Too Much Cool - Embedded.com

Too Much Cool

I used to date an architect. A very artistic nature complemented her intellectual brilliance. Art mostly eludes me; as an engineer I think mostly about raw functionality: is it fast enough? Does it consume too much power? Are all of the features working properly? So it was an education to view the world through her perspective. A building is functional: it provides shelter, heat and light. But is it beautiful? Does it gracefully merge into the surrounding cityscape?

She often commented about electronic products, and how some evolved from merely functional things to objects of beauty. She was fascinated by the stunning marriage of form and function embodied in the Macintosh. Compared to the stark plastic towers housing Wintel machines, Apple has indeed created a thing of beauty.

Not long ago bare functionality was enough to sell consumer products. Later form became equally important. Today there's a surprising (to me) business in cell phone snap-on covers. My Spockish engineer's brain can discern no benefit from these accessories, yet they are undeniably hot items.

Today cell phones are pretty much all the same. Maybe a cool design is the only tool vendors have to differentiate their products. So I figured the pendulum, which once rested entirely in the functionality camp, would stop its swing at the current level of great features and a pretty shape, with a few extra features tossed in to grab consumer's cash.

Maybe not. The pendulum continues to swing, into a new arena where cool is more important than functionality. Now some phones are so small that people with normal-sized fingers have trouble pressing buttons. Small is cool — but at what cost?

A New York Times article reveals even more tradeoffs made in the pursuit of “cool”. Antennas buried inside the phone give a sleek look but yield more dropped calls and less clarity on the rest. Silly features like built-in cameras that chase the dream of “cool” suck power too rapidly, draining batteries much faster than phones a year or two less advanced.

Extra features also result in buggier code and more customer frustration. I used to have a cell phone that had 100 different functions, ranging from a handful of important setup commands to frivolous games. A group of developers obviously spent a lot of time implementing all of those capabilities. But the phone itself was horribly unreliable and buggy, dropping calls, not acquiring signals properly, and generally driving me batty with frustration. Its primary feature — making calls — wasn't reliable. The developers focused too much on the frills and not enough on the basics.

Yet it is possible to combine coolness and functionality. Apple is probably one of the great engineering companies extant. Their iPod is a wonderful product, an MP3 player that is a thing of beauty, with great battery life and almost infinite capacity to store tunes. It's hard to imagine a better music machine; my hat is off to Apple's engineers and their industrial designers. I've yet to succumb to the siren call of cool that explodes from the iPod, but it's near the top of my wish list.

In a beery discussion one evening at an Embedded Systems Conference, a developer from a major consumer electronics company confided that features and coolness are far more important to them than reliability. He admitted that their products were rife with firmware bugs, but claimed they were merely responding to consumer demands for more features faster. Cool sells. When was the last time you saw “and, it actually works like we claim” plastered over an ad for a new electronic gadget?

Not recently, I'm thinking.

Cool features and an elegant look are, well, cool. Seductive. They make us instinctively reach for our wallets. But when cool comes at a substantial cost in performance, I think we're making the wrong tradeoff. A built-in camera is indeed at least sort of cool. But don't compromise the ability of the device to function like a really great phone. I think too many of the alleged “consumer demands” are really the product of some 23-year-old marketing weenie's late night brainstorm. Or nightmare.

Jack G. Ganssle is a lecturer and consultant on embedded development issues. He conducts seminars on embedded systems and helps companies with their embedded challenges. He founded two companies specializing in embedded systems. Contact him at . His website is .

Reader Feedback

Hi Jack,
I agree that we should not compromise on basic functionality at the cost of more (cooler) features. I am sometimes amazed at the complexity of software that goes inside these devices and can also see that they are a lot buggy.

But, on retrospect, many features that are considered de-facto were once cool and have stabilized only over months or years. A company cannot wait too long to make their software perfect in this competetive market. So, I feel the trend is to come up with products with great features and later stabilize the features. I dont want to discuss if this is ethical or not, but I feel there is no other way out in these competitive markets.

Regards,

Sriram VIyer
Sr. Software Engineer
Philips Semiconductor


One of the companies that I worked at 15-20 years ago had an interesting structure. It was a company with mutliple divisions, but only one development group. That group “charged” the other groups for development. We would develop products, but the other groups would either pay directly for development (NRE) or would pay a royalty to the R&D division for each widget made.

This had the effect of (IMHO) of making the other groups think about what features would be in a product, and justifying them (hopefully through higher sales) rather than just filling up the boat with ideas.

This has not been the case with any other comapny I've worked for. In most cases we rely on the Sales or Marketing departments who don't always give (again IMHO) the scrutiny warranted some customer features (demands). In many companies, the cost of R & D is buried (damn engineers are there anyway) as part of overhead – the only costs that anyone sees are tooling charges and prototypes. The actual design time is usually not captured on a per project basis – and when it is, even rarer is it compared to sales of the widget itself.

While 'cool' can sell an item (hey man, pet rocks are cool), in the long run I would tend to stick with the basic 10 features that most products have. I always avoid consumer items where a review says “hard to use”. Chances are there wasn't a lot of thought put into a product like that.

And no, I never owned a pet rock…

Tom Mazowiesky
Director, Product Development
Global Payment Technologies,Inc


You are right Jack !

Shivamallu
Senior Engineer
TATA ELXSI LTD

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.