When It’s OK to Be a Guinea Pig

This article is part of the On Tech newsletter. You can sign up here to receive it weekdays.

If you bought a book and it came with a bunch of blank pages and no ending, you would be super annoyed, right?

I want to explore when it’s fine for tech companies to release half-baked products, like what the company behind the Snapchat app is doing with its Spectacles glasses, and when it might not be, like what Tesla is doing with its Autopilot driver-assistance software that does not safely pilot cars.

What separates worthy real-world laboratory tests from reckless ones is how much risk and cost they ask people to endure. Here’s another, relatively overlooked, factor: How well a company sets expectations for what a product can and can’t do. Many of us are fine being guinea pigs — as long as we know it.

Snap is an example of both bungling expectations of an imperfect product and then managing them expertly.

When Snap released the first version of its Spectacles in 2016, it was an experimental product that the company led us to take way too seriously. The company went all-out for attention, including with sales stunts that made it tantalizingly hard to buy the $130 sunglasses with built-in video cameras.

It turned out that not many people wanted a pair. So many Spectacles went unsold that the company had to write off a bunch of inventory and unused parts.

As with the similar Google Glass a few years before that, one of Snap’s errors was overhyping Spectacles. Setting off a frenzy was bad for Google, Snap and anyone who bought an intriguing but janky new thing that didn’t have a compelling reason to exist. Google Glass has come back, quietly, but I’m not sure it will ever recover from its first overhyped act.

Lesson learned, at least for Snap. The company has continued to release updated versions of Spectacles including the fourth iteration that it announced last week. The new Spectacles allow users to not only record short videos but also add computer-generated images to their surroundings, much like the features in the Snap app. Imagine looking up at the stars and viewing a guided animation of the constellations.

Lots of people and companies are justifiably excited about the potential of this technology called augmented reality. (Snap also is buying the start-up that made the Spectacles display for more than $500 million, The Verge reported.) We’ll see if augmented reality becomes as big a deal as its proponents hope. But Snap is smart this time around to make it clear that Spectacles are essentially an unfinished book.

Snap’s chief executive says that no one has yet totally figured out augmented reality, and the technology won’t become widespread for at least a decade. The lack of hoopla is appropriate and refreshing. This experiment is fine because the company is telling everyone that Spectacles are half-baked.

And then there’s Tesla’s experiment. Regulators have repeatedly said that the company is misleading the public about its half-baked Autopilot, a driver-assistance system for tasks like steering and braking on certain roadways. But the name and the ways that Tesla talks about the feature implies that drivers can safely turn their attention away from the road.

Again, part of the problem is that expectations are out of whack with reality. Unrealistic promises about self-driving cars also generated public enthusiasm and investment dollars, but also made slow development of the technology seem like a failure, my colleague Cade Metz wrote in a new article. Of course, a half-baked car on public roads is in a different league from experimental glasses that let you giggle at dancing hot dogs.

As long as the stakes are low, it’s neat to see companies tinker out in the open, and learn and improve with people’s feedback. The danger is when companies aren’t honest with themselves or us about the limits of their products.

Tip of the week

You — YES, YOU — can revive that old computer you have shoved in a closet. Here’s how Brian X. Chen, the New York Times personal technology columnist, did it, with expert help.

Recently, I took on a personal experiment: I dusted off my 11-year-old iMac, which I had abandoned long ago because it felt too slow, to see if I could put it to use again.

Be the first to comment

Leave a Reply

Your email address will not be published.