126 million people saw Russian Facebook ads

Facebook is preparing to go before Congress later today to confess to investigators that a lot more people saw Russian-funded advertising in the lead-up to the 2016 US election.

For its part, Facebook has tried to shirk responsibility until now. In November 2016, Mark Zuckerberg said at a conference "personally I think the idea that fake news on Facebook, which is a very small amount of the content, influenced the election in any way — I think is a pretty crazy idea."

Just a month ago, the social network claimed 10 million people saw Russian advertising. Today Facebook has a different narrative: 126 million users saw posts and "other content" created by the Russian government and says that organic surfacing (the algorithm showing it to others based on engagement) may have helped it go wider.

It's going to be a tough session for Facebook in Congress, with the company planning to highlight how small the ads are in contrast to the size of overall sharing. That narrative, however, isn't likely to stand up anymore and the reality is Facebook is trying to dodge any type of regulation that's likely to come out of this.

Washington isn't Facebook's only issue: The New York Times reported yesterday that the service is also playing a core part in destabilizing countries around the world as people come online for the first time. The company is scrambling internally, but it's damage control, not an attempt to change.

The crux of the issue appears to be we are fundamentally not wired to be connected 24 x 7, and the science behind fake news' effect on your brain says a lot about why it spreads like wildfire. Algorithms are optimized for building the perfect echo chamber but those same organic sharing numbers also happen to be a core performance metric for Facebook.

Facebook's VP of News Feed said last week that "In the end, I don’t think we as a human race will regret the internet" but there's a risk we'll regret not regulating a company that can reach more than two billion eyeballs in a few seconds every single day.

Why I'm even covering this and why it matters: Writing about this is an issue I've struggled with. I've wondered: Is this technology news? Will it be relevant to people? Should I try ignore it, given we're overloaded with news about Russia already? Do I know enough about what's going on?

I came down on the side of trying to help cut through it because of the immensity of the problem at hand: we all suffer in a world where it's hard to distinguish fact from fiction and Facebook has repeatedly tried to downplay the scale of its role despite the enormity of its impact. 

We're close to a turning point, I believe, where technology companies have historically tried to act like  innocent innovators but are the harbingers of massive social changes without taking much, if any, responsibility for their actions. The free market is important, but maybe a little regulation here is long overdue.

The UX of machine learning

A lot of people are, er, surprised today to learn that Apple's machine learning is automatically categorizing photos of bras, cleavage and more in a category called 'brassiere' without them knowing. 

I've got no idea where this trend started, but Twitter is alive with angry, confused people who don't understand how their phone has a 'folder' full of their nude photos. Because it's apparently happening on its own, it's inherently creepy to people who had no idea Apple's iOS 10 release started crunching images to understand the contents of them.

This is an interesting side effect of abstracting the 'magic' of machine learning away from the technical details: people don't really grasp how this could be happening. Apple performs all of this analysis locally on-device, but it doesn't make it all that clear what's going on – or that it's crunching everything as well as your cat photos.

There's this uncomfortable moment where many are realizing the convenience of being able to search "beach" and have it find the results also comes with a side effect of your personal, private photos being somewhere in that index too. 

Apple censored many of what it considered 'inappropriate' searches early on, and the iPhone ignores them, but if you know enough synonyms you can generally surface it anyway. 

Weirdly enough, you can't disable this feature. Presumably Apple assumed that as it's on-device, people will feel safe and it should always be enabled but now there's a part maybe the coders didn't consider: what if I'm uncomfortable with a computer sifting through my unmentionables?

It's a weird UX problem: how do you make it easier for users to understand that it's not creepy to index everything in this manner? I don't think there's an easy answer and I don't think people realize this is happening.

Tab Dump

Samsung gets three new co-ceos
If you thought being CEO of two companies was weird (look at you, Jack Dorsey), Samsung's got something for you: it just appointed three new CEOs at the same time. The business is big enough that it warrants this (over 500,000 employees) but it's looking for a new direction as its scandal in South Korea careened out of control.

HomePod can't play Spotify, other rival services
In typical Apple fashion, the upcoming $349 home speaker can't play Spotify music at all, and there's no plans to add support. This approach works fine on the iPhone, but in the home environment where each member of a family has their own preferences good luck convincing everyone to switch.

Square made a dedicated $999 point-of-sale
First, I didn't think I'd ever even consider a cash register interesting but the work Square is doing here is impressive. You're used to seeing an iPad on a swivel stand, but now the company's built its own entire device with dual displays to cut Apple out entirely and take control of its own destiny.