Sculpting a new Windows CE

by Reads (6,132)

We hear it every day: Windows CE needs a facelift, and maybe even some liposuction. Apple Newton users, who’ll seemingly condemn anything birthed in Redmond, and Palm owners take contant swipes at Microsoft’s palm-size operating system, often citing its complexity and girth.

The good news is that Microsoft acknowledges Windows CE’s flaws, hinting that it’s working on a whole new baby, code-named Rapier. However, don’t count on a new Ally McBeal-sized Windows CE, sexy and skinny. Based on Microsoft’s history with new-and-improved versions of Windows, Word and Excel, it’ll probably be more of an Anna Nicole Smith-sized CE–sexy but downright hearty.

It may seem odd that the company founded by Bill Gates, self-proclaimed creator-of-tight-code, can’t seem to develop a small operating system for handheld devices. In fact, in the early 1980s, Gates is credited with making the bold statement that no user would ever need more than 640K in a desktop computer. He believed that optimizing code would continue to be an integral part of software development. Yet, as brilliant and forward-thinking as he is, Gates did not anticipate how incredibly fast processor chips would become and how the capacity of memory chips would grow exponentially, while the cost per megabyte would drop just as dramatically. Ultimately, Microsoft realized that the human costs associated with extensive tweaking of code didn’t make good business sense. I’m not saying that optimizing code isn’t important, it’s just not as important as it once was.

A case in point is the new Casio Cassiopeia E-100 and E-105. The engineering wizards at Casio gave it an amazing tune-up by pumping up the memory (the E-105 has 32 megabytes) and the processor speed (the E-100’s NEC chip runs at 131 MHz), which made Windows CE perform as we’d always hoped it would. They did this and still remained cost competitive with the high-end Palm Computing organizers, like the Palm V and Palm VII.

My point is that processor speeds, along with the miniaturization of memory and storage devices, will continue to far exceed the rate and extent to which code can be optimized. So I’m not really that swayed by the Palm purists who gleefully point out to me that, currently, the Palm OS is very small (less than 40K in size) and runs extremely fast on a meager 16 MHz processor.

I will concede, however, that Palm does have one substantial benefit by running on that low-power processor–it conserves energy, enabling Palm devices to run on standard batteries for weeks.

But let’s not forget that we are still in the handheld Stone Age. Processors will become faster, memory will get denser, and batteries will improve. What I’m really concerned about is, what is the future of Palm OS and what’s the future of Windows CE?

Basically, which operating system can offer me an acceptable level of simplicity, while providing a wonderfully expanding array of functions? And which operating system is more poised for the future?

Don’t get me wrong. I’m not saying that Microsoft shouldn’t go ahead and substantially improve Windows CE. Far from it! In fact, here a just a few of my own personal suggestions for Bill & Company’s consideration.

Suggestion #1: Change the name

Now I’m no marketing guru, but my first suggestion for Microsoft is to rename the Windows CE product. That’s right, Microsoft should drop their strict association with the Windows brand and develop a catchy new name for their palm-size operating system and devices.

Remember, they did this when they moved from DOS to Windows, and Explorer doesn’t bear any scars of the Windows legacy. So why not Windows CE?

I fully understand Microsoft’s motivation. They simply want to leverage both the Windows user interface and the Windows brand. However, this alignment is equal parts good and bad for Windows CE. Palm OS doesn’t have the same cross to bear–the heritage of a pre-defined GUI and brand–and exploits both of these to its advantage.

Imagine if Palm were forced to abide by the same branding strategy that 3Com uses for its network products, like its hubs and routers. I don’t think it would be as hip and edgy as their current Simply Palm ads, do you?

Maybe Microsoft should seek advice from Proctor and Gamble. After all, P&G has a successful history creating brands. They’ve got at least 20 top brands in the household cleaning area alone. For example, P&G didn’t invent Tide and then decide to call everything in the household cleaning category Tide. Imagine Tide soap, Tide dishwashing detergent, Tide floor polish.

New product, new brand.

Maybe I’m way off base on this one, but Microsoft’s palm-size operating system is as different from Windows 95/98/NT as Windows was from DOS. And it deserves a unique name.

Maybe then Microsoft won’t have to be so stuck on not enabling it to communicate with other desktop operating systems, namely MacOS.

Suggestion #2: Create more magic

Almost every user, even hardcore programmers, expects some form of magic from their computer. Now I’m not talking about mind reading or crystals or tarot cards, what I’m referring to is programming magic–essentially, hiding complexity from the user.

Indeed, over the years I’ve begun expecting, and getting, more and more magic from my computer. Just look at Explorer 5.0.

Of course, complexity is relative and incremental. For some of us, applications and macros are complex. While, for others, it’s languages, or even machine code. The primary goal of a user interface (UI) is to make the operating system (often a complicated beast in its native state) easy-to-use. A UI strives to hide whatever is deemed "too complex" for the current user.

This can be further enhanced using two common techniques: user modes and scripting languages. User modes (beginner, intermediate, advanced) allow magic at the appropriate level. They hide functions and parameters that may be too complicated for the user, often using default values. Scripting languages, meanwhile, offer another method for hiding code and parameters behind a selection or icon. However, scripting languages can be used by software developers and power users alike to create their own magic.

Palm OS does a much better job at creating magic. The first magic I’d recommend that Microsoft create would be to hide the process needed to connect to an ISP or server to get email.

That would be a start.

Suggestion #3: Reduce the clicks

One click is great, two is okay, but three or more clicks to get to a function os truly unacceptable.

I call it my three-clicks-and-you’re-out rule.

For example, in Windows CE it takes four clicks (Start, Settings, System, Tasks) to view the current running tasks. Even starting an application can take three or four clicks.

Still, this is a tough nut to crack.

On one hand you have progressive disclosure, which in UI parlance means showing only what is pertinent at the time. I call it information on a need-to-know basis.

Sometimes, however, this can get a little carried away. For example, how many screens do you go through at an ATM just to get $20 for lunch? Too many, if you ask me.

I call this progressive disclosure gone berserk. So reducing the clicks takes a lot of work and requires an understanding of user interface patterns.

Other vendors, such as BSQUARE with bTASK, ThumbsUp Software with StarTap, and Sticky Software with Sticky Buttons, have stepped in to create add-ons to address the shortcomings of Window’s CE’s UI. But who wants to buy additional software for things that should be there in the first place?

Again, much of this can be traced back to Microsoft’s insistance on sticking with its Windows interface strategy.

Suggestion #4: Embrace color

Sorry all of you Palmists and Newtonians who send me emails stating, "Who needs color!" It’s a fact: color dramatically increases recognition. And it’s not just those of us from Pleasantville who say this. Color conveys information. And this is one area where Microsoft has gotten it very right. Bravo!

Children at an early age learn the implied meaning of red, yellow and green. And again, the name of the game in small computers is conveying as much information as quickly as possible using a small amount of space.

Sure, there are the oft-cited technical speed bumps (like battery usage) associated with using color, but these will inevitably be overcome.

So let me ask you, Would you rather have rainbows or clouds?



All content posted on TechnologyGuide is granted to TechnologyGuide with electronic publishing rights in perpetuity, as all content posted on this site becomes a part of the community.