Snapshot: Important

  • Similar
    Today (!(2018–09–30 13:30)) I found this https://patrickcollison.com/questions, that is in a similar style. It was quite enjoyable to read. I think, I have asked questions of similar depth yet, because I have far less reach have not been as able to generate interesting conversation. I am certain I am not alone in this. I think we can do a better job at integrating parallel threads. My own personal wish is to be able to engage in deep and philosophical conversations with many people regardless of standing. You just have to be able to push us all in interesting directions. Hmm, I would like to email this Patrick Collison person but my confidence is much too low for that. I am especially interested in the successor of the book, general person programming and the future of research but assume it is a dull thing for most. It would be exciting to be able to hang out in such a community.
  • Water
  • Scaling sewage treatment with dense populations
  • Drainage and floods
  • Potable
  • Food
  • phosphorous
  • Meat alternatives
  • Factory Farming
  • Farming
  • increase efficiency
  • nitrogen run off
  • crop scheduling
  • Ecology
    It’s important to stop associating environmentalism solely with pollution and $$\mathrm{CO_2}$$. The maintainance of ecosystems and biodiversity is important. Even from a human centric point of view, more biological diversity is better due to dynamical robustness, dynamical systems which the planet is an interlinking set of.
  • co-existance with nature, maintaining biodiversity
  • over fishing
  • enroaching upon habitats — don’t punish animals for our growth, build with them in mind
  • AI Spirits?
    Humans (with some indigenous exceptions) have not shown themselves as good stewards of nature, often pulling things in the worst possible way: eg not good for the environment long term and not good for people (such as protesting nuclear plants). Meanwhile, actually damaging things like oil pipelines more often than not, get completed because the true corporate citizens demand so. Can an AI be a strong representative for rivers and forests, without either enroaching upon human rights? What would a river choose if it could choose?
  • Reforesting
  • Sprawl reduction
  • Honorless Hunting (trophy hunting)
    self-limiting to require use (eg food, local materials) might be a good compromise
  • No hunting endagered, protected and intelligent species
    Why prioritize Intelligence? Because intelligence controls the capacity to suffer, the duration and richness of trauma and the ability to anticipate and imagine variations of suffered events.
  • Pollution
  • Plastics
    plastics are damaging to the environment and are overused. can we find an alternative in all cases? Better accounting of the full lifetime of a product and eliminating the ability for corporations to externalize the costs will be important. The problem of plastics is not new; as we see with plastics, lead, CFCs, nitrogen and gasoline. Worse, manufacturers sometimes knew the true risks and yet continued. A more responsible way of handling the full lifecycle of materials is important. A simple ban of a single material isn’t a general enough solution.
  • Sequestration
  • Ocean Acidification
    Like the maintainance of ecosystems, habitat loss and somewhat, the effects of pollution; the problem of ocean acidification gets little attention. There is ongoing research into extracting CO2 from oceans. Large or mobile nuclear powered platforms could power such extractors. What are better reactions, materials and processes to do this effectively at scale? How to find them better? https://pubs.acs.org/doi/pdf/10.1021/es0701816
  • ocean polluting waste
  • Energy
  • long distance transmission
    not only a tech problem
  • safer compact nuclear
  • breeder reactor designs
  • storage
  • ships, planes, shipping
  • fast transitioning off polluting sources
  • enough to increase wealth for all
  • Planning
    In addition to corruption, one aspect that makes projects fail is that planning is difficult. Especially in developing nations, where many processes are still by pencil and paper; memory, coordination and planning are lacks which make the large scale projects needed for progress less likely to succeed.
  • Central Planning
  • Free Markets
    Free markets have incentivized global cooperation but at large human and environmental costs. Additionally, markets are not as good at long horizon planning partly because those are impossible to predict and markets are essentially aggregated predictors to effect allocation. How can better be done? What other consensus, concession, coordination and distribution mechanisms are scalable?
  • Prediction Market
  • Patreon, kickstarter and GoFundme show a different disaggregated way of funding is possible but are still sensitive to popularity and preferential attachment. Not everything important will get funded. Ponzi schemes and blockchain mania are proof of principle that a great deal of money can be gathered outside traditional finance. How to take this further and do it properly, to spend on important schemes in a sustainable manner?
  • When people are no longer paid, money will make less sense. What will replace it? One thing is moving to economies where the primacy of information, energy and attention as the true flows are recognized.
    Video game tokens as a subset of attention tokens?
  • City infrastructure
  • maintainance
  • Housing
  • predictable Timely Non-car based Transportation
  • subways
  • light rail
  • electric busses
  • regreening
  • biomemetics
  • waste mining
  • input/outputs
  • Industrial symbiosis and ecology
  • fast safe maintainable/sustainable building
  • Full Systems Thinking
    Currently, most approaches to problem sovling only consider solutions in isolation. More needs to be done on full-scale modeling to at least attempt to anticipate some consequences of solutions. For example, what are the biological, ecosystem and other down-stream effects of some solutions to ocean acidification?
  • Resources
  • Things are thrown out or burnt, need to create more closed cycles in resource use
  • Accounting for and harmony with ecosystems even while extracting
  • GDP does not track efficiency gains and epheremalization. Chasing after it with exponential growth indicators is not sustainable. What positive alternatives can be introduced?
  • Demos
  • Preferential Attachment, Yule Process
    Many inequalities of wealth, social, access and oppurtunities etc are a somewhat an inevitable consequence of the underlying dynamics of how we choose to run society. Even without any nefarious actors, we can expect convergence towards highly centralized, winnings to winners type outcomes. Centrality of winnings is inevitable in any Yule type Process but society need not run in a manner abstractable to a Yule type Process. How to do this correctly? One important aspect of this is human limitations regarding attention and computation. Heuristics that do everything but actually analyizing the content at hand will favor dynamics which lead to such centralization.
  • The conceptually related Matthew Effect — credit to the most famous — is a problem of information flow in credit assignment, with a second order effect: persitent under-representation where people will naturally want to be associated with existing prestige.
  • Hiearchies
    A tree structured information aggregation mechanism where every node has the same computational power (more or less), will suffer from lossy compression especially if the coarse grainings are suboptimal. What are other data structures, compression mechanisms and credit assignment which will help information flow and better decision making?
  • De-escelation and Tolerance for Errors
    Many problems occur because people become backed into a corner and saving face is more important than admitting fault. People then search for and iterate down a path that commits and believes them to this explanation beyond all sensible levels. Perhaps ridding the culture of blame might help in some instances. How to identify and de-escalate?
  • Culture of Blaming
    If no one can admit responsibility for an error early, due to large costs placed by society, they might cover and ignore a problem till it gets too large. At which point many otherwise good people will confabulate an unlikely interpretation of reality. Leading to situations like ill-maintained infrastructure poisoning citizens. The notion of accepting responsibility might be too crude as there are many interactions at various scale and there is an arbitrary decision in where to stop. Identify the problem and take steps to cut its re-occurence probability.
  • Voting
  • AI reps
    Similar to above on Nature AI Reps, can an Optimizer combine your preferences in a way that takes into account long term and systemic effects? Why would you trust it?
  • Liquid Democracy
    finding the most knowledgeable person on a topic that most shares your recommendations on media as a proxy for shared beliefs.
  • Exiting corruption
  • Trust
  • equilibrium is not pro-social
  • coordination/cooperation
    global scale cooperation holds back all kinds of important progress
  • Without trust and cooperation then complex projects cannot be undergone. Without coordination and trust it is difficult to compose and build complexity by building on existing substrates.
  • AI for development. ~Optimal planning trajectories
    Observe Tesla, where it seems the advantages of a fresh start and leap frogging do not counteract a lack of tacit knowledge and experience. Developing nations have a similar problem, in addition to negative equilibriums and debt. What are the steps of sets of improvements that optimally can be taken to significantly raise the quality of life for the average citizen in a wholesome manner? China managed but at great cost to the environment (and alas, by placing a lot of restrictions to simplify the decision space). Singapore did but these things might really be scale dependent. Most important, these nations lacked the disequlibrium and had more lenient debt treatment
  • it is possible to build things that promote toxic behavior and bring out the worst in people, can we build things that do the opposite? Where the equilbrium is prosocial at the worst?
  • When lives have low value such as from ubiquitous poverty, then violence is in turn also more likely. Which creates a feedback loop depressing the value of lives, in turn making violence less costly. In this environment, long term planning, coordination, cooperation and trust are very hard to take hold. This seems like an equilibrium. How to exit? This is an important aspect of corruption and general poverty that gets less attention than other more visible problems that are actually less central (gangs, warlords etc which form to fill the vacuum)
  • Until Equality is Respected Everywhere: AI that takes your age, gender,orientation, sex, race, etc. and tells you whether it is safe to travel somewhere
  • Communities
  • As the need for work reduces, people will need alternative ways of finding meaning
  • (Distant Future) People assume robots will only be producers but what if there were consumer robots of all types which lent computing power in return for resource then used on products they are fans of? They could even help the human creator improve in skill or volunteer for polish.
  • Virtual worlds
  • Art, Craft, Fashion, Cooking, Music,Performance, and more
    How to share this so that those at all levels can find their passion, grow and share with others no matter what it the area?
  • Loneliness
    Paradoxical seeming, as connectivity grows so do reports of loneliness (and superficiality). The centralizing and global/non-local aspect means attention becomes concentrated and only extreme personalities are most selected for. How to build more wholesome, inclusive yet non-isolated communities? Improve interactions? Perhapse debate nodes, sessions, events across ideological oposed nodes?
  • The current system is what can you do for me? Or, how does supporting you reflect on me? It should be: how can I help lift you given my personal constraints on time? Done recursively, it ends up at a good equilibrium. Another take: This is interesting so I will boost it, I don’t care about it not being absolutely correct because creative consistency is more important than exact concordance with common the view.
  • How to promote and teach reaching for two-way interactions where one tries to accomodate while the other tries for understanding?
  • Current policies dehumanize non-citizens
  • Perspective Sharing tools
    Can games help? I have had eggs thrown at me (luckily dodged all) while being called a monkey. It still hurt (maybe even more) when slurs were thrown against my cat and lizard person — and we were not allowed to shop because of their species
  • Answer Boards
    In Vernor Vinge’s Rainbow’s End, answer boards allowed distributed and collective just in time teams to solve challenging problems. Imagine a world where after some area was hit by flood, it would be possible to pool resources beyond just attention and lamenting. Where even the smallest contribution leveragable was findable. Other example: research groups springing up around a rare disease including very cheap and accesible synthesis, genetic, computational and research/hypothesis generation work.
  • Better means of Educating both Collective and Individual
  • Intelligence Amplification to go with and beside better Educational Tools
  • Collective Intelligence
  • Open Research MOOCs
    The ability to learn up to undergraduate university level topics is available. While these distributed learning programs do not replace traditional universities they are already decent option lacking any other alternative (with more work to be done). Distributed Companies also exist, with tools enabling remote work continue to improve rapidly. Yet for the situation of workable Distributed Graduate school and research groups, no such uplifting examples exist. That solutions exist for both undergrad and work suggests that this should be possible at some capacity (eg not all scientific work requires local access to expensive machinery). Indeed we wish to move towards a world where such services need not be the exclusive purview of a lucky few. More locations, more specialized labs performing services might go some way towards filling this gap. Furthermore, one can note massive collaborations between universities have been done and one assumes a certain expectation of standard and quality by being restrictive. Yet arguably many Open source projects of no more triviality manage somehow. Beyond self-sorting, the difficulty might be in proper skill adjusted delegation to which being closed and exclusionary is a hacky solution.
  • https://early.khanacademy.org/open-ended/
  • Boosting collaboration instead of the toxic signalling cascades of attention based markets.
    In a large (global) network, preferential attachment will promote behaviors where people calculate personal benefit, reducing oppurtunities for collaboration; consequently, people become more likely to display trollish behavior, being more memetically fit, and where preening cleverness is more likely than patience and helpfulness
  • better distributed cognition
  • How to engender more positive sum scenarios?
  • Boosting cooperation and fair concessions
  • Freedom
  • Cyclic Core Beliefs vs Right to Exist as Oneself
    Cyclic core beliefs are beliefs that are core but can’t be interrogated. They are I believe, a result of the fact that each human participates as part of a collective and the ability to conditionally update locally in a consistent manner is new, a result of incidental generalizability of a whole host of interacting systems, including one that evolved to aggregate as part of a distributed intelligence.
  • Generally, they are beliefs that are shared by many people and simply accepted. In smaller, slower moving society, the convergence speed relative to the environment might have been fine. The noise and disaffected would also have been ignored. But as population has grown, the absolute number of disaffected has too grown. Meanwhile, high connectivity has not interacted well with such old modes of reasoning. Continuing such Core Beliefs — which are an important part of so many identities, though they might infringe or interact oppressively on the freedom of some — creates an environment where it is more acceptible and probable, to attack that other’s agency and being. Speed and high connectivity means messages amplifying and confirming beliefs against certain types of person can infect a group and make them act in ways that they see as rightfully cleansing. The conversations around, in typical human limits (seeming lack of being able to take more than one side at a time) fails to recognize such interactions, remaining polarized at large cost. What is a way to communicate with this newly augmented but highly-ill adapted system that seeks to oppress so many? Competing beliefs might evolve according to a replicator dynamic, taken across human minds. Not all equilibriums will have good global welfare values.
  • Concentration of Intelligence
    It might not seem obvious but centralizing AI to a few companies results in a large unbalancing of power. Renting AI services for one’s use is in effect renting one’s augmented intelligence. If we accpet intelligence is not limited to be within one’s brain, a useful analogy would be: How happy would you be if having only a hippocampus, you rented your cortex from some corporation? A corporation whose only limit against exploitation is the public outcry that would ensue (meanwhile, technology proceeds, accruing asymetric gains to said corporations and other large bodies at a rapid clip), how happy would you be with such a situation? As things increase in sophistication, dependence and connectedness, a malicious overseer could, in essence, rewrite part of your brain and have a handy form of mind-control.
  • Asymetric deployment and utilization of AI tech such as on surveillance
    Of all the important things on this list that could be targeted, why are methods that concentrate power, aggregate attention to purchases, increase the ability to generate misinformation, and applicability to warfare getting the bulk of resources and attention?
  • Concentration of Wealth
  • Targetted Genetic Warfare
  • Reducing gene pool’s entropy
    Overly broad applications of genetic screening such as: get rid of autism. Or increase intelligence with a greedy approach based on polygenic risk scores
  • Materials
    Many energy related and sequestration technologies largely require developments in this area
  • Health
  • Animal Experimention
    Animal experimentation has medical benefits but are not humane. Certainly some kinds are not needed. Alternatives include yeast, simpler organisms if the pathway is preserved, organoids.
  • Free and Affordable Health Care
    The basics should be free, scaled by disability needs. Complex and rare needs should be affordable. See answer-boards, not non-profits but not profit focused ad-hoc groups to tackle problems in a decentralized way. Local and disaggregated, so developing notions are not beholden.
  • Mental Health
    Many have the misaprehension that only physical blows can cause damage but this is not true for any information processing system.
  • artificial distinction between drugs vs medicine creates negative associations that may be costing un-necessary lives and missed oppurtunities to help people that have been discalibrated by drug dependencies. Similarly, due to suitcasing so many concepts into one thing, drugs with positive effects such as psychedelics, the oppurtunities to make safer, plasticity promoting versions of these has been forgone.
  • Moving to a method that views cancers, viruses and bacteria as strategies and populations where a rational approach (as a search problem with complexity concerns for any of the pathologies cannot overcome) is feasible
  • Fuller more systemic understanding of biology
  • Viruses
  • Cancer
  • Autoimmune
  • Genetic diseases
  • Can developmental errors be corrected if caught early enough?
  • A human can live up to 130 years, how to compress health detioration and remove the luck in reaching that number?
  • Eggs
    Whether overpopulation or under, increased effective lifespan without extending the time any woman has available if they wish a child, without having to plan ahead or require extensive capital, creates a precarious equilbrium and thus increasing the probability of broad human rights violation. If the population is high, wealth inequality becomes more pressing (and there’s a feedback loop). If this is over-corrected into too much negative population growth, a woman’s reproductive rights becomes increasing likely to be violated. Socities might feel justified to undo many of the hard won gains in women’s rights.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store