Isys Information Architects

Interface Hall of Shame

- RealCD Reader Comments -

(Visitor Bryan Oakley's response to the RealCD Review)

I just read your review of IBM's RealCD software, and think you have completely missed the point. You seemed to review the software as if it was destined to be a well-behaved Windows95 client, instead of being the design study that it is.

That RealCD runs under Windows95 is only an accident of nature -- it's the most widely available platform and therefore the most accessible to the most people. That should in no way contrain the designers in their goal of creating a new type of interface. In fact, IBM says as much on their web page, stating

RealCD is not intended to be a conventional Windows95 application. It is more of an experiment in user interface design intended more for a new real-world user interface style we are currently exploring.
So, to criticize the interface for not using standard Windows95 controls is to miss the point of the design altogether.

Let's look at the interface in another light. What if, a few years from now, a physical CD player was no bigger than a CD case in all dimensions. Plus, it was able to be embedded with a couple of LCD displays and touch panels. Now imagine that the interface is RealCD. It would be very simple, very straight forward, and would work exactly like one would expect after experiencing the "old" CD style. That is, to see the list of songs you open the device up. To view "help" you open (click or tap) the "liner notes" inside the cover. Not interested in that information? Close it and the device is no bigger than a CD. Wouldn't that be great? Now, include a wireless link so you can sit on the couch and hold the controls in your hands while electronically browsing the CD information and we have a killer product.

The point is, RealCD isn't about making the world's easiest to use *Windows95* application. The point is to break the shackles of existing interface design principles and try something new. In that regard I think they have mostly succeeded. I'm willing to bet that during the design phase their target machine was *not* windows, but a mythical device that had no requirements or bounds.

Let me address some specifics of your review. One of the very first things you write is:

"At the very least, IBM certainly has great copywriters. They've managed to fit nearly every contemporary buzz- word into the two sentence description: "
I fail to see how using buzzword-complient descriptions in any way affects the design of the software. Yet, you seem to imply that because they use the buzzwords that they do, there is no reason to expect their design to work. At least, that's the basic premise I inferred from your web site.

Shortly thereafter you write:

"The problem is, there is no indication from the main interface that these features exist, nor does the interface provide any clue as to how the user can access them."
I would argue that the ease in which one intuits the features and functions of a design is _not_ necessarily indicitive of a good design. Sure, controls must provide reasonable affordances, but sometimes you just have to learn something new to use something new. In fact, if one's goal is to create a completely new design, one can't help but use design elements no one has seen before.

Now, you can talk affordances and accessibility, but they aren't universal maxims, and depend a lot on context. Sure, if they were creating a commercial Windows95 application they should probably use Windows95 controls, but that wasn't the goal of RealCD.

As a perfect example, you criticize the exit control because it's not a standard control:

"The Exit Application control is perhaps the least intuitive control we have seen in a graphics user interface...Clicking on a green light to exit an application is about as intuitive as clicking on the Start button to exit Windows95"
I would have to disagree completely with this assessment. Clicking on the green light seems like a much better method of closing the application than, say, clicking on a button that looks like an X embedded in the title frame of an application. Why is RealCD's control so good? Take a look at many (most? All?) decent stereo equipment. How do you turn it on? Press the power button. What happens next? Likely, a green or red led turns on to tell you the thing is powered up. What do you do to power it down? Press the button with the green light on it. At least, that's how my Nakamichi equipment works. Makes perfect sense to me, and and I can see why the IBM designers decided to stick with this concept for RealCD.

Continuing on, you write:

"The remaining control is perhaps the most important control in RealCD, and true to the design team's tendencies, it bears the least resemblance to a control. It is the only means to access the "Help Book", and the only means to view the contents of the CD. This control is the "Open the CD Case Button", represented by the 2 gray squares."
I think you are blinded by your use of existing computers. If I were to show this application to my mother who, while extremely intelligent, is a computer neophyte, she might very well intuit this control quite readily. Maybe not, but once you tell her "to make the application open, press the button that looks like an open case" she'll remember it.

Furthering your bias towards "if it's not Windows95, it's not easy to use", you write:

"RealCD does not provide a status bar nor tooltips to indicate the functions of the various controls, nor for that matter, to identify which screen objects are indeed controls (IBM considers such interface elements as "clutter"). Since the "Open the CD Case" control is so completely un-apparent, our guess is that most new RealCD users will never realize that the application offers more than the basic playback functions. "
Again, I'll have to agree with IBM on this one. Tooltips and statusbars are a crutch. Interfaces that are easy to use and easy to learn are, IMHO, superior to interfaces which might be able to be "figured out", but which rely on cruthces to jog your memory. I bet that after spending 5 minutes with a user's guide, no user, expert or novice, would ever need tooltips or a statusbar in this particular application. The controls are very easy to learn and, once learned, make sense with their functions. I would argue that the RealCD control is a lot easier to figure out and remember than existing Windows metaphors like "go to the view menu, pull down the menu and select "View Songs".

I could go on, but I'm not sure what it would buy us. My bottom line critique is, your evaluation of the interface was created using the wrong perspective. You seem to think that unless the interface uses standard Windows95 widgets it's not a good interface. I think you completely missed the point of the RealCD design, in that it was an experiment in creating a whole new, easily usable interface. Window95 just happened to be the platform they chose to make the design available.

Personally, I think they largely succeeded. Does it make a good Windows95 application? No. Is it a good user interface design in general? In my opinion, Yes.

Bryan Oakley
Software Engineer
Healthcare Communications, Inc.

© 1997-1999 Isys Information Architects Inc. All rights reserved.
Reproduction in whole or in part in any form or medium without express written permission is prohibited.