The Internet of Things

What’s Hard Today is Temporary

There was a time when flat panel displays were impractically expensive.

“Once priced at $20,000 and up, more and more superlarge, 40-inch-plus plasma displays are dipping below the $10,000 mark. No, they won’t replace standard TVs–or rear-projection systems–in the near future, but upscale consumers who love home theater are already taking them seriously. According to industry observers, plasma’s audience should broaden by 2005, when prices could sink to $4000 or less.”
CNN, 2001

A decade later the concern shifted to screens becoming too cheap to turn a profit. A great development for consumers.

“The price of LCD panels fell by 80% between 2004 and 2008, while the manufacturing costs declined by 50%”
— The Economist, 2012

Disqualifying ideas based on current cost of tech is silly, especially when you’re planning for the future. The best ideas will even drive down costs as they pick up steam. Even in arguably the highest cost industry out there, SpaceX is well on the way to reducing launch costs to a tenth the previous gold standard.

Technology has a wonderfully way of getting cheaper over time, especially when smart people work on it. There are some exceptions like graphing calculators that are inexplicably immune to market forces, but generally you can expect today’s “rich person toy” will become tomorrow’s consumer commodity given some time.

Software developers face this intersection in every feature they plan. Availability of cheap storage and on-demand cloud servers shifts product focus to “What’s the best way?” instead of worrying that a single misstep will put the company under. Failing to optimize is a relatively cheap mistake to make today.

There are many reasons why technology doesn’t hit adoption, but relying on parts that are a little too expensive for a little bit longer shouldn’t be one of them. At the right speed you can even get a head start on the future.

When we build software with this in mind, we roadmap features that will exist once reality catches up. Right now, I’m bullish on thin clients in IoT. They were an early flavor of connected devices in some senses, especially those that relied on a central server to work. Today, they’ve started to ship as plug-n-play computers that cost barely over $100, fit in your pocket, and can use an HDMI port.

This is great news for those people who develop apps for the real world where digital/physical converter boxes are needed. The more devices capable of running apps, the broader the adoption we’ll have. Right now office will spend upwards of $500 for an iPad mounted to a wall to mange a conference room. Upcoming technology can (and will) do better. There’s no reason a single thin client running the right software couldn’t replace most hardware in the room.

What’s hard and expensive today is a temporary condition. If everything was too cheap to matter, what would be worth having?

Subscribe to Robin at Work

Writing about improving the work day, product updates, and what we learn building a company along the way. Straight to your inbox.