Some Things Defy Easy Answers

I was reading this article about the settlement Facebook reached with it’s moderators and I think this serves as a really good example of something where many people want an answer that fits neatly in a Tweet, or a slogan, but the reality is much more complicated.

Take the Facebook example as explained by TechDirt:

Content moderation creates all sorts of downstream impacts, and many people seem to want to attack companies like Facebook both for not employing enough content moderators and for the mental harms that content moderation creates. And I’m not sure how anyone can square those two views, beyond just thinking that Facebook should do the impossible and moderate perfectly without involving any humans in the process (and, of course, I’ll note that many of these same people laugh off the idea that AI can do this job, because they’re right that it can’t).

This, like many other issues we are living through, gets the the crux of the matter. Users want Facebook to “clean up” the platform, without really understanding what that means. All they know is that a huge tech company with all those smart people should be able to figure it out. But, the things they ask for, don’t compute.It’s not just a matter of one side being right, and the other being wrong, it’s that they both are right, and wrong.

AI is not going to solve the problem by itself. It hasn’t shown the ability to keep up with ever-changing definitions and subtleties of language and other media. It requires some human oversight. That oversight is going to end up with some bias, and having to weed through all of that stuff is going to cause mental health harm to the people doing it. In the end you have three sides to the debate:

1. Do nothing and let social networks get over run with spam and crud.
2. Use AI and just deal with the inaccuracies.
3. Hire more human moderators, and to heck with their mental health

None of these are really great, are they? There’s no magic answer. The only way through the problem is navigating all of these realities and adjusting as you go along and mistakes are made. It’s serious, and grown-up work. I’m not sure we as a society are up for that. Let me give you another couple of examples that are similar in my opinion.

First, COVID-19 lock downs.

1. If we do nothing and continue to just have mass gatherings and take no health precautions, a lot more people will probably catch it and die.
2. If we remain in lock down for the next year waiting on a vaccine, the number of people who will die as a result of poverty (from violence, suicide, and other health complications), will likewise, be very high.

Many people don’t want to hear this, but both of those things can be true. I believe they both are true. So, how to you solve the problem? You try and find the more narrow path that avoids both of those outcomes. It’s not easy. If it was as easy as some of your friends and family on Facebook make it sound, it would have been done before.

Speaking of the lock downs, one of my least favorite “takes” that I’ve seen is that what this lock down has proven is that we can, and always have been able to, cut down on greenhouse gas emissions at an extreme level.

But again, I’m not sure that the opposite side of that argument is wrong either. The rational counter-point to the argument that we have to cut emissions a great deal has always been that the economic damage created by that, the levels of poverty, and starvation, would be a terrible price to pay. As much of a price, if not more than the price, of smaller cuts and some environmental damage. (For the sake of brevity I’m ignoring the less realistic takes on all of these issues, including provably false conspiracy theories.)

Yes we are watching our emission levels take historic drops in a short time. We are also watching whole industries, and potentially economies, collapse, putting many, many millions of people out of work and into poverty. If you’re a full-fledged supporter of a Green New Deal, you should probably accept that will be the result of those policies.

Medicare for All? Free college? Flat Tax? Everyone work from home? No one work from home? Banning “hate speech” on social media? Gun control or gun rights? Facial Recognition?

I can go on, but I think you get the point. When we hear from advocates on any side of these questions, we usually hear about how that solution will fix “this problem”, but rarely do they address the downstream effects or the difficult choices that will need to be made, let alone how the change in incentives will change how people behave.

Pick your poison. There’s no easy answer.

It reminds me of the chapter in Freakonomics about the public schools in Chicago. They wanted to fix the problem of poor test scores coming out of the schools by measuring teacher’s performance using the test scores of their students, and were shocked, SHOCKED, to find teachers only teaching the kids just to pass the test, or outright cheating on the tests.

Huh, who could have seen that coming?

If you want to implement a massive change, like an all autonomous vehicle transportation network, for example, you need to explain what happens to the tens (hundreds?) of thousands of people who drive for a living, right? Or the hundreds of thousands of people who work in the health insurance field if you do away with that. When solving for one problem creates three other ones, you need to convince people why it’s worth it, and how those will be addressed. Simple slogans and tweets don’t get that done.

When it comes to “cleaning up” social media, it’s hard, and anyone who says it isn’t is not being honest with you. Remember that the next time you see someone saying it is, or any other issue for that matter. If it were easy, someone would have done it by now, but really, social media is not the place for difficult discussions. It requires much more than what we normally see.

Follow these topics: SocialNetworking