Tuesday, June 30, 2015

Mark Thoma continues to shine

Nobody talks about the assumptions behind free markets better than Mark Thoma, who has a new piece on the subject.  It really should be required reading, because it sets the pattern of regulations that all society's develop in context and makes it hard to have an absolutist view on a lot of subjects.

At the same time, it appropriately acknowledges just how powerful this process can be when the necessary preconditions exist.

Cracked, on the other hand, is very good at talking about math

I tend to avoid Cracked.com. Their articles are thoughtful, informative and enormously entertaining. They can discuss complex, even technical subjects in a clear and engaging way. All of this makes the site a dangerous time sink and I just can't afford to hang ouot there.

From 6 Small Math Errors That Caused Huge Disasters
#4. A Huge Walkway Collapses Due to a (Seemingly) Inconsequential Design Change

When designing their newest hotel to be built in downtown Kansas City, the fine people at Hyatt Regency wanted all the bells and whistles in it. The architectural firm in charge of the building design came up with a series of aerial walkways suspended from the ceiling so that guests could people-watch from a heightened vantage point. All in all, it was a pretty nifty feature. Until it suddenly collapsed and killed more than a hundred people.

The Laughably Simple Flaw:

One long rod was replaced with two short ones.

If there's one principle consistent across all human nature, it's that we will always prefer the path of least resistance (i.e., "if you can get away with a half-assed job, do it"). The original plan was for two walkways that were directly on top of one another to both be supported by one very long rod that would anchor into the ceiling. Like so:

Easier to work with, easier to install, works exactly the same. Right?

That little change killed 114 people, injured 216 more and cost $140 million in lawsuits.

Look at the first image again.

One rod, two nuts. Each nut only has to carry the weight of its own platform. Which is good, because each nut (and the welded beam it's screwed to) is only rated to carry the weight of one platform.

Now look at the second image. See the nut we've labeled "OH SHIT"?

That one single nut now has to carry the weight of BOTH platforms, and all the doomed tourists standing on them. Look obvious? Congratulations, because none of the professionals at either company caught it.

And so, one night during a dance competition, the stressed "OH SHIT" nut cleaved clean through the beam and the walkways collapsed.

During the ensuing lawsuits, it came out that neither the steel company nor the engineering firm in charge of construction had even bothered to do a back-of-the-envelope calculation that would have shown them this glaring flaw.

Monday, June 29, 2015

There may be a quality control problem here -- nearing the end of the EngageNY thread

At the teaching blog, a follow-up to the language of Eureka post using the same lesson:

Eureka Math Tips for Parents -- worst SAT prep question ever

Here's the offending passage this time.

[And before you ask, the answer is no, a rectangle and a triangle can't have corresponding parts.]

I have lots more material -- hell, that same page had a different problem that screwed up by omitting the AAA similarity theorem -- but I think that we've already uncovered an unacceptable number of major errors after only a few spot checks. More examples would just be piling on.

Sunday, June 28, 2015

On the teaching blog

More on appropriate and inappropriate use of formal language in math lessons,

Eureka Math Tips for Parents -- well, that clears up everything

Friday, June 26, 2015

Third, when you call yourselves "Great Minds," you're just asking for trouble

A bit more background on the Eureka Math thread (see here, here and here).

First off, I volunteer as the math adviser for a urban after school program here in LA and based on that not-at-all random sample, Eureka Math is widely used in the LAUSD and both kids and tutors (including those with math backgrounds) hate it.

Second, as this Quartz article by Rachel Monahan shows, Eureka Math is very much the accepted choice for schools looking to align their instruction with the Common Core Standards.
Louisiana published a review of Common Core curricula last year, and gave EngageNY’s Eureka Math a top ranking, leading a large number of districts in the state to adopt it, according to officials with the nonprofit Great Minds Inc., which developed Eureka Math for EngageNY. Parts of Core Knowledge also ranked at the top along with one textbook series.

Favorable reviews there and elsewhere of the free EngageNY materials have helped expand interest. Achieve, a nonprofit group that backs the Common Core standards, gave three Expeditionary Learning units its highest ranking.

“Do we want as many districts as possible adopting our stuff? Sure. Am I more interested in our setting a new standard for the quality of instructional materials in reading and math? Yes,” said Great Minds executive director Lynne Munson. “I’m vastly more interested in seeing that more students are learning these subjects well than in selling anything to anyone.”

Thursday, June 25, 2015

What high school students are learning about statistics

Yet another post up at the teaching blog on the Common Core affiliated lessons from EngageNY. Most of it focuses on the language we use to teach math, but I would like our regular readers to check this out (you can find the original lesson here).
Suppose newspaper reporters brainstormed some headlines for an article on this experiment.  These are their suggested headlines:

A. “New Treatment Helps Pericarditis Patients”
B. “Colchicine Tends to Improve Treatment for Pericarditis”
C. “Pericarditis Patients May Get Help”

7. Which of the headlines above would be best to use for the article?  Explain why.

Headline A would be the best because this is a well-designed experiment.  Therefore, a cause and effect relationship has been established.  Headlines B and C talk about a tendency relationship, not a cause and effect relationship.
I've read this over multiple times and have run it past an old professor of mine who teaches experimental design and we keep coming to the same conclusion. In a lesson on drawing inferences from experimental data, the authors seem to believe that 'causal' means 'deterministic."

That would be bad.

I have more concerns in my original post, and those represent only a fraction of the problems I found in the section. I recommend you take a look for yourself, and when you do, remind yourself that this is widely considered the gold standard of the new wave of instructional materials.

Wednesday, June 24, 2015

Defining dysfunction

This paragraph from a recent Paul Krugman post reminded me of something I've been meaning to dig into for a while.
A brief aside: I don’t think it’s right to call this a case of Washington “dysfunction”. Dysfunction is when we get outcomes nobody wants, or fail to do things everyone wants done, because there doesn’t seem to be any way to package the politics. In this case, however, people who oppose TPP voted down key enabling measures — that is, they got what they wanted. Calling this “dysfunction” presumes that this deal is a good idea — and that kind of presumption is precisely what got successfully challenged yesterday.
We hear “dysfunction” thrown around a lot (often by me), so it might be a good idea to pin down some definitions. Krugman is definitely on the right track, but statements about “nobody” and “everybody” are obviously unrealistic. Every scenario makes somebody happy, up to and including the rise of Cthulhu and his dark, chthonic host. A workable definition will have to take that into account, as well as considering differing intensities of opinion.

A system is dysfunctional if there is no consistent weighting of preferences that corresponds to its actions. (I'm going to be careful not to let this drift into a discussion of voting paradoxes because a good portion of this audience knows a great deal about the subject and I would have to do serious research to make sure I didn't make a fool of myself.)

For example, a group could do what the plurality wants, or it could use some sort of weight by rank (first choice is worth five points, second is worth four...), or it could take into account strongly held positive or negative opinions.

Let's use restaurants. A group might go to an Armenian place because three out of seven listed it as their first choice, or they could go to Chipotle because six people listed that as their second choice, or they could take Chipotle off the list because one person refused to go (I'm with that guy. Living in LA and going to Chipotle is like living in Rome and going to Pizza Hut). All of these decisions are consistent with a functional organization.

If, on the other hand, the group ends up going for Thai when everyone would have preferred burgers, that's dysfunctional. I can’t think of a reasonable and consistent weighting scheme that can produce that result.

A political party is more complex than a group of friends, but in some ways it may not be that much more complex. I’ll try to flesh this out later, but for now, while you have to be careful talking about what a large group “wants,” I suspect that there are a lot choices that the GOP would “like” to make (infrastructure spending, for instance) in the same sense that those friends would “like” to be having burgers now.

Based on these definitions, for large chunks of the Twentieth Century, I'd say that the Democratic Party was the more dysfunctional. The Republicans, however, do seem to be making up for lost time.

Reading over this, it's pretty clear that I have a ways to go before I have a proposal for something coherent and measurable and usable, but I do believe there is something out there, What's more, I suspect that, in 2015, it's probably more important to worry about dysfunction than about ideological extremism.

Tuesday, June 23, 2015

A genuinely funny conspiracy theory

Maybe I'm just getting around to noticing, but it seems that Cracked has been pushing their videos more into the sketch category. The results have been a bit mixed, but some of the clips have been very good and the site has built up a great of good will over the years.

I may be biased -- I have a fondness for satiric conspiracy theories and a bad feeling about the practice of making big budget films out of TV shows (particularly the recently departed) -- but, with that caveat, I think you'll enjoy this one.

Monday, June 22, 2015

Captain Outrageous

My big complaint with business journalism in the Ithuvanian Age is not just that journalists mistake luck and hype for genius, but that they miss genius when they see it. Reed Hastings can get on the cover of anything; Neal Sabin, who essentially invented a large and highly profitable media industry, doesn't even get a Wikipedia page.

There is an aesthetic to business just as there is to sports. The great players do look good on paper, but there is a pleasure in watching their style and their brilliance that the stats simply can't capture.

And nobody ever had more style than Ted Turner.

Friday, June 19, 2015

Uber and rules

From the comments at Cathy O'Neil's page:
Uber is a worldwide company; NYC type medallions are found in just six US cities. There’s absolutely no evidence that inflated medallion values in these six cities had anything to do with medallion owners pocketing obscene taxi profits at the expense of riders and drivers (there are no obscene taxi profits in any city; medallions are a financial instrument, and inflated in line with comparable instruments). The “heroic innovators versus corrupt protectors of the status quo” framing is designed to prevent anyone from actually examining whether Uber can produce taxi service much more efficiently than a reasonably run Yellow Cab, or where these massive efficiency gains might come from, or whether they are sustainable over time, or why no one else had ever thought of them.
This was generated by a question that Cathy posed:

On the one hand, it does seem to be a different act to raise your hands on Broadway versus using an app on your phone. But by the time we have chips implanted into our heads, just thinking the words “hail a taxi” might do the trick, and that’s where the grey area lives. Or, put it another way, yellow taxis might also want to have hailing apps, and in fact they really should. 
I can attest that yellow cabs have hailing apps in some cities already, Seattle for example.  Should that allow those companies to evade transportation regulation?  This is very much like the sales tax issue with Amazon -- slight differences in business model lead to a discussion as to whether the generally accepted rules apply at all.  In general, I think there are a lot of pernicious regulations and there is a lot of things that can be done to make things more efficient. 

But I also believe in treating groups fairly.  I don't think that technical exceptions to rules should be the goal -- rather we should overturn bad rules.   I'd have more sympathy if the new entrants were selling their app to all comers (pure software company) or if they were pushing for the rules to be revised wholesale into a new regulatory regime that all of the relevant players could participate in.  It isn't always the case that the status quo is good, but exception-based regulation seems to be a dangerous way to proceed as that can create a less competitive market.  Even Ayn Rand thought that was bad (see the first part of Atlas Shrugged and Taggart Transcontinental). 

Thursday, June 18, 2015

The New Yorker's culture desk could use a fact checker

I recently came across an uncredited piece that drove my inner film geek crazy. About halfway into the review of Ace Records’s new compilation, “Come Spy With Us: The Secret Agent Songbook,” I came across this.

[emphasis added]
The world of spy themes doesn’t stop at Bond (or at Bond offshoots or Bond antidotes), and neither does Ace’s set. Lalo Schifrin’s immortal “Mission: Impossible” theme is here, along with the Challengers’ version of Hugo Montenegro’s “Theme from the Man From U.N.C.L.E.” Both of those illustrate the relationship not only between spy music and surf music—similar in instrumentation, similar in insistence—but also between spy music and the music of spaghetti Westerns.


There are far too many good selections here to list them all: Billy Strange’s “Our Man Flint,” Nancy Sinatra’s “The Last of the Secret Agent” (Flint and Sinatra would collaborate on the theme song for the Bond film “You Only Live Twice,” which isn’t on the set) [That should be "Strange and Sinatra," Derek Flint being fictional and all. It should also be noted that the version of “You Only Live Twice” that most of us are familiar with is by Barry and Sinatra. Billy Strange had nothing to do with it -- MP], and Matt Monro’s “Wednesday’s Child.”   
Billy Strange was an arranger and session musician now best remembered as a member of the legendary Wrecking Crew. Hugo Montenegro was a minor film and TV composer (other than I Dream of Jeannie, I doubt any of his compositions would register if you heard them) who was best known for cheesy but popular cover arrangements.

Both released albums of covers of soundtracks of popular movies and TV shows. As far as I can tell, neither had anything to do with the original scores. Those came from composers such as Ennio Morricone, John Barry and, in this case, the man who wrote the theme for the Man from UNCLE and composed most of the music for the show's first season and who scored both Flint films, Jerry Goldsmith.

For movie people, Goldsmith is kind of a big deal:
Jerry Goldsmith has often been considered one of film music history's most innovative and influential composers.[8] While presenting Goldsmith with a Career Achievement Award from the Society for the Preservation of Film Music in 1993, fellow composer Henry Mancini (Breakfast at Tiffany's, The Pink Panther) said of Goldsmith, "... he has instilled two things in his colleagues in this town. One thing he does, he keeps us honest. And the second one is he scares the hell out of us."[65]  ...  In a 2001 interview, film composer Marco Beltrami (3:10 to Yuma, The Hurt Locker) stated, "Without Jerry, film music would probably be in a different place than it is now. I think he, more than any other composer bridged the gap between the old Hollywood scoring style and the the [sic] modern film composer."[67]
For someone writing about film music, crediting Montenegro or Strange with a Goldsmith composition is the kind of mistake that makes you wonder how much of the writer's expertise came from the liner notes. Perhaps worse, it is such an easily avoidable error. Thanks to Wikipedia, it takes so little time to get the facts right.

In fairness to the author, some of the critical points are valid (such as the relationship between spy films and surf music. For example, check out the arrangement from this sequence from Our Man Flint,

But even good arguments are difficult to take seriously when they come with careless mistakes.

p.s. I didn't want to go full nerd in the middle of a post, but if you feel like releasing your inner spy geek, I recommend checking out these discussions of the various arrangements of Man from UNCLE themes (including the revelation that Goldsmith hated Lalo Schifrin's new arrangement).

p.p.s. I ran this past an actual authority, Brian Phillips. He pointed out another one I should have caught: "Though Bill Cosby starred in “I Spy” as early as 1965 (the brassy Roland Shaw theme is included)..."  The I Spy theme was, of course, by Earl Hagen who was, in Sixties television, also kind of a big deal [Andy Griffith Show, Dick Van Dyke Show, etc.].

Brian also questions whether the bassline to "Come Spy With Me"  is really James Jamerson.

Wednesday, June 17, 2015

I don't have time to comment on this one

But I really want to.

The David Brooks files: How many uncorrected mistakes does it take to be discredited?

Eugen Weber explains the origin of window shopping

I've got a big MOOC thread (actually a MOO? thread) involving Dr. Weber and his seminal video course, the Western Tradition (previously discussed here).

Whenever I research one of these megathreads, I end up learning all sorts of interesting things that don't quite fit in with the thread. For example, I'd never associated the Industrial Age with peering through a shop window.

Tuesday, June 16, 2015

Over at the teaching blog

I review Gary Rubinstein's deep dive into Common Core-aligned assignments. What he finds is not pretty.

FYC -- AMC really, really likes Emmy Awards

More notes from LA.

As I believe we've mentioned before, certain parts of town (particularly Hollywood, North Hollywood and Studio City) are carpet bombed on an  annual basis by ads bearing the phrase "For your consideration."

Though a great deal of money goes into chasing these little statues, it is by no means spread out evenly across the industry. For the major networks, Emmys are almost entirely a social currency; they have almost no effect on ratings or syndication packages (it may not be a coincidence that I haven't seen any ads for CBS). For the pay services (HBO, Netflix, etc.) and for certain cable channels, however, the awards can do a lot to raise profiles.

Probably no channel has gotten more of a reputational bump than AMC starting with their critical breakout Mad Men. Perhaps it's not surprising that the company is willing to make an expensive push to get a few more statues from the show, despite there not being an even theoretical chance of a ratings boost. .

Monday, June 15, 2015

What could possibly go wrong?

Cory Doctorow writes:
Nathan Myhrvold's Intellectual Ventures has received a patent for a DRM [Digital rights management -- MP] system for 3D printers, to stop people from printing out trademarked and patent objects. Like other DRM systems, this won't work (it will either have to be so broad in its parameters for recognizing prohibited items that it will balk at printing innumerable harmless objects, or it will be trivial to defeat by disguising the objects beyond the system's ability to recognize them).
From Tech Review's Antonio Regalado:

“You load a file into your printer, then your printer checks to make sure it has the rights to make the object, to make it out of what material, how many times, and so on,” says Michael Weinberg, a staff lawyer at the non-profit Public Knowledge, who reviewed the patent at the request of Technology Review. “It’s a very broad patent.”

The patent isn’t limited to 3-D printing, also known as additive manufacturing. It also covers using digital files in extrusion, ejection, stamping, die casting, printing, painting, and tattooing and with materials that include “skin, textiles, edible substances, paper, and silicon printing.”
Because if you're going to approve a really sweeping patent, who better to give it to than Nathan freakin' Myhrvold?

Friday, June 12, 2015

Another convoluted IP tale

Mark Evanier spells out the strange sequence that led from the creation of Mickey Mouse to this:
I was in the hospital when it was announced that Universal and Disney had concluded a deal that would send sportscaster Al Michaels to NBC while Disney would reacquire title to Oswald the Lucky Rabbit.

UPDATE: Right on cue, Mark Evanier provides another excellent IP post, this time on residuals.

Thursday, June 11, 2015

Josh Marshall on Scott Walker's anti-tenure push

I've got to get out the door, so I don't have time to dig into this question, but many reform advocates such as Jonathan Chait have argued in very broad terms that educational systems which grant increased job security based on seniority only attract deadwood; good, productive academicians have no interest  in tenure since they would always be the last to be fired.

If this is true, why haven't elite private schools replaced tenure with a bonus system?

From TPM:
So I want to take a look at a different part of this. The crown jewel of the Wisconsin university system is the University of Wisconsin at Madison. It is one of the top research universities in the country and the world. And with this, you will basically kiss that jewel goodbye. To me this is the more salient reality than whether you think academic tenure is a good thing or not in itself.

If this happens, over time, the professors who can will leave. And as the top flight scholars and researchers depart, so will the reputation of the institution. So will graduate students who want to study with them, the best undergrads, money that flows to prestigious scholarship. Don't get me wrong. Not in a day or a year or even several years. But it will. If you don't get this, you don't understand the economy and incentive structure of university life.

Over the last couple decades, especially in the humanities, we've seen develop what increasingly looks like an aristocracy of tenure. The lucky PhDs land tenure and they've got a pretty good gig. In some cases they have a great gig. But the system is sustained by an army of TAs, adjuncts, other non-tenure track positions and assistant professors fighting for tenure. In some cases realistically, in some cases not at all realistically, all those folks are fighting for the hope of landing tenure at some point. Take that away and the whole system of sweated academic labor comes crashing down.

But again, that's bigger picture. Let's look at the medium picture. Take tenure out of the University of Wisconsin and the people who can will - over time - leave. If we had a single national system, that would be one thing, as it would effect all equally. And the departees wouldn't have any place to go. But it won't. Private universities, with the most outrageously high tuitions, will not do this. And the top academics will go there. The net effect of all this will be to kill off or bleed dry great state universities which are yes, still pricey, but not as crazy expensive and hard to get into as the prestige private universities.

Cracked has some fun with ad campaigns

The 7 Most Blatant Lies Famous Brands Based Entire Ads On

Slightly NSFW

Wednesday, June 10, 2015

Land of the Pyramids

I can't quite give this TPM/Slice article on multilevel marketing of dietary supplements an unqualified recommendation (for this and other reasons), but it's worth reading.

You can find a better treatment of multilevel marketing from the good people at This American Life (specifically Bianca Giaever and Brian Reed). And, if you really want to get deep into the weeds, check out this video from the Internet Archive...

Then follow the link to see what became of that Nutrilite.

Damn you, Jessica Williams!

God as my witness, I was working on a post about how open carry laws like this have have huge de facto exceptions.

Tuesday, June 9, 2015

Coming attractions

This is Joseph

Mathbabe (Cathy O'Neil) has been on a roll lately.  Expect to see some comments on her recent articles coming up in the weeks to come (Mark has an impressive queue at the moment).  In the meantime, one could do worse than to check her out.

Good cop, bad cop

Quick recommendation. I thought this TPM analysis of the McKinney, Texas incident by a former policeman turned law professor did an excellent job comparing the responses of the different officers on the scene. Obviously there's a limit to what you can tell from this video, but the difference certainly appears stark.

The issue isn't (just) excessive CEO pay

Nothing particularly new here, but it does reinforce the argument that current levels of CEO pay are impossible to justify even if you ignore the inequality question and limit yourself to value added to the company. 
The best-paid chief executive of a large American company was David Zaslav, head of Discovery Communications, the pay-TV channel operator that is home to "Shark Week." His total compensation more than quadrupled to $156.1 million in 2014 after he extended his contract.

Les Moonves, of CBS, held on to second place in the rankings, despite a drop in pay from a year earlier. His pay package totaled $54.4 million.

The remaining four CEOs, from entertainment giants Viacom, Walt Disney, Comcast and Time Warner, have ranked among the nation's highest-paid executives for at least four years, according to the Equilar/AP pay study.

One reason for the high level of pay in the industry is that its CEOs are dealing with well-paid individuals.

"The talent, the actors and directors and writers, they're being paid a lot of money," said Steven Kaplan, a professor of finance at the University of Chicago Booth School of Business. "In industries where the talent makes a lot of money, the CEO makes a lot of money as well."

Pay packages for CEOs overall grew for the fifth straight year in 2014, driven by a rising stock market that pushed up the value of executive stock awards. Median compensation for the heads of Standard & Poor's 500 companies rose to a record $10.6 million, up from $10.5 million the year before, according to the Equilar/AP pay study.

Peer pressure is another factor driving up executive compensation. The board members responsible for setting CEO pay typically consider what the heads of similar companies are making. If pay for one goes up, it will likely go up for others.

For the chieftains of media, there are also other factors boosting pay.

Several work at companies where a few major shareholders control the vote.

The media magnate Sumner Redstone controls almost 80 percent of the voting stock at CBS and Viacom. Because of his large holdings, Redstone can easily override the concerns of other investors about the level of CEO pay. Discovery's voting stock is heavily influenced by the brothers Si and Donald Newhouse and John Malone, another influential investor in the media industry.

At Comcast, which owns NBC and Universal Studios, CEO and Chairman Brian Roberts controls a third of his company's voting stock. That means he has substantial influence on the pay that he is awarded.

Comcast had no comment when contacted by the AP for this story.
What strikes me in these cases is the lack of correlation. You can make a good case that Moonves brings more than $54.4 million of value, but I'm not sure you can say that for any of the other names on the list. Zaslav is still being paid top dollar for a run that appears to be over. Iger got enormously lucky with Marvel (and you have to wonder how much longer that run will continue). Those, in turn, are more defensible than Roberts of Comcast, a company that proves that no amount of bad management can sink a de facto monopoly.

Monday, June 8, 2015

What happens when someone actually tests one of Nathan Myhrvold's culinary tips?

Myhrvold's wine tip doesn't appear to hold up.

From over at the food blog,

Privacy in and out of the classroom

This is Joseph.

From Dean Dad's comments section on a discussion of a taped conversation at Kennesaw showing an advisor acting poorly towards the student:
You will have a hard time getting and keeping good employees if you always assume that a selected video clip like that one indicates a pattern of behavior rather than a response to repeated misbehavior that even extensive training cannot always eliminate. As a man, you might not be aware of how often women on a campus are disrespected by male students when they do the same things that men do in a classroom or an office.
I am not in complete agreement with this perspective.  But I do agree that it is dangerous to use a minute long video clip (that one party knew was happening and the other did not) as a proxy for the entire interaction. 

This sort of "selective sampling" is likely to become a bigger and bigger problem as the ability to record in public becomes more and more trivial.  What do you do when it's not just phones that can record but eyeglasses as well?  But the ability to edit out previous interactions and to control the timing of the recording can be a powerful tool to drive the conversation. 

Just one more point to ponder.

Saturday, June 6, 2015

Revisiting the SNAP challenge

We've already established that Trader Joe's is not where you want to shop if you have to live on a food stamp budget, but is it even possible to stave off hunger for a week with $28 worth of Trader Joe's groceries?

Yes, but just barely.

Friday, June 5, 2015

The internet has made historical revisionism so much easier

[UPDATE: Brad DeLong found an arguably more embarrassing example from the National Journal.]

This may be the best example of New York Times political reporting you will see you all day.

It started as a standard narrative journalism/puff piece. Amy Chozick and Trip Gabriel used a handful of anecdotes and a couple of well-received speeches to build a breathless account of political underdog Carly Fiorina surging toward the lead.

Hack political writers love this narrative. They also gravitate toward positive stories about candidates with whom they are comfortable. When I say "comfortable" I am talking about culture not politics. I will try to back this up in future posts, but I have long argued that left/right biases are far less common than more significant biases involving class, race, religion, region, education, etc. While the New York Times probably disagrees with most of Fiorina's politics, they are more than comfortable with almost everything else about her, from her prominent family to her CEO background to her wealth and extravagant lifestyle.

So far, all of this is just another day at the office for the New York Times election beat. Soon after the piece ran, however, people started to notice that the writers had really buried their lede. Deep in the story, it was revealed that Fiorina's surge was not quite as substantial as the headline suggested.

From paragraph 8 (as pointed out by Duncan Black):
While supporters in Iowa noted that she had doubled her standing in state polls, it was a statistically insignificant change from 1 percent to 2 percent, according to a Quinnipiac University Poll released May 6. (That may seem piddling, but the same poll had Mr. Santorum, who won the Iowa caucuses in 2012, also at 2 percent, while 5 percent supported Mr. Bush.)
It is one thing to have a paragraph in the middle of your story that completely undercuts your premise; it is quite another to have people point out a paragraph in the middle of your story that completely undercuts your premise. A quick rewrite was definitely in order.

The resulting headline doesn't make a lot of sense -- if the polls are a reflection of the state's voters, Iowans appear to be swoon-shrugging over Fiorina -- but it does partially inoculate the story from further mockery.

Of course, the NYT has standards. They don't just rewrite a published story without even acknowledging it. The original headline is right there at the bottom of the page.

In small print and pale gray letters.

Thursday, June 4, 2015

Yes, I am taking the pro side here on autonomous vehicles

In a comment on my recent relatively positive post on autonomous vehicles, Joseph points us to a Megan McArdle article that takes a different view.

While it is good to see that the conventional wisdom is starting to acknowledge some of limitations with driverless cars, I still have quite a few problems with the piece. McArdle makes some good points about the labor implications, but she does not seem to have a strong grasp of the technological or the implementation issues involved with using autonomous vehicles for long-haul trucking. We can get back to implementation later; for now let's talk about tech.

Here's McArdle:
You hear a lot about how Google cars have driven an amazing number of miles without accidents. You hear less, however, about how they have achieved this feat: by 3-D mapping every inch of those roads so that the car has a database of every stationary object, from traffic lights to guardrails. That allows the car to devote its processing power to analyzing the movement of objects that aren't in its database.

Such mapping is incredibly labor intensive, which is why, according to Lee Gomes, those amazing mile counts that Google's driverless cars are racking up "are the same few thousand mapped miles, driven over and over again." Most of them are near Google's headquarters in Mountain View, a place that gets only 15 inches of rain a year and never has snow or ice -- three common weather hazards that long-haul truckers must frequently contend with.

Just getting Google's technology to a point where we could have self-driving trucks would require mapping every inch of the nation's more than 164,000 miles worth of highways. But then what do you do with the truck? You're probably going to have to map some of the roads that connect to those highways too. And then constantly remap them, because things change all the time. You'll also have to teach the computer system what to do in a blinding snowstorm on Wolf Creek Pass. As we wrote in January, "The technology giant doesn’t intend to offer a self-driving car to areas where it snows in the near term."
McArdle makes a couple of common mistakes: assuming that, because Google dominates the coverage of driverless cars, it also dominates the research (which we'll get to later); and assuming that what is difficult for humans is difficult for robots and vice versa.

Rain and snow are problematic for us humans both because they can limit visibility and because they tend to create very complex physics problems that have to be solved in a fraction of a second. Bad weather visibility is much less of an issue with autonomous vehicles* than it is with human drivers while classical physics problems are the sort of thing that roboticists are very good at.

Along similar lines, McArdle observes [emphasis added] "But it seems like getting from there to fully automated trucks--necessarily huge, heavy, and  capable of horrific damage, with handling capabilities that change depending on the load, and a stopping distance almost twice that of a car at high speeds, will probably take a while." Yes, this will take a while, but not for the reasons McArdle imagines. Load effects and long stopping distance do make truck driving much more difficult for humans, but for computers they just represent simply another set of parameters. Furthermore, the biggest factor in real-life stopping distance is often reaction time, an area where computers have a distinct advantage.

Nor does the fair-weather testing complaint hold up. It is true that Google has largely limited its testing to clement conditions, but you certainly can't say the same for the upcoming Volvo test in, you know, Sweden.

Google's PR department has done a masterful job identifying the company with autonomous vehicles. This is not simply a matter of corporate ego. As I said earlier:
Google has a lot of reasons to want to be seen as a diversified, broadly innovative technology company, rather than as a very good one-trick pony cashing in on a monopoly (possibly two monopolies depending on how you want to count YouTube). A shiny reputation helps to keep stock prices high and regulators at bay.
It is enormously telling that McArdle cites Google ten times in her article while she doesn't mention Daimler by name and she never refers to Volvo at all.

* As far as I can tell, Daimler's prototype is doing its mapping independently in real time. While impressive, I'm sure the production models will share data and will also rely on existing maps.

Wednesday, June 3, 2015


Dean Dad makes a tremendously important point about MOOCs:

As a commenter correctly noted, there’s nothing stopping someone now from taking a MOOC in a “gen ed” area and then taking a CLEP exam to get credit.  CLEP fees are often lower than even community college tuition.  The ASU model is a more expensive and clunkier version of CLEP.  The MOOC-to-CLEP option has existed for a couple of years now, but students haven’t taken advantage in significant numbers.  
In case you're not familiar:
The College Level Examination Program (CLEP) is a group of standardized tests created and administered by College Board. These tests assess college-level knowledge in thirty-six subject areas and provide a mechanism for earning college credits without taking college courses. They are administered at more than 1,700 sites (colleges, universities, and military installations) across the United States. There are about 2,900 colleges which grant CLEP credit. Each institution awards credit to students who meet the college's minimum qualifying score for that exam, which is typically 50 to 60 out of a possible 80, but varies by site and exam. These tests are useful for individuals who have obtained knowledge outside the classroom, such as through independent study, homeschooling, job experience, or cultural interaction; and for students schooled outside the United States. They provide an opportunity to demonstrate proficiency in specific subject areas and bypass undergraduate coursework. Many take CLEP exams because of their convenience and low cost (typically $15) compared to a semester of coursework for comparable credit.

I plan to spend a lot of time this summer writing about better using CLEPs  and improving MOOCs. For now though, I want to get a couple of big points.

The Internet has a way of producing deceptively large numbers. This has certainly been true with MOOCs. Articles have breathlessly reported huge enrollments despite the fact that for online classes enrollment is an almost meaningless statistic. When we have tried to assign meaningful metrics to online classes, they have tended to do very poorly. CLEP-usage would appear to be another example.

CLEP exams are a well-established, easy, and cheap way for students to get college credit for taking online courses, but very few seem to be taking advantage of it. That's a bad sign, but it does suggest a way forward.

Now, just to be clear, I am not saying that CLEP tests are a perfect solution for this problem – – I am certain we could come up with a better system, particularly once we have some experience to build on – but for the time being these exams are probably our best option and the fact that we're not seriously exploring them indicates a deeper lack of seriousness about MOOCs.

Tired blood and Nazi superman vitamin pills

One of the reasons I enjoy going through old pop culture ephemera is the perspective it can give on the way popular thinking has evolved or, in some cases, stayed the same. Check out the topics a popular comic book superhero chose when making conversation in 1945.

Having spent a lot of time recently going through food related ads and instructional films, I've noticed a strong mid-Twentieth Century fascination with the scientific basis of nutrition. Much of this expressed itself as pseudo-science, but it was driven by a string of real breakthroughs. Remember, he very concept of a vitamin was largely a Twentieth Century discovery.

From Wikipedia:

The discovery dates of the vitamins and their sources
Year of discoveryVitaminFood source
1910Vitamin B1 (Thiamine)Rice bran
1913Vitamin A (Retinol)Cod liver oil
1920Vitamin C (Ascorbic acid)Citrus, most fresh foods
1920Vitamin D (Calciferol)Cod liver oil
1920Vitamin B2 (Riboflavin)Meatdairy productseggs
1922(Vitamin E) (Tocopherol)Wheat germ oil,
unrefined vegetable oils
1926Vitamin B12 (Cobalamins)Livereggs, animal products
1929Vitamin K1 (Phylloquinone)Leaf vegetables
1931Vitamin B5 (Pantothenic acid)Meatwhole grains,
in many foods
1931Vitamin B7 (Biotin)Meatdairy productseggs
1934Vitamin B6 (Pyridoxine)Meatdairy products
1936Vitamin B3 (Niacin)Meatgrains
1941Vitamin B9 (Folic acid)Leaf vegetables
It is easy to see how all of this got people thinking about vitamins as an almost mystical cure-all. Even instructional US govt. films included lines like "The Nazis are supposed to have a superman vitamin pill." Of course, the mystique of the vitamin combined with the marketing power of television presented huge profit potential.

After vitamins, food advertisements loved to talk about 'energy.' They were always vague about exactly what they meant by the word but they wanted you to know their products were packed with it.

Companies even promoted dextrose as healthy.

It's easy to mock, but you have to remember how fast nutritional science was advancing. We've had decades to process this information and I'm not sure we're doing that much better.

Tuesday, June 2, 2015

Dining tip – when eating out with Maureen Dowd, don't share an appetizer

Given her attitude toward people in the service industry, I would be nervous about what goes on in the kitchen.

Brad DeLong handles today's garbage collection.

Why Don't New York Times Writers Possess Any Awareness of Their Presentation-of-Self?

Before we talk about 21st Century technology...

...we need to spend some time thinking about the 20th Century technology that got us here.
A controlled-access highway provides an unhindered flow of traffic, with no traffic signals, intersections or property access. They are free of any at-grade crossings with other roads, railways, or pedestrian paths, which are instead carried by overpasses and underpasses across the highway. Entrances and exits to the highway are provided at interchanges by slip roads (ramps), which allow for speed changes between the highway and arterial roads and collector roads. On the controlled-access highway, opposing directions of travel are generally separated by a median strip or central reservation containing a traffic barrier or grass. Elimination of the sources of potential conflicts with other directions of travelers dramatically improves safety, fuel consumption, and travel times.
It turns out that many of the innovations mid-20th Century engineers came up with to maximize throughput (controlled access, divided lanes, etc.) also tend to make the jobs of today's engineers much easier when it comes to autonomous vehicles. This relationship between technologies will be a big factor when we get back to this.

Monday, June 1, 2015

Jack Shafer, poster child [repost]

You might have heard about Bruce Bartlett's recent paper on the effect that Fox News has had on the Republican Party. I don't entirely agree but it makes some interesting points about the way information flow affects politics.

The paper raises serious questions not just about Fox but about the entire journalistic ecosystem. For a long time and for a variety of reasons, major mainstream outlets like the New York Times and Politico have been reluctant to call Fox out on its most egregious lapses (these days, the news media is pretty much one long row of glass houses).

It's not surprising that Politico media apologist Jack Shafer took umbrage. If you really want to read Shafer's comments, you can find a link in this piece by Eric Boehlert, but, before you do, I would recommend that you take a few minutes to review Shafer's rather... flexible standards for journalistic quality and ethics.

More on journalistic tribalism
from WEDNESDAY, APRIL 3, 2013

Having brought up the charge in a previous post, I should probably take a minute to spell out exactly what I'm talking about. I'm using a very broad reading of the term 'tribalism' (perhaps so broad I should say something like 'tribalism and other social psych phenomena'). The traits I'm thinking of include:

1. Us/them mentality;

2. Excessive reliance on in-group social norms;

3. Deferring to and preserving hierarchies;

and as a consequence

4,   A tendency to use different standards to judge interactions based on the relative positions of the parties.

There is inevitably going to be a degree of subjectivity when deciding who goes where in the hierarchy, but I think it's fairly safe to say that Maureen Dowd and (till his death) Michael Kelly were in the innermost circle with writers like David Brooks and most prominent, established Washington and, to a lesser degree, New York journalists fairly close.

In this tribal model, it makes perfect sense that Politico would view Chris Hughes' (outsider) request for a small change in the copy of Timothy Noah (insider) as a major affront. It also explains Politico's attacks on Nate Silver (outsider) when his work started making established pundits (insiders) look bad.

The press corps's treatment of Al Gore in 2000 is another case in point. Following the lead of Dowd and Kelly and reinforced by a general dislike of the candidate, the group quickly established social norms that justified violating the most basic standards of accuracy and fairness.

The poster child for this kind of journalistic tribalism is Jack Shafer, or at least he was a few years ago when I was first experimenting with blogging. One of my main topics was the press's inability to face up to its problems and Shafer was the gift that kept on giving (I haven't read him much since). That blog is gone now but I still have my notes so here are some highlights.

Shafer was openly disdainful of readers and generally dismissive of their interests which is an extraordinary starting point for a journalism critic. Consider this passage from the aptly named "Why I Don't Trust Readers"
I'm all for higher standards, but I draw the line when journalists start getting more complaints about less serious professional lapses. Serious: Plagiarism, willful distortion, pattern of significant errors, bribe-taking. Not serious: campaign donations in the low three-figures for reporters distant from that beat; appearance of conflict of interest; a point of view; friendships with the rich and powerful.
First, notice the first item on the list. Plagiarism is certainly a serious offense, but the other serious offenses are the sort of things that can destroy people's lives, conceal crimes and enable corruption. Even more interesting is what didn't make the list: unintentional distortion due to laziness or bias; patterns of minor errors; isolated cases of serious errors due to negligence; selective reporting (as long as it doesn't rise to the level of distortion); failure to dig into important aspects of a story; cozy relationships with subjects as long as it doesn't involve the quid pro quo of a bribe.

What's important here was the victimology. In plagiarism, the primary victim is a fellow journalist. In all of these other cases, the primary victim is either the subject or the reader. Shafer was a tribalist and his main objective was almost always the defense of his tribe and its hierarchy.

There's a remarkable inverse correlation between the rank of Shafer's subjects and the harshness with which he treats them.  This is particularly apparent when different subjects of the same article have different positions. Shafer provided an excellent example when he wrote a post complaining about liberals writing books that actually called conservatives liars in the titles.

The books were Al Franken, Lies and the Lying Liars Who Tell Them,  Joe Conason's Big Lies and David Corn's The Lies of George W. Bush. Of these three, Conason was something of a pariah (Shafer dismissed him as a Clinton apologist) and Franken was clearly a journalistic outsider. Corn, on the other hand, was very much an insider in the Washington press corp (Shafer even described him as a friend in the post).

Under these circumstances, it's not surprising that Shafer finds a way to shield Corn from much of the blast.
This criticism applies more to Franken and Conason than it does Corn—you can't expect a book about Bush's lies to also be about Clinton's lies. And Corn acknowledges in his intro that Bush isn't the first White House liar and that Clinton lied, too. 
Of course, you could easily make a similar but more persuasive argument in Franken's behalf. Lies was largely focused on the relationship between the GOP and conservative media and since the book was published in 2003 when there was no Air America and MSNBC was just starting to experiment with liberal programming, there was no way to provide similar examples on the left.  Just to be clear, I'm not making that argument; I'm only saying that it's just as viable as the one makes for Corn.

For an even more dramatic bit of paired data, consider two obituaries Shafer wrote, separated by only a few months. The first was for Walter Annenberg, best known as a philanthropist and founder of TV Guide. The second was for Michael Kelly, journalist and former editor of the New Republic. Once again there's a clear hierarchical distance between the subjects: Annenberg, though decades earlier a power in publishing and to his death a major force in philanthropy, was not a journalistic insider; Kelly, on the other hand was about as inside as you can get.

As you've probably guessed by now, Shafer's approach to these two obituaries differs sharply. Though they don't fully capture the difference, the epitaphs give a good indication of the respective tones:

Michael Kelly: "Husband. Father. Journalist"

Walter Annenberg: "Billionaire Son of Mobster, Enemy of Journalism, and Nixon Toady Exits for Hell—Forced To Leave Picassos and van Goghs at Metropolitan Museum."

The contrast is sharpest when Shafer addresses journalistic scandals and cozy relationships with controversial right wing politicians, areas where there are definite parallels between the two men. Shafer actually explains away the New Republic/Glass scandal as an instance of Kelly being too loyal for his own good.

Shafer often judges figures on the periphery of the journalistic establishment based on a much higher standard than "Plagiarism, willful distortion, pattern of significant errors, bribe-taking." For someone like Larry King, a few disputable errors and minor discrepancies (such as changing the date of an incident from 1972 to 1971 when retelling an anecdote) merit an entire column. (It's worth noting that this column ran in the middle of 2009, a period when the coverage of politics, the economy and the European crisis were raising all sorts of journalistic questions, questions that didn't get a lot of space in Shafer's column. This raises the issue of trivialism in media criticism -- see On the Media for a myriad of examples -- but that's a topic for another thread.)

If marginal figures committing minor offenses are treated harshly by Shafer, what happens when someone at the top of the hierarchy does something that Shafer normally considers a serious offense like plagiarism? We got an answer to that one when Maureen Dowd was caught lifting a passage from Josh Marshall.

Here's her explanation in Bloggasm:

“i was talking to a friend of mine Friday about what I was writing who suggested I make this point, expressing it in a cogent — and I assumed spontaneous — way and I wanted to weave the idea into my column. but, clearly, my friend must have read josh marshall without mentioning that to me. we’re fixing it on the web, to give josh credit, and will include a note, as well as a formal correction tomorrow.”
And here Shafer explains why it's not so bad:
1. She responded promptly to the charge of plagiarism when confronted by the Huffington Post and Politico. (Many plagiarists go into hiding or deny getting material from other sources.)

2. She and her paper quickly amended her column and published a correction (although the correction is a little soft for my taste).

3. Her explanation of how the plagiarism happened seems plausible—if a tad incomplete.

4. She's not yet used the explanation as an excuse, nor has she said it's "time to move on."

5. She's not yet protested that her lifting wasn't plagiarism.

6. She's taking her lumps and not whining about it.
And here was my response at the time:
1. 'Responded.' Not to be confused with 'confessed,' 'owned up,' 'took responsibility,' or any phrase that uses a form of the word 'plagiarism.'
2. "[A] little soft"?
3. Yeah, near verbatim quotes make it through convoluted processes all the time.
4. "[M]y friend must have read josh marshall without mentioning that to me." -- What exactly would an excuse look like?
5. No, she just implied it wasn't plagiarism. That definitely gives her the moral high ground.
6. What a trooper.
(I apologize for the tone. I was in a snarky phase, but I'm trying to play nicer these days.)

I've spent a lot of time on Shafer because he's a good example,  I was familiar with his work and, as a media critic, he has an important role in journalism's self-correction process, but he's is not an isolated case, nor is he the worst of bunch (particularly not since the rise of Politico).

The point of all this is that journalism has a problem with tribalism and other social dynamics. These things are affecting objectivity, credibility and quality. What's worse, journalists seem to have so internalized the underlying mindset to such a degree that most of them don't even realize what's going on.

When the data runs contrary to what people claim

This is Joseph.

We all know that correlation is not causation.  But one theme that Mark likes to bring up is that when the data on an association shows the reverse of the claimed causal mechanism there is a high burden of explanation.  One very good example was brought up by Noah Smith.  There is often a causal assumption that big government impedes economic growth.  The problem?  Rich countries appear to have big governments:
Are we supposed to believe that rich countries are rich in spite of the fact that they all have big governments? Should we believe that government is a parasite that always, without fail, finds a host in the body politic of every single country that reaches first-world status?

Or should we conclude that big government is a necessary ingredient for countries to get rich?
 The plausible mechanisms for this vary -- including the not trivial point that losing wars is bad for a country (just ask the Huron Indians, if you can find any).  But it is not a small point that the empirical relation is the opposite of what proponents of small government propose (nobody says "let's shrink the government to reduce the level of prosperity in our country and give other countries a chance to experience being rich instead"). 

What also bothers me about this discussion is that the discussion on the size of government is mostly orthogonal to the discussion about the efficiency of government.  No matter what size government happens to be, I want it to be an efficient use of resources.  I am more worried by the political burden of improving infrastructure created by things like environmental review of bike lanes.  That is an inefficient process that costs extra resources and it doesn't matter whether we spend a lot of money on infrastructure or just a little -- it is just a bad way to do things.  I have the same opinion of universal health care (say the Canadian or the British approach).  It is not that I hate markets but that there are compelling theoretical efficiency advantages to having the state handle this form of risk pooling.  Heck, there is evidence that workers still covered by their parent's insurance (thank you ACA) are able to be more flexible employees.  That is an efficiency gain for employers like Uber and potentially an important one as we transition into a more flexible work force. 

As for what the real issues are behind the push for small government, well, that is complex and we'll need to wait for a follow-up post.