Traces Newsletter #23: The Mobilization State

Last night’s newsletter today. If you like the newsletter you should sign up here, as sometimes I post these here and sometimes I don’t.

No main story this time. Please note a new edition to our format — some stories are marked “evolving”. These are stories which have caught my interest, but where the story is nascent enough that I haven’t seen it reported on a solid site yet. I’ve struggled with these stories, because one of the things I like to do is share early-stage stories here: this tag is my compromise. “Evolving” means simply “There’s a story out there that were keeping an eye on, treat with caution.”

Killing the town square to “save” it

Apple calls its stores town squares, which makes us all throw up in our mouths a little.

Uber won’t be fined by Portland for flagrant violations of the law.

YouTube may be banning LGBTQ+ gaming videos from making money on the platform, just for mentioning bad words like “lesbian”. [evolving]

Paleofake Persistence

Warm water doesn’t help clean your hands, but no one can kill zombie advice, so we pour tons of carbon into the atmosphere just because.

Global Potemkin Village

Fighting fake news in Germany (with success, actually).

An alt-right emerges online in South Korea.

BBC sorts out the “fake news” issues in the Myanmar conflict – short answer: yes, there is some exageration and fakery, but when journalists are banned what do you expect?

Blue Whale hysteria hits India

Kenya continues to investigate fake news aimed at stirring ethnic tensions, with an eye toward prosecution.

1984, Inc.

According to a new-ish report, Vladmir Putin doesn’t so much oversee a totalitarian state as a “mobilization state” – private, individual actors are mobilized for state purposes, without being directly controlled by the state. “The government is willing – within certain bounds – to accept the presence of civil society, a free press, independent economic activity, and even some political pluralism. However, in keeping with its general philosophical belief that it is at (political) war and faces an existential cultural and political threat from the West, it reserves to itself the right to co-opt any individual or organisation when it feels the need.” This is the new face of authoritarianism.

ACLU is suing to stop warrantless phone searches at the border. Probably a lot better use of their time than defending Nazis.

The EFF has resigned from the W3C over the W3C’s reckless push for a standard of web-based DRM that makes everyone less safe.

WikiTribune was going to have a new wiki approach to news, and I’m interested in that. And I like founder Jimmy Wales, who has been a great influence on the web, of course. But I’m very worried that the first article they’ve put out is a puff piece about an organization Jimmy Wales is working with. Actual lede: “An unusual coalition of figures from the movie and music industries, fantastically wealthy philanthropists, human rights leaders, politicians, and diplomats are pushing against the nationalism and gloom evident since Brexit and the U.S. elections to push for bold new goals to combat poverty, health and security.” And it’s a crowd Wales will be talking to. We need better than this, please. (And is it just me or is that sentence really hard to parse?)

Facebook is becoming the internet, and when seen through that lens it looks more like China than a modern democracy. [NYT video] Also, from April, We Need More Alternatives to Facebook.

There’s a conservative case to be made for busting the Google-Amazon-Facebook monopolies.

In light of recent events, you probably want to read Bruce Schneir’s old post on Data as a Toxic Asset. The crazy but true thing proposed there – companies hold onto toxic data because it’s the only way to justify out of whack valuations. They risk your life and our democracy not even for direct profit, but for investor hand-waving. They are risking everything to preserve their elevator pitch.

BTW the world is burning

It turns out we may have a chance – just sliver of a chance – of avoiding the catastrophic levels of global warming. No secret how: stop pumping carbon into the air, abide by current agreements. Under another administratiion this would be amazingly hopeful news. [evolving]

Crapjects for a Post-Truth World

If you teach this stuff to students you should probably know what “prank” site generators are.

Polarization and the Twilight of the Elites

Hillary Clinton finds that de Tocqueville got there first. According to his treatment of the French revolution, revolts start not in the places that have it worst, but in places with the biggest expectation gaps.

This paper argues that engaging a frame of “civic duty” leads to more rational policy preference formation when processing information. That would be great if true, and could be a path to saving the world. Too bad the article’s paywalled.

Justin Murphy argues that the data we have (at least presently) shows very little movement towards the conservative or liberal poles in the U.S. in the past few decades, and that the only long term trend is increasing social liberalism.

If that’s true, then how do we explain polarization? I’m going to come back to the point I keep pressing: polarization is driven by elites. And where I am right now on it is that that elite polarization is largely driven by strategic concerns and outcomes, both in media and political life. I’m not 100% sure on this, but that’s where I am at the moment. Elites polarize for strategic reasons, partisans follow the cues. That’s not to say that some ideas on the “fringes” aren’t right — but when we look at why polarization seems asymmetric, for examples, we should find our answers as much in strategy as psychology.

Related to that, an interesting paper that I haven’t dug into yet that shows the impact on partisan cues on reasoning.

Via Maria Popova, a reminder from Karl Popper: “Since we can never know anything for sure, it is simply not worth searching for certainty; but it is well worth searching for truth; and we do this chiefly by searching for mistakes, so that we can correct them.” This confusion – that truth and certainty are the same – underlies much disinfo and conspiracy theory.

The Rock 2020

Conspiracist Alex Jones has said he might support The Rock if he runs. I won’t link – you can look it up. I think Johnson’s better move is to run on the Democratic side if he’s running in 2020: the math on presidential same-party challenges is pretty bad. (Note: I’m not saying that it’s good or bad he runs, I’m just saying there’s a good chance he’s going to run and do fairly well).

Advertisements

How To Participate In Digipo (September 2017 version)

Every time I say I can’t make it easier to participate in Digipo, I find a way to make it easier.

The current process involves no skills greater than knowing how to work a word processor, and (more importantly) allows students to participate anonymously if they wish, without having to sign up for Google accounts or have edits tracked under pseudonyms. We accomplish this through a Microsoft Word template and by submitting the files into public domain.

You can of course use a more complex process, sign your name to the article, and use Google Docs as your central tool. Depending on your needs and skill level you may want to do that. It’s just not required anymore.

Here’s the steps.

  1. Read (at least some) of the book.
  2. Pick a question to investigate from our list of 300+ questions, or make up your own.
  3. Have your students download this Microsoft Word template that guides them through an investigation of a question. Apply the skills from the book.
  4. Do whatever sort of grading, assessment, or feedback you want.
  5. Take student reports where the students have agreed to submit them into public domain, and zip up the word documents. Mail them to michael.caulfield@wsu.edu. Make sure you introduce who you are, what the class is about, and a bit about your experience as I do not open zip files from random people. Also give me a blurb about how your class would like to be identified on the site (they have the option  of remaining anonymous too). For verification purposes, send it from your university account. I may email back to verify.
  6. I’ll put them on the Digipo site in a subdirectory with a bit about your class and give you a password that allows them to edit online going forward.
  7. At a later point we’ll assemble a small panel of professors who will go through the student work and choose ones to “promote” to the main directory based on quality. The key question reviewers will ask is whether the document provides better information than at least one of the top ten Google results for the question.

That’s it!

 

 

That Watson for OER Pitch Is Classic Information Underload

Saw this today. Watson is going to solve OER!

1000

Wait what? Only a thousand lessons?

Divide that by six grade levels and that’s about 166 lessons per grade. Figure you probably teach 80 lessons a year…

So Watson, with its supposed brain the size of a planet, is going to do what for you exactly? Whittle down 200 lessons to 80?

Not exactly HAL, is it?

We keep doing crap like this, proposing “discovery” solutions to content production problems. To find out why we have to stop focusing on the wrong end of this, read this from me and this from David Wiley.

When Fox News Is Not a Bad Choice

Misinformation is an asymmetrical phenomenon, occurring more in sources followed by Republicans than Democrats. There are historical reasons that explain this: the creation of a right-wing media system was heavily funded and subsidized by corporations and donors in a way that left-wing media never has been. This isn’t to say that the left couldn’t become much like the right in this respect, and there are indications it may go that way in time. It just isn’t there yet.

This asymmetry puts educators in a bit of a bind, since skills training presented to students can come off as partisan, even when it isn’t. And if it is perceived as partisan, many students will dismiss it. No one wants to learn a set of skills that seem to threaten their core identity. Not conservatives, not liberals, not leftists, not centrists. As such, it’s important to look for examples that demonstrate the lessons students need to learn but that at least occasionally upend partisan expectations.

Here’s an example of something along these lines. In this set of Google results, to a person not knowing these sources, Fox News is the best choice:

seventh.PNG

Can you see why? Here’s a close up:

admiral

We can see in the snippet that there are sources to this story, called out by name (or at least position). Now I wouldn’t trust that headline (you should never trust headlines from anybody, really) and I wouldn’t take the Navy’s word for it. But it’s clear here from the snippet that this article will give you some sources you can follow up on (or, as we say, allow you to go upstream to the source).

And when you click on that article, you do find it is well sourced, and that it draws most of its info from a McClatchy news service article.

McClatchy2

And you can take that information and find the article Fox is using (which apparently it finds a valid source since it is citing it) and find it with a simple search modification (adding McClatchy to the search):

mcclatchy.PNG

And it turns out the McClatchy article that Fox is quoting has a great deal of research behind it (and, notably, contradicts what the Navy is saying). But what Fox News’s print articles do (at least a portion of the time) is follow some of the journalistic norms which allow you to understand where information is coming from and, if possible, track it to its source.

I’ll repeat, before I get slammed in the comments. This is not meant to be a defense of Fox News. But if you want to help people get better at navigating their information environments, you have to start where they are. And part of that might be talking about times when Fox might be a decent choice for a click, and how tracing Fox News claims to the source can provide a deeper, more informed reading experience.

A State Sales Tax on Personal Data

I live in Washington. If I go and buy a USB drive for $10, Washington gets about 80 cents. If I buy a copy of Microsoft Word for $100, Washington State gets more than $8.

Services aren’t taxed, but probably should be. If I buy a year of Microsoft Word online, it’s not clear to me why I shouldn’t also pay a sales tax on that product. In both cases, I am merely licensing the software from Microsoft. It seems to me as we’ve moved from products to services we’ve given a real unfair advantages to services here. And since the future of software and media is services, we are creating an untaxable set of products, and slowly starving states of revenue.

But the bigger problem is this — when I use Facebook or Google Docs I don’t pay with cash, I pay with data.  Google gives me use of the Google Docs platform in exchange for collecting valuable information  about the sites I visit, the products I seem interested in, my daily location (tracked by my phone) throughout the day. With a little settings jiggering I can reduce the amount of data I provide, just as coupon clippers can reduce the amount of money they spend. But I can’t ever get the price to zero.

The fact that data is revenue that is not taxed creates perverse incentives to collect it and is also partially behind the downward pressure on state budgets. The data economy exists not only unregulated, but untaxed, and it’s probably no surprise that in a time of massive market capitalizations (based in many cases on accumulated data held by companies) state budgets are dying.

So I propose that states tax the data you provide to corporations such as Google and Facebook. It would be a seller-paid tax, not visible to the customer. But it would capture the value of the data collected (and maybe held as well) on us all.  Doing this would apply at least some brakes on surveillance capitalism, especially if the tax was set relatively high — you can certainly make the case that the data risk created by companies is as great as the damage caused by alcohol sales, which in Washington State are taxed at 17%. And while it may seem absurd, I might even tax the data by the kilobyte.

Why? Because, properly constructed, while such a tax wouldn’t ban the collection of data, it would stop the gathering of data by default. Companies would have to make decisions about whether it was worth collecting and storing every location I go to during the day, every website I visit, every search I plug into a search engine. Heck, it may be the first time that companies would be forced to publicly account the amount for the amount of data they hold on citizens.

And while kilobytes are not a precise measure of either profit or risk, for that matter neither is the alcohol tax. An expensive bottle of wine does no more damage to society than a cheap one, but people pay more taxes on it, and a $10 bottle with a profit of a couple bucks is taxed the same as a $10 bottle with a profit of 20 cents.

In other words, don’t get hung up on the precision of targeting here: look at the potential effect: companies slow down collection of data, state budgets get a desperately needed new budget stream, and as a bonus people who charge money for products might actually have a chance to compete with the surveillance companies like Google. I don’t claim that this proposal has all the details worked out. But it sees to me to be the right thing to do.

Let’s Tax the Collection and Storage of Personal Information

The effects of data theft have been enormous this past year, but the situation is likely to get worse.

Why? Because right now we are looking at the theft of only slivers of fairly traditional information. But we now live in a world where every device and service you consume is collecting data on you. That data can help others rob your house, hijack your bank account, destroy your reputation, and spy on your children. It can out your teenage child as gay, or reveal your psychological issues. It can help others punt you out of a job, or blackmail you over things you did, or things that can be made to look as if they were done by you. Pull multiple sources together and you’ll find that companies store near perfect surveillance of you throughout your day. In turn this information can be leveraged into access which allows others to be able to do things on your behalf, whether it’s activating anti-theft on your car or posting anti-semetic comments to all your friends. Or nude pictures from your Amazon Look.

And if (as?) the government tracks towards authoritarian rule, this information all goes to a state that can punish you for whatever crimes the current regime deems criminal. You realize that Google and Apple keep a near perfect record of who visited Planned Parenthood in a pattern that indicates an abortion, right? That that DNA testing agency you used to see if you were Scottish knows all your pre-exisiting conditions, right? That the affair you and your wife worked through privately is meticulously documented by Uber?

Think about it: 2016 was the year of ransomware,  where malicious software was used to lock up people’s computers unless they paid a small fee. Future ransomware will not lock up your computer — it will lock up your life, threatening to destroy it pending payment or a promise to not be making trouble for the wrong people.

Tax the Data

In light of this social threat, you’d think that companies might reduce the amount of risk they pass onto you by limiting the amount of information they collect. But the ad-fueled surveillance model of the modern web has pushed them in the other direction.

To halt this trend, we could regulate data collection, and we could punish data loss to theft more harshly. We should explore both options. But I would like to propose another option as well.

Tax the data.

Tax, yearly, the data that these companies hold on you. Put a sales tax, payable by the companies, on the data they collect from you. Tax them yearly for holding on to that data.

This may sound absurd, but it shouldn’t. Companies have admitted that data is their greatest asset. And data is actually what we are paying for these systems instead of cash. We give them our personal data, they give us services.

Now I’m not a tax genius. But I know that if I did consulting work for you and you paid me with a car, that car is income. It’s taxable. It’s also potentially taxable as property. If at the end of the year I have a bunch of cars sitting around from clients, that’s profit. The IRS doesn’t say, well these are cars, they are not payment. So why when I give Google my personal data in exchange for services is that not income?

So tax it. And because it has worse effects on society than holding on to money, tax it at a higher rate, and in more ways.

Taxing personal data collected from individuals would force companies to make decisions about what data to collect and keep with the externalities priced in. We could imagine pricing data on individuals at about $10/MB, and taxing it yearly at 10%. To remind you: a megabyte is quite a lot of data. The entire text of Shakespeare’s Hamlet is less than a fifth of a megabyte. If Google and Equifax won’t pay $1 a year to hold a megabyte on you, then clearly the social risks of holding that data outweigh the benefits. No one doubts that holding a megabyte of data on someone confers more more social risk than a dollar a year. If that data is not worth that to them, they should let it go.

Will it make tech more expensive? Maybe. But it will make it expensive in the right ways. It will force tech to account for the massive risks they are pushing back on society. And it would fairly tax assets that have flown below the radar of traditional public policy.

How the Independent created a fake news Facebook card out of a real story

Here’s a thing going around Facebook today: Chicago mayor Rahm Emmanuel banned Trump from Chicago!

So did Rahm just go Rahm-bo? Did he ban Trump from the city? Clicking through and seeing the headline on the actual article suggests a less dramatic story:

And the quote in context?

[Chicago Mayor Rahm Emmanuel] added: “Chicago, our schools, our neighborhoods, our city, as it relates to what President Trump said, will be a Trump-free zone. You have nothing to worry about. And I want you to know this, and I want your families to know this. And rest assured, I want you to come to school … and pursue your dreams.” [italics mine]

So in other words, the Facebook headline is a complete lie. Emmanuel was simply saying he was not going to spend resources to enforce federal law in Chicago. Trump himself has not been banned from anything at all.

So where did that initial headline come from? The clickbait one on Facebook which tells a flat-out lie? Was it made by Facebook? Or added by a deceitful user?

Nope. It was written by the Independent.

Let me repeat that. The Independent wrote the fake headline.

So why do we not see it on the Independent site? Well, a little known fact about newspapers and other websites is they embed code in invisible HTML “meta” tags that provide different headlines to different platforms, when the content is shared. And if we look in those meta tags we see that someone at the Independent coded the false headline in the meta tags, even though they would never dare publish such a headline on their web site.

Here’s the title you’ll see in your browser bar or tab (and the title that used to be shared with sharing services) as it appears in the HTML:

<title>Chicago mayor declares city ‘Trump-free zone’ after US President declares he will scrap DACA immigration programme | The Independent</title>

Here’s the title that people see, again, as it sits in the HTML on the actual page:

<h1 itemprop=”headline” class=” “>Chicago mayor declares city ‘Trump-free zone’ after US President declares he will scrap DACA immigration programme</h1>

And here is where the HTML tells Twitter and Facebook what to use:

<meta property=”og:title” content=”Chicago just banned Donald Trump from the city” />

<meta name=”twitter:title” content=”Chicago just banned Donald Trump from the city” />

The only places in that code where the Independent even mentions a “ban” are visible in Facebook and Twitter but not on the site, so the site can publish clickbait into social platforms while still retaining a shred of respectability on its website. And if people complain about the Facebook headline, they can always point to the headline on their site as being more or less valid (although it’s still horrible, tbh).

Does this sort of deception work? Why, yes it does. This story has been shared on Facebook almost half a million times in 24 hours. That likely makes it one of the top shares, if not the top share of the day. And they accomplish this by abuse of the platform-specific headline codes. The whole thing is shameful, and an insult to the good work that the Independent’s reporters do. And it’s time for it to stop.

Solutions

Normally I don’t offer bullet-pointed solutions to things. But the solutions are almost ridiculously simple here. It’s just a matter of will, ethics, and incentives to get them done:

  • Papers: stop doing this. Apply procedures and oversight to meta tags.
  • Facebook: stop tolerating it. Scanning the semantic difference between og:title and <h1> title is an easy fact check. Write code to do that and flag offenders. This is Spam 101.
  • People: Click through before you share. Always. And demand better from established papers. The “reverse mullet” headline (party in the front, business in the back) must die, once and for all.