Blog Ipsa Loquitur

Hey, remember that story from back in December about Seattle’s new tunnel project, and how poorly construction is going? A 500,000 pound machine custom-built to dig this one tunnel broke down after making it about one-tenth of the total distance it’s scheduled to dig. Ominous!

It started out as a boondoggle, and it’s only gotten worse from there. I find this whole thing baffling beyond comprehension, but Karen Weise wrote a wonderful story for Bloomberg Business about how it all went so wrong. Basically, this particular project isn’t special or remarkable at all. Gigantic public works projects are always way more expensive and way more time-consuming than the contractors are willing to say. Seattle is no different.

Bent Flyvbjerg, a professor at Oxford’s Saïd School of Business, has followed [Seattle’s problems] from afar. His research on megaprojects has been cited by both backers and critics of the tunnel. Nine times out of 10, massive infrastructure jobs go over budget, he says. Tunnels on average cost 34 percent more than anticipated. No region is better at predicting costs, and estimates over the past century haven’t become more accurate, his data show.

Wow. Those are some pretty earth-shattering numbers. How can experts get these so wrong almost all the time? Whether that means that big public works contractors are lying or simply inept, it’s hard to tell.

Just kidding!

The [Washington State Department of Transportation] WSDOT awarded a $1.4 billion design-build contract to STP, a joint venture between Tutor Perini, a California-based construction company with $4.5 billion in annual revenue, and Dragados USA, the local division of a Spanish company with an expertise in tunneling. […]

STP beat out another consortium in part by estimating it could finish the project by December 2015, 11 months ahead of the state’s schedule. It was time to stop hyperventilating, the state said. “With this contract, we are confident that the tunnel will be built within budget and delivered on time,” declared Paula Hammond in 2011, then the head of WSDOT. (She left in 2013 for the engineering firm that led the reviews.)

Got that? The company that Seattle paid to double-check whether the project would be a disaster gave it a thumbs-up. And then they gave the official in charge of the contract a paycheck. Oh, and by the way, that whole “eleven months ahead of schedule” bit might have been off a little; the tunnel is roughly two years behind schedule. Nothing shady here, though! This is definitely a very large surprise to the reviewing firm. Yes.

Aside from the conspicuous appearance of corruption, Seattle’s ongoing tunnel problems are impressive. According to Weise’s article, the machine hadn’t even begun digging through the difficult part of the terrain when it broke down. I’m sure this story will keep spiraling out of control into a cautionary tale that another local government will ignore at their own peril. I can’t wait!

Published on under Procurement Hell

Shot: Texas court throws out “upskirt” photo law.

The state’s highest criminal court on Wednesday tossed out part of a Texas law banning “improper photography or visual recording” - surreptitious images acquired in public for sexual gratification, often called “upskirting” or “downblousing” - as a violation of federal free-speech rights and an improper restriction on a person’s right to individual thoughts. […]

In an 8-1 ruling, the Texas Court of Criminal Appeals said photos, like paintings, films and books, are “inherently expressive” and, therefore, are protected by the First Amendment.

Chaser: Texas Bill Would Make It Illegal for You to Film a Cop Beating You.

Section 38.15 of the Texas Penal Code makes it an offense to interrupt, disrupt, impede, or otherwise interfere with “public duties,” including those being exercised by a police officer. That’s the law pretty much everywhere, of course, but the question that has arisen in recent years is whether you are “interfering” (etc.) with a police officer just because you are recording what he or she is doing. The Texas statute doesn’t say anything specific about recording, although it does say a person can’t be prosecuted if the interfering acts “consisted of speech only.” […]

Okay, now along comes Rep. Jason Villalba (R-Dallas) with H.B. 2918. This bill would amend section 38.15 to expressly include within the definition of “interference” the conduct of “filming, recording, photographing, or documenting the officer within 25 feet of the officer,” or doing so “within 100 feet of the officer” if you are also carrying a concealed handgun.

Sure, maybe the upskirt law was a little overbroad, and it stepped on the toes of the first amendment. I get it. I don’t know that I would have tossed out the whole statute, but hey. Texas doesn’t mess around with free speech.

But come on. If you have the right to film kids in bathing suits (no, really, that’s what that case was about), then you should at least be allowed to film a cop. Especially one who’s beating you.

Published on under Not the Onion

Mike Fabio wrote a great post about Spotify a couple weeks back. It accompanied the relaunch of Tidal, a streaming music service owned in part by Jay-Z. Tidal had an elaborate press event in which a parade of artists took the stage to decry the pittance they’re paid by other streaming music services.

Here’s the gist:

It’s scary for artists to learn how many people have listened to their music, and compare to the fractional royalty statements they’re being sent. Thing is, it’s not Spotify’s fault.

It’s the labels.

See, all those artists on the stage are signed to labels. Their contracts dictate that the music they record is owned by those labels, sometimes in perpetuity. And most of those artists have publishing deals that take a chunk out of their performance and mechanical royalties.

The reason artists don’t get paid from streaming services is that they don’t own the music that they record.

I’d take issue with that “music is owned by the labels in perpetuity” bit, but otherwise, Fabio’s post is great and you should read every word.

As an aside, copyright lasts for 70 years after the artist dies, which probably feels like an eternity, but it’s not. Let’s just agree that record companies own the copyrights to the music for almost forever.

That’s important, because when you own the copyrights, you get the money. And there’s a lot of money here. Take Pandora. They paid record companies over $440 million last year. Subscription music services as a whole paid out $1.57 billion in the year 2014.

That same Guardian story quotes the same music industry trade group, the International Federation of the Phonographic Industry, which says the global music industry as a whole generated just under $15 billion last year. So streaming music is just about one-tenth of the music industry.

Well, for now. Streaming music revenue was up 39% from 2013 to 2014. That’s a pretty insane jump. Artists complaining about streaming music services are looking to the future.

But really, if artists are upset at the size of their royalty checks, they ought to be looking at record companies, not streaming music services.

Published on under The Digital Age

Walter Scott was shot in the back three days ago by a police officer in South Carolina. The police officer claimed that Scott ran (on foot) from the police, seized the officer’s Taser stun gun, and attempted to use it against him.

But that’s not what actually happened. The New York Times has video of the shooting, and it’s horrifying. The officer uses his Taser on Scott, and Scott runs from the officer. The officer fires wildly at the unarmed Scott and kills him. And then it gets worse.

The officer walks over to Scott’s body, handcuffs him, walks back to the spot where he dropped the Taser, picks up the Taser, and drops the Taser next to the corpse. This is horrifying. This should really, really upset you. This is shooting an unarmed man in the back, this is tampering with a crime scene, this is lying about whether or not the unarmed guy was armed, and so on.

I’m not an expert on South Carolina laws about self defense, but I don’t have to be. In 1986, the Supreme Court ruled in Tennessee v. Garner that it’s unconstitutional to simply shoot fleeing unarmed suspects. The police cannot use deadly force (firing a pistol is always deadly force) unless it’s necessary to prevent the escape of a fleeing suspect who the officer has probable cause to believe “poses a significant threat of death or serious physical injury to the officer or others.”

The Worst Part(s)

Walter Scott is dead. But here’s the second-worst part.

Who was the highly dangerous Walter Scott? What violent felony were the police attempting to halt his flight from? If he is such a hardened criminal, why did he forget to bring his own flamethrower?

It was a traffic stop. Scott’s Mercedes had a taillight out. It wasn’t a felony. It wasn’t even a misdemeanor.

It actually wasn’t even a violation. South Carolina only requires cars to have a single taillight. Scott wasn’t even breaking the law. The officer was apparently as ignorant about the legality of shooting unarmed fleeing suspects as he was about the niceties of the local traffic law.

If this taillight business sounds familiar, that’s because the ink is barely dry on a Supreme Court opinion about the consequences of confusion around a broken taillight. That case happened in North Carolina, where a police officer was mistaken about the legality of having a broken taillight. During the traffic stop and subsequent search, the police found drugs in the car. That used to be an illegal search, but the Supreme Court ruled that reasonable ignorance of the law is okay.

Of course, reasonable ignorance of the law usually doesn’t lead straight to wild shootouts where only one of the parties has a gun. That’s a special kind of ignorance. I can’t wait to see how the local prosecutor fails to indict this one.

Published on under Nuke Us From Orbit

Adam Steinbaugh goes full law nerd on the Revenge Porn (I hate that term) industry’s first prison sentence. It couldn’t happen to a nicer guy.

Kevin Bollaert was sentenced by a San Diego court to 18 years in prison following his February conviction on twenty-seven counts of extortion and identity theft. Bollaert was the chief operator of (later, which published the nude photos of over ten thousand men and women, almost uniformly without their permission.

Bollaert, with his partner Eric Chanson, monetized his site with both advertisements and a more devious plan. He launched, which ‘advertised’ on YouGotPosted as an independent service which could remove the photos from YouGotPosted in exchange for a few hundred dollars. […]

Kevin Bollaert is the first revenge porn site operator to be convicted and sent to prison. Hunter Moore, widely viewed as the progenitor of revenge porn sites, will be sentenced in June after taking a plea deal on CFAA charges. Craig Brittain, the patron saint of harnessing revenge porn as a means of extortion, somehow escaped with only a light bruising on the wrist from the FTC, despite having the most bizarre (and malicious) plot.

This is great news, and hopefully this is the first of many jail sentences for people like this. Al Franken would like to see a world like that, too. However, Steinbaugh isn’t just reporting, he’s analyzing. He examines two potential routes for Bollaert’s appeal, and actually finds one of them a probable winner. Which sucks. We’re both left hoping that the appellate court finds any excuse to keep Bollaert in jail, even if bad facts make for bad laws.

Published on under Motion to Point and Laugh

Yesterday, Apple said another thing about its Apple Watch; it previously announced when people could get one, but now we know when Apple will officially grant permission for money to be thrown in their general direction. The Apple Watch is probably going to make a lot of money, but Apple’s other big development in the last month is ResearchKit.

Michael McConnell wrote last month about ResearchKit, but the gist is that it makes at much easier for people to volunteer to participate in universities’ medical studies. For example, eleven thousand people enrolled in a Stanford University study virtually overnight. It usually takes 50 medical centers a year to enroll even ten thousand folks, so this is kind of huge.

It’s huge, but not because an iPhone ought to be a gateway to participating in medical research. That’s called selection bias and it’s bad. Rather, ResearchKit could serve as a model for using technology to improve medical research, no matter what kind of phone someone has.

ResearchKit is still in very early days, but I have sky-high expectations for it. Using smartphones to provide medical researchers with the information that will help them cure disorders caused by overuse of smartphones is beyond poetic. It represents the best and worst and silliest in humanity, and that’s something I can get behind.

Published on under Disrupt Everything