IEEE Spectrum November, 2015 - 8

11.15

Technological Progress and the
Perpetual Learning Curve
As technology evolves at warp
speed, so must engineers

ta lk to m y iPhone 6 Plus to comPose text messaGes , and

I'm not alone. Many people are abandoning typing, which can be
hard on the hands. Why let your fingers take a pounding when you
can get what you want by speaking?
Thanks to steady improvements in voice recognition, orality-
vocal expression-is unexpectedly reviving. But we aren't speaking
more with other people; telephony between humans is collapsing.
Instead, we speak to our digital devices as never before.
The shift is part of a broader transformation in everyday life. Our artifacts are in upheaval. Compact discs are being creatively destroyed
by streaming music. Tickets to sports and other entertainment events
are now images on a smartphone screen. Keys and ID cards are vanishing, replaced by biometrics.
Even learning to drive an automobile is on the endangered list. Neither of my two 20-something children has a driver's license. My son says
he may never get one. "Between Uber and driverless cars," he says, "I
won't ever learn to drive."
Typing and driving-both courses I took in high school in the 1970s-
are no longer part of the skill set that defines adulthood. For engineers
who design and build the platforms upon which human life depends,
rapid and radical changes demand a response. Engineers, to stay rele-

Nov 2015

North americaN

SPectrUm.ieee.orG

vant and employable, must change along
with the rest of us.
Consider the shifts in coding, a lucrative field that, despite high pay and status, seems addicted to change. Take the
size of coding teams. When I wrote a
book 20 years ago about the making of
Windows NT, the 250-person team that
wrote the massive program at Microsoft
seemed to foretell the size and shape of
software work to come. Success as a programmer in megateams depended on
finding a niche, on hyperspecializing and
fitting in. Today smaller, agile teams of as
few as 15 coders instead value versatility.
Solo coding, meanwhile, was once the
stuff of legends. Gates, Brin, and Zuckerberg all wrote initial versions of popular
programs. Today, the lone-wolf coder,
fueled by a me-against-the-world sensibility, is an endangered species.
Pivotal, a large software house in San
Francisco, has halted solo programming,
opting instead for two-person teams that
enable each programmer to monitor the
other. The buddy system seems to boost
reliability and reduce the tendency of
coding "divas" to hold projects hostage.
Programming languages also go in and
out of style and shape career trajectories.
Some languages rise and fall on the fortunes of particular hardware platforms,
but whatever the cause of the fluctuation in popularity,
it's vital to engineers to be able to see the trends. It's
no surprise that IEEE Spectrum's Top Programming
Languages app has tens of thousands of views each
month. Coders want to know what languages are hot.
Development paradigms also shift. Consider the
DevOps paradigm, which seeks to overcome barriers between development and operations. Since its
debut in 2009, DevOps has rapidly gained followers.
The imperative to keep pace with change unites
engineers with other humans. Change isn't always
a journey into the unknown; at times, the experience returns us to the familiar, to older, even dormant practices. By enabling our new digital tools to
harvest best practices from the past, we all benefit.
Whatever the speed, we can still go home again-
at least until the next adaptation is impossible to
resist. -G. Pasca l Zach a ry
G. Pascal Zachary, a professor of practice at Arizona State
University, is the author of Showstopper! (1994), on the making
of Windows NT.

Ian SpanIer/Getty ImaGeS

SPectraL LiNeS_

http://SPectrUm.ieee.orG

Table of Contents for the Digital Edition of IEEE Spectrum November, 2015