This is the second half of our interview with Laura Arlov. If you haven't read the first part of this Interview then you may like to do so before continuing. This conversation took place by telephone on Friday, March 10th, between Oslo and Dallas. The participants were David Anderson, Ed Kennedy, Sonny Lacey and naturally, Laura Arlov.

In this second part, we talk about Laura's book, "GUI Design for Dummies". We ask the piercing question, "What one lesson would you like people to take from this Interview?". We ask Laura for her predictions for the next 2 or 3 years in the UI Design field, we talk about her new book which is available only in Scandavian languages and finally she gives us her "Silver Bullet" for an improved User Experience over the coming years.

GUI Design for Dummies was a great book. So many well respected people have commented so positively about it, and rightly so.

I must say that the reviews that the book has received have been a great pleasure for me. I feel very grateful to my colleagues all around the world who have spoken well of the book. They have given me a sense of confidence in my consulting work.

You may have noticed that the book isn't easy to get hold of. Commercially, the book wasn't a big success. I am proud of the work and I stand for it.

Nowadays, you can get it at Half Price Books. It will save me a lot of email if you could publicize that for me.

I really wanted to write a Dummies Book. I find they are really accessible to the readers and I enjoy the fun in them. I have to admit that it probably shouldn't have been a Dummies book, because it doesn't fit the publishers marketing activities for that line [of books] which is why it hasn't sold as well as I might have liked.

You asked me [in the script], if there was just one lesson that I wanted people to take from GUI Design for Dummies, what would that one lesson be.

"You can't make a good User Interface just by following a good standard, but you can make a good User Interface by following a good design process."

One Goal for this series of master interviews is that the reader should take away the one nugget of advice from the expert. If the reader takes the one lesson from Alan Cooper, and now that one lesson from Laura Arlov and next month, the one lesson from Larry Constantine, then those three lessons combined start to turn them into good designers.

I don't expect the world to change overnight but I do want to take part in a change. I want to be one of the seeds of a change. I want better User Interfaces for all of us computer users.

One of the ways for me to contribute to that is to help your average programmer to think that there are things that he or she can do to have a better user interface in their projects.

That's maybe where I differ from others in the field. I'm not so intellectual. I'm not so sophisticated. I tend to try to make things easier to do, also for programmers. I like to have them think that making an improvement to the user interface is something which they can do without being really really difficult.

With every improvement, I hope that they will want to make more and more improvements.

I have brought out a newer book, here in Norway and I am working on a Web UI book for the Norwegian market. There is a Danish publisher who is looking to publish this across Scandanavia.

Many programmers I've worked with don't like to do UI coding because it is so fluid, so subject to change. How do you think we can approach the goal of better defining a UI Design and reducing the rework programmers have to do?

Well better tools will always be important for that.

Another pre-requesite is that we [ the UI Designers ] need to have extremely good technical knowledge or we have to find that percentage of the programmers who are actually interested in working on the User Interface Design. I don't believe that I can make a good User Interface design without knowing which tool is going to be used to develop it. I need to understand the technical architecture that is going to constrain my User Interface.

To me that means that a User Interface Design project of any size is really a multi-disciplinary sub-project. So, if we haven't got one of the good technical people in that team, then we at least need one that we can go talk to and will work with us.

Just like there are programmers who are really good at database programming, so there are programmers who become great at User Interface development and can become really great User Interface Designers. The best ones!

I wanted to recommend a new book. It's called "Programming Industrial Strength Windows" by Peter Hesselberg. Peter is one of my favourite techies. When he solves a problem, it stays solved. This book's great strength is that it details how to do error handling and exception handling in the UI and how to architect that into the design. Error handling is so important and is often done so poorly or not at all.

The big challenge is picking the work you want to do. There seems to be more work available than UI Designers to fill the positions.

I'd like to see us convince consumers that they deserve good, comfortable User Interfaces. I'd like to see more demanding, more radical consumers who send stuff back if they are not happy with hard to use software.

I'd like to see designers becoming a more integral part of the software industry especially in "custom software" projects.

The Web is helping a lot in this. The web has convinced software people that they need to talk to graphic designers. It is no longer difficult to make this happen. They can easily compare one website with another and see which is better.

I thank the Web for making computer people more aware of Visual Design.

Also there are so many sites which are dependant on User's choice. On the web, User's vote with their feet or their "clicks". Web Statistics are readily available to web site owners and this making them realize that they have to act.

We keep hearing that everyone in Scandanavia is getting a WAP Phone. How about you?

I guess I'm waiting in line but the dates just seem to get pushed back and back. I can guarantee you that Finland is ahead of Norway but really we just don't see phones here yet.

I will almost certainly be getting one. I like technology but I won't be the first in line.

I'll buy one when a service I use a lot is easily available on the WAP Phone. I use directory assistance a lot. If that becomes available then I would get a WAP Phone. I also order a lot of Taxis, so if I could do that by phone then I would buy a WAP Phone immediately. The trigger for me is the service not the technology.

I don't really feel qualified to answer this as I'm doing my first WAP application at the moment, however, I will share with you a few things.

In the past, I can always take as a given that the User is sitting down. I have some context for how the User is working. With WAP, I just don't know that any more. I can't say what they are doing: working; playing; or something else.

I also find that the physical devices are often a bad fit for my body, they aren't comfortable. I find it hard to hold them and look at them simultaneously.

I find that I am having to compensate in the software for the misgivings of the physical hardware design. I find that the hardware ergonomics are causing problems.

So I have to take all this into account.

I am planning extra Usability Tests to investigate how Users interact with the device.

In my own experience, the decoupling of the WML markup from the hardware design is actually very limiting with devices of this limited modality.

And yes, I too have concluded that additional Usability Testing will be needed and that the cost of Usability Testing as a percentage of the total project will probably be disproportionately higher with WAP applications.

I feel that there was really no thought put into the concept that it is the hardware and software combination which together make an "information appliance" for the User.

Yes, the industry suffers from the obsession with the notion that somehow software should be designed to be independent from the physical device or infrastructure such as Screen Size or Web Browser. This notion of code it once and it will run anywhere. That really is Nirvana for technologists.

However, this is a nightmare for the User Interface Designer because presentation is such a major part of the experience. It is how we communicate the functionality to the user.

Wireless gives us the worst of two problems. We have a very limited medium which forces us into very simplistic interaction and "baby face" conversations, whilst our clients are coming with more and more features. My reaction is to cut back. "Cut Back!" I keep telling my clients to "take more out".

You have to start small.

We are going to market more and more quickly and we are not going to get client acceptance for more and more usability testing. The way to go is to come out with smaller products and observe them in the field and extend them.

I am focusing on making User Interfaces which are easily extendible. I am making complete models then only realizing part of them. I'm building in the room to grow.

I have a Silver Bullet for this!

I believe that you can start with Web Designers, Programmers, Technical Writers. Just about anyone can become a User Interface Designer.

The Silver Bullet is - Teach People to do User Testing!

I like to see a development group doing its own usability testing. I am perfectly happy to see them bring a Usability Engineering Professional. If this people are available then bring them in. However, ultimately it is the developers who are responsible. Even if they don't organise the tests, they ought to be administering and conducting them.

For me, observing my own designs being used and tested is the best and fastest working education.

Immediate feedback is what helps Users understand how Interaction works! Immediate feedback is what Programmers need to understand what is a good design.

So I no longer object when a client asks for only a one day training session. I go to the client for just one day and all I teach them is how to do User Testing. If we are going to do only one day on UI Design, then I will teach you to do informal Usability Testing.

My final piece of advice on this is that, if you are running a test with a test leader, a paper prototype, a User and a number of observers, then one observer should concentrate on the test leader and observe everything they do which might be influencing the test outcome. This builds a self-correcting feedback loop which helps you improve the test quality as you work through more users.