AskTog: Interaction Design Solutions for the Real World
Interaction Design Section   Living Section   About Bruce Tognazzini
NN/g Home > AskTog > ReaderMail > July, 2000 Ask Tog, July, 2000

Ask Tog Reader Mail

 Cellphones to remain obnoxious?
 Microsoft-style MDI interface passé?
 IT manager confuses Excel with Oracle
 Suggested Mac enhancements
 Bill Gates invented toolbars and margarine
 Hierarchical Menus: Mac vs. Windows
 Microsoft to release OpenDoc 2.0?
 Apple's new Rockwellian icons
 One-finger Qwerty. Right or wrong?

Cellphones to remain obnoxious?

Dear Tog,

From Cellphone Solution in Search of a Champion (May, 2000):

Add detector circuitry to all new cellphones that will receive and respond to a simple coded signal telling them not to ring. Then, sell transmitters to all the restaurants, theaters, churches, and bosses. Turn on the transmitter and it will transmit a very local signal that will force the phones to pipe down.

Your proposed solution to the spreading nuisance [of] loud public cellphone ringing (and often ensuing and no less public conversations by so-equipped beings) rings hollow to my ears, due to one big logical flaw to its, ehmmm, logic:

The implementation rests squarely on the telephone producers and carriers’ willingness to cripple current functionality level of their products/ services. This runs counter to their intent, making as much money off their customers as possible.... pardon, maximizing the profits OF COURSE. The less obnoxious the phones are, the less they’d remind their owners of their existence, the less they’d be used.

Assume for a moment—we can dream, can’t we?—that the powers that be mandate that all phones HAVE to have such remote- switching-of-audio-ringing capabilities. Do you know how long it’d then take city commuters, say, to demand that public transport companies (a concept still viable in Europe) install such “switchoff black boxes” in all busses and trains, other enclosed public spaces?

“Like, no time at all.”

This can hardly be unknown to cell phone producers and operators --they conduct various usage scenarios as part of their forward planning-- so your humble proposal stands little chance of being implemented. In fact, as long as there is no widespread public outcry over annoying use of cellphones --I’ve seen people attending meetings who, rather than switch off the call, responded and engaged in “hushed” conversations while crouching in between stalls, as if that automatically *silenced* them to others-- as long as there is no “cellphone rage” over the lack of respect of others’ silence when conducting private one-party dialogue in presence of others, I don’t see a viable solution in sight.

But, once a few “happy-cellphone-campers” get their phones forcibly taken off and smashed and (in America, probably) killed, voices will rise to legislate “counter-annoyance” changes into industry standards.

Remember, you read it here first.

Ian Feldman
Stockholm, Sweden

I made my proposal in anticipation of exactly the sort of cellphone rage you mention. Such a backlash is indeed gaining momentum in the United States. In fact, this particular solution came to be after listening to hours of debate over what to do about cellphone abuse. Not one person ever mentioned a technological solution to the problem. Debate, as usual, centered around the consumers involved, rather than the manufacturers of the offending products.

The recent success in reigning in our most popular purveyors of death, the tobacco companies, is encouraging lawyers and lawmakers everywhere to embrace new Draconian measures to attack other forms of air pollution, as well as moral pollution. Cellphone abuse is climbing ever higher on the list. Before that spotlight turns on, and the cellphone industry finds themselves fighting a battle to banish cellphones from public spaces altogether, they had better start rethinking their products.

Microsoft-style MDI interface passé?

Dear Tog,

Our GUI group was unable to reach consensus in a design review. The proposed design was based on MDI (Multiple Document Interface).

Some of our group considered MDI to be a poor design choice. It was observed that Microsoft, who invented MDI, have moved away from it in their recent products. Also, that they have posted this warning on their website:

Many new and intermediate users find it difficult to learn to use MDI applications... Therefore, you may want to consider other models for your user interface.

When we argued it, valid points were made on both sides of the issue. It was not clear whether the advantages of MDI outweighed the disadvantages. Do you have any experience or information that would tilt the scales?

Thanks sincerely,

Heman Robinson.

I must confess to having never discovered any advantage to MDI. It is modal, it robs the user of freedom, and it is confusing to learn and to use. The only reason I can see for its creation was to make the interface different from the Macintosh. Unfortunately, the Mac got it right.

When I use a typical Windows application, unless I make the application window full screen size, I feel as though I'm working through a little porthole. The problem is magnified as soon as you open a second document. Instead of being able to spread the two apart, there is usually no room to maneuver. You end up having to increase the size of the application window just to move a document within it.

Because this window is missing entirely on the Macintosh, this extra step simply doesn't exist. You can move application windows as far apart as you want without having to cover up everything in-between with a big, hulking window.

IT manager confuses Excel with Oracle


I have been working with computers since I was about 6 years old, and as such, have a pretty good understanding of why they are so unfriendly to users. Anyway, my dad recently signed up with a new company which is very good in most respects except for one thing: The guy in charge of computer stuff just does not know what he’s doing.

Recently, my dad asked for help sorting through a database listing companies and various information about them, such as name, customer number, etc. All my dad needed me to do was to sort the database according to criteria, e. g. name, address, or whatever.

I figured that this should be pretty easy, until I discovered that some bonehead had decided to input the database into a Microsoft Excel file. I tried explaining that Microsoft Excel is for financial/math info, and that software is best when used as a specialized tool instead of a swiss army knife approach.

My dad found this hard to believe and asked me to help anyway. Reluctantly I attempted to do this operation, which would be relatively simple with a good database, and I soon discovered is very annoying under a spreadsheet. At my school, I often volunteer to help with their computers, and they use a proper database for student information. There, doing what I was struggling with now is a trivial operation that takes roughly two seconds to do.

How can I explain to the clueless IT person at this company that Excel isn’t a good database program? More to the point, do you agree with me that Excel shouldn’t be used for databases?


I have seen people do the most amazing things with Excel, often with disastrous results. An old expression suggests that when you only have a hammer, everything looks like a nail. Excel, of course, is not a hammer. It is a Swiss Army knife. Unfortunately, all the blades except one are broken. However, because Excel has so many extraneous "features," people get focused on it, diving deeper and deeper into Excel, while turning a blind eye to alternative solutions.

Of course, your problem is not with the boneheads that crammed 32768 features into Excel. Your problem is with the bonehead that thought it would be a good idea to use them.

First, accept the fact that you may lose. These people may simply not care that Excel is a foolish tool for databasing. They will receive their paychecks nonetheless, and that may be their only real concern.

Second, work on the assumption that they would do the right thing if only they had a clue what the right thing was.

I used to go through life surrounded by people who were stubbornly doing the wrong thing on a regular basis. And I would point it out to them by yelling. My wife finally pointed out to me (while not yelling) what was really happening: I'm smarter than average. Therefore, average, to me, seems kind of stupid. People weren't purposely being stupid. It just came naturally.

I suspect you are, likewise, smarter than average. You will likely go through life, as I have, surrounded by people who just can't see things are clearly or as quickly as you can. They are not being stubborn. They just don't share your abilities. Since I have stopped yelling and started helping, things have gone a whole lot better.

If you really want to change their minds, you need to demonstrate the superiority of a real database. That means you need to get ahold of FileMaker or some other real database, transfer over some of their information, and show them the advantages, through side-by-side comparison. You also need to sell them on the ease with which they can learn the new tool, as well as the ease with which they can import the data.

Which returns us again to point one: They may "get it" and still not care. And that's OK, because while they remain condemned to circle in place in their inefficient ways, you will striding forward, winning far more important battles.

Two Mac Enhancement Suggestions


Re: your wishlist for future Mac or PC enhancements, please consider the following:

1. Multiple Clipboards. The edit menu is one of the most powerful features of PCs. Without it one would have little way to get data between apps. You could jot things down and retype them, or just rely on memory. The Clipboard has got to be one of the most important tools, with some of the least enhancement in the past twenty years. Having only one store for the clipboard data is too limiting. For example, you can’t swap the contents of two fields since you can only hold one item on the clipboard at the same time. I propose that we allow multiple stores for the clipboard. The way the Newton did it wasn’t so bad. Drag things to the side of the screen and they hang out there with a visual representation consisting of a summary or some such. When you want to paste it, you drag it over. Of course have keyboard shortcuts for this task as well.

2. I would like it very much if when clicking on the scroll bar area to do page downs (not scroll by one line, but a whole page at a time), and are about to click one last time to scroll to the last page, if the application actually scrolled one whole page. Right now, the last page only scrolls enough to put the last line of the document on the last visible line of the window. This creates a big problem for our eyes/brain which expect to find the next line to read at the top of the page, not n lines down. I have asked countless people and everyone says that they end up having to skim over the text to find the line they were at. NewsWatcher on the Mac is the only application I know of (except for our own network management app) that scrolls in such a way that you know exactly where the next line appears after a page down or up. The way NewsWatcher does that is to add blank space to the bottom of the page, just like the last page in a chapter in a physical book may have blank space after the last line of text.

I spend hundreds of hours on the web reading information, and have found that I would rather scroll one line at a time or drag the thumb, than to let the last page confuse my eyes.

Mahboud Zabetian
AG Group, Inc.

Excellent suggestions. The last-page page down has driven me crazy on a daily basis for years. I have come to the conclusion that all new revision efforts should apportion 20% of the design effort to correcting what is already there. Almost universally, people become so wrapped up in the exciting new features that they ignore the current problems.

Bill Gate invented toolbars and margarine

(OK, he doesn't claim he invented margarine—yet.)

Hi Tog,

Love the site; it has been an incredible help to me and has broadened my interests into an area I barely knew existed until I started reading your columns. (you must almost be sick of hearing this story from dozens of people a day, so I will get on to the subject of the email)

I read this interesting (nay, appalling) article:,3266,44557,00.html on TIME’s website, written by Bill Gates, in which he defends the claims that MS could not innovate as two separate companies, and along the way also claims that MS invented toolbars with the advent of MS Office. I’d love to hear your two cents on the subject in next month’s column.

Jonathan Heron

1) I am not now and will never be sick of hearing this story.

2) Microsoft does not invent. It copies. Always has. It's the secret to their success. They look around for good ideas and appropriate them. Sometimes they have to pay money. Even exorbitant money, like the $50,000 they got held up for for MS DOS. Took minutes for that to finally pay off. Other times, they just take the ideas outright. Much better for the bottom line. Now, they seem to be adding a new element: Claiming they invented that which was appropriated.

The first time I saw toolbars was on Apple's LisaPaint, which Bill Atkinson wrote in the late 1970s, years before Word 1.0 even existed. My best recollection is that LisaPaint was based on a high-end graphics system that Bill had seen at SIGGRAPH or its ancestor, and that that graphics system likewise was toolbar based.

What those boneheads at Microsoft did (to borrow a phrase from John) was to move the toolbar away from the left edge of the screen, where it drew advantage from pinning (see: Fitts column). Instead, they put it in about the worst place on the screen. Typical.

Hierarchical Menus: Mac vs. Windows

While playing around in my menu bar, I noticed something very interesting about the way the Mac handles hierarchical menus.

The code for maintaining the menu apparently defines two regions, with the border between them a diagonal line drawn between the top-left corner of the hierarchical menu item in the parent menu and the lower-left corner of the submenu. As long as the cursor stays within the blue region, the 1.5 second delay applies before the next item is opened. If the cursor ventures into the right-hand region, however, the submenu immediately closes and the next row is highlighted, all with no delay whatsoever. Thinking about it, that’s the most intelligent possible way of determining the user’s intention -- not many people aim lower than the last item on the list if they want to hit one of the items in the list.

Now, Windows, on the other hand, does something different. I can’t quite remember the exact way it works (I think I repressed most of that -- it’s too painful to an HI designer), but I believe that if you wait long enough to actually let the submenu come out and then move onto another menu item, Windows does a little of both; it leaves the submenu open for a second or two so you can come back to it if you accidentally left its boundaries, but it also selects the next menu item. Of course, taking a screen shot of that particular situation and showing it to someone would produce a puzzled look; it would seem as if a submenu had popped into existence out of nowhere, with no visible parent menu.

When I first started writing this, I had no idea when, how, or from whose mind that particular trick originated. It seems now I have the answer to that question, and I would like to personally thank you for creating possibly the most subtle, efficient, and elegant human interface coding I have ever seen. It’s such a simple thing, but it’s something I and likely millions of other people use every day. So thank you for that, and my wish that you could have an hour in a locked room with the designers of OS X just became quite a bit more fervent. I don’t know what Apple’s gotten into their collective heads, but if I see one more brushed metal application I am seriously going to scream.

Brian Ellis

Eric Hulteen share equal credit. I don't know for sure why the Microsoft solution is so clunky. It could be they just didn't do a good job of copying the Mac. More likely, they never tested and iterated. When Eric and I finished our first prototype, it was at least as clunky as Microsoft's solution. We, however, put ours through another four or five rounds of testing and design before we elected to ship it.

Microsoft to release OpenDoc 2.0?

Recent rumors indicate that the upcoming version of Microsoft’s Office for Macintosh will include the ability to share tools, resources, and data within documents in ways which haven’t been previously possible.

For years now, one of the most frustrating things about working on my Macintosh has been the lack of integration. Each program has its own spell checker and they all work differently. My email address book doesn’t talk to my contact manager, which in turn doesn’t integrate with my word processor. And the OS itself hardly talks to any of them. I had high hopes that OpenDoc would eliminate all these incompatibilities, along with the whole concept of focusing on the tools rather than the product. But since Steve killed OpenDoc, what’s left?

Now, in steps Microsoft, true to form, taking a good idea and implementing it poorly, but with enough muscle that it becomes the standard. If Microsoft includes some sort of file management, I imagine that users could put Office in their Startup Items folder and run nothing else. In effect, Office becomes the OS. It’s got email, browser, word processing, etc. -- and it’s all integrated. A lot of people will jump on it because it’s a single solution. Unfortunately, it’s probably not a good solution.

While all the talk about the upcoming MacOS is focused on transparent icons, docks, and the like, the real goal of the computer as an integrated, intuitive tool remains science fiction. Watch any 35-year-old episode of Star Trek for confirmation.


I've been explaining integrated documents for almost fifteen years now, and yelling about them even longer. They are the obvious solution for most document creation tasks. Applications are modes, even when they don't trap documents in an MDI window (see above). When you create a document with words and a picture in the real world, you don't have to use specialized stationery for each medium. You doodle and you print. You should be able to do the same thing on a computer.

Unfortunately, OpenDoc finally arrived on the Apple when the party was winding down. With Apple down to less than 10% of the market, where it remains today, the R&D money has simply dried up. Apple can no longer compete. Microsoft seems to take around five years to copy Apple, so if they follow and OpenDoc approach, they will be just about on time.

Of course, OpenDoc was open, with all developers invited to join the party. It will be interesting to see whether Microsoft really adds value by letting other developers in, or whether this is to be yet another Microsoft enclave, designed to harm the competition.

On Apple's new Rockwellian Icons


Can I assume that Apple’s policy with Aqua/Quartz, to have 128x128 icons in millions of colors, conflicts with your philosophy?

One has to ask, What is the purpose of an icon?

From what I gather, its original purpose was to communicate an idea as clearly and simply and minimalistically as possible. Reducing uncertainty was a high priority.

Take the icons used in road signs for example. A Pedestrian Crossing road sign has changed very little, if at all, in the 50-75 years or so it has been around.

And though science and technology has made quantum leaps in the last 75 years, the Pedestrian Crossing sign hasn’t changed a bit.

Today, rather than taking on the detail and shading of a Norman Rockwell painting, the Pedestrian Crossing sign is still comprised of simple stick figures to communicate as cleanly and coherently as possible.

This is a classic case of less is more. The less information the road sign icon contains, the more it communicates and the less the uncertainty it creates.

So I am curious about your opinion on Apple’s Aqua icons where coherency seems to have taken a back seat to artistry.

I would also like to point out a new phenomenon that is occurring in the wake of these new detailed Aqua icons: sub-icons. These highly detailed Aqua icons have created the need for tiny minimalistic icons (more in the tradition of the original icon concept) to narrow down uncertainty.

Mike Donahue

You do not understand, Grasshopper. The purpose of OS X's giant icons, the purpose of its whole new flashy look, is to keep the customer happy during that critical period between the time of sale and the time the check clears.

Seriously, though, Steve has never had respect for interaction designers and has always held graphic and industrial designers in the highest respect. When the Mac came along, the Mac team, grudgingly, accepted the design effort of the Lisa team, which was dominated by interaction designers. Today, no one is providing balance, and features that are demonstrably deleterious to the interface are being fervently embraced just because they look pretty.

One-Finger Qwerty. Right or wrong?

I’m continually amazed at how many times the only keyboard offered for one-finger entry or stylus entry is the Qwerty, which is definitely suboptimal. I have found that the Fitaly layout ( really does work, but even so I have to >>add<< it to my Palm.

And recently I had occasion to try a Rocket eBook—very nice, in that you can make annotations, but when you do you have to use a stylus on a Qwerty keyboard.

Qwerty may be okay for a seldom-used key entry (e.g., touchscreen keyboard at a shopping mall kiosk), but if the user has to enter text repeatedly and frequently, I wish usability engineers would get device manufacturers to include a Fitaly option.

Thanks for listening... er, reading.

Michael Ham
MetaWare Incorporated

The Fitaly keyboard is indeed a nice design for one-finger entry. However, you must realize that we have a long, long history of superior keyboards losing out. Indeed, it is rare that anything new happens in input devices. In the past 40 years, only the trackball, mouse, and touch screen have made serious inroads. Dvorak keyboard layout continues to be an also-ran, in spite of its proven superiority. (Or is it proven. Check out

I can't fault Palm for not including Fitaly. They included something that was at least as courageous, if not more so, in the form of Graffiti, a handwriting system based on something other than normal handwriting. Graffiti differs only subtly from normal handwriting, but Dvorak differs only equally subtly from Qwerty. (I'm told it only takes three or four days to get up to speed; I'm still learning special characters with Graffiti after six months.) People just hate to learn new things.

The other reason I don't fault Palm is that Graffiti, once learned should be much faster for most uses on the Palm. That's because Graffiti can be entered anywhere, anytime without having to enter a special mode to do so. When I do want to really write on my Palm, I use Palm's fold-up full-sized keyboard. And that will be five or ten times faster than Fitaly, too.

I can see building a device around an actual, as opposed to virtual, Fitaly keyboard. The big task, however, will remain getting people to take the time to learn it. It can be done. It just requires the kind of painful effort Apple put in with the mouse and Palm put in with Graffiti.

In the meantime, palmtop devices now have memory to burn, and little reason remains not to include a Fitaly option.

Don't miss the next action-packed column!
Receive a brief notice when new columns are posted by sending a blank email to

return to top

Contact Us:  Bruce Tognazzini
Copyright Bruce Tognazzini.  All Rights Reserved