is that it is so close and easy of access to Earth while the asteroids which require the lowest delta v to get to from Earth are also the ones that phase in and out with our orbit most slowly, typically easisest to access once a decade.Perhaps the most practical idea for the most easily accessible asteroids with low relative delta v is to grab a chunk and bring it back to the Earth Moon system for further reprocessing.But the Moon has done that already by catching them in the impact craters. I think we need to explore both and see what happens but if I were a betting man my money would be on the Moon to be the lions share of the economy if we do get a space economy in the near future exporting to Earth.The asteroids do have advantages especially for autonomous mining, the spin can be used to deliver materials back to Earth since many are spinning, the low gravity may be an advantage in some ways, the idea to bag an asteroid with heated carbon monoxide - then heated by the sun 24/ 7, solar power continuouslyu which you only get at the poles on the Moon. But the latency for controlling operations from the Earth may be an issue plus distance to go for servicing. Hard to say as we may get rapid progress in autonomous operation and in how easy it is to get to the asteroids - but both those would benefit the Moon too.SWIMMING ON THE MOON AND RUNNING ON WATERI’ll just finish with some fun thoughts. Did you know that in the luanr surface - if we ever have habitats there with large amounts of water - humans could run on the water like the Basilisk (“Jesus”) lizard?Running on water - a possible future lunar Olympics sport :). Four out of six subjects were able to run on water in simulated lunar gravity using small rigid fins - similarly to the way Basilisk lizards can run across rivers. So running on water could be a future lunar sport (in an air filled habitat of course) paper hereAnd a strong swimmer could easily leap out of the water like a dolphin? See Hilarious XKCD about lunar swimming :). You'll be amazed by what humans could do on the Moon if or when we have our first swimming pool there. Lunar Swimming on XKCDThat might be somethign we can explore sooner than you expect. After all a lunar habitat will need water stores for both living there and perhaps for rocket fuel too. Why not have some of it in the form of a swimming pool? And especially if the water can be mined locally on the Moon then it might be quite easy (comparatively) to set up a lunar swimming pool.CASE FOR MOON FIRST BOOK AND OUR FACEBOOK GROUPYou can find the book here: Case For Moon First - has more on nearly all the things I’ve covered here (though not on the dolphin leaps and running on water, I have yet to add a section about that).Also my Touch Mars? Europa? Enceladus? Or a Tale of Missteps? and MOON FIRST Why Humans on Mars Right Now Are Bad for Science,Also see our facebook group hereCase for Moon for Humans - Open Ended with Planetary Protection at its CoreIt is for anyone interested in a Moon first approach - not necessary at all that you agree with my views on everything or anything :). There are still so few places for discussion of a Moon first prespective, at least compared with Mars colonization discusisons.