academia | advice | alcohol | American Indians | architecture | art | artificial intelligence | Barnard | best | biography | bitcoin | blogging | broken umbrellas | candide | censorship | children's books | Columbia | comics | consciousness | cooking | crime | criticism | dance | data analysis | design | dishonesty | economics | education | energy | epistemology | error correction | essays | family | fashion | finance | food | foreign policy | futurism | games | gender | Georgia | health | history | inspiration | intellectual property | Israel | journalism | Judaism | labor | language | law | leadership | letters | literature | management | marketing | memoir | movies | music | mystery | mythology | New Mexico | New York | parenting | philosophy | photography | podcast | poetry | politics | prediction | product | productivity | programming | psychology | public transportation | publishing | puzzles | race | reading | recommendation | religion | reputation | review | RSI | Russia | sci-fi | science | sex | short stories | social justice | social media | sports | startups | statistics | teaching | technology | Texas | theater | translation | travel | trivia | tv | typography | unreliable narrators | video | video games | violence | war | weather | wordplay | writing

Monday, April 23, 2018

There are dozens of reasons to fear AI, and only one mistaken reason not to

James Vincent just wrote a review in The Verge of Chris Paine's new documentary Do You Trust This Computer?, which features noted AI alarmist Elon Musk (and is being distributed for free online thanks to a donation by Musk).

The review leaves a lot to be desired. Yes algorithmic policing deserves criticism and attention... but that's a totally different film. This one seems focused on the existential risk. There's a very good chance animal life on earth will end in the next 100 years because of AI. Vincent has objections like (paraphrasing) "But a lot of people are publishing papers about possible ways to rein in AI!" I think that deeply misses the point.

There are thousands of different, independently developed AI programs running right now that play MMORPGs and are programmed to mine resources, forge weapons, seek the inhabitants of their world and kill them. Lots of these were made by teenagers. No one had to program them to be evil--they're not evil!

How far is Boston Dynamics from being able to perform all of those steps in real life? How far is the US military? There will probably be an arms race with killer robots, there will be tons of private sector research, there will be script kiddies, we already have homebrewed armed drones... and there's no real way to reliably keep software from taking actions in the real world that they would take in a virtual world.

The rhetorical task in front of the alarmist camp is merely to point out that there are many reasonable scenarios in which we inadvertently destroy the human race (and perhaps most animal life on earth) in the course of improving AI.

The rhetorical task in front of the optimists is to prove that all such scenarios that anyone could ever conceive are all unreasonable and all unlikely. I've read a lot of optimists, and I haven't encountered one that seems to be thinking about the problem thoroughly and rigorously.

They mistakenly operate from the presumption that only one body of researchers needs to come to agreement about restrictions on AI research. Or they mistakenly assume that evil is necessary for a machine to kill. Or the act is that we are still in the mainframe or personal computer era, and AI agents will be physically unpluggable. Or they misunderstand how easy it would be to kill hundreds of millions of people (just get a few dozen sheets of uranium together in various places and clap them together, or set up a network of autonomous factories that disproportionately use up the atmosphere's gases, or make tiny drones that seek and burrow into necks...) Or they are missing the aggressive direction of major power militaries towards creating software and hardware for self-perpetuating, autonomous killing machines. Or they misunderstand how opaque the inner workings of trained machine learning algorithms are.

When I hear someone say that they are certain there's no risk that this ends with the destruction of humanity (and probably most animal life on Earth), I think they are experiencing a profound failure of imagination.

Labels: , , ,

Thursday, April 19, 2018

Comey is dead wrong: mass incarceration is real

Reviewing James Comey's new memoir in The Intercept, Peter Maass focuses on an episode in which Comey complained to Barack Obama that the presidents use of the term mass incarceration was offensive:
Comey did not hide these views while at the FBI, and after making a speech in Chicago in 2015 that was not well received by the civil rights community, he was summoned to the Oval Office by then-President Barack Obama. Comey describes that session in his book, and he seemed to double down, telling the country’s first black president that the law enforcement community was upset at the way Obama had used the phrase “mass incarceration.” It was offensive, Comey told the president.

“I thought the term was both inaccurate and insulting to a lot of good people in law enforcement who cared deeply about helping people trapped in dangerous neighborhoods,” Comey writes. “It was inaccurate in the sense that there was nothing ‘mass’ about the incarceration: every defendant was charged individually, represented individually by counsel, convicted by a court individually, sentenced individually, reviewed on appeal individually, and incarcerated. That added up to a lot of people in jail, but there was nothing ‘mass’ about it.”

Maass makes a number of good points in his counterargument, including the point that more than 90% of cases on the state and federal level are settled with a plea bargain and not given their day in court.

A few more points I would add:

  1. There are hundreds of thousands of people sitting in jail, some for months or even years, who are there not because they were convicted but because they have a pending case and can't afford bail.
  2. Police departments routinely sweep up and jail groups of people with little knowledge of whether or not they have individually committed any crime, and release them without charge. (This is most visible at political protests when it happens to middle-class professionals, such as the dentist on his way to work who was arrested in New York City when police just grabbed everyone in the vicinity of a protest against the 2004 Republican convention. But it happens far more often than that.)
  3. Numerous police departments have faced revelations in recent years that their police and prosecutors routinely incarcerated people on charges they knew were false.
  4. Prison management and construction firms have lobbied, both legally and illegally, for increased imprisonment in multiple states. In Pennsylvania, a judge who secretly received a $2 million "finder's fee" from a private prison jailed kids for *years* for such offenses as having been given a bike by his parents that they bought on Craigslist that turned out to have been stolen.
  5. In numerous police departments, whistleblowers have revealed the persistence of quotas for arrests and citations.

In all of these aspects, people who did not commit crimes or who have not been convicted of crimes are incarcerated for large amounts of time due to political and economic dynamics far beyond any formal demonstration of their guilt.

Mass incarceration of the guilty is also an issue; but at the very least, any decent human being with a modicum of honesty must concede that the widespread, systematic incarceration of those whose guilt has not been determined and demonstrated by the system constitutes mass incarceration. In this country, some people get individual consideration and are presumed innocent; others are not, but rather treated a significant amount of the time as an undifferentiated mass of poor people and people of color.

Maass mentions a separate part of the book where Comey is surprisingly frank about this distinction. He quotes Comey as writing, of former general David Petraeus's slap on the wrist for lying to the FBI and mishandling classified information:

I believed, and still believe, that Petraeus was treated under a double standard based on class... A poor person, an unknown person — say a young black Baptist minister from Richmond — would be charged with a felony and sent to jail.

Comey is right about that, but of course there's even more than those two levels of standard. For many, arrested and jailed en masse, nothing close to the level of offense and amount of evidence in Petraeus's case is necessary to jail you for months or years.

Labels: , ,

Wednesday, April 11, 2018

The real problem with scientific papers

An article by Richard Goerg in the Atlantic arguing that the scientific paper is obsolete has been making the rounds.

There are too many lazy assumptions in this (commonly repeated) line of analysis to count. I do appreciate the skepticism of Wolfram that the author briefly discusses, but I wish he thought to apply it more broadly. Yes, it's definitely really cool that you can use Mathematica to do all kinds of math. But from my experience playing with Wolfram Alpha, the whole "analyze the rainfall during the battle of the Somme" stuff is way overblown. When you get away from cherry picked Wolfram-provided examples, the data sources are thin and spotty and most importantly, they presume a single universal source of truth as if we haven't learned from the past 25 years that any data or facts you assume to be true without looking at them carefully turn out to be at least somewhat false.

Bret Victor is a genius, but like Wolfram he suffers from a smugness so deep that it gets in the way of his brilliance. This research paper rewrite is a perfect example. It's great, of course--but it's great on its own and after a tremendous amount of work by an unusually brilliant programmer.

Compare the technologies the original authors used--a word processor plus some kind of spreadsheet with graphing capabilities--to the technologies Victor used--custom javascript, css, and html. How many researchers have access and capabilities to producing rapidly with the former versus the latter? How many people in the world could quickly and reliably produce the latter? How likely are the products of each to be obsolete in 5 years, 10 years, 20 years? (When was the last time you ran software from more than 10 years ago on anything but its original dedicated hardware?) How replicable/applicable is Victor's work here to other papers in a domain even a few degrees away from this particular type of graph analysis?

Yes, Mathematica points to a direction where non-programmers could produce complex interactive widgets and such, but given that centuries of person-hours by brilliant people has gone into it and it still isn't being used for that, how broadly achievable is that vision, really?

I think there's a much more fruitful direction for explanatory publishing: the clarity, and focus on what matters, that a good explainer brings when a companion asks them to explain something on a napkin at a bar. (I call this a "Gawker explanation", because Gawker was founded with the intention of telling you the New York media-related news that an informed insider would share with you off the record at a bar.) There are ideas too complex to explain this way, but most published papers aren't that complex. This can be done visually with text and images--think how easy would be to convert Victor's fun interactive sliders into static images and text that would be nearly as good (much like he already did in the bulk of that redesign)--and it can be done in the great explanatory medium of our era, video! That Victor, and this author, don't seem to seriously consider video as an explanatory medium tells you, I think, that they still have a lot of work to do in their work to improve the ways we inform others.

The replication failure crisis is a result of the medium of publishing, yes, but that medium is mostly social, economic, and institutional, not textual. These forces award surprising new findings and reinforce academic power structures, and discourage work to double check, declare null findings, and explore alternative hypotheses, and tell the truth about research labor and credit. Similarly, video explainers such as TED talks tend to be mediocre because their medium, viral social media, awards fame to speakers who can line up emotional zingers and simplistic narratives that reassure their audience by obscuring complexity and providing easy answers.

People don't tell the truth about what they do and don't understand. They create easy answers and positive narratives. They worship power and obey authority. They go along to get along. They respect confidence more than skepticism and see honest self-doubt as weak. They construct pyramid schemes of authority wherever they go. That's the problem with academic papers, not "print".

Labels: , , , , , , ,

Thursday, March 29, 2018

The Cosa Nostra of cryptocurrency?

This profile of the founder of the cryptocurrency exchange Binance is jaw-dropping. Binance is a totally new kind of startup: according to the article, it isn't officially based in any country, and it doesn't even have a bank account!

I'm skeptical. How do they pay for server time? Does this only mean that there's no publicly announced bank account?

At any rate, if this claim isn't entirely true yet, it will be possible soon. When you exist purely in the cryptocurrency realm, the barriers to entry are way lower. I think that might be one of the more profound long term effects of cryptocurrencies. It might even mean that governments lower regulatory barriers to forming businesses, to hold onto tax profits from complying businesses that might otherwise choose to form entirely out of encrypted bytes in the cloud.

Which raises the question: who will be the Stripe/Braintree of cryptocurrencies? It's theoretically feasible today to set up a company in a single day that offers some SaaS, charges using cryptocurrencies, and pays for its server time using traditional currency. But those last two are still somewhat headaches, and ground the company in some jurisdiction, given the need to convert.

And what does this mean for money-laundering? At some point only foolish criminals will demand their palms be greased with legal tender. The smartest criminals will do their racketeering in cryptocurrency land itself--removing any need to launder it.

Labels: , , ,

Tuesday, February 27, 2018

A progressive national agenda

An online discussion has emerged about what the Democrats agenda should be, for when they next enter power.

Here is my wishlist:

  1. Politics:
    1. Make Washington DC a state

    2. Aggressive federal involvement in any state that Gerrymanders to prevent statewide representation reflecting the population. Require maps to fall within formulas for compactness and representation matching.

    3. Abolish the electoral college.

    4. Replace the party balloting system nationally with one where all candidates with signatures from at least some percentage of voters (say, 1%) from some percentage of states (say, 25%) are listed in all states

    5. Replace the single choice presidential vote system with approval voting, allowing for each single voter to choose as many candidates as they approve of, so as not to act like a spoiler.

  2. Health:
    1. Single-payer healthcare system built on expansion of Medicare for all

    2. Still allow supplemental insurance, to cover doctors not part of Medicare

    3. Outlaw drug, gun, lawyer and military advertising except for 2-color, non-pictorial text ads

    4. Make drug approval process for drugs already tested extensively in other countries extremely expedited

    5. Expand FDA testing and regulation of health supplements, medicines. Create grade system reflecting company's history of their products proven to contain what they say they contain; new entrants can buy into this by putting up their own capital in a bet that their products will prove pure

    6. Federal rule to suspend drivers' licenses in all cases where injury occurs and the driver is partially to blame, on increasing scale

  3. Poverty:
    1. UBI, including accounts that are inheritable, earn interest

    2. Attack poverty with the Harlem Children's Zone model: increased federal funds in poor areas for schools, hospitals, health clinics, job training, home visits by nurses

  4. Social justice:
    1. Make May 31st (date of 1921 Greenwood Race riot) a national day of remembrance for the victims of slavery and Jim Crow, with ceremony in Washington reading the names and brief description of all known victims

    2. $1T of reparations for slavery and Jim Crow to all black descendants of US slaves, paid over 25 years into UBI accounts (about $1k/person/year).

  5. Violence:
    1. Federal arms restrictions based on testing for rate of lethality in different scenarios (like auto safety)

    2. Fund and conduct firearms and other public health research. May require permanently budgeted, independent public health research office

  6. Environment:
    1. Emissions and pollution cap & trade system

    2. Aggressively pursue international pollution/carbon pacts

    3. Federal gasoline tax of $1/gal to force Saudi Arabia to effectively subsidize our shift to renewable energy

    4. Tax environmental damage at the source. Whatever you produce or import, you need to pay or account for the cost to the environment. This should make single use plastic bags unaffordably expensive.

  7. Taxes:
    1. Wealth tax of 0.5%/year

    2. Increase the earned income tax credit, the standard deduction, and the per child tax credit by 50%

    3. Taxes can be paid by estimate by each taxpayer. If IRS challenges, can pay $500 to make taxpayer do taxes. Taxpayer owes 5x any discrepancy.

  8. Foreign policy:
    1. Open all foreign policy and domestic surveillance archives pre-1990, except for info that could be used to identify agents still in the field

    2. Recall most overseas troops

    3. Multiply foreign aid budget by 10

    4. Withhold military aid and trade to all countries that officially discriminate on the basis of sex, religion, ethnicity, or sexual orientation, or who use their military to colonize lands outside their country

    5. Demand that all countries that officially discriminate on the basis of sex, religion, ethnicity, or sexual orientation, or who use their military to colonize lands outside their country, and specifically call out nations that violate this

    6. US pays reparations to all countries with democratic gov'ts overthrown or civilians murdered through US funds

  9. Immigration:
    1. Abolish ICE's immigration lawbreaking division

    2. Greatly increase rate of admitting refugees

  10. Education:
    1. Federal rule, with funding, for minimum of 1 full time school psychologist per 200 kids, 1 full time teacher per 20 kids

  11. Economics:

      End the mortgage interest deduction, the health insurance and benefits deduction, and most other specific deductions

    1. End most farm subsidies

    2. National and 50-state quasi-governmental funding body to make 7-year, $20k/yr grants to journalists

    3. Deregulate predictive betting--let's let smart people make money by sharing what they're right about with the world, and punish consistently wrong pundits

    4. Rescind DMCA and other overreaching big gov't intrusions into the internet, and restore net neutrality

    5. Break up Google (YouTube, Waymo, Maps), Facebook (Insta, Messenger, WhatsApp); regulate Apple (allow competitors to App Store)

    6. Do not allow social networks to acquire each other

  12. Criminal justice:
    1. Outlaw for-profit jails

    2. Require prisons to allow unannounced spot inspections by federally approved NGOs

    3. Outlaw solitary confinement for more than a small percentage of prisoner-hours

    4. Direct federal prosecutors to focus on crimes that erode the public trust, such as fraud, embezzlement, false advertising, false representation, and violent crime

    5. Require all firearms to carry a serial number and to be registered in a federal database; failure to report a firearms transaction and removal of serial number are federal crimes

    6. All firearms transactions require a background check and a waiting period of at least 7 days

    7. Treat issuing guns like issuing drugs or poisons: partial criminal liability for all past gun owners including importers and manufacturers when guns are used in a crime, suicide or accidental death; liability is small per incident, but accumulates over time

    8. Make perversion of justice punishable by up to life imprisonment and full asset seizure; prosecutors, and judges and police given public trust who abuse it should go to jail

    9. Clarify/revise federal perjury rules: stating false information, even if it cannot be demonstrated that you knew it was false, is criminal.

    10. Prosecute all perjurers aggressively, even if they claim a good reason for perjury.

    11. Prosecute for perjury and contempt of Congress those who refused to answer questions, like Steve Bannon and Hope Hicks

    12. Prosecutors, police and judges who knowingly participated in prosecuting people they had reason to believe were innocent should be prosecuted for being accessories to crime after the fact, since they helped the truly guilty avoid justice.

    13. DOJ white collar crime commission to figure out why Manafort/Gates crimes were not investigated until Mueller; prosecute it with max sentences

    14. Make false advertising a federal crime punished by both fines and imprisonment

    15. Make false representation a federal crime punished by both fines and imprisonment; bring charges for false representation against banks and their employees such as Wells Fargo and Goldman Sachs

    16. Federal legislation to forbid judicial settlements for criminal activity that do not contain criminal punishments and admissions of guilt

    17. Criminally prosecute those who enforce discrimination by gender, ethnicity, or sexual orientation in public facilities such as buses in Williamsburg

    18. Impeach Mitch McConnell for refusing to fulfill Senate duty to advise President Obama's Supreme Court nomination

    19. Impeach and criminally prosecute Clarence Thomas for perjury, fraud and/or obstruction of justice

    20. Impeach and criminally prosecute Cyrus Vance for obstruction of justice in dropping solid cases against Harvey Weinstein and Trump children

    21. Prosecute many members of Bush administration for roles in torture, Iraq War

Labels: , , , , , , ,

Wednesday, February 14, 2018

Last night, yesterday morning: a Wheeler family story

My father's patrilineal family line, the Wheelers, have a particularly great family story that's been passed down from generation to generation.

The story takes place in the south, probably in Kentucky, sometime before the Civil War; my great-aunt, who knew her grandparents' generation well, placed it around 1950.

County fairs were the big entertainment in those days, huge week-long extravaganzas full of pumpkin growing contests, horse pulls (seeing whose horse can pull the greatest weight), bake-offs, and various carnival games. In the days before movies, let alone television, if you weren't in a city and near a vaudeville theatre, this was just the best it could get.

My great great grandfather, Charles James Merriweather ("C.M.") Wheeler, born around 1830, loved the carnival games as a boy, and as he got older he was especially fascinated by the unconvetional games--where instead of throwing hoops around a nail, you were trying to outwit a carnival veteran.

His favorite booth each year was the memory expert: a man who swore he could remember any story under a hundred words, which had to be grammatically correct of course, after only one hearing. To challenge him, you had to obtain or come up with the story, and read it aloud off a piece of paper; the crowd behind you would follow the words, so there would be a record to compare the carny's response to.

The fireworks came after you'd finish reading; there would be a pause, when it would seem to the crowd perfectly impossible to remember every word; and then the memory expert would rattle it off, the whole thing, in one or two breaths, slap his hand on the money you'd wagered, and add it to the bank, while the crowd gasped and clapped and shook their heads.

Oh, and that wager was a whole quarter, a good day's wage in that time. The Wheelers were schoolteachers who had recently immigrated from Scotland, and a quarter was no small amount to part with on a lark.

If by some miracle you won--something C. M. had never seen, in several seasons of hanging around the booth--you would win a "whole ham". In other words, the cured hindquarter of a hog, a massive two hundred pound piece of meat that would feed a family through the winter--an enormous payoff.

We all know this story in my family because C. M. saved up, wrote the most nonsensical story he could think of, and one summer day plunked down his quarter, waited for the crowd to stop shouting, pulled out his paper and read aloud the following:

Last night yesterday morning, about one o'clock in the afternoon, a hungry boy about 40 years old bought a custard for a levy and threw it through a brick wall nine feet thick, and jumping over it, broke his right ankle above his left knee and fell into a dry mill pond and was drowned. Ten years later on the same day, a cat hatched nine turkey gobblers and a high wind blew Yankee Doodle on a frying pan in Boston, where a deaf and dumb man was talking to his Aunt Peter.

The memory expert got half way through the mill pond, and no farther. C. M. won the ham, became a local hero, and (with plenty of help) brought home a huge surprise for his parents and brothers and sisters.

I asked my father to comment on this, and he said: "Of course he won the prize! This ditty has been handed down from father or mother to son and daughter across five enerations now, in the 150+ years since."

Labels: ,

Friday, February 09, 2018

The certainty of the education reformer

In a recent interview with Reason, economics professor Bryan Caplan made the case against our education system.

Watching it sent me down and epistemological rabbit hole about social change and how we know what we know about its workings and consequences.

There's a ton in this interview that I agree with. I certainly agree that most students don't seem to be getting from their education anywhere near as much as I got from mine.

And I have a lot of trust in the assessment that for most people, much of the value of college, and even high school, comes from the signaling that a degree provides.

At Columbia University, I was often taken aback by how impersonal and anonymous the learning experience was in many classes. In one particularly frustrating computer science class, the professor refused to reveal answers to homework assignments when they were over (so students could, you know, learn from our mistakes), because then he felt he'd have to write new problems for future semesters. And after all, he wasn't really paid to teach; if he was the greatest teacher on earth but not a research asset for the university's brand, he wouldn't be there.

So to some degree, Caplan's right that even top universities are something of a diploma mill. And of course, many colleges are little but a diploma mill.

Still, more of the classes I took in my public and private education taught me something significant then ones that didn't, and I turned out as the sort of student I think we want to produce. Of course, that means analyzing education in general a strange and difficult thing for me--the experience of most people is much worse than mine, and I don't know how much of that difference is me and my background (and genes), how much is a difference in the schools I went to (in progressive, intellectual, diverse Cambridge, MA), and how much is chance.

When Caplan talks about people not coming away from their education with knowledge of politics and literature and such, it seems wildly out of step with what education was for me. I certainly have tons of complaints about the quality of my teachers, and there were lots of times that I wasn't learning much in this or that class, but by and large I did learn in a direction and rate that I would happily apply to society as a whole if I could. Thanks to my elementary school, I discovered Katherine Paterson, J. R. R. Tolkein, Agatha Christie, Roald Dahl, creative writing, and the BASIC programming language. Thanks to my high school, I discovered political activism, The Autobiography of Malcolm X, Louise Erdrich, Eugene Ionesco, Man Ray, Huston Smith, Charles Dickens, Stephen Jay Gould, playwriting, theater production, electronic music production, photography, sexual orientation and gender identity, evolutionary psychology, the Pascal programming language, and reading the Bible as literature. Thanks to my college, I discovered newspaper writing, teaching programming, how to write a good essay, the nation of Georgia, the history of Iran, the Shahnameh, Virgil, Rachmaninov, Hume, Octavia Butler, Darryl Scott, Thaddeus Russell, my wife Kate, and how to tell an intellectually acute historian from a complacent one. (Bonus points if you can tell me which kind Alan Brinkley and Simon Schama are.)

Of course, I would have learned a lot and might have had just as much (or more!) smarts and skills if I had had less of a traditional education, or if it had been just mildly supervised, personally directed learning. Likewise, the kids who he's talking about who are graduating without skills or knowledge, or dropping out of college, are mostly people who would not be pursuing knowledge independently.

That's what I feel is most missing from his tone and approach: a recognition of the full landscape of the problem and all of the foolish traps that lie in the direction Caplan is pointing to. It's easy to look at the current direction and point out the existence of failures. And it's fine, even necessary, to make causal inferences, to some degree, that connect the current direction to those failures. But if you're not anywhere near the stage of actually trying out alternatives and trying to leverage their lessons for large numbers of people, then at some point these denunciations move past intellectual inquiry into something more like polemical masturbation.

Of course in the self-branding, publishing world, you need to be a promoter of a simplified view of your ideas, so I'm not saying Caplan can't have a more circumspect and skeptical approach in his actual work than he does in a brief online video! It's just that an awareness of these limitations feels glaringly absent in this interview, and in most loud denunciations of the education system.

That said, I basically agree with the incremental direction he's advocating: towards more student self-direction and focus on apprenticeship.

(And I think it should probably be easier for kids to drop out, and in fact to be kicked out, from school at a younger age; I think the simple fact that students haven't actually chosen to be in school, or at a particular school, or in a particular class, is a huge barrier to their engaging at the level necessary to improve themselves.)

I just think school in that apprenticeship-and-play direction, or in a more radically decentralized and non-coercive direction, has almost all of the same problems with outcome as school now. (I don't think it's the fault of school that there is such a high correlation between books a family has at home and long-term earnings of the children.) I've visited, or sent to my children too, many schools that have massive amounts of undirected time. I've also visited "no excuses" schools with a much more constantly active, regimented focus. My personal sense is that kids were about as happy and learned at both, and I don't know how much of a blessing it would be to liberate students from no excuses schools and send them to self-directed schools instead.

I appreciate Kaplan's looking at the outcomes of students with different numbers of years of college education in order to illustrate how much of the value is just in the signaling of the degree. But I think he could similarly look at the spectrum of educational approaches currently in play. I don't think there have been better outcomes from schools, or state policies, that are incrementally more in the direction that he recommends. He might argue that much of the value of his direction won't show up in earnings and test scores; I would say the same about college.

I'm not a big fan of Diane Ravitch's overall polemical writing, but a few years ago I read and enjoyed her book The Death and Life of the Great American School System. The thesis is that there have been an endless stream of hyper-confident, publicly lauded reformers who have no shred of doubt that they know precisely which changes will obviously improve the education system. One after another, they have done their best to massively intervene in local interactions between children and the adult community members that attempt to educate them; and one after another, these reforms have flamed out and left a wake of destructive disruption.

Ravitch advocates much more caution before presuming that we know that revolutionary new education systems would be better than the systems they replace. This humility is frustratingly rare; there is seldom any reflection from the reformers or their backers that the confidence of the eternal reformist project itself may have been part of the problem. The conversation is instead about which other reform must be the unquestionably true, obviously correct direction.

This is all to say that, warts and all, the collective knowledge and wisdom that has gone into constructing the current system is much greater than your wisdom or mine. And the collective flaws and foolishness are much greater than yours or mine. I don't just mean that in aggregate, since obviously the number of people I'm comparing is wildly different! But I mean taken as a whole, there are thousands of tiny pieces of wisdom and policy correctives to incorporate empirical observation and adjust for unintended consequences.

(The same could also be said of our deeply flawed political and legal system--we might be quite unpleasantly surprised to find out what a popular revolution in America would give us instead, as we are somewhat finding out with the Trump quasi-revolution.)

There are certainly better ways we could be doing this, especially if you afford those alternative ways their own history of accumulated wisdom; such as when you compare our system to other countries'. The inertia of a current system is not a reason in and of itself to avoid changing it. But, that inertia contains information about the nature of attempts to change it that we are being foolish if we do not use to inform those attempts.

If Caplan has an inkling of these complexities and the importance of that sort of wisdom, I don't detect it in this interview.

Labels: , , , , ,

Your own Chris Hemsworth pet

In the Black Mirror 4th season episode USS Callister, it was surprising to me that they used DNA alone to create simulations of people when you could just do what they did in the early Black Mirror episode "Wish You Were Here", and scan social media to build a simulation of a personality and body.

What's especially interesting to me about that scenario is how it doesn't really need to perfectly mimic the person. As long as it seems like a person and it matches them closely enough, I suspect our empathetic brains would fill in the rest and we would perceive the full human. Of course we will need to develop AI to the point of general (human equivalent) intelligence before we could really fall in love and reason with such a simulation, but if all you want to do is use them as a prop and exploit them, we are probably very near that technological point already.

Which means, if you think about it, you could have your own Chris Hemsworth or Jessica Alba VR pet very soon. Some of these might be deliberate commercial products, but others will be totally unauthorized products of social media stalking. Have an old flame for an ex? You're not far off from being able to create a simulation of them convincing enough to have virtual sex with or even abuse. Checking out someone on the subway with your hacked HoloLens glasses? Now you can own them forever. (Especially easy since you don't even have a real sense of their personality to compare against.)

I think this sort of thing may enable a flourishing of inhumanity and socially dysfunctional rabbit holes.

Labels: , ,

Thursday, February 01, 2018

Biopics that would blow people's minds

Biopic screenplays that would blow people's minds:

* Oney Judge, who escaped from George Washington and was hunted by his goons

* Josephine Baker spying by night

* Paul Robeson singing opera, receiving law degree from Columbia, being persecuted by US gov't... while playing in the NFL

Labels: , ,

What skills are needed to be a Quant?

On Hacker News, a user asks, what skills are needed to be a quantitative trader?

My answer:

From many people I have seen succeed and fail at being quants (in the high frequency trading realm, which is different in many ways than the derivatives analysis world), you don't really need any financial background, besides being a thoughtful and reflective thinker who has naturally wondered and thought about how finance works on the national, business, and personal level.

In fact, I don't think you necessarily need to have background knowledge of anything in particular. What you do need, absolutely, is the ability and interest to learn complex concepts and areas of expertise in a diligent and meaningfully insightful way. That is, you need to be something of a Feynman-type thinker, learning statistics and programming and the math of data analysis and algorithmic analysis truly from the inside out, so that if you taught any of it to other people you would be a phenomenal teacher.

If you're not quite sure what that means, consider teaching a statistics class by having the students work their way chapter by chapter through a statistics textbook. Now say that you randomly insert 10 errors in to the textbook: you switch one word for another, you misuse Bayes' theorem in an example, you forget to adjust sample standard deviation to student sample standard deviation, you leave out a crucial paragraph of explanation in a lecture, etcetera. And you don't tell students to anticipate this and point it out.

How many students at a place like Stanford would catch most of these errors or omissions and speak up about their confusion? How many students would already be putting in the consistent, focused, diligent effort so that they could be reasonably confident the problem wasn't just in their laziness or inattention? How many would care so much more about understanding the material than about potentially embarrassing themselves to interrupt you in class?

If you taught that class 10 semesters in a row, in all that time I doubt there would be more than a handful of students who met that standard. If you took those students, with no particular background in finance, math, computer science, or statistics, and put them to work as a quant, it's highly likely they would succeed.

Whereas if you took students who never would have raised a question about any of those errors or omissions and gave them years of experience in all those areas, it's highly likely they would not succeed as a quant.

Of course, you certainly will need to rapidly develop a background in these disciplines, but you'd be surprised how quickly a focused study of finance allows you to surpass the financial knowledge of many professionals.

(Background: I worked as an algorithm developer at a major high frequency trading company for 6 years. Some of the most valuable employees, who generated tens of millions of dollars of value for the company, started with only spotty knowledge of finance, statistics or computer science.)

Labels: , ,

"Innocent until proven guilty" is a lie

One common misconception of the comfortable political center is that our justice system presumes innocence. But innocence until guilt is proven is not the practical case for most offenders.

For one thing, when a defendant pleads guilty, guilt is *never proven*.

For another, there are a large number of guilty pleas by innocent people.

And for yet another thing, many criminal convictions rest on no evidence except for police eyewitness testimony, which does not constitute proof.

And for yet one more thing, police testimony has been shown to be likely false in at least hundreds of thousands of cases.

What's most interesting to me is that the assumption that our justice system presumes innocence is taken as an article of faith, quite unlike the assumption, say, that politicians represent their constituents. That is taken to be an ideal principle, sometimes met, but it's not presumed to be met, by and large; even in polite society, everybody knows that politicians frequently represent their donors, entrenched wealthy interests, organizations that get out the vote (like unions), and themselves.

I consider this sort of misleading assumption a "Brinkleyism", after centrist Columbia history professor Alan Brinkley, who muddles all sorts of forgiving assumptions about the workings of power in the United States with the official story that the U.S. power structure has constructed to guide thinking away from understanding its workings. Although you could also call them "The Economist-isms" after the venerable, self-congratulatory, and frequently obfuscating London publication.

For instance, the assumption that police must get a search warrant to force entry to a property, or the notion that Americans didn't like being ruled without representation by King George, or the idea that the U.S. government has long fought against terrorism, or the story that the U.S. government opposed socialist governments taking power for humanitarian and democratic reasons. The more you learn about how politics and justice play out in real life, the more you realize how misleading, inadequate or flat-out wrong these centrist presumptions are.

Labels: ,

Thursday, January 25, 2018

Thoughts on using Airpods on Android

I have an Android phone--a Pixel 2--but I recently got Apple Airpods, after failing to get Bluetooth or wired headphones to work properly with my pretty good so far.

Observations:

Pairing is so much better than with my old crappy Aukey BT headset. I've never had this good an experience pairing anything with Bluetooth ever.

You can barely hear them on the subway. I suspect the speakers would be better if the design pushed them a bit into the canal (or, obviously, given a rubber closed canal sheath, but I understand that's a distinct direction with its own tradeoffs). My guess is that if they could have that functionality without doing what everyone else does and making them look like something from eXistenZ, they would :) On the other hand, the open air design makes it easier to just leave them in.

The double tap thing is totally inconsistent. A lot of the time it just doesn't work. Could be my pixel's fault, though it very consistently responded to the Aukey's single button.

Taking them on and off is a bit of a chore compared to the pixel buds-type necklace design. Does anyone just pocket them, without using the case, and leave the case plugged in at home?

The charging case does make running out of juice way less of a thing. It's just nicer to plug them in than the direction specific, fiddly micro USB port on my Aukeys. Apple does power capacity so well (when it isn't secretly using it to prematurely render its devices obsolete)

Sound sometimes flickers in and out on one or the other, which makes me think it's the Airpods' fault.

They produce that "duh-dumm" event sound at kinda random times sometimes, trying to figure out why. Might be my pixel's Bluetooth coming in and out.

Also, I ordered a black matte wrap for them for $15 so I can pretend I'm not a glasshole :)

Labels: ,

Wednesday, January 17, 2018

A few questions about basic physics

Here are a few questions I have about physics after reading (or attempting to read) several books.I can't seem to find consistent answers to these. I don't have the base of understanding necessary to ask good questions yet, but I hope most of these questions are at least interesting!

(Apologies for typos! My voice recognition seems to insist that molecules are "made up of Adams"...)

Section 1: About general relativity:

One of the starting example situations that often seem to be used in general relativity is this idea of someone who is traveling on a train or other moving base, who shines one light forward in the direction they're traveling and another light backward. 


There are two targets that are equally distant from them on the train, in the forward direction and backward direction. If a rider on the train turns both lights on at the same time, naturally the rider perceives the lights hitting those targets at the same time.


But if there is an observer who is stationary on the ground outside the train, watching the train pass them by, then because the speed of light will be absolute relative to them, the light can't be traveling forward at the speed of the train *plus* the speed of light. Here's where standard explanations say something like "To our outside observer the beams will be going at a different speed relative to the train than what the rider sees." The outside observer will observe the light hitting the backward target first, which in these examples is supposed to demonstrate the weird implications of the absolute speed of light relative to any observer.


There are a lot of different assumptions here that aren't being addressed, which I don't understand. First of all, the example above seems to mix subjective experience and objective descriptions of reality. What does it mean to say that "to" the outside observer, the light hits the backward target first? The outside observer doesn't have magic, objective-time goggles that can instantly communicate when the light hits a target. 


But they can see when the light that hit the target bounces to them and strikes their eye or camera or whatever. All they can do is receive reflected light. 


But suppose it *was* possible for light to travel faster than the official speed of light. That would mean it would be possible for there to be one objective reality between the two observers: the light is moving forward at the speed of light that the observer on the train observes, plus the forward speed of the train. The light moving backward is moving at slightly slower than the speed the observer on the train perceives. And the two targets really do got struck at the same time. But because the forward Target is farther away from the outside observer than the backward Target, the light that reflects towards the outside observer would not reach them at the same time. They would perceive the backward target to have been struck first. That's the order in which they perceived the targets to be struck by light actually does not demonstrate the absolute speed of light limit.


In this scenario, it really only depends on the relative position of the train to the outside observer when the targets are struck. If the train is moving towards the outside observer, and is relatively far away, they will see the light from the forward target first. If the train is past the outside observer, they will see the light from the backward target first.


But, dropping my counterfactual, that's the same outcome that's predicted by the general relativity example. Only in a very tiny range of starting locations for the train, almost but not quite to the position of the outside of server, do my counterfactual and general relativity differ on which target the outside observer will appear to observe first.


I ultimately Einstein is right, and I think I've argued myself back to a position of agreeing with him, but I'm just doing a ton of extra scenario building and going way past the point where I have any idea if what I'm saying is right or totally wrong. The simplified scenarios that are illustrated in example after example seem to me not to actually demonstrate the difference that general relativity predicts. Sometimes I wonder with some scenarios if, for information to return to all of the original observers, it's like it has to undo the effects of general relativity.


I read a quick explanation of the Michelson Morley mirror experiment, in which (so I've read) it was observed that light traveling in the direction that the Earth is spinning or moving around the Sun or something didn't move faster than light traveling some other direction. I'm sure I am wildly misstating this. But the way it was explained to me seemed to have nothing surprising at all. Because if the light was indeed moving at the suppose it speed of light plus the speed of the Earth, but we are ourselves moving at the speed of the Earth, how would we perceive that extra speed?


Again, I'm sure that from the perspective of someone who really understands this stuff my questions are totally misinformed, I just can't fill in that context myself with the books I have.


So my overall question is, what is a scenario in which our conventional expectations are subverted by what we observe, and we can only conclude that general relativity is correct?


Question 1: is my description of the train scenario correct?


Question 2: what is a scenario in which are conventional expectations are supported by what we observe, and we can only conclude that general relativity is correct?

Section 2:

Regarding atomic resistance to contact:


I read in several places that as electrons in different atoms (which are each magnetically balanced) approach each other, they exchange photons. That is, photons are the mechanism by which electrons communicate their proximity to each other. 


Or maybe photons are the mechanism by which atoms communicate their proximity to each other. I don't really begin to understand. 


I've read that you could think of electrons using a metaphor of two people passing a dodgeball back and forth, where they're able to do it much faster as they get closer together, which means in effect their resistance to each other is higher. And as atoms get too close, their electrons push each other away and that pushes the whole atom away. 


But another thing I've read is that atomic nuclei just really don't like to be too close together, that there's a sort of proton exclusion principle or something, and that's why two atoms don't like to be pushed together. 


The source of my question is why is it, fundamentally, that two objects don't just merge as they collide? What is changing within them as they get close together? When I stand on a floor, I understand that the floor is pushing back at me with 200 lb of force. What I'm surmising is that that force comes from the compression of the floor, whose atoms are getting pushed together more than they like, and something is happening within them that repels those atoms from each other, and repels my shoe atoms from the floor, and those atoms repel my foot atoms, Etc. But I can't figure out which law of physics is the main one that's at work here.


A related question: I've been told that water is essentially incompressible, that it doesn't shrink and heat up as you compress it the way gases do. 


But say two people are standing on top of two big plungers, where the space under the plunger for one person is filled up with water and for the other person is filled up with air. I assume that what's happening within the chamber in each case is that the material is being compressed and creating resistance that pushes back against the plunger with an equal force to the weight of the person, at least as soon as the plunger initially depresses to adjust to the person's weight. What is the source of the force in each case? 


I've been told in a gas it is the rate of Brownian motion collisions from the gas atom bumping upwards against the plunger. But I gather that's not the case with the water. If not, what is it? Are there two totally different laws operating in the two cases? Or is it more a matter of degree, where the sort of atomic explosion I've been talking about is indeed still at work in the gas and that's why gas atoms bump each other and communicate motion, and similarly there is a bit of Brownian motion in the water as well but it's just not a significant way of describing the water's resistance to compression? Is there anything relating to that electrons playing dodgeball with photons idea that makes sense here?


Question 3: why is it, fundamentally, that two objects that are pressed together don't just merge or pass through each other?

Question 4: is my description of the Brownian motion of the gas producing force correct?


Question 5: by what atomic-level mechanism is the water producing upward force against the plunger?


Question 6: what happens within a single atom as another atom (which it doesn't need in the sense of forming a molecule with it) approaches it?

Section 3:

My next question is about reflection of light. I know there are some aspects of quantum mechanics that it's foolish to expect to understand intuitively, and maybe this is one of them. From what I (shakily) understand, incoming light is being reflected by a mirror at all sorts of angles, and the only reason we perceive an angle of light shining on a mirror at 10 degrees as also reflecting off the mirror at 10 degrees (or, rather, 170 degrees) is that the average of the possible angles of the reflected light is 10 degrees. I'm sure I have said that wrong, I'm not sure if this is an issue of quantum mechanics or of the wave/particle nature of light, or both.


I'd love a good explanation of that aspect of reflection, but I'm trying to keep my questions a bit more focused. So instead, I want to ask:


Question 7: when a photon hits an object, say a speck of green paint, what happens within that within that atom that causes a photon to be sent back out? 


Question 8: in a very light absorbent substance like vantablack, what is happening when one of its atoms is struck by a photon?


Question 9: is it correct to assume that a surface a single atom thick would still, in theory, be capable of reflecting light at a complementary angle to the one the photons came in on?


Question 10: on the atomic scale, why should a photon that makes contact with the atom at one angle be sent out at the same angle?


Question 11: say you had many such single atom think surfaces, reach oriented at a different angle. You shoot a stream of photons at each one, striking each one at a single atom. How do the photon and atom know what angle to send the photon out at? Isn't the scenario identical, for all intents and purposes, for each of these atoms being struck by a photon?


About the two slit experiment, I don't understand enough to ask a good question. I guess I fundamentally don't really understand what we're talking about when we say light acts as a wave. Sometimes it seems as though the concept of it being a wave is something far beyond our ability to have an intuitive understanding of. Other times, it seems practically Newtonian and straightforward. For instance, I get that some substances are transparent to some wavelengths of light and opaque to others; and that's like a measurable thing, you can imagine AM radio waves hitting all kinds of obstacles in all there going back and forth that higher bandwidth FM waves sail past. Or with polarized filters, you can say only light that is moving back and forth along some particular rotation angle will get through. All this has given me the sense that light waves involve light photons essentially moving up and down as they travel forward, varying their position relative to the average center of their direction of travel by inches or even feet.


But this understanding doesn't seem like it can be right. Because you can't like catch a photon several inches away from the center of its vector of travel, right? Because it seems to be that would mean our eyes would have no idea where any of the photons it receives are coming from.


And then you go to the two slit experiment and suddenly we're talking about light waves not being a phenomenon centered along one direction of travel, but something where the waves are a more elusive concept than the somewhat concrete sense I get from the ease of polarization.


Anyway, I'm far from being able to articulate good questions about the two slit experiment, or even just polarization. I just know there's a lot of background that I don't understand!

Labels:

Wednesday, January 10, 2018

Infinite music hasn't been written yet

I can appreciate Radiohead's plagiarism case against Lana Del Rey (or rather, their publisher and production team's case against her publisher and production team). (Though the offer of 40% of the writing proceeds seems like a more than fair offer, considering how changed the song is AND the degree to which they likely borrowed for their original. Rejecting that and using our tax money to try to get more is disgusting.)

A friend of mine, skeptical of the series of court cases where musical writers claim ownership of similar songs, wondered "how many pleasing combinations of chords or notes can be created before everything that follows might be considered evocative or a remix of its predecessors".

But a musical composition is so much more than just a sequence of notes on paper. There are so many dimensions along which compositions can be placed: multiple instruments, voices, pacing, things which brilliant artists understand on a subconscious level but which we are a long way from even being able to formalize. It's akin to the breadth of expression possible with letters. Any order of letters seems obvious in retrospect, and billions of words are written every day, yet no one has accidentally recreated the plot of Don Quixote. I mean heck, two generations ago there are entire genres of music that had never even been imagined. There are surely hundreds more musical genres that haven't been imagined yet.

Lots of people have directly and indirectly taken inspiration from Don Quixote, much like, say, Yankee Hotel Foxtrot takes inspiration from "Revolution no. 9". But the space of possible stories isn't so crowded that anyone has written a story that accidentally also features a novel-besotted wannabe warrior leading a series of misadventures where he concocts patently false scenarios with the help of a more clear-eyed assistant in his thrall. Some people, of course, purposely write an updated DQ! That's my point: the world of stories is wide open, not crowded; accidental similarities are tenuous. When the similarity is higher, as with DQ's influence on The Idiot, City of Glass or The Moor's Last Sigh, it turns out to have been intentional.

When you hear extended musical similarity like Billy Joel vs. Beethoven, it's not an accident.

When someone has a musical innovation like, say, Teddy Riley's New Jack Swing, and it sweeps through the musical landscape, you see how much new musical composition draws from the field of existing ideas (and as opposed to coinciding with existing ideas through a sort of pigeonhole principle). You can practically see an idea like using voices as percussion in R&B sweeps its way through the memeosphere, from Bone Thugs 'n Harmony to Destiny's Child.

And much like with plagiarism in the written word, there's actually less gray area in practice than you might expect. So many turns of phrase and rhythms and patterns are borrowed from the landscape of subconscious reference, and brief deliberate copying. But then you have times that someone says the same eight words in a row that were in someone else's speech and it's 100% clear that the line was lifted altogether.

Similarly, in music there are tons and tons of times a sequence of 5 or 10 notes is going to match some existing song closely. I imagine algorithms are finding these for some clever troll as we speak! But the similarity in Lana del Rey's case goes way beyond the 40-odd notes (modulo an arpeggio or two) that match Radiohead's. The relative length of those notes, the relationship to the chords in the accompanying instrumentation, the story they tell--you'd need millions of parallel universes with no Creep before you'd get a song this similar.

It can seem strange that building specifically on others' ideas is so accepted in, say, the startup world and the world of novels, but so punished in music. Uber doesn't own anyone else's "Uber for X" (unless they get a BS patent). But think how much easier it would be to make, say, a good punk album where you just do a reworded version of a Bon Iver song here, a Cat Power song there, than to really write those songs from scratch.

The world of possible music composition isn't crowded. The world of possible stories isn't crowded. Most great music and most great stories haven't been written yet!

Don't be fooled by the relative stagnation of pop music in the last 15 years. Infinite music hasn't been written yet. More spookily, infinite music will never be written. It's not crowded in there--it's lonely!

Labels: , , ,

The communications tool I wish existed

I've daydreamed for a while about building a Swiss Army Knife messaging-management software-as-a-service.

Basically, I've been paid to develop messaging-manager software by several clients, and approached by other clients who I didn't have time to work for. They're using Sheets, Excel, Google Contacts or Salesforce, and that's all fine for storing lists of contacts, but they want to be able to track who got which email or SMS message when, how they responded, etc.

I've heard "Why can't I email all my active clients and tell them about the new website? I have them right here in Salesforce!" And, "Why can't I text all the seniors who haven't turned in their graduation requirements that they're due in 3 weeks? I have the list right here in Google Sheets!"

I'm not really sure what v1.0 would make sense for this, but I've had multiple clients swear to me that they would use it even if they had to import their contacts every single time they wanted to send a message. (It feels like I'm taking crazy pills when I hear this--surely it exists? But it doesn't, as far as I can tell... and I've tried a bunch of Mailchimp-family SaaS's to try to help them find one that works.)

What I've tried so far:

  • Salesforce: AFAIK, has no built-in ability to systematically contact the contacts; expect a GUI nightmare, or a lot of coding from scratch, if you wish to take actions like adding a field using the output of a query
  • Mailchimp: when I asked them how to maintain the information I'm gathering about communications with my Mailchimp contacts, they suggested regularly exporting them to a spreadsheet, updating columns, and re-importing the list.
  • Sendgrid: works well for what it is, which is very limited. Also, they confirmed to me that they discard your entire history each week, and won't stop doing this at any price
  • Mixmax: I'm a paying user and their customer support is great, but I find the Gmail integration janky, contact syncing inconsistent, and the various campaign configurations impenetrably confusing
My feature wishlist:
  1. Gmail: I can still use gmail as my primary email client
  2. Google Contacts: it syncs with my google contacts
  3. Google Sheets: I can point it to a google spreadsheet and it can figure out how each row connects to a contact (eg, by email, or name, or whatever)
  4. Static fields: I can add arbitrary fields to contacts, like “customer lifetime value” or “has been sent invite” and it’ll remember the values for each contact forever
  5. Tags, not lists: Instead of each list/group maintaining its own info about a given contact (as Malchimp does), the assumption is that a given contact belongs to many groups/lists/tags. Google contact groups are just treated like tags; google spreadsheet columns can also be tags, and I can create custom tags.
  6. Static tag info: each tag has its own properties, including associated “why you got this” text
  7. Exploding addressee lists: I can compose an email to thousands of people and it’ll make each one appear to have received a personal email from me, without anyone else included
  8. Safe templates: my email can use a template like "Hey there ,", and it will either let me define what to show if there is no firstname, or flag all that don’t have an obviously good entry for each field used (so there’s no awkward “Dear Mr. ,” blanks)
  9. Exploding phone number lists: I can compose a text and it’ll send to 1 number for each person
  10. Detect bounces: it’ll detect bounces from email, SMS and mark those addresses/numbers to not be used (and tell the difference between "address not found" and "mailbox full")
  11. Detect dupes: it’ll detect duplicate addresses, so that each person only gets one email total (using Google Contacts lists in Gmail violates this, I believe)
  12. Why you got this (email): it adds “why you got this” text to bottom of email
  13. Unsubscribe (email): it adds unsubscribe text and link to bottom of email
  14. Why you got this (SMS): it adds “why you got this” text to bottom of initial SMS
  15. Unsubscribe (SMS): it adds unsubscribe text and link to initial SMS
  16. Read receipts: it has email read receipts to show me if someone read it
  17. Templates depend on medium: I can make distinct email text and sms text for the same message
  18. Failover to SMS: I can send a message once without specifying email or SMS, and it can send SMS if email is not available or bounces
  19. Salesforce: it can sync with salesforce
  20. Live Salesforce queries: tags can have associated salesforce queries that pull in a set of contacts live
  21. Process responses: I can include questions in the message, and it can store responses
  22. Response parsing: message responses are automatically processed when relevant, like “Hell no” = no, “6 out of 10? maybe 7?” = 7, “*** 1/2” = 7, etc.
  23. Saved searches: search results can be stored as a column: eg search result list for “have received 0-2 emails from me” becomes: new member == 1
  24. Live saved searches: search results can be stored as a live/self-updating column (updates after every message send or data change)
  25. Responses are a column: message responses become a column that can be used in logic, like “all people who haven’t responded or responded ambiguously or responded with a no"
  26. History lives forever: I can look up the history of communications with any recipient, I don't need to worry about storing it myself
  27. Set math: I can get a set of all contacts in various combinations of groups/tags, using logic like or, and, in B but not A, etc.

Labels: ,