We Need A New Interface

Some Software Companies Are 
Making A Profit

 

9/28/01

The pull-down menus, overlapping windows, and point-and-click interface familiar to any Windows user is now quite old.  It dates back to the mid-70’s research that preceded the 1980 launch of the Xerox Star, to be followed by Apple’s Lisa (1983) and the earliest versions of Windows (also 1983) and the Apple Macintosh (1984).  The problem is computer hardware has become amazingly more powerful and applications much more interesting and complex, but we’re still relying on the same interface style.  Its age is showing.

Of course, the Windows interface has been enhanced (or modified, depending on your point of view), by introducing elements of the Browser style.  One could make a strong argument that many users live entirely (or nearly entirely) inside their browsers, with many users mainly employing a computer to access email, web sites and web-based applications.  It’s understandable that Microsoft might think users would prefer to make Windows more like a Browser to provide a more homogeneous and familiar environment.  Unfortunately, many users also want to use that interface environment (Windows) to interact with applications that are too complex (offer too many choices) to be appropriate for a browser interface.  Dialogue boxes, pull-down menus, and other interfaces may be more appropriate for complex applications.

You can see the symptoms of an interface stretched to the breaking point and trying to provide the best service possible to each of its wildly differing constituencies:

Menus, which based on our understanding of cognitive psychology, should be limited to seven items, now reach to the moon.  Often, they generate sub- and sub-sub-menus which are both difficult to navigate and hard to mentally map and remember.

Because menus are so long, Microsoft has attempted to employ a variety of devices to help.  Unfortunately, they are inconsistently applied and bewildering in their ultimate effect. 

• Long menus are often shortened by removing or reordering items not used recently, but some users have memorized (deliberately or by long usage) the positions of menu items and are confused by their seeming disappearance.
  

• Icons, often substitutes or alternative instantiations of menu items are also relocated by the operating system depending upon its view of what the user is doing and how often he does it.  Users may have different preferences and often grow weary of trying to figure out how to negotiate with Windows to leave their interface alone.
  

• Helpful alternative systems, such as the panels (actually dialogue boxes) that appear in Office XP take up valuable screen space and while useful to novices are extremely annoying to skilled users.  They are, of course, the children of Clippie, the oh-so-helpful Office Assistant, now hidden as a result of a wave of user complaints (for which I thank Microsoft).

It’s hard to evolve or change an interface.  Small changes may be hardest – users complain that they can’t find what they’re used to and the incremental value of the change may not seem compelling.  We suspect this may be the case with the Aqua interface of Apple’s OSX operating system.  It’s spiffy and Apple OS certainly needed a face lift, but it’s unsettlingly different.   It was hard enough that I, a user of the original Mac – and of the NeXT machine which is really the source of Aqua) had some difficulty with initial navigation.

We would suggest that there are other possibilities.

We could leave the interface as is, with minor tune-ups and offer alternative navigational modes which operate without changing the look-and-feel of the interface for mainstream users. 

Skilled users may prefer a command interface to point-and-click.  This permits them to keep their fingers on the keyboard and minimize need interruptions to their speed (and concentration) by the need to find and use a mouse or other pointing device.  It’s partly a matter of when they learned to use a computer and partly a matter of who they are – we’d note, for instance, that programmers often prefer command code interfaces, as do skilled users of some complex packages.

• Many of the command codes for menu items which permitted users to request actions without accessing menus, have disappeared.  They may still work, but you’d need to learn them from another source. 
  

• New functions which were created after the Windows interface took hold, often never had a command, frustrating the keyboard-oriented user.

There are some independent software products that let you do this.  We’ve been playing with one called Active Words http://www.activewords.com/ that will let you access anything you do via a command code you create.  The tradeoff, of course, is that you have to create the codes.  You can do this up front or on-the-fly as you use the product.  It will help you by noticing things you do frequently and offering to make them into commands.  If you’re a keyboard-centric user, you’ll want to check this out. Remember the rules of delegation; they work here, too.  If you do something frequently, it’s worth the time investment to recruit some help and train them; if you’re doing something once, it’s probably easier to do it yourself.

Very casual users, by the same logic, could be offered an interface that required little or no knowledge of the computer but simply gave them access to what they requested through natural language processing of keyboarded or spoken requests. 

• The user could simply say “Send an email to Ellen” and the appropriate program would open and an email would address itself.  If “Ellen” is an ambiguous address (if the user knew three Ellen’s, for instance), the system could offer a list of all three, with a request to make a selection.  Text could be dictated or keyboarded, as the user preferred.  Other functions would be handled in similar ways.
  

• For casual users, who may be presumed to require more help, the system could default to offer dialogue box panels, spoken, or other help, always offering to turn it off if the user prefers.
  

The Big Bang:  A Whole New Interface 

A second possibility is the “big bang” theory.  That is, to completely reinvent the interface but to offer an advantage so compelling that users are willing to learn a new way of interacting with their computer.  I suspect that this is unlikely to happen on current PC’s although once it happens (and that may be sooner than you think)an appealing new interface may move backwards – somehow or another – to as many existing systems as possible 

It’s hard to imagine anyone but Microsoft with sufficient clout to change the PC interface and they have no incentive to do something revolutionary rather than evolutionary – revolutions threaten market breaks and if you’re the monopolist in control the last thing you want is a market break. 

We’re more likely to see a whole new interface as part of new technology – interface appliances, handheld computers, computing designed to be purpose-built for something that current interfaces are really not appropriate to support.  But this REQUIRES that someone new be behind such efforts because otherwise what we get is an effort to create an interface that fits into the current mold.

I don’t know what this interface looks like.  I like the idea of real world metaphors but I think we’ve pretty much used up the piece of paper/file folder idea.  Maybe we should move to communications metaphors.  I also like the idea of a simple interface that unfolds and shows more of itself as the user needs to use more of the system’s facilities.  I don’t know what the interaction to do the unfolding looks like, though.  Kai Krause, the designer of some of the best graphics software I’ve ever seen or used (Kai’s Tools) had some ideas along these lines.

The best hope here is that we get some new blood into the game – design engineers from the consumer products world or game designers who build interfaces because they’re fun to use not because they’re similar to what we already have.  You’ll know you’re looking at something with possibilities when

The Palm was the last interface and device that I saw do that.  It was great but not great for enough people fast enough.  So we need someone as smart as Jeff Hawkins (now that’s a tall order) to do it again.  I’ll know by my traditional method.  It will pass tests one and two, of course, and the hair on the back of my neck will tingle – that happens about once every few years and it’s always worked.  Hurry up guys, I’m waiting.

(back to top)  

==========================================

Some Software Companies Are Making A Profit

==========================================

We note with pleasure that in a tough market some software companies are continuing to make a profit.  This week saw, for example, Corel Corporation of Ottawa, Canada, announce profits for the third quarter in a row.  Admittedly quite small ($102.7 million year-to-date), nevertheless, it’s good to see that by refocusing it’s efforts on its core competency (the graphics market) and by very careful cost controls they can continue to be profitable – and to take advantage of the many fine software technologies that are now available for sale.  They’ve bought several recently, including Micrografx and SoftQuad. 

Symantec (nearly eight times Corel’s size) has also been holding up well in a tough market, continuing to be profitable by maintaining its tight focus and careful cost controls.  Recently it announced that it will be looking at the need for a Security Suite rather than separate security offerings (their focus has been on their famed anti-virus software, an industry standard), but still strongly focused on the security and data integrity market.

We suspect that in a market where many firms are available for a fraction of their real worth, it is very tempting for companies with cash to go on a buying spree.  It takes real discipline to maintain focus and buy only companies whose products and other values (channel strategies, partners, in-progress research) have a good fit with the buying company’s strategic focus.  Growth through acquisition is a tough strategy and it takes a strong hand at the helm to make it work.

(back to top)

Comments or Questions: Send Email to opinions@wohl.com

Home/ Search / 2005 Articles / Issue Archive / Free Newsletter

Entire contents © 2001  by Amy D. Wohl. All rights reserved. Reproduction of this publication in any form without prior written permission is forbidden.