The War on Data Continues

Andrew Gelman and I wrote a piece for the ASA a while back called "The War on Data." It discussed what appears to be a disturbing trend of powerful interest groups trying to discredit and/or defund major sources of important, high-quality data, ranging from the Census Bureau to the National Oceanic and Atmospheric Administration.

Now we can add the CBO to the list. Jonathan Chait spells out the ugly details. You should read the whole thing but here are a few key paragraphs.

The Congressional Budget Office is a 40-year-old institution that has acquired enormous clout within Washington by virtue of its reputation for ideological neutrality. It furnishes Congress and the public with budgetary estimates that, if necessarily imperfect (as all predictions must be), are arrived at fairly. It is also a perfect modern expression of an old Progressive Era–ideal: that policymakers should be informed by the work of impartial experts. That the conservative majority has set out to corrupt this institution as one of its first major acts is, therefore, perfectly fitting.

The old methods CBO used to measure legislation would account for changes in behavior that a new law might create. (Say, higher cigarette taxes would lead to less smoking.) They did not attempt to measure legislation’s impact on the economy as a whole. This is because the two parties disagree completely over what policies make the economy grow faster. Democrats, for instance, believe that tax rates on the rich have little effect on economic growth, but that investing in public infrastructure or education has a lot. Republicans believe the opposite. Congress voted yesterday to require the CBO’s measurement of the budgetary cost of legislation to incorporate assumptions about how it will affect economic growth. Specifically, the GOP's assumptions.
...
The whole reason the Republican Congress is instituting dynamic scoring comes as a response to its attempt to write a tax reform bill last year. The idea was to lower tax rates while eliminating loopholes and preferences. But Republicans discovered that, while lowering rates is easy, eliminating preferences is hard. After Representative Dave Camp produced a tax reform bill that failed to cut tax rates for high-income taxpayers enough for their liking, Republicans abandoned it en masse. Paul Ryan openly declared his plan to change the forecasting rules so that Republicans could cut tax rates without having to pay for every dollar by ending preferences. The first step was kicking out Douglas Elmendorf, the CBO director widely respected by both sides. The second step was yesterday’s vote.
...
The new, “dynamic” CBO will be systematically biased to make conservative proposals appear misleadingly cheap and liberal proposals misleadingly costly to the public fisc. This would be true even if the Republicans were soliciting a fair range of forecasting perspectives. By its design, the dynamic scoring rule allows the party in power to game its effects. It applies “dynamic scoring” only to legislation affecting 0.25 percent of Gross Domestic Product. As Chye-Ching Huang and Paul Van de Water point out, congressional leaders can manipulate this requirement easily: They can break up large pieces of legislation into smaller bills to avoid dynamic scoring, or combine smaller pieces into a major bill, if needed to make their agenda appear more affordable. Dynamic scoring is subject to abuse by its very design.

“History of the memo”

I've been arguing for a while now that the shiny-object school of technology reporting isn't just annoying; it does real damage. We waste a tremendous amount of time talking about technology that will have a trivial impact or will probably never come to fruition at all. At the same time, genuinely important tech stories go all but unnoticed.

Shininess is a poor indicator of importance. Standardized shipping containers aren't glamorous, but you can make the case that the economic impact of containerization is greater than that of the combined output of Silicon Valley over the last decade.  

Paul Krugman has always been good at separating the real impact of technology from the gee-whiz aspect (I think he slightly overrates Paul David, but that's a small complaint and a topic for another post). Here's an excellent example.

Vox has a nice article about the origins of the business cubicle, which includes a discussion of filing cabinets – which is fine as far as it goes. But there is actually quite a lot more to say, as you’d know if you’d read a wonderful book (with a rather drab title), Control Through Communication, written by my friend JoAnne Yates of MIT’s Sloan School.

JoAnne’s book – which her husband calls “history of the memo” – is about the coevolution of information technology and the business world before the digital age. Of course, back then people didn’t talk about information technology – but IT did exist, indeed developed in fundamental ways, and changed everything. And filing cabinets are a window into those changes.
...
Organizations have always needed a record of their communications. No doubt ancient Roman merchants had slaves making copies of letters, the way Tom Standage (in another great book) tells us aristocrats did, turning senatorial correspondence into a form of proto-Facebook. Much later, JoAnne tells us, some businesses used a pretty amazing system: outgoing letters would be dampened, placed between the pages of a big book, and squeezed with a screw press to create a sort of reprint.

Then came a revolution: carbon paper! Or actually carbon paper plus typewriters. Suddenly, everything was in triplicate, and keeping a record of all correspondence became easy.

The next question, however, was how to find the relevant correspondence. When damp letters and screw presses were the limits of technology, there was no choice: a chronological record, to flip through when needed, was it. Carbons offered new possibilities: copies could be filed by subject instead or as well. But how should they be filed?

Boxes or drawers were one possibility, but they still involved a lot of shuffling, and relevant letters could easily be overlooked. The answer? The vertical file, with a tab indicating the contents of each folder.

To complete the revolution, however, you needed a behavioral change. Previously, businessmen wrote letters, narratives that might touch on multiple subjects. With the coming of the filing cabinet, however, they had to be disciplined to write each individual document on one and only one subject, so that it could be filed properly. The memo was born.

I love this story on multiple levels. For one thing, I always love reminders that many of the technologies that made the modern world were humble and inconspicuous – one of the great things about Daniel Boorstin’s The Americans vol. 3 is its account of things like the invention of the flat-bottomed paper bag. 

The Road to Bakersfield

A few weeks ago I drove up to Bakersfield for the day. I'd been meaning to make the trip for a while, but the immediate impetus was a series of post I'm working on about Elon Musk's less-than-credible proposal for an almost supersonic train. One of the issues that has not gotten much coverage is the terrain of the proposed route.

Press releases tend to show a the elevated track passing over small, gently rolling hills.

While the actual route often looks more like this.

Snow closures are not unusual on this stretch of I-5, nor are high winds, fires and earthquakes. With more than three-hundred miles of proposed elevated high-tech track that needs to be built to extremely tight tolerances, these factors add up.

It might be helpful if the California-based reporters covering this story would drive some of the road for themselves just to get a feel for what's being proposed. It is also a good excuse to visit Bakersfield, which, as Jonathan Gold explains, is a good way to spend an afternoon.

When I worked at the L.A. Times in the 1990s, my fondest desire was to get Bakersfield on the cover of the travel section, for just one week to replace the Rhine castles and Parisian cityscapes with a picture of the Bakersfield sign that then arched over Union Avenue, and it was a proud day when I succeeded, although I think the editors at the Bakersfield Californian thought I was making fun of them. (The unfortunate headline was "Achy Breaky Bakersfield.") I wasn't.

Bakersfield, a scant two hours away, offers the not-inconsiderable pleasure of being in a place that is neither Los Angeles nor part of greater Los Angeles, a town that is thoroughly Californian but can also feel a lot like the good parts of Oklahoma. It's the home of the Bakersfield Sound, the Merle Haggard/Buck Owen/Rose Maddox thing that brought a bit of grit back to country music, and without it the radio now would probably sound even more like Taylor Swift.

But mostly, at least for me, there is the old-fashioned cooking at one of the city's Basque dining halls, huge, multicourse feasts originally intended for the Basque shepherds staying at the local boardinghouses. They have become so popular that the few sheep men who show up are treated like local celebrities.

If you are sitting down at a long, oilcloth-covered table and there is a tin bowl of beans in front of you, a tureen of thin vegetable soup and a bowl of mild tomato salsa, you know you're at a Basque restaurant even without looking at the maps, the paintings of sheep-protecting dogs and the reservation books in which Echeverria is a more common name than Smith.

The perils of being overly tax aggressive

Wow.  

Just wow.

Josh Marshall:

Which brings us back to Roger Ver, variously known as a "Bitcoin entrepreneur" or the "Bitcoin Jesus." Ver is now a citizen of Federation of Saint Kitts and Nevis. He was so excited about avoiding taxes that as soon as he became a Nevisian he set up yet another start up that would allow you to use bitcoins to buy a Saint Kitts and Nevis passport so you too could avoid US taxes. Alas, it folded after a few months, apparently because the St Kitts government disavowed it.

Unlike Facebook billionaire Eduardo Saverin who renounced his citizenship to avoid US taxes back in 2012, I don't get the impression that Ver is remotely that rich. He may be worth a few or even many millions of dollars. But he does not seem remotely in the category of 100s of millions, let alone billions. In any case, now he wants a visa to return to the US to speak at a Bitcoin conference this weekend in Miami. But the US has repeatedly denied his requests. And he's extremely upset at "the tyrants [who] won't allow me to attend #CES2015, #TNABC or anything in the US."

 This is actually a limitation of trying to become an island unto yourself.  When you cut the cords with everyone else, it isn't shocking that they might actually reciprocate the feeling and cut cords with you.

Visualizing a different kind of data

I particularly liked the way he handled the dynamics. Definitely one for the Monday videos.


First Arabesque, by Claude Debussy, performed and animated by Stephen Malinowski.

This could almost get me to start listening to the TED radio hour

Another sharp and well-observed clip from College Humor.

Plagiarism -- repeat offendees

Alex Toth was perhaps the artist's artist in the field of comics and animation. For decades, if you asked people in the industry to name their favorite, Toth would show up with (and sometimes above) names like Will Eisner, Jack Kirby, and Neal Adams. Jules Feiffer famously said that Eisner was the artist other artists stole from. According to this 2006 piece by comics writer and historian Mark Evanier (who worked for most of the major animation houses and comics publishers), Toth was the artist other artists traced.

After a show was sold and they began making episodes, Alex would usually do model sheets to design the characters — both the regulars and the "incidentals." Incidentals are new characters that appear in only one episode…and there would also be model sheets for vehicles, major props, key pieces of scenery, etc. He produced hundreds of these, perhaps thousands, and they are avidly collected by appreciators of fine comic-style illustration. They were a bit more controversial within the studio where some felt that Toth was the wrong guy to have setting the parameters of what everyone else would then have to draw. Everyone admired his drawing ability but there were those who argued that he was either too good or too special.

In most cases, the artists who would have to then draw the characters based on Alex's models had nowhere near his skills, or at least nowhere near the skills for that kind of illustration. H-B did adventure shows but they also continued to produce shows in what one might call the Yogi Bear style. Depending on what kinds of shows were sold each year and how many, it sometimes happened that a "Yogi Bear" artist would get assigned to an adventure show and many of these otherwise skilled artists struggled to replicate the kind of thing Toth was doing on the model sheets. Even a few artists who were solid in adventure-style art found his work too angular at times. Some Toth model sheets were worked over by others — traced and simplified (some would say, "watered-down") — before they went into production. But many artists were also stimulated by the challenge they presented and found that the designs were solid and, as is necessary with animation, designed with an absolute economy of line.

As I mentioned, animators still hoard and trade copies of them. It is not uncommon that an artist assigned to do up a model sheet of a policeman for some new show will haul out a Toth model sheet of a policeman and trace it, making just enough changes to pass it off as new…or not. I have seen Toth-designed characters appear without modification in shows he had nothing to do with, and I expect we always will. It's just part of the grand legacy that the man leaves us.

I'm a bit nervous about posting this at the teaching blog

As previously mentioned, I've been putting together short STEM videos for this weekly feature at You Do the Math. I've also been wasting quite a bit of time at the decidedly not-safe-for-work College Humor site. You might not expect much of an overlap between those two but this clip proved to be the exception.

I'd love to post this on the math education site, but I can't help worrying about the teacher who plays this for his or her class then after it finishes accidentally clicks on the link for... well, pretty much anything else from the site.

Crowdfunding research and the threshold of reportability

Marketplace recently ran this story

"As funding drops, scientists turn to the crowd"

Here's an excerpt:

Susan Nagel from the University of Missouri studies the health impacts of chemicals used in fracking. Last year, Nagel found remnants of fracking chemicals in Colorado streams near locations of previous fracking spills.

“This was an initial study, and we found this kind of strong association,” Nagel says. But she wanted to go farther and confirm her results with more testing.

Her grant application with the National Institutes of Health, however, sat in limbo for months, so she turned to crowdfunding. Nagel set up a project page on the site experiment.com.

“We spent a lot of time on the actual site, developing a video, developing the content to be short but explicit, to be understandable to a broad audience,” Nagel says.

It worked.

Nagel raised $25,000 for a follow-up study. Research like hers is the latest destination for online donors looking to back projects they like. Brian Meece of the crowdfunder RocketHub says science that strikes an emotional chord does better on his site. “Research for animals, research for the environment ... things that are curious, things that are quirky, things that are fun” all do well, Meece says.

There is nothing howlingly bad about this story -- no mangled explanations, no obvious factual errors -- but it's troubling nonetheless because, simply by virtue of occupying a chunk of prime journalistic real estate, it leaves readers with the impression that crowdfunding is playing a substantial part in driving research.

We are talking about a tiny number of people and, in the grand scheme, a trivial amount of money. If you are looking for a philanthropic outlet, experiment.com might be a worthwhile place to send a few dollars, but it doesn't have a place in a serious discussion of how we fund science.

There's a certain natural tension between the desire to cover the unusual and unexpected versus the desire to talk about things that are important and have a large impact (the very fact that something doesn't happen that often tends to limit the number of people it affects ). We therefore cannot expect most stories to be unusual and important, but all too often we see stories that are neither.

Crowdfunding is one of these topics that should often fall in the uninteresting middle. It has been around for years and people use it to raise money for all sorts of things. The fact that someone is using something like Kickstarter to fund a project is no longer, in and of itself, newsworthy.

Crowdfunding also fails the newsworthiness test when we approach it from the other side and look at its impact on society and the economy. With a few niche exceptions (independent filmmaking for example), the role of these sites is almost invariably minor, if not vanishingly small.

We see stories like this partially because they involve what are seen by reporters and, more to the point, editors and publishers as sexy topics that fit nicely with standard narratives. This one involves the internet, the "new economy," and it uses 'crowd' as a prefix. More importantly, it falls squarely in the ever-popular easy-solution genre. No one wants to talk about the tough choice between the nasty political fights and higher taxes needed to fund research adequately on one hand and on the other, the technological and economic stagnation that comes with neglecting the investment. Crowdfunding is a comically inadequate solution but it allows us to put off unpleasant conversations.

Why I am rooting for Netflix (no, really)

Over the past few years, Netflix has become one of this blog's go-to examples for bad business writing. This is partially because there has been a tremendous amount of coverage focused on the company and partially because so much of that coverage has been either logically or factually flawed. After cutting through all of the hype, I find it difficult to be optimistic about Netflix, but I would very much like to be.

Right now, there are three major unlimited streaming services (four if you choose to include Google's YouTube). In addition to Netflix, we have Amazon and Hulu . Amazon has what strikes me as the best business model; unfortunately they also already have a dangerous level of economic power and a history of abusing it (if we include YouTube, the same comments apply to Google). Hulu is a walking antitrust violation in an industry that is already plagued by media consolidation. With competition like this, it is difficult not to root for Netflix.

My suspicion is that CEO Reed Hastings is playing a short-term game, trying to keep all of the balls in the air until he can either find a buyer or make a sufficient killing on the stock. A number of the company's policies feed this impression: the big one is the decision not to build a content library; then there is the tremendously expensive marketing that often seems more focused on building the company's brand among journalists and investors than among  potential consumers; there is the apparent disinterest in controlling costs, something that a company without a deep-pocketed parent cannot manage for long; and there are the disturbingly opaque metrics and bookkeeping practices. These things, along with the general conditions of the market, make me fear that we will not have an independent Netflix for that many years to come.

I do hope I'm wrong though.

Patent Law and Progress

This is Joseph.

This post on Google+ got me thinking about the relation between patent law and technological progress.  Consider:

But 3D printing is an extremely simple concept. Why was 2014 the year it rose to prominence? 

The short answer is that patents are expiring. Many major patents on 3D printing, especially laser-sintering, an inexpensive yet accurate method, expired last year, leaving the concept open to development by any willful person. 

So I can't necessarily vouch for the full accuracy of the statement, but it creates an intriguing explanation for why this technology languished for so long.  Ironically, instead of spurring innovation, patents seemed to completely stymie it in this case.

Episodes like this definitely weaken the "public good" argument behind patent law.  

There's a flying pig joke here somewhere

One of the many annoying things that tech journalist are prone to do is to start out with a somewhat overexcited but generally reasonable account the new innovation then suddenly jump to a speculation that is technically impractical and in some cases theoretically almost impossible.

Case in point (from the Daily Mail but I've seen similar quotes elsewhere):

Greg Henderson, the creator of the Hendo Hoverboard, has created a working prototype of the device which works by using electromagnets to glide an inch off the floor.

The boards will currently only work over metal surfaces, so Mr Henderson is also designing the world's first hover park - similar to a conventional skate park but with an aluminium floor.

However he has also vowed to develop the technology so that it works over every surface, including water, allowing people to recreate the opening sequence of Back To The Future II where Marty uses a hoverboard to escape his pursuers by skating over a lake.

This is so bad I have to wonder if it's a misquote.

Hopefully I will get the physics right on this one (if this gets confusing Wikipedia has good write-ups of Diamagnetism and Eddy Currents), but the hoverboard demonstrated here represents a clever application of very well-established physics and engineering principles. Specifically, if you pass magnets over a conductor like copper they produce currents which in turn produces magnetic fields. The interaction between these fields can be used, among other things, to levitate the magnets over the conducting surface. This is one of the technologies that is used for maglev trains.

Historically, one of the problems with this technology is that the magnets have to be in motion relative to the conductor. Henderson appears to have gotten around this problem by having the magnets rotate in the hover board so that the board can levitate while stationary. That is a clever piece of engineering and it has certainly produced a cool piece of technology. Furthermore, it is easy to imagine lots of potential applications ranging from recreation (skate parks) to manufacturing (moving loads across a factory floor).

What is difficult to imagine is how this could work "over every surface." It is true that all materials have diamagnetic properties which means that a magnetic field can repel substances that we normally think of as non-magnetic, but the magnitudes are tiny compared to the effects you can get with the conductor and moving magnets arrangement mentioned above. It takes incredibly powerful magnetic fields to do this...

...far more powerful than anyone will manage with a skateboard in the near future.

All of which raises the question, why exaggerate? We see this all the time in the tech world. Developers come up with a fun, exciting, potentially useful innovation, then for no apparent reason, they feel compelled to make outlandish claims about cost and functionality. If I were building actual, honest-to-god levitating skateboards, I don't think I would feel the need to embellish the cool factor.

The real challenge of infrastructure

This is Joseph

Scott Lemieux points out that the city of Seattle is now planning another major tunnel project.  What makes this unfortunate is that Seattle is currently home to a fairly unsuccessful tunnel project at the moment.  But it was not inevitable that the current project would work out so poorly, nor is it unclear that things could not be further improved upon in the current project.  Boston's big dig project had a lot of problems, was way over budget, but ended up providing what look like quite good infrastructure results.

What worries me more is that these sorts of ineffective infrastructure projects undermine public support for doing more of them.  To be a rich and successful country, it seems that infrastructure is a requirement.  Or at least I cannot think of any examples of countries with crumbling and defective infrastructure that are doing especially well.  The Eastern bloc, for example, has these issues and doesn't seem to be a powerhouse of economic dynamism at the moment (I happily solicit  contrary opinions from readers).

What has changed to make infrastructure projects slow, expensive, and subject to a lot of problems in execution?  I could see any two of the three if we had major pay-offs on the third axis (e.g. slow and problem-ridden but cheap could make a lot of sense).

And how do we change this?

Innovative business models

Mark Evanier has an interesting observation about an extended warranty for an external hard disk he bought at Amazon.

A 2-Year Data Recovery Plan for $9.95. This proposition gives me more information. It says, "If your drive stops working, the Rescue data recovery plan will recover the data from the failed drive and return it to you on a new piece of external storage." That sounds interesting until you get to this line: "If your data isn't recovered, you get your money back."
...

Okay, so follow me on this. Let's imagine that I have no technical capability to recover data from a damaged drive. None whatsoever.

I offer this deal. When you buy a new hard disk, you send me ten bucks. If the drive never fails you, I keep the ten bucks.

If the drive does fail, you send it to me and I send it back to you with your ten bucks and say, "Sorry, I couldn't recover your data." I don't even have to send you that new piece of external storage I send you if I do recover your data, which I don't even try to do.

What is my potential loss here? Well, I haven't investigated far enough to know but I suspect when you send me your drive for possible data rescue, you have to pay a postage and handling fee for its return. If you do, I'm out nothing. I might even make a few more bucks off that postage and handling fee. If you don't, then I'm out the cost of sending you your refund and returning your drive.

Charles Pierce shows how journalists should acknowledge a mistake

No excuses. No "if anyone was offended." Just take responsibility and apologize to everybody.










Then compare yourself to a character in a Monty Python sketch.