Werd I/O

Aaron Chimbel, over on PBS MediaShift:

At most universities, students are required to take English composition courses, and at many others speech and/or foreign language classes are also required. Yet in the debate about teaching code in journalism programs, code is often reduced to a shiny toy.

I've argued before that learning to code is not the same as being a coder, and that some degree of digital literacy is useful in a world that is slowly being eaten by software. Most recently, that was shown by the Silk Road trial, where a single investigator found the marketplace's founder using a simple Google search, two years into the investigation.

An understanding of how to put a live website together is handy. For a long time, I was a subscriber to the NICAR-L list, a mailing list full of journalists discussing computer-assisted reporting. It's often about things like embedding an OpenStreetMap map using Leaflet, or otherwise wiring up a simple dataset to a visualization. A little light JavaScript hacking, and perhaps some HTML and CSS.

It's actually pretty similar to the kinds of things Erin did when she visualized her Known checkins. (Known exports checkins as both KML and GeoRSS.) And while it's fun to do this kind of tinkering, and is quite a long way away from coding, I don't think it's good enough.

Journalists, and people like them, need better software, which makes these links more obvious. Taking location data from a platform like Known and bringing it into Google Maps or OpenStreetMap should be a one-click (or drag) operation. For us, that'll become more important later this year, when we release more data-centric tools. But it should be a given for everyone. You shouldn't need to know an arcane URL parameter, or understand that KML exists, to be able to manipulate your data in the way that you need.

We spend so much time talking about data that's locked in through business models and terms and conditions that sometimes we forget about data that's locked through design decisions. Letting your data flow freely is part of an open web.

The musician Zoë Keating (who I admire) has written a very transparent, and a little sad, post about the decision she needs to make about YouTube:

They were nice and took time to explain everything clearly to me, but the message was firm: I have to decide. I need to sign on to the new Youtube music services agreement or I will have my Youtube channel blocked.

So what's in the YouTube music services agreement?

It turns out to be a five-year contract that requires her to make all of her catalog available for free on YouTube, with an ad on it. In other words, YouTube requires her to relinquish control over what she releases where.

Is such control too much for an artist to ask for in 2015? It’s one thing for individuals to upload all my music for free listening (it doesn’t bother me). It’s another thing entirely for a major corporation to force me to. I was encouraged to participate and now, after I’m invested, I’m being pressured into something I don’t want to do.

Yet this is the web! Zoë should be able to post her content to her own website, as well as services like Bandcamp, SoundCloud and the Pirate Bay. And she does do all those things (including, progressively, making her content available over BitTorrent). But YouTube gives her something that she feels she can't get simply by sticking her shield out on the web at large: audience.

This is the same reason that people are choosing to post to Medium rather than their own blogs. It's why photographers initially flocked to Google+, and why you'll find so many on Instagram now. It's because there are ready-made network effects that artists can harness in order to obtain greater reach, and ultimately get paid for what they do.

When a single entity controls the audience for a particular medium, as YouTube now effectively does, they can leverage control over the artists. The result is a worse situation both for artists and their audiences, as the activity of both is shaped to fit the platform owner's interests. From an artist's perspective, the terms demanded by sites like YouTube can feel predatory and invasive. In particular, why would mass market listeners pay for Zoë's album on her terms, when they can listen for free on YouTube's site?

The solution is to build the audience-generating network effects employed by YouTube to the web itself. It's not just about decentralized conversations: it's about driving traffic to artists' own websites, and allowing for organic, seamless discovery of media wherever it lives online. Ironically, it could be argued that this is what Google - still the web's go-to search engine provider - should really be doing.

So, fine. Google is choosing not to do that, and to build value into its own platform. In the best tradition of Internet technology, that leaves room for someone else to fill the gap. And in the best tradition of Internet technology, they will. It's just a matter of time.

If you've read this far, you should go get yourself a free Known site. Publish on your own site using a variety of media, and share it across the web.

This is a summarized version of my talk at Hacks and Hackers tonight at Matter in San Francisco.

This last summer, Alberto Guzman, a hairdresser in New York, uploaded a picture of him and his husband sharing a kiss on their wedding day. He tagged it with the hashtag #happy, as well as #gaypride and #equalrights. It was a wonderful day for them both.

And then Instagram removed it - for being inappropriate.

It turned out that it had been flagged as inappropriate, which is very easy to do on Instagram. You just click flag, and then tell them why. There isn’t a lengthy procedure to go through, and the moderators typically remove items very quickly.

This happens all the time. Famously, it isn’t safe to upload photos containing breastfeeding to Facebook, because they’re flagged as pornography. The company changed its policy a few years ago, but as recently as last year, mothers were demonstrating outside of Facebook’s headquarters because their photos were still being removed.

Late last year, drag queens had their profiles frozen and removed from Facebook because they weren’t using their “real names”.

In each and every case, these deletions were caused by organized groups of users who wanted the content gone because they didn’t like it.

It’s not just photos of motherhood and same-sex marriages that are being removed.

As the British blogger Elliot Higgins noted last year, Facebook pages about the sarin gas attacks in Syria have also almost all been removed. History is being rewritten.

In fact, the problem is so bad that when Mark Zuckerberg posted in favor of free speech following the awful events in Paris last week, a legitimate question about freedom of expression in the comments was flagged and removed.

These aren’t government requests. All of these were due to ordinary people: civilians making a decision about what people should be allowed to publish and how they should define their identities.

I'm picking on Facebook because they're big. 835 million people access Facebook alone every day. It’s how they get their news, how they talk to their friends, how they learn about the world. Smartphone users check Facebook 14 times a day on average.

We’re all familiar with Edward Snowden, and his revelations about illegal NSA surveillance. But the truth is that we’re all spying on each other, too. The content standards that Facebook sets, and its policies regarding inappropriate content, have a real impact on how people learn about the world.

And it’s a real impact. The PEN American Center found that the number of writers in democracies who report that they self-censor the topics they write about is approaching the number of writers in non-democracies who self-censor. Which is to say that one third of writers in so-called “free” countries self-censor because of surveillance.

I would argue that we have the cloud to thank for this. This famous slide from our friends at the NSA describes the best place to intercept data being stored in Google’s cloud. “The cloud” sounds fluffy and nice, but it actually means that you’re storing your stuff on someone else’s hard drive. If you store your data on Google’s cloud, you’re storing it on Google’s hard drive. If you store your data on Facebook’s cloud, you’re storing it on Facebook’s hard drive. Their hard drives, their rules.

And as we’ve seen, it’s easy for someone to get your content removed if they don’t like it, whether they’re the government or just a person who disagrees with you. In a world where reach is everything, no wonder writers self-censor.

And yet, the Internet is amazing. It’s the most powerful engine for communications and learning the world has ever known. It’s an important driver for free speech and it’s changed the way we do business. And we shouldn’t have to give up any of those things.

The early Internet was designed to be resilient: the opposite of the giant siloed stacks we now pour all of our content and conversations into. The idea was to connect up universities and military labs to share resources, in a decentralized way.

That decentralized structure allows us to use services like Google, Twitter and Facebook through a single browser window, but what if we rethought how we shared data online? What if each of us had our own service? What if our conversations and ideas lived on our own devices, in our own living rooms and in our newsrooms? And what if these devices were as easy to use as an iPhone?

We’re beginning to see this future emerge. The Intel Compute Stick is a tiny computer that costs just $89, while platforms like Sandstorm turn publishing and talking to people online from your own server into a one-click operation. Sandstorm, by the way, just announced $1.3 million in funding.

In the old days, Microsoft disrupted the tyranny of mainframes and timesharing by imagining a world with a computer on every desk. Today, I believe we should be imagining a world where we all own our content and conversations online.