Could A Neuroscientist Understand A Microprocessor by Rationally Speaking—“Eric Jonas, discussing his provocative paper titled ‘Could a Neuroscientist Understand a Microprocessor?’ in which he applied state-of-the-art neuroscience tools, like lesion analysis, to a computer chip. By applying neuroscience’s tools to a system that humans fully understand he was able to reveal how surprisingly uninformative those tools actually are.”
Reasonable Doubt New Look Whether Prison Growth Cuts Crime by Open Philosophy—Part1 of a four part, in depth, series on Criminal Justice reform. The remaining posts are linked below. “I estimate, that at typical policy margins in the United States today, decarceration has zero net impact on crime. That estimate is uncertain, but at least as much evidence suggests that decarceration reduces crime as increases it. The crux of the matter is that tougher sentences hardly deter crime, and that while imprisoning people temporarily stops them from committing crime outside prison walls, it also tends to increase their criminality after release. As a result, “tough-on-crime” initiatives can reduce crime in the short run but cause offsetting harm in the long run. Empirical social science research—or at least non-experimental social science research—should not be taken at face value. Among three dozen studies I reviewed, I obtained or reconstructed the data and code for eight. Replication and reanalysis revealed significant methodological concerns in seven and led to major reinterpretations of four. These studies endured much tougher scrutiny from me than they did from peer reviewers in order to make it into academic journals. Yet given the stakes in lives and dollars, the added scrutiny was worth it. So from the point of view of decision makers who rely on academic research, today’s peer review processes fall well short of the optimal.”
L Dopen Thread by Scott Alexander—Bi-weekly public open thread. Berkeley SSC meetup. New ad for the Greenfield Guild, an online network of software consultants. Reasons to respect the society of friends.
Book Review Mastering The Core Teachings Of The Buddha by Scott Alexander—“Buddhism For ER Docs. ER docs are famous for being practical, working fast, and thinking everyone else is an idiot. MCTB delivers on all three counts.” Practical buddhism with a focus on getting things done. buddhism is split into morality concentration and wisdom. Discussion of “the Dark Night of the Soul” which is a sort of depression occurs when you have had some but not enough spiritual experience.
The Best Self Help Should Be Self Defeating by mindlevelup—“Self-help is supposed to get people to stop needing it. But typical incentives in any medium mean that it’s possible to get people hooked on your content instead. A musing on how the setup for writing self-help differs from typical content.”
Out To Get You by Zvi Moshowitz—“Some things are fundamentally Out to Get You. They seek resources at your expense. Fees are hidden. Extra options are foisted upon you.” You have four responses: Get Gone, Get Out (give up), Get Compact (limit what it wants) or Get Ready for Battle.
In Defense Of Unreliability by Ozy—Zvi claims that when he makes plan with friends in the bay he never assumes the plan will actually occur. Ozy depends on unreliable transport. Getting places 10-15 early is also costly. Flaking and agoraphobia.
Why Attitudes Matter by Ozy—Focusing on attitudes can be bad for some people. Two arguments: “First, for any remotely complicated situation, it would be impossible to completely list out all the things which are okay or not okay. Second, an attitude emphasis prevents rules-lawyering.”
Humans Cells In Multicellular Future Minds by Robin Hanson—In general humans replace specific systems with more general adaptive systems. Seeing like a State. Most biological and cultural systems are not general. Multi-cellular organisms re tremendously inefficient. The power of entrenched systems. Human brains are extremely general. Human brains may win for a long time vs other forms of intelligence.
Prediction Markets Update by Robin Hanson—Prediction markets provide powerful information but they challenge powerful entrenched interests, Hanson compares them to “a knowledgeable Autist in the C-suite”. Companies selling straight prediction market tech mostly went under. Blockchain platforms for prediction markets. Some discussion of currently promising companies.
Srisk Faq by Tobias Baumann (EA forum) - Quite detailed responses to questions about suffering risks and their connection to AGI. sections: General questions, The future, S-risks and x-risks, Miscellaneous.
===EA:
Reasonable Doubt New Look Whether Prison Growth Cuts Crime by Open Philosophy—Part1 of a four part, in depth, series on Criminal Justice reform. The remaining posts are linked below. “I estimate, that at typical policy margins in the United States today, decarceration has zero net impact on crime. That estimate is uncertain, but at least as much evidence suggests that decarceration reduces crime as increases it. The crux of the matter is that tougher sentences hardly deter crime, and that while imprisoning people temporarily stops them from committing crime outside prison walls, it also tends to increase their criminality after release. As a result, “tough-on-crime” initiatives can reduce crime in the short run but cause offsetting harm in the long run. Empirical social science research—or at least non-experimental social science research—should not be taken at face value. Among three dozen studies I reviewed, I obtained or reconstructed the data and code for eight. Replication and reanalysis revealed significant methodological concerns in seven and led to major reinterpretations of four. These studies endured much tougher scrutiny from me than they did from peer reviewers in order to make it into academic journals. Yet given the stakes in lives and dollars, the added scrutiny was worth it. So from the point of view of decision makers who rely on academic research, today’s peer review processes fall well short of the optimal.”
Paypal Giving Fund by Jeff Kaufman—The PayPal giving fund lets you batch donations and PayPal covers the fees if you use it. Jeff thought there must be a catch but it seems legit.
Against Ea Pr by Ozy—The EA community is the only large entity trying to produce accurate and publicly available assessments of charities. Hence the EA community should not trade away any honesty. EAs should simply say which causes and organizations are most effective, they should not worry about PR concerns.
Demographics Ii by tee (EA forum) - Racial breakdown. Percent white in various geographic locations. Political spectrum. Politics correlated with cause area, diet and geography, employment, fields of study, year joining EA.
===Politics and Economics:
Raj Chetty Course Using Big Data Solve Economic Social Problems by Marginal Revolution—Link to an eleven lecture course. “Equality of opportunity, education, health, the environment, and criminal justice. In the context of these topics, the course provides an introduction to basic statistical methods and data analysis techniques, including regression analysis, causal inference, quasi-experimental methods, and machine learning.”
Speech On Campus Reply To Brad Delong by Noah Smith—The safeguard put in place to exclude the small minority of genuinely toxic people will be overused. Comparison to the war on terror. Brad’s exclusions criteria are incredibly vague. The speech restriction apparatus is patchwork and inconsistent. Cultural Revolution.
Deontologist Envy by Ozy—The behavior of your group is highly unlikely to effect the behavior of your political opponents. Many people respond to proposed tactics by asking “What if everyone did that”. Ozy claims these responses show an implicit Kantian or deontological point of view.
Peak Fossil Fuel by Bayesian Investor—Electric cars will have a 99% market share by 2035. “Electric robocars run by Uber-like companies will be cheap enough that you’ll have trouble giving away a car bought today. Uber’s prices will be less than your obsolete car’s costs of fuel, maintenance, and insurance.”
What We Didn’t Get by Noah Smith—We are currently living in a world envisioned by the cyberpunk writers. the early industrial sci-fi writers also predicted many inventions. Why didn’t mid 1900s sci-fi come true? We ran out of theoretical physics and we ran out of energy. Energy density of fuel sources. Some existing or plausible technology is just too dangerous. Discussion of whether strong AI, personal upload, nanotech and/or the singularity will come true.
Unpopular Ideas About Children by Julia Galef—Julia’s thoughts on why she is collecting these lists. Parenting styles, pro and anti-natalism, sexuality, punishment, etc. Happiness studies. Some other studies finding extreme results.
The Margin Of Stupid by Noah Smith—Can we trust studies showing that millennials are as racist as their parents, except for the ones in college who are extreme leftists?
Role of Allies in Queer Spaces by Brute Reason—The main purpose of having allies in LBGTQA spaces is providing cover for closeted or questioning members. Genuinely cis-straight allies are ok in some spaces like LBGTQA bands. But straight allies cause problems when they are present in queer support spaces.
The Wonder Of International Adoption by Bryan Caplan—Benefits of international adoption of third world children. Adoptees are extremely stunted physically on arrival but make up some of the difference post adoption. International adoptions raises IQ by at least 4 points on average and perhaps as much as 8.
===Misc:
Coin Flipping Problem by protokol2020 - Flipping coins until you get a pre-committed sequence. You re-start whenever your flip doesn’t match the sequence. Relationship between the expected number of flips and the length of the sequence.
Seek Not To Be Entertained by Mr. Money Mustache—Don’t be normal, normal people need constant entertainment. You can get enjoyment and satisfaction from making things. Advice for people less abnormal than MMM. What you enjoy doesn’t matter, what matters is what is good for you.
Comments For Ghost by Tom Bartleby—Ghost is a blog platform hat doesn’t natively support comments. Three important use cases and why they all benefit from comments: Ex-Wordpress blogger who wants things to ‘just work’, Power suers care about privacy and don’t want to use third party comments, The Static-Site Fence-Sitter since the main dynamic content you want is comments.
Prime Crossword by protokol2020 - Can you create a grid larger than [3,7],[1,1] where all the rows and columns are primes? (37, 11, 31 and 71 are prime).
===Podcast:
Reihan Salam by The Ezra Klein Show—Remaking the Republican party, but not the way Donald Trump did it. “The future of the Republican Party, the healthcare debate, and how he would reform our immigration system (and upend the whole way we talk about it). ”
Conversation with Larry Summers by Marginal Revolution—“Mentoring, innovation in higher education, monopoly in the American economy, the optimal rate of capital income taxation, philanthropy, Hermann Melville, the benefits of labor unions, Mexico, Russia, and China, Fed undershooting on the inflation target, and Larry’s table tennis adventure in the summer Jewish Olympics.”
Hilary Clinton by The Ezra Klein Show—Hilary’s dream of paying for basic income with revenue from shared national resources. Why she scrapped the plan. Hilary thinks she should perhaps have thrown caution to the wind. Hilary isn’t a radical, she is proud of the American political system and is annoyed other’s don’t share her enthusiasm for incremental progress.
David Remnick by The Ezra Klein Show—New Yorker editor. “Russia’s meddling in the US election, Russia’s transformation from communist rule to Boris Yeltsin and Vladimir Putin, his magazine’s coverage of President Donald Trump, how he chooses his reporters and editors, and how to build a real business around great journalism.”
Gabriel Zucman by EconTalk—“Research on inequality and the distribution of income in the United States over the last 35 years. Zucman finds that there has been no change in income for the bottom half of the income distribution over this time period with large gains going to the top 1%. The conversation explores the robustness of this result to various assumptions and possible explanations for the findings.”
Could A Neuroscientist Understand A Microprocessor by Rationally Speaking—“Eric Jonas, discussing his provocative paper titled ‘Could a Neuroscientist Understand a Microprocessor?’ in which he applied state-of-the-art neuroscience tools, like lesion analysis, to a computer chip. By applying neuroscience’s tools to a system that humans fully understand he was able to reveal how surprisingly uninformative those tools actually are.”
Rational Feed: Last Week’s Community Articles and Some Recommended Posts
===Highly Recommended Articles:
Why I Am Not A Quaker Even Though It Often Seems As Though I Should Be by Ben Hoffman—Quakers have consistently gotten to the right answers faster than most people, or the author. Arbitrage strategies to beat the quakers. An incomplete survey of alternatives.
Could A Neuroscientist Understand A Microprocessor by Rationally Speaking—“Eric Jonas, discussing his provocative paper titled ‘Could a Neuroscientist Understand a Microprocessor?’ in which he applied state-of-the-art neuroscience tools, like lesion analysis, to a computer chip. By applying neuroscience’s tools to a system that humans fully understand he was able to reveal how surprisingly uninformative those tools actually are.”
Reasonable Doubt New Look Whether Prison Growth Cuts Crime by Open Philosophy—Part1 of a four part, in depth, series on Criminal Justice reform. The remaining posts are linked below. “I estimate, that at typical policy margins in the United States today, decarceration has zero net impact on crime. That estimate is uncertain, but at least as much evidence suggests that decarceration reduces crime as increases it. The crux of the matter is that tougher sentences hardly deter crime, and that while imprisoning people temporarily stops them from committing crime outside prison walls, it also tends to increase their criminality after release. As a result, “tough-on-crime” initiatives can reduce crime in the short run but cause offsetting harm in the long run. Empirical social science research—or at least non-experimental social science research—should not be taken at face value. Among three dozen studies I reviewed, I obtained or reconstructed the data and code for eight. Replication and reanalysis revealed significant methodological concerns in seven and led to major reinterpretations of four. These studies endured much tougher scrutiny from me than they did from peer reviewers in order to make it into academic journals. Yet given the stakes in lives and dollars, the added scrutiny was worth it. So from the point of view of decision makers who rely on academic research, today’s peer review processes fall well short of the optimal.”
Deterrence De Minimus by Open Philosophy—Part 2.
Incapacitation How Much Does Putting People Inside Prison Cut Crime Outside by Open Philosophy—Part 3.
Aftereffects Us Evidence Says Doing More Time Typically Leads More Crime After by Open Philosophy—Part 4.
===Scott:
L Dopen Thread by Scott Alexander—Bi-weekly public open thread. Berkeley SSC meetup. New ad for the Greenfield Guild, an online network of software consultants. Reasons to respect the society of friends.
Meditative States As Mental Feedback Loops by Scott Alexander—the main reason we don’t see emotional positive feedback loops is that people get distracted. If you do not get distracted you can experience a bliss feedback look.
Book Review Mastering The Core Teachings Of The Buddha by Scott Alexander—“Buddhism For ER Docs. ER docs are famous for being practical, working fast, and thinking everyone else is an idiot. MCTB delivers on all three counts.” Practical buddhism with a focus on getting things done. buddhism is split into morality concentration and wisdom. Discussion of “the Dark Night of the Soul” which is a sort of depression occurs when you have had some but not enough spiritual experience.
===Rationalist:
Impression Track Records by Katja Grace—Three reasons its better to keep impression track records and belief track records separate.
Why I Am Not A Quaker Even Though It Often Seems As Though I Should Be by Ben Hoffman—Quakers have consistently gotten to the right answers faster than most people, or the author. Arbitrage strategies to beat the quakers. An incomplete survey of alternatives.
The Best Self Help Should Be Self Defeating by mindlevelup—“Self-help is supposed to get people to stop needing it. But typical incentives in any medium mean that it’s possible to get people hooked on your content instead. A musing on how the setup for writing self-help differs from typical content.”
Nobody Does The Thing That They Are Supposedly Doing by Kaj Sotala—“In general, neither organizations nor individual people do the thing that their supposed role says they should do.” Evolutionary incentives. Psychology of motivation. Very large number of links.
Out To Get You by Zvi Moshowitz—“Some things are fundamentally Out to Get You. They seek resources at your expense. Fees are hidden. Extra options are foisted upon you.” You have four responses: Get Gone, Get Out (give up), Get Compact (limit what it wants) or Get Ready for Battle.
In Defense Of Unreliability by Ozy—Zvi claims that when he makes plan with friends in the bay he never assumes the plan will actually occur. Ozy depends on unreliable transport. Getting places 10-15 early is also costly. Flaking and agoraphobia.
Strategic Goal Pursuit And Daily Schedules by Rossin (lesswrong) - The author benefitted from Anna Salamon’s goal-pursuing heuristics and daily schedules.
Why Attitudes Matter by Ozy—Focusing on attitudes can be bad for some people. Two arguments: “First, for any remotely complicated situation, it would be impossible to completely list out all the things which are okay or not okay. Second, an attitude emphasis prevents rules-lawyering.”
Humans Cells In Multicellular Future Minds by Robin Hanson—In general humans replace specific systems with more general adaptive systems. Seeing like a State. Most biological and cultural systems are not general. Multi-cellular organisms re tremendously inefficient. The power of entrenched systems. Human brains are extremely general. Human brains may win for a long time vs other forms of intelligence.
Recognizing Vs Generating An Important Dichotomy For Life by Gordon (Map and Territory) - Bullet Points → Essay vs Essay → Bullet Points. Generating ideas vs critique. Most advice is bad since it doesn’t convey the reasons clearly. Let the other person figure out the actual advice for themselves.
Prediction Markets Update by Robin Hanson—Prediction markets provide powerful information but they challenge powerful entrenched interests, Hanson compares them to “a knowledgeable Autist in the C-suite”. Companies selling straight prediction market tech mostly went under. Blockchain platforms for prediction markets. Some discussion of currently promising companies.
===AI:
Focus Areas Of Worst Case Ai Safety by The Foundational Research Institute—Redundant safety measures. Tripwires. Adversarial architectures. Detecting and formalizing suffering. Backup utility functions. Benign testing environments.
Srisk Faq by Tobias Baumann (EA forum) - Quite detailed responses to questions about suffering risks and their connection to AGI. sections: General questions, The future, S-risks and x-risks, Miscellaneous.
===EA:
Reasonable Doubt New Look Whether Prison Growth Cuts Crime by Open Philosophy—Part1 of a four part, in depth, series on Criminal Justice reform. The remaining posts are linked below. “I estimate, that at typical policy margins in the United States today, decarceration has zero net impact on crime. That estimate is uncertain, but at least as much evidence suggests that decarceration reduces crime as increases it. The crux of the matter is that tougher sentences hardly deter crime, and that while imprisoning people temporarily stops them from committing crime outside prison walls, it also tends to increase their criminality after release. As a result, “tough-on-crime” initiatives can reduce crime in the short run but cause offsetting harm in the long run. Empirical social science research—or at least non-experimental social science research—should not be taken at face value. Among three dozen studies I reviewed, I obtained or reconstructed the data and code for eight. Replication and reanalysis revealed significant methodological concerns in seven and led to major reinterpretations of four. These studies endured much tougher scrutiny from me than they did from peer reviewers in order to make it into academic journals. Yet given the stakes in lives and dollars, the added scrutiny was worth it. So from the point of view of decision makers who rely on academic research, today’s peer review processes fall well short of the optimal.”
Deterrence De Minimus by Open Philosophy—Part 2.
Incapacitation How Much Does Putting People Inside Prison Cut Crime Outside by Open Philosophy—Part 3.
Aftereffects Us Evidence Says Doing More Time Typically Leads More Crime After by Open Philosophy—Part 4.
Paypal Giving Fund by Jeff Kaufman—The PayPal giving fund lets you batch donations and PayPal covers the fees if you use it. Jeff thought there must be a catch but it seems legit.
What Do Dalys Capture by Danae Arroyos (EA forum) - How Disability Adjusted life years computed. DALYs misrepresent mental health. DALY’s Miss Indirect Effects. Other issues.
Against Ea Pr by Ozy—The EA community is the only large entity trying to produce accurate and publicly available assessments of charities. Hence the EA community should not trade away any honesty. EAs should simply say which causes and organizations are most effective, they should not worry about PR concerns.
Ea Survey 2017 Series Qualitative Comments Summary by tee (EA forum) - Are you an EA, how welcoming is EA, local EA meetup attendance, concerns with not being ‘EA enough’, improving the survey.
Demographics Ii by tee (EA forum) - Racial breakdown. Percent white in various geographic locations. Political spectrum. Politics correlated with cause area, diet and geography, employment, fields of study, year joining EA.
===Politics and Economics:
Raj Chetty Course Using Big Data Solve Economic Social Problems by Marginal Revolution—Link to an eleven lecture course. “Equality of opportunity, education, health, the environment, and criminal justice. In the context of these topics, the course provides an introduction to basic statistical methods and data analysis techniques, including regression analysis, causal inference, quasi-experimental methods, and machine learning.”
Speech On Campus Reply To Brad Delong by Noah Smith—The safeguard put in place to exclude the small minority of genuinely toxic people will be overused. Comparison to the war on terror. Brad’s exclusions criteria are incredibly vague. The speech restriction apparatus is patchwork and inconsistent. Cultural Revolution.
Deontologist Envy by Ozy—The behavior of your group is highly unlikely to effect the behavior of your political opponents. Many people respond to proposed tactics by asking “What if everyone did that”. Ozy claims these responses show an implicit Kantian or deontological point of view.
Peak Fossil Fuel by Bayesian Investor—Electric cars will have a 99% market share by 2035. “Electric robocars run by Uber-like companies will be cheap enough that you’ll have trouble giving away a car bought today. Uber’s prices will be less than your obsolete car’s costs of fuel, maintenance, and insurance.”
What We Didn’t Get by Noah Smith—We are currently living in a world envisioned by the cyberpunk writers. the early industrial sci-fi writers also predicted many inventions. Why didn’t mid 1900s sci-fi come true? We ran out of theoretical physics and we ran out of energy. Energy density of fuel sources. Some existing or plausible technology is just too dangerous. Discussion of whether strong AI, personal upload, nanotech and/or the singularity will come true.
Unpopular Ideas About Children by Julia Galef—Julia’s thoughts on why she is collecting these lists. Parenting styles, pro and anti-natalism, sexuality, punishment, etc. Happiness studies. Some other studies finding extreme results.
The Margin Of Stupid by Noah Smith—Can we trust studies showing that millennials are as racist as their parents, except for the ones in college who are extreme leftists?
Role of Allies in Queer Spaces by Brute Reason—The main purpose of having allies in LBGTQA spaces is providing cover for closeted or questioning members. Genuinely cis-straight allies are ok in some spaces like LBGTQA bands. But straight allies cause problems when they are present in queer support spaces.
The Wonder Of International Adoption by Bryan Caplan—Benefits of international adoption of third world children. Adoptees are extremely stunted physically on arrival but make up some of the difference post adoption. International adoptions raises IQ by at least 4 points on average and perhaps as much as 8.
===Misc:
Coin Flipping Problem by protokol2020 - Flipping coins until you get a pre-committed sequence. You re-start whenever your flip doesn’t match the sequence. Relationship between the expected number of flips and the length of the sequence.
Seek Not To Be Entertained by Mr. Money Mustache—Don’t be normal, normal people need constant entertainment. You can get enjoyment and satisfaction from making things. Advice for people less abnormal than MMM. What you enjoy doesn’t matter, what matters is what is good for you.
Propositions On Immortality by sam[]zdat—Fiction. A man digresses about philosophy, the nature of time, the soul, consciousness and mortality.
Comments For Ghost by Tom Bartleby—Ghost is a blog platform hat doesn’t natively support comments. Three important use cases and why they all benefit from comments: Ex-Wordpress blogger who wants things to ‘just work’, Power suers care about privacy and don’t want to use third party comments, The Static-Site Fence-Sitter since the main dynamic content you want is comments.
Prime Crossword by protokol2020 - Can you create a grid larger than [3,7],[1,1] where all the rows and columns are primes? (37, 11, 31 and 71 are prime).
===Podcast:
Reihan Salam by The Ezra Klein Show—Remaking the Republican party, but not the way Donald Trump did it. “The future of the Republican Party, the healthcare debate, and how he would reform our immigration system (and upend the whole way we talk about it). ”
Into The Dark Land by Waking Up with Sam Harris—“Siddhartha Mukherjee about his Pulitzer Prize winning book, The Emperor of All Maladies: A Biography of Cancer.”
Conversation with Larry Summers by Marginal Revolution—“Mentoring, innovation in higher education, monopoly in the American economy, the optimal rate of capital income taxation, philanthropy, Hermann Melville, the benefits of labor unions, Mexico, Russia, and China, Fed undershooting on the inflation target, and Larry’s table tennis adventure in the summer Jewish Olympics.”
Hilary Clinton by The Ezra Klein Show—Hilary’s dream of paying for basic income with revenue from shared national resources. Why she scrapped the plan. Hilary thinks she should perhaps have thrown caution to the wind. Hilary isn’t a radical, she is proud of the American political system and is annoyed other’s don’t share her enthusiasm for incremental progress.
David Remnick by The Ezra Klein Show—New Yorker editor. “Russia’s meddling in the US election, Russia’s transformation from communist rule to Boris Yeltsin and Vladimir Putin, his magazine’s coverage of President Donald Trump, how he chooses his reporters and editors, and how to build a real business around great journalism.”
Gabriel Zucman by EconTalk—“Research on inequality and the distribution of income in the United States over the last 35 years. Zucman finds that there has been no change in income for the bottom half of the income distribution over this time period with large gains going to the top 1%. The conversation explores the robustness of this result to various assumptions and possible explanations for the findings.”
Could A Neuroscientist Understand A Microprocessor by Rationally Speaking—“Eric Jonas, discussing his provocative paper titled ‘Could a Neuroscientist Understand a Microprocessor?’ in which he applied state-of-the-art neuroscience tools, like lesion analysis, to a computer chip. By applying neuroscience’s tools to a system that humans fully understand he was able to reveal how surprisingly uninformative those tools actually are.”