Many tend to over-complicate this issue. GUI's are an intensely practical thing. It is sort of a chicken and egg thing.

First off, do we have GUI standards or De Facto GUI Standards?

I contend we have De Facto Standards. AFAIK, no ANSI GUI standard exists. No IEEE GUI Standard exists.

Second, which came first, the thoery or the gUI?

The GUI said glibly as she was smoking a cigaratte...< bg >..

Seriously, the thoery, such as it is, is the result of lots of trial and error. I am not sure how much thoery really exists since GUI's don't exist in the abstract. It simply is not an abstract concept.

I love folks that over-intellectualize the issue of GUI's.

The right GUI is the GUI your client is asking for. If it feels good to them, it is the right GUI. A grid or listbox? An option button that looks like the old radio buttons or graphical radio buttons? A drop down combo or a checkbox?

Through the body of experience, the wheat gets separated from the chaff. The cream rises to the top. Pick your cliche... What works, works. What does not, does not.

Sure, there are a lot of studies out there on GUI-Design. Ultimately, it does not mean an awful lot. My experience and the experience of my clients is what counts. To do anything else, is to over-think, over-intellectualize, is to be ineffective, is to be inefficient, is to cost your client more money, is to not get much done....

Simply put, look at what works and do that. I know a lot of people refer to Cooper. I wonder how many people have actually read his book. I think I spent about a half-hour with it. Finally, I said DUH!! This guy is the master of the incredibly obvious. Secondly, I asked myself if I could practically conduct the type of analysis this guy calls for. The answer - no. Would the client pay for it? No. Should the client pay? No.

The thing about books like Coopers that gets me is that the authors love to make up new words, reinvent the jargon. Out of hand, rejects things. A good example are confirmation dialogs. Cooper hates them. What if your client wants them? Are you going to say sorry, Cooper's law says no. Of course you are not going to do that and I am not suggesting you would. IMO, the book could have been cut in half in lieu of taking a sub-microscopic view of the world.

I did not mean this to be a diatribe on slamming Cooper's book and the like. Still, folks quote this puppy, and I did as well for a while. There are some good ideas. However, on the whole, folks should rely on their own experience and the need of their clients.

Just my 2-cents...

< JVP >








>>Boy, I am afraid I couldn't disagree with you more here...I couldn't care less about some ambiguous "standard" written up in a Microsft MSDN library book (MS is not the only company to build a GUI operating system, so I don't necessarily care for Windows standards)...
>
>We're probably going to have to agree to disagree here- I think that our difference of opinion is a difference in philosophy. My philosophy is that the way that people get comfortable with 'computers' is by getting used to patterns in software. They know that they can close a window by clicking the 'x' or hitting escape, they know that they can move from field to field by hitting the tab button. Eventually, they learn that a check box signifies a yes/no choice, a dropdown list contain several choices, and that you can use the scroll handle to scroll a window or grid. Developers often take it upon themselves to 'improve' the interface by changing what a control is used for, because they personally (or an interviewed user or two) think the app would look or work better that way. Often the interviewed user is one that is not familiar with windows software, or the developer thinks that the average user is incapable of learning patterns. Well, eventually the user gets comfortable with
>Windows software through experience- they're using Office, Photoshop, OE, IE etc. They understand the patterns that are consistent across software, and then come back to the software that was 'custom designed' for them, and it looks plain goofy. It turns out that the special non-standard uses for standard controls are no easier than if the proper control had been used, and now they have a piece of software that looks out of place in windows, and works differently from all the other software that they use.
>
>The optiongroup/checkbox debate is not an extreme enough example to make the point clearly, but I have seen _lots_ of this type of software- the developer thought he was going out of the way for the user, but ended up leaving out functionality that is vital for quick navigation or creating a confusing form, all this as a side effect of creating a UI using controls and employing usage patterns that are not time-tested. If you have not done so already, I suggest you check out The User Interface Hall of Shame and/or read 'About Face' by Alan Cooper. The book, and especially the website are chock full of stupid interface decisions that were made by developers trying to 'think out of the box', trying to create a slick interface, or simply creating a confusing app because of their ignorance of UI standards. When I say that the research has been done for you, I mean that there are millions of users world wide using millions of software
>programs that are employing these very standards.