Ethics Co-Processor

Maciej Ceglowski, on the Internet of Things, Amazon’s Mechanical Turk, and the role humans might play as AI takes over the world:

My favorite Internet of Things device is a fan called the Ethical Turk that subverts the whole idea of scriptable people.

This clever fan (by the brilliant Simone Rebaudengo) recognizes moral dilemmas and submits them to a human being for adjudication. Conscious of the limits of robotkind, it asks people for ethical help.

For example, if the fan detects that there are two people in front of it, it won’t know which one to cool.

So it uploads a photograph of the situation to Mechanical Turk, which assigns the task to a human being. The human makes the ethical decision and returns an answer along with a justification. The robot obeys the answer, and displays the justification on a little LCD screen.

The fan has dials on the side that let you select the religion and educational level of the person making the ethical choice.

My favorite thing about this project is how well it subverts Amazon’s mechanization of labor by using human beings for the one thing that makes them truly human. People become a kind of ethics co-processor

January 13, 2017

Catching Up

About a year back, I blogged about the interesting relationship between science and practice in the health, fitness, and wellness worlds. On the one hand, a lot of what passes for ‘best practices’ in the trenches – from professional athletic teams’ weight rooms to your local Gold’s Gym – is completely unscientific garbage. But on the other, there’s also a long history of well-executed journal research simply lagging behind new and effective innovations that have already gained traction in the real world.

So it was particularly interesting to stand in the middle of that process, when a recently published meta-analysis of 27 weight-loss studies fully endorsed several of Composite’s key ideas that we’ve been developing over the past year and a half.

As the name implies, Composite is built on a multi-faceted approach; in building fitness, we think that the parts of your life that happen outside of the gym – things like nutrition, movement throughout the day, and lifestyle – are just as important as what happens in class.

We also think that coaches – the highly-trained leaders of those classes – have a role beyond teaching and guidance, as accountability points for those outside-of-class factors.

And we know that the community we build in classes can similarly reach beyond the gym, to provide support, encouragement, and motivation that helps people build and sustain healthy habits over the long term.

So it’s no surprise to us that the paper’s authors conclude precisely the same thing:

“Programs supervising attendance, offering social support, and focusing on dietary and lifestyle modification have better adherence than interventions not supervising attendance, not offering social support, and focusing exclusively on exercise.”

As I said, that’s not a surprise. Even so, it’s nice to be right.

Mobilized

With yesterday’s ten-year anniversary of the iPhone, I spent a bit of time marveling at how far pocketable technology has come in the past few decades.

At the time the iPhone launched, I was happily toting a Blackberry Pearl, a big upgrade at the time after years of carrying a series of Treo’s.

But my mobile nerd-ery extended far further back than even that, to 1999, when at the Consumer Electronics Show in Las Vegas, I stumbled across a small booth where a Canadian startup called Research in Motion was selling a new email-enabled two-way pager they were calling the Blackberry 850:

rim-850

At the time, the company I was running built software for hedge funds, and I remember the bankers staring at my Blackberry, incredulous that anyone might need – or even want – to check email away from their desktop.

Of course, the Blackberry existed in a parallel universe to my phone, a trusty and chic Motorola StarTAC. So at the same point, I also owned a Starfish ClipOn Organizer, which piggybacked thickly on top of the flip-up screen, stored a “whopping” 1000 contacts, and allowed you to – get this bit which blew my mind at the time – dial your phone just by clicking a contact’s number on the Starfish:

233_01

The Starfish also boasted a calendar, and a notes app. Though as there was no data-entry mechanism, the notes were read-only. And as the Starfish synced by popping the thing off the phone, and then sticking it into your PC’s PCMCIA card slot, the calendar was perpetually out-of-date.

Today’s a far cry, indeed, from those primitive times back in 2007 or 1999. Back when we couldn’t check Google Maps any time we got lost, couldn’t search the web to find relevant information on the fly, had to spend dinner actually paying attention and talking with the real people physically surrounding us.

Where’s the Beef

Here’s an interesting question to ponder: what aspects of normal life today will future generations look back upon as moral failures on our part, much as we look back upon things like slavery in the past?

The list is, sadly, probably quite long, as we don’t seem to learn the lessons of history particularly well. As a young Jew attending Hebrew school, I was shown more Holocaust films than I can count, yet I’ve taken essentially zero action to help prevent the wholesale slaughter of nearly a half million people in Syria in the past few years.

Recently, I’ve been studying up on the effective altruism movement, and thinking about things I can do – small and large – that would positively impact the world. And while I have a number of other ideas brewing, I wanted to share at least two small things I’m going to try to do differently in 2017:

First, I’m going to eat only humanely-raised animals (and only eggs from humane egg farms).

Second, I’m going to eat less chicken, and more beef instead.

For over a decade, I’ve followed – albeit rather loosely at times – a Paleo / ancestral approach to diet. Meat plays a part in that diet (though, often, a smaller one than a caricature of the approach might imply). So inherent in my diet is killing animals.

While I’ve considered becoming a vegetarian for ethical reasons, I don’t believe that diet is optimally healthy. Nor do I believe, for most people, that it’s sustainable. After 18 months, about 85% of vegetarians and vegans return to eating meat, which is why the percentage of vegetarians in the US has held steady at about 5% for the past thirty years.

But I do think that, if I’m going to eat meat, I should do it in a way that kills as few animals as possible, and that leads to those animals being raised in the kindest way possible.

While most packaged food descriptions are so loosely regulated and third-party verified that they’re essentially meaningless (e.g., ‘natural,’ ‘free range,’ or ‘humanely raised’), a number of independent organizations now exist to ensure that animals raised in farms that achieve their designations live much better lives. In particular, the Animal Welfare Approved, Certified Humane, and Global Animal Partnership (especially with ratings of 4, 5, or 5+) are now fairly widely available, and hold farms to very high standards. Yes, these animals are also killed, but, as rancher / activist Joel Salatin puts it, they get to live good, healthy lives in accordance with their animal nature, and then have one very bad day.

Or consider eggs. In factory farms, chickens are painfully de-beaked, then confined in cages and filthy, overcrowded barns. Conversely, a producer like Nellie’s Free Range Eggs (which earns a Certified Humane rating) has no cages anywhere, ample space in clean, well-ventilated barns, high quality feed without hormones or antibiotics, gentle handling, and much greater full-time access to the outdoors and grass. A dozen Nellie’s eggs at my local grocery store costs about a dollar twenty more than factory-farmed eggs. And, in short, I’d rather eat a few eggs less each week from a producer like Nellies than use my dollars to support animal cruelty for a few more.

Secondly, as I said above, I’m going to try to eat less chicken, and more (humanely-raised) beef. The reason is simple: you have to kill many, many more chickens to get the same amount of meat as you get from a single cow. In fact, the beef from a single steer is equal to about 200-250 chickens. Which means that, simply by switching from chicken to beef, in a year of my standard meat consumption, I can be responsible for a single death, rather than several hundred.

Additionally, by choosing grass-fed beef, I can get healthier food (the high omega-3’s of grass-fed beef fat are far healthier than the omega-6 laden fat in chickens or grain-fed beef), and help the environment (as, due to carbon sequestration by the grass in areas they fertilize as they graze, grass-fed cows are carbon-neutral, whereas methane emission from factory-farmed animals account for some 10-15% of all carbon increases annually).

Here, too, grass fed beef costs more. But, again, I’d rather eat less of something I can feel good about, rather than simply default to supporting something terrible with my dollars and then trying not to think about it too much.

Sure, it’s not helping slaves escape by the underground railroad, or hiding Jews during the Holocaust. But it’s something small I can do, day in and day out, that helps make the kind of change I want to see in today’s world, and that I think future generations will look back on with respect.

January 10, 2017

2017-01-09

Yiddish phrase of the day: “Itlecheh shtot hot ir meshugenem.” Every village has its idiot.

January 9, 2017

Totally Random

With the start of the New Year, I’ve been trying to audit my approach to productivity, thinking about ways in which I can be more effective in the year to come.

I tend to tag my tasks with relative priorities, and I keep a log of completed work from each day. So, last week, I sorted through the log entries for 2016, to see if I could find any patterns to my efficiency.

In that analysis, something quickly emerged: my daily efficiency had a clear barbell distribution, with stretches in which I banged out huge volumes of work (including my most important tasks), and then stretches in which I barely got anything done (and what little I did accomplish tended to be menial, low-priority stuff).

Thinking back to the floundering, low-productivity days, I could highlight at least one obvious, unproductive behavior: I’d look at my list, feel overwhelmed, and then simply procrastinate by avoiding my list entirely for much of the day. Whereas on days in which I was able to get cranking, I’d simply hop in and get to work, moving directly from one task to the next.

The problem, then, appeared to be one of momentum. Once I was getting things done, it was easy to keep getting things done. But if I felt stuck, I had a hell of a time getting unstuck.

This morning, I was clearly in that ‘stuck’ mode. So, picking up an older idea from British productivity author and tinkerer Mark Forster, I set out to try what he calls the ‘random method.’

In short, I went over to the Random Integer Generator, and created a table of 100 random numbers from 1 through 16. (The 16 there being somewhat arbitrary – it’s my lucky number.)

The first number on the table was an ‘8,’ so I started at the top of my list, and counted down to the eighth task. Whatever it happened to be (in this case, processing a pile of paperwork), I hopped in, did it, and crossed it off the list. Then I went back to the table – this time a ‘6’ – and counted six more tasks down, and did that task next.

If my count took me to the end of the list, I’d simply loop back and keep counting from the top. And if I landed on a task that had already been crossed out, I’d ‘slide’ down until I reached the next uncrossed task.

In that way, I randomly looped my way through a first dozen tasks by noon, accomplishing more this morning than I thought I’d likely get done all day.

I’m not entirely sure why this works, though I suspect it has to do with removing the element of choice – if I had to decide what I wanted to do next, I could simply avoid making a decision and do nothing. But if the integers ‘told’ me that I had to do something, that was just enough external pressure to spur me into action.

Admittedly, this is a slightly ridiculous approach. I’m not sure if it will similarly be able to jump-start me on future ‘stuck’ days, and it’s certainly more cumbersome than simply trying to choose the next most important or appropriate task on days when I don’t need the extra push.

But it’s something I’ll certainly be testing out over the weeks ahead. On the chance that you, too, sometimes avoid your to-do list entirely in favor of scrolling through Twitter or diving into the rabbit-hole of Wikipedia, perhaps it’s worth similarly giving it a try.