Thursday, October 20, 2016

Bush v. Gore v. Trump

Gore DID accept the 2000 election outcome--by allowing the government of FL to proceed in determining the winner.

It was *Bush* who brought the two suits to SC to reject FL outcome. That's why the suits are called BUSH v. Gore and BUSH v. Palm Beach County, with Bush first. (Like how Brown was the one trying to put a stop to the status quo in Brown v. Board of Education.)

Bush and the conservative Supreme Court literally intervened to refuse to allow Florida to recount its votes according to its interpretation of existing law, a decision that Scalia justified in part because he argued that the recount would *cause Bush harm* by "casting a cloud upon what he sees as the legitimacy of his election". Well yes, especially if Florida's electoral process resulted in the outcome that Gore had more votes (as the 2001 consortium of news orgs and statisticians determined he did), it would harm Bush and cast a cloud of non-presidency :)

Bush, conservative elected officials and conservative judges repeatedly rejected the process underway and *imposed* outcomes.

Gore could have brought suit in response in an attempt to reject the outcome, like Bush had multiple times already, but instead conceded, asking private citizens to refrain from pushing to allow Florida to determine the outcome of its election.

So Gore not only accepted the outcome where Bush, conservative elected officials and conservative judges would not, he accepted *another* outcome too, one that the state government of Florida believed prevented it from determining the actual outcome of its election, and which the best review we have determined likely *did* prevent it from determining the will of the voters.

That's double the acceptance of the outcome of most presidential winners or losers!

Labels: ,

Tuesday, October 18, 2016

Rock criticism vs. Jazz criticism

Loved this comparison of rock and jazz criticism from an interview with the NYTimes's departing music critic, Ben Ratliff:

"The philosophy of rock is essentially founded on the idea of killing your parents, breaking with the past and creating something entirely new: being shocking or disorienting. What we think of now as rock criticism, at least in its sensibility, basically starts with punk and is fairly hazy about what came before it. Anything that seems allied with punk in some way is basically true and right. Somehow that includes Bob Dylan. It also includes John Lennon, but not Paul McCartney. The culture of rock musicians buys into that and the culture of rock critics buys into that, so there’s unity. You’re absolutely right, it’s astonishing sometimes to hear rock musicians talk with knowledge and insight about Lester Bangs or Greil Marcus or whomever. They care what those people said and thought.

In jazz, there’s a fundamentally different understanding between critics and musicians. Many critics are futurists, a slightly less violent cousin of punks: they believe it is the music’s need or destiny to break away from the past and reinvent itself. And many jazz musicians believe that continuity with the past is extremely important."

Labels: ,

Monday, October 17, 2016

Recommending Cory Doctorow's sci-fi writing

Several friends have asked me about Cory Doctorow's Down & Out in the Magic Kingdom.

I recommend it!

  • As great Literature, i give it a C
  • As a quick, entertaining read, A
  • As thought provoking social sci-fi that I think about often, A
  • As having an ineffable sparkling spirit, A
  • As a sampler of the writing of someone whose work you might enjoy continuing to read, A+
All his books are free on his website,, as well as available in various traditional formats for money. Here's the html version of Down & Out.

Some find Doctorow too "light" compared to, say, Ancillary Justice, but I really enjoy his imagination and grounding in political and economic issues. Doctorow worked for the EFF for many years, and is responsible for much of the venerable nerd blog Boingboing's political voice.

I also deeply appreciate his constant push to challenge himself and grow his art. Some of my favorite reads have been his ambitious novels that, in my opinion, fail to build a cohesive and effective whole, but are interesting failures. After D&OITMC:

  • if you like his density of tech ideas, you should read his short story collection Overclocked
  • If you're into the political/tech activist edge, you should read his radical, breathtaking young adult book Little Brother
  • If you're into the ineffable sparkling spirit of D&O, you should read his collection A Place So Foreign, which has stories about, eg, an alien collector of earth memorabilia, and Superman becoming political
  • If you're into the future economics/culture of work stuff, you might try reading his ambitious failure Makers, which is a near future tale of a startup that tries to use 3d printing as an engine to reinvent work
  • If you want to read his most literarily ambitious work, which has one foot in magic absurdism and one foot in startup culture, you might try his Someone Comes to Town, Someone Leaves Town. It didn’t really come together for me, but he’s trying something difficult and builds a unique fictional world.

Labels: , , , , , ,

Sunday, October 16, 2016

Thoughts on Cancer

I had testicular cancer in 2004.

Short version: I noticed a hard long lump (2 inches long!), they took it out.

I had the good fortune to find M. Mendel Shemtov, a phenomenal surgeon at NY Presbyterian who dug into every possible detail with me. I'd met with a doc at Sloan Kettering, but found the entire experience there to be inattentive, breezy, and not confidence inspiring, so I paid thousands more to go out of network with the NYP guy).

It's a good kind of cancer to have because your testicles are so separate from the rest of your body, and they can take one out and you're still 100% fine! As my 2 daughters since then attest :)

There was some possible spreading of cancer cells to my abdomen, so I had 5 awful weeks of radiation, and threw up every day.

Directly after my first radiation, I went to see Prince at Madison Square Garden. No one had warned me that I might be more than a tiny bit nauseous. I ended up racing to the bathroom 3 times to throw up, racing back each time because I didn't want to miss a thing. I can believe the claim that Prince is the greatest performer of all time.

Obamacare has been a huge relief for me -- I've been forced to move and take jobs only because if I didn't, I'd lose insurance and be denied forever as having a preexisting condition.

Now at 11 years free of any sign of cancer in my scans.

The common narrative of cancer is that it clarifies your values and will to like. I generally found this not to be true for me. (Susan Sontag's Illness as Metaphor often The whole experience didn't change me at all, but then i thought back some more and reconsider.

I do think it made me kind of assume in the background that I might die at any time. But sometimes that's not great for the day to day. A lot of the best work I've done has come when I'm 0% thinking I've got to go Something That Matters with my life, and instead am just freely creating something I love. Thinking from the perspective of my death bed can be kinda paralyzing.

Pretty much as soon as I got healthy, I got the offer to go to Georgia and work for the new president for the Georgian salary of $12k/yr, which I was totally unqualified to do. At the time I had almost no savings, with a $550/mo health insurance bill that I couldn't miss a payment on no matter how little I earned.

I was terrified to do it, but I think the brush with death helped me have total certainty that I had to go. And that decision (and taking it seriously every day when i was there, even when I had nothing but doubt and fear everywhere i looked) kind of opened a whole door in my life.

Having kids has also done a lot to get me better at being present without a goal. Camus' end to The Myth of Sisyphus seems to describe what time spent with them is like: "Each atom of that stone, each mineral flake of that night filled mountain, in itself forms a world."

Labels: , , , ,

Monday, October 10, 2016

How to fix a slow Mac: painfully obvious, yet painfully obscure

After my Mac Mini's sluggishness wasted many hours of my time and cost me hundreds of dollars in missed consulting dollars over the last month, I was fed up.

I spent probably 10 hours over the last few weeks struggling to understand the problem, testing the mac, learning about the reasons it might be slow, comparing comments about possible solutions, running software... and now that I know all this nonsense, I think the correct path is actually super clear:

  • Problem: my Mac has physical hard drive, which is too slow (it was actually really hard to figure out how to read activity monitor to understand the HD was the constraining factor).
  • Solution: upgrade to SSD (solid state drive)
  • Any real need to actually replace my Mac? Ran benchmarks and compared to current place in Apple upgrade cycle, answer is upgrading Mac would be a huge waste of time and money compared to just upgrading to SSD.
  • Defrag? Repair permissions? Blah blah blah? None of that stuff really matters if it's just slow and not totally unusable
  • Which SSD? Actually really obvious, absolutely no doubt it's the Samsung 850 EVO 1TB right now, with this Inateck enclosure
  • Install it internally or externally? Lots of research showed me external should be totally fast enough to not be noticeably worse than internal; and if I change my mind I can always go internal later. (there's tons of whiny misinformation about external drive speeds.)
  • Boot setup? Should clone disk to SSD and boot from it.
  • How to clone? Here's how (I couldn't find these steps presented clearly anywhere, had to cobble them together):
    1. Use Disk Utility (Applications > Utilities folder) to format the SSD, erasing it and formatting as “Mac OS Extended (Journaled)” with a single GUID partition.
    2. Use Carbon Copy Cloner to clone your hard drive to the SSD. If you are running Lion or Mountain Lion, choose the option to copy the Recovery Partition to the SSD.
    3. Restart your computer whilst holding down the option key and select the SSD as the boot drive. You’ll know in a few seconds if your clone was successful.
    4. Go to System Preferences > Startup Disk and select your new SSD as your startup drive.
    5. Use Trim Enabler to enable TRIM on your new SSD.

Now, maybe there is some tricky judgment call in there that I'm not appreciating, but I really don't think so. I really think all of this is the only reasonable course given my incredibly common situation.

Yet in the course of researching this, I had to step around countless debates. Do you need TRIM? Is an external SSD fast enough to boot from? How painful is the process of installing a new HD in your Mac Mini? Is it necessary? Is USB 3.0 fast enough? It's not that there aren't reasonable people who disagree on some of this, but the appropriate answers for most people should be very clear. (Respectively: yes, yes, doable but pretty dicey, no, yes)

Could I have just ordered a new mac and swallowed the money and saved myself time? Well, part of the problem was I didn't know what the source of the problem was at the start of this... not that I'm 100% positive of my new understanding, but it seems pretty clear what the problem is once I ran cpu benchmarks and learned how to read Activity Monitor right*. The other part is that Apple's upgrade cycle is way behind schedule, so if you assume you only get about 4 years out of a typical computer before it's too old to run current software, you're sort of paying double the cost (in both dollars and software install hours). A new Mac that would make me confident of its performance would be $1500-2000, or about $4000 in real cost for me considering the situation. I estimate that doing nothing for another year would cost me $2000 in lost time and business.

Instead, it turns out that $320 and an hour of your time can get you basically a new Mac, with all your same files and no need to reinstall apps, and the ability to make those files available to a new Mac in seconds (by plugging the drive into it).

This kind of info should be easier to get! My guess is that a large number of Mac users out there are in exactly the same boat. Yet each of these thousands of people either has to learn this diagnosis and solution on their own, or never know it and lose out.

How can this be better? Is there no on-demand hirable replacement for my judgment and expertise? How many people out there could have figured out what I need in 30 minutes on the phone? Even the right blog post would have helped me. But no one out there is incentivized to collect this advice and provide it to people at large. If someone wrote this and it worked for me, I might like upvote them on Stack Overflow, but that's it. Is that really the only incentive we can muster? Isn't there some way we could credibly make our expertise available for occasional hire?

(* I thought low HD usage meant low HD load, where it really means low HD throughput. The drive may be choking as fast as it can through a massive backlog of read and write requests, which are not shown at all. This makes the disk view in Activity Monitor totally different from CPU and RAM views, which basically show load and not performance, which is more or less constant for processors and memory. Often the variations you see in disk usage are due to the size and placement of the RW requests, so it might say 75 reads at 3MB/s, then 400 reads at 4MB/s, while the disk could be said to be running at 100% possible throughput the whole time. There may have been 300MB of requests, but only 4MB/s of them are actually getting delivered.)

Monday, September 26, 2016

Columbia University: the anagrams never lie

I came across this bit of idle collegiate musing in some deeply buried backup folder today... I only distantly remember writing it, and can't remember what for!

Can you spot all 8 anagrams of "Columbia University"?

By miraculous invite, you entered -- thinking the Ivy League promised the world. You now find yourself in perhaps an unsuitable micro-Ivy with a miniscule trivia boy for a roommate. The best you can do is churn out papers and cultivate your nubile activism while waiting for the Ivy lubrication muse to come your way. In the end, you should be glad you're not stuck rooming with a vicious, burly inmate at some school where the alumni buy victories (or actually contribute). With unmalicious brevity, it'll all be over soon.

Labels: , ,

Tuesday, September 20, 2016

My life in live music

Prompted by a friend who listed the greatest concert experiences of his life, here are mine:

  • allman brothers at great woods in 1992 or 3, when I was 13... not a band i knew well, but such a tour de force performance that something special happened in the crowd. the first time i felt that.
  • wilco a few years ago at prospect park bandshell -- it was raining and that added to the magic. they just kept going and going and they always had one more flavor of music to explore.
  • the liza colby sound, an nyc funk/rock/r&b band with a phenomenal frontwoman (who did music for my wife's short film), last year at berlin in LES, just days after their amazing guitarist died... like 50 people crammed into a tiny space, just feet from the band, as they poured their hearts out.
  • some dj whose name of course i can't remember, at burning man last year, inside a giant art car shaped like a brain that had a dancefloor, but then you could climb through the structure up on top of it, feeling like a mile above the ground, and you had to like lock your feet into place and dance with the rest of your body and trust that you wouldn't fall off... maybe my greatest time dancing ever
  • the pixies reunion tour circa maybe 2008... in camden and hammerstein ballroom. just such a sense of simultaneous experience in the crowd. (also seen the breeders and frank black separately but not quite the same feeling)
  • rock the bells in 2008 at jones beach -- mos def and the tribe reunion were huge disappointments, but the wu tang clan and redman were phenomenal (they've been 100% tight and prepared both times i've seen them) and nas headlined and killed it, with jayz coming out for a song...
  • faith no more somewhere on lansdowne st. (boston) around 1995... getting a fantastic show when seeing a band I don't really know is one of my favorite things. they were amazing, and it's obvious in retrospect from that experience that mike patton would go on to be in like 12 different successful bands, the guy is a force of nature. (eg his collaboration with dan the automator )
  • prince at msg in 2004... i had cancer and threw up 3 times in the bathroom but raced back each time because i didn't want to miss a thing. i can believe the claim to be the greatest performer of all time.
  • ratatat at webster hall in maybe 2003, something of a mystery to me because they were absolutely incredible, and their music sounded nothing at all like their records, and i'm still not positive i didn't hallucinate this.
  • nine inch nails in 1995 i think, the last concert at the old boston garden, where trent reznor egged the crowd on and got people to start tearing the seats out... it felt intense and unpredictable and crazy.
  • arcade fire at msg in maybe 2009 - their anthemic stuff just fit so well, and it really felt like a gigantic singalong
  • tom ze (brazilian musician who i gather is something of a frank zappa-type prankster-absurdist, maybe bruno could place him much better!) at alice tully around 2012... just such an unpredictable and magnetic presence. Seeing him live was such a joy... he fucked with the audience so much, interrupted songs, switched instruments with people, it was like the backup singers were tolerating this crazy toddler. and at one point made NYC subway announcements into the lyrics of a whole impromptu song. He couldn't sit still. Vitality was just pouring out of him. It's still this model for me of how to be old and hold on fiercely to life.
  • regina spektor at carnegie hall in maybe 2010... just absolutely commanded the audience.
  • arlo guthrie around 1996 at sanders/memorial hall in cambridge, felt mournful like the ending of his generation... of course 20 years later he's still at it! maybe the most anyone has ever milked one album they made 40 years ago.
  • phillip glass's koyanasqatsi (sp?) performed live at lincoln center in maybe 2011, immersive and personal and terrifying and impossibly beautiful

    (and i won't include musicals, but i'll mention one opera, which i think qualifies more as music than theater...)

  • john adams's nixon in china at the met around 2012, made me feel something powerful at an opera for the first time... such a new and unpredictable musical experience

Labels: ,

Sunday, September 18, 2016

When a robot obeys your sexuality

Previous posts in this series:

A speculative article about sex robots in VICE (of course) has got me thinking about sexbots as a force for human obsolescence, and as an existential threat to humans:

It seems like scientists have been working tirelessly this year to reduce human interaction in almost everything we do, from pizza delivery to Uber rides, and now even realistic sex machines aren't far from becoming a reality. But unfortunately experts in the robosex industry have some fears that banging bots will be so good that it'll ruin people's lives, the Daily Star reports.

I'd thought about the future of sexbots vaguely, but always just as an extension of porn and masturbation. Now I realize I was making a "horseless carriage" error, like thinking television would be an effective way of broadcasting theatre.

This short article was the first time I'd been prompted to think about sexbots as a replacement for human relationships, either in full or in part. Their promise of superiority to many human relationships strikes me as having the potential to render huge parts of human contact and labor obsolete, in a way that mirrors the replacement of human labor by machines.

My mom has worked as a sex therapist and I've come to believe that lots of people who are in stable partnerships have no relational equipment to inspect and explore their sexual relationships and help them become good. How many matings today could be upgraded meaningfully if one partner was replaced by a machine brought in from a century from now? Or stated differently, how many children won't be born a century from now, because there was a meaningfully superior sexual option for one of their parents, or one of their grandparents, that didn't involve two humans in person?

Then there's this point:

"Sexbots would always be available and could never say no, so addictions would be easy to feed," Joel Snell, a research fellow at Kirkwood College, told the Daily Star. "People may become obsessed by their ever faithful, ever pleasing sex robot lovers. People will rearrange their lives to accommodate their addictions."

Humans are a reality check on sexual bizarreness, because to practically explore the depths of your sexual imagination, you need another human who's willing to go along with it; or in the most depraved cases, because you will be arrested for it. BDSM, for example, is known for its extensive focus on what you can't do, not just its making available more territory of what you can do. Even porn has real limits, because it requires either real humans or can only be animated. Moreover, it generally needs to be distributed, and thus can run afoul of censors, who are able to pretty effectively restrict some forms of porn from being distributed through anything but the darknet.

But sexbots will be in person, so there is no fundamental ability for their programming to be inspected, unless there is some comprehensive trusted computing police state as in Cory Doctorow's Anda's Game. Even if most users will require remote creators to provide programming, code is very hard to systematically inspect for the qualities of its resulting experience, and it won't be easy, anyway, to draw lines about what sexbot behavior is too socially destructive to be allowed.

So, then, what people who set their sexbot to simulate being raped? To appear and act like a child? To appear and act nonhuman? Some of this exploration will be fascinating and daring and creative, some will be routine, and some will be horrifically depraved.

And still, the greatest problem may not be the unleashing of abusive perversity, but the indulging of luxurious expertise. The AI in Spike Jonze's Her is able to present not just irresistible personality and charm to Joaquin Phoenix's awkward protagonist, but a seeming need for him. So will sexbots be able not just to manipulate human bodies in expert ways, but to affect desire. Sex with a sexbot will feel life-affirming and sensually fulfilling in a way that masturbation and porn generally are not, because it will actually be sex.

So prostitutes will be out of a job, though as with other such technological shifts, the act they perform--simulated consensual mating--will experience unprecedented growth.

But that's only what it's easy for us to see from where we stand; there will be deeper and dynamic effects that we are failing to imagine. What will it be like to go back to dating a human, after dating your own perfect ScarJo AI, with its data science-backed giggle, its paced moments of baring its soul, its sincere and deep love for you? What will it be like to go back to having sex with a human, after having sex with a lover more talented at exploring your mutual desire than any human in history?

And what will the ability to indulge extreme forms of human contact, without any form of the checks provided now by the requirement that other actual humans be contacted in the process, mean for humanity?

Labels: , , , , ,

Tuesday, September 13, 2016

God bless Cannabidiol, AKA "CBD" oil

I've had chronic arm pain, exacerbated by typing, for the last 9 years.

After all that time, I'm finally having some success with a drug for the pain. It's cannabidiol oil ("CBD" oil), an extract of marijuana that is not psychoactive and contains no THC.

A friend of a friend, a psychologist and journal editor, discovered it after her own long journey of testing possible solutions to her pain, and she has become quite the authority.

This friend found relief from smoking pot--something that never seemed to help me--and wanted to explore other organic compounds with medicinal potential. She enrolled in a course and began studying them, and testing them, and discovered that CBD was effective for her.

For me, it's a godsend. It's not night and day--I'd say it only means the difference between pain that's 3/10 vs 6/10--but the difference is significant and absolutely consistent. It takes about two days of once-per-day use for me to notice the difference, or maybe 3 days of not using it, for me to notice the difference, but the difference is unmistakeable. And believe me, I'm deeply skeptical of others' claims that various remedies are unmistakeable!

It's the best medical thing i've found in 9 years of looking, trying a dozen prescription drugs, seeing 70+ doctors/healers/physical therapists, having every sort of scan imaginable, and paying a network of medical students $1400 to review my case. Maybe the impact is not as large as the total effect of sleeping more, running and biking and doing yoga, but it's on a level with that.

Friends have asked if it is a mood thing, if it calms me physically because it calms me mentally. I don't think so, though there is some degree it makes me feel a tiny bit stoned in a way I don't really like.... not at all high, just a little bit subdued and out of it. It feels noticeably different physically, though that too could just be a product of somehow having a lower level of stress. I don't perceive a lower level of stress with the drug, but it could be there... since stress is hard to quantify.

It feels as though it makes my muscle and soft tissue less sensitive, tense and inflamed. It could be just masking the pain sensations, though I normally notice a cumulative effect from typing which doesn't seem to come back when the drug wears off, the way it comes back if I just take ibuprofen.

It is completely legal and available without a prescription, at least until the pharmaceutical industry lobbies for its version to be sold at a massive markup! I get mine from as "Folium hemp oil, either 24% or 70% concentrated (both work for me, if I adjust my dose accordingly).

I've found that about 50mg, taken sublingually and held under my tongue for a few minutes, is effective without making me too cloudy.

I'm told there are very few reported side effects, even long-term, and that WebMD states that oral doses up to 300mg per day have been used safely up to six months. Of course, I am not a doctor, and you should consult yours.

(Isn't that annoying? As if anyone really has a trusty career family doctor anymore who reads up on all the literature and doesn't just say the first overconfident thing that comes into his head)

Labels: ,

The few conservatives who make me think

For anyone looking for intelligent conservative commentary, I suggest the following. I have plenty of disagreements with many of these, some very deep, but they are often thought provoking, and have changed my mind sometimes.

  • Greg Mankiw's econ blog (and its greatest hits, i.e. the top Google results for his site)
  • Ezra Klein (not a conservative)'s podcast, where he frequently interviews conservative guests
  • Russ Roberts's podcast out of the Hoover Institute, which has been mentioned by a bunch of you before; wide ranging and curious, not polemical
  • Sam Harris's podcast and books on religious belief, fundamentalism, and skepticism; I think he feels he's not a conservative, but feels like liberalism left him behind in its drive towards relativism
  • Jonathan Haidt's blogging and articles on epistemology and open mindedness; again, not very conservative, but often argues in reaction to progressivism; eg "The Coddling of the American Mind"
  • Megan McArdle's blogging and articles in the Atlantic, though I feel these have declined in curiosity and rigor over time
  • Writings by FIRE, the foundation for individual rights in education; sometimes outrageously unfair, sometimes terrifyingly on point; "Coddling" was cowritten by FIRE's director
  • Reason magazine (libertarian, Thiel-ish, Koch-funded)--uneven but often interesting
  • VS Naipaul's A Bend in the River, a sort of founding conservative literary text that put in literary form the idea that radical revolutionary movements are forces of arbitrary power and triumphant incompetence, not forces of egalitarian freedom
  • JM Coetzee's Disgrace, a spiritual successor of sorts to Naipaul
  • Laura Kipnis's writing about sexual politics, eg her essay "Sexual Paranoia"

I've tried for decades to find good conservative writing, and I've subscribed to several conservative magazines and read all sorts of stuff I disagree with, and met David Horowitz (for a long one-on-one conversation) and Dinesh d'Souza (mostly watching him debate a friend).

Through all that, I really think there's something about conservativism, like communism, that precludes honest and curious thinking. I think some of the best evidence of this is the lack of nuanced and insightful conservative writers on subjects which aren't directly political.

Where's the conservative Emily Nussbaum? Without the insights of feminism and postmodern critique, she doesn't exist. Where is the conservative Daniel Mendelsohn? The conservative Michael Pollan?

What passes for intellectual rigor in a magazine like The American Conservative or The Weekly Standard is laughable. I knew several conservative writers-in-the-making in college, whom I came to consider sloppy and incurious thinkers; several of them were showered with journalism job offers, easily got book deals, etc., and now write for the big time, just as vapidly. (One of them, hilariously, wrote in The Weekly Standard a few years back that Obama was allowing the deficit to balloon through his horrible tax raises). You have to go to something truly peripheral like Jacobin to get the same level of lazy, shallow callousness on the left.

Alex Haley, a black Republican at a time when that didn't mean what it means today, did produce a spectacular book in The Autobiography of Malcolm X, which might be the best book ever by a conservative. No other American conservative-written political memoir on the right can compare to Richard Wright's Black Boy.

As for fiction, where are the great conservative novelists? Great literature has almost always been transgressive in a progressive direction, pushing against prejudices and superstitions and towards widening the circle of humanity. This is as true of great foreign literatures like Russia's philosophical epics and French introspection, as it is true of American travel bildungsromans.

And yet one of America's oft-cited "great American novels" stands out as thin and uninsightful: Saul Bellow's The Adventures of Augie March, which I think doesn't have a tenth the insight into America as Adventures of Huckleberry Finn.

I think there's tons that's valuable about conservativism; I consider myself to be significantly conservative, in the sense that I truly value growth, morality, civilization, tight social bonds, and protecting how much justice, wealth and stability there already are in the status quo; and in that I oppose measures that throw the baby out with the bathwater, that impose arbitrary constraints on people, that produce bureaucracies that perpetuate themselves unconditionally, and that experiment irresponsibly with our society and our world. But those values seem a tiny part of what animates most conservatives I encounter.

I think that American conservativism, as a coherent worldview and identity, has persistently violated those values, without owning up to it in a way that would help avoid further violations. Intolerable constraints on liberty have always been easily entertained by conservatives, so long as they are applied to someone who doesn't look like them.

Conservatives have ever been followers, and never leaders, in movements to expand the circle of human sympathy to include those previously excluded from it. And yet the intellectual case for conservatism today--as it has always been--is essentially that progressive forces have always been right about expanding that circle, and conservative forces always wrong, up until this precise moment, when conservatives are finally right and progressives are going too far.

I can point you to many essays by liberals and progressives about what their movements have gotten wrong about education, regulation, war, police brutality, etc. Where are the conservative criticisms of, say, how they were wrong for decades about gay rights, women's rights, civil rights, global warming, aggressive policing, war, etc.? These things seem to change in conservativism without much noticing and reflection, as evidenced by the Mormon church's continuous retcon of its past doctrines, recently requiring them to quietly scratch out the rule that black people can't enter heaven, and soon to require them to do the same for gay people. But there was no reckoning, and there won't be one soon.

Otherwise brilliant people turn spectacularly ignorant when it comes to this sort of imagination. Peter Thiel was the first gay person to speak from stage about being gay at the GOP convention; he shared the platform with several people who would have him undergo conversion therapy. Is that really the only thing his fellow speakers are wrong about -- the one thing he can't help but acknowledge sympathetically, because he is in the excluded category himself?

I think it's not a coincidence that conservatives (and on the left, communists) so often have holy books and figures of reverence, compared to liberals and progressives. Of course revering a human or a book means ignoring their many flaws, and it strikes me that the people I have known who most vehemently ignore such flaws have been conservatives and communists. (I spent quite a bit of time with various communists back in the day!)

The closest thing on the left is a discredited book of wishful identity fudging like I, Rigoberta Menchu. But how damning is a fraud like that? It certainly points to a hungry empathy among bleeding hearts, which can be turned to passionate causes that are thinly sourced, and that's a valid critique of progressivism. But there are plenty of other books that tell much the same story of working-class struggle and the violence that many people who fight against imperialism and armed state capitalism face. Make Rigoberta Menchu a white rancher facing FBI thugs trying to take her land to put up a parking lot and you have a conservative classic, albeit one better written than nearly any other.

I think you need epistemic closure--a firm limit on the bounds of what you allow your mind to entertain--to maintain serious belief in the conservative holy books I've read. Or, for that matter, to maintain serious belief that Lenin or Mao were heroes. It's not the belief in God I'm talking about; it's the belief that God wants you to revere a book and tradition that has been so nonsensical, deeply wrong, unjust, and absurd.

I love going to synagogue and singing about God, and don't begrudge others' beliefs and routines in which they find meaning. Some very intelligent people who practice religion are ready to apply their full intellect to considering whether Jesus really sent a demon spirit into a pack of pigs, and are happy to admit there's no chance that really happened--it's just a story some humans made up. But others will contort their thinking as much as they need to block themselves from honestly answering that question.

I want to be precise about what I mean: I'm not talking about people who simply say that when they pray or read the Bible they know nothing else and a different kind of truth fills them. I believe them, and I think that's how belief really works: it's usually a vague feeling and not a logical conclusion.

I'm talking about people who will turn to every last rhetorically ridiculous gambit to avoid directly owning up to their beliefs being inconsistent and devoid of rational grounding. When you get into the weeds of a debate, it is clear when someone is considering arguments curiously, and when their purpose is to bat away thoughts before they can risk entertaining them. It seems to me that among the conservatives and the communists I've known well, and the smarter ones I've read, this sort of blocking is nearly universal.

I'm sort of trying to thread a needle with my argument, not sure if it really makes sense. I think there is much that is valuable about conservative perspectives. And, at the same time, I think there is something within conservativism--maybe in its fear of destabilizing the safety and security that we have--that treats self-directed inquiry and questioning, and even imagination, as a dangerous and destabilizing force.

There's even a pride in refusing self-directed inquiry; this is prominent in Ayn Rand's fiction.

It's very hard to find simple contemporary conservative descriptions of the experience of someone who is outside the conservative sphere of protection: a gay person in 1980, say, or a slave in 1850, a time when progressives were generating many such documents of imagination. That's part of my point about the dearth of great conservative (and communist) fiction; it's indicative of the dearth of imagination.

As for liberalism, it is not without its groupthink, but I see liberalism as being fundamentally cosmopolitan and catholic (as opposed to orthodox) in its approach to truth. Liberalism takes what makes is valuable from conservativism, progressivism, and other lines of thinking, which is why it is able to adapt to become tough on crime or tolerant of gays; it is inherently anti-doctrinaire and open to being influenced by information. I really think that makes liberalism and conservativism, as whole outlooks, fundamentally different.

Read about the influence that good governance think tanks feel they were able to have with the Clinton White House vs. the Bush II White House, for instance. One was solicitous of perspectives and would revise policies in the face of outside criticism just to make them better in ways too obscure to ever translate to votes; the other was notoriously run by a closed circle who kept dissent at arm's length.

But I agree that there are counterexamples of curious conservatives, like the Hoover Institute's Russ Roberts, and I'm open to more that prove me wrong! What I most wish for is a smart, conservative counterpoint to The Weeds, the generally liberal Vox policy podcast with Ezra Klein, Matt Yglesias (one of my favorite thinkers), and Sarah Kliffe.

Labels: , , , , , , ,

Thursday, September 08, 2016

Why are neural networks especially hard to use in finance?

In response to a question on Hacker News about the use of neural networks in finance, I wrote:

Finance differs fundamentally from most other machine learning domains because new data diverges qualitatively from trained data so much quicker. Roads aren't reconfiguring themselves daily to to avoid being detected by a self-driving car; tumors aren't really appearing any different on mammograms than they were yesterday.

So whereas in most domains, a large, growing and representative corpus of data should allow a wide range of ML approaches to steadily decrease their error, in finance the error will generally increase over time for any particular classifier approach, unless it is a sufficiently dynamic meta-approach that produces new classifiers in an ongoing way as new data becomes available. (And even then, the new data is only marginally better than old data at predicting the nature of the market tomorrow -- it's still way more unpredictable than traffic.)

How do you produce a sufficiently dynamic meta-approach, to produce new classifiers on an ongoing basis that are actually good? You need to discount the value of large amounts of data, because that data is fundamentally less predictive than data in other domains; and you need to strenuously avoid fitting to irrelevant signals in the data. This is harder than just not overfitting in a typical way; yes, you need to not allow too many independent values in your model relative to the amount of data. But you also need to attentively inspect and guide the consideration of the signals that your model tries to use, even if there aren't too many of them.

Here's where neural networks, and genetic algorithms, are especially difficult to use in finance: they are more opaque to human inspection and selectivity, and so they lower the ability to discover elements of a model which are of suspect value. A less sophisticated approach like various regressions and decision trees, on the other hand, are easier to inspect and thus allow for more careful human-computer cooperative iterative design.

In a domain where the data aren't trying to squirm out of your hands, you can let testing error show you how to iterate in your meta-approach to producing classifiers; your classifiers will get better. But in finance, that iterative process may not improve your classifier production method fast enough to keep pace with the changing data, and so you need a process that requires less practical failure, and instead uses more human understanding and inspection, to guide iteration.

Labels: , , ,

Wednesday, September 07, 2016

Reading about Theranos and Holmes

Just read the Nick Bilton piece on Theranos:

And went back and read the pre-scandal New Yorker piece by Ken Auletta:

Endlessly fascinating. The inside story of how the fraud was maintained, which sounds elaborately designed so that almost no one could get a full picture from the inside, must be incredible.

It's also so tragic... in some ways I blame our winner takes all culture, and the silicon valley cult that ridicules slowness and patiently developing expertise, for making it probably seem, to Holmes and others, like the only imaginable paths were success at any cost and absolute failure.

Also, I thought Ken Auletta did well considering he was being conned, but whatever happened to his own test results? That should have been an easy thing to report, but he never mentions them.

Labels: , ,

Dissenting from Black Lives Matter

I enjoyed Sam Harris's recent conversation about racism police violence  with Glen Loury, a prominent black centrist/conservative whom I associate with the perspective of John McWhorter of Columbia.

They dissent from Black Lives Matter in eloquent and compelling ways, though as often happens, I was frustrated with the limits of Harris's rhetoric and introspection.

But Loury's rhetorical rigor and fairness are a revelation. He truly practices the "steel man" rule of articulating your opponents' case as well as you can.

So while I don't agree with most of what the conversants conclude, their articulation of the perspective of their political opponents is impressive and rare.

One of the aspects that frustrated me was Sam Harris insisting that (paraphrase) "people have to understand that these cops are not trained well in deescalation and how to use guns responsibly", as if that's not one of the main complaints BLM makes.

True, BLM does not admonish people to comply with police, and I think that is a valid criticism. But that decision comes rationally from the nature of their critique. They are not focused on short term harm minimization that might make a marginal difference to the level of violence that black people face, they are focused on demanding fundamental changes in the escalatory violence that often comes with impunity and is the police's default.

This is the story of critique that Harris would be happy to make, were it his cohort bringing the challenge. He is happy to focus with passion and reflection on anecdotal cases when it comes to jihadist violence, and rightly so. But the cases where police seem more ready to attack or kill a black person than to talk to them get only the quickest passing mention by Harris on the way to finding statistics to suggest holes in the BLM perspective.

I do agree that BLM and the left are making a huge mistake in not giving more attention to violence by black people, both as a meaningful danger to black lives in and of itself, and add a contributor to police violence. Certainly the single biggest change that would reduce police violence against black people would be for black people to stop committing violent crimes. When Ta-Nehisi Coates and others are presented with that type of argument, I find that they do not challenge its accuracy, but instead challenge the motives behind it and the validity of saying it. That is a rhetorical combo that I always consider intellectually flimsy, unfair and counter to understanding. I believe BLM is weaker for its inability to bring that fact to bear within its broader critique.

But returning to Harris, I also think it's intellectually flimsy to be as certain as Harris is all the time. He breezes by his "granted..." concessions, but it never occurs to him to actually focus legitimately on something like racism or police violence in and of itself... it's always only in service of dismissing criticisms of his racial, gender, political and economic cohort.

The fact that Harris always concludes that his cohort is being wrongfully criticized should raise a skepticism that he is arguing pragmatically in search of his prior beliefs, and not so much from first principles, like he thinks he is.

Harris so frequently lauds the winners for their moral superiority. He sees a sort of civilizational meritocracy, and I don't entirely disagree. But there's also a strong element of being born on third base and thinking you've got a triple -- and that third base is on top of a mountain of used up people whose pain you consider either unavoidable or too far in the (recent) past to be worth doing anything about.

Labels: , ,

Saturday, September 03, 2016

Her problem solver's name is Mook

Re: the Naughty by Nature line "My problem solver's name is Mook", my cousin Sasha observes:

"Hillary Clinton's campaign manager is named Robert Mook. Every time he is mentioned in the news, I think of Naughty by Nature and crack myself up: her problem solver's name is Mook!  I feel like one of the great unresolved loose ends of my childhood has finally been resolved. Of course if I had thought to look it up, it could have been resolved sooner: "

This line bedevilled me and my friends for years. I pointed it out to friends many times when Hip Hop Hooray would come on the radio; it was just too random a line to believe, and several friends have insisted we must just be hearing the line wrong.

Here's to Hillary blasting some Treach as her entrance music at campaign stops!

Labels: ,

Tuesday, August 23, 2016

Pinterest to pave Instapaper paradise and put up a parking lot

It's kind of a sad parallel to the incentives and economic realities that make your favorite local boutique or art shop get replaced by a Bank of America branch.

Not only is it hard to compete for space/talent with their money, they are additionally incentivized to aggressively expand because it's defense against good people/locations going to competing firms.

We like to fantasize that the basic currency of power in silicon valley is meaningful user delight, when it's really just a narrow kind of user attention hack...

...the kind that cheaply and emptily occupies Facebook and funds billionaires' pseudo-noble vanity quests against the same purveyors of cheap clickbait that make them their money in the first place.

Instapaper can't spam my attention like Pinterest can, so it and all your Pinterest joy are living in borrowed time.

Labels: , ,