Welcome to the 18th issue of Pineapplesoft Link. I welcome comments, suggestions and ideas at [address removed, the newsletter is no longer published thank you for your support].

Book Selection

For the first time, an issue of Pineapplesoft Link is entirely dedicated to one book. Of course, the book is available in our online shop at http://www1.fatbrain.com/FindItNow/Services/home.cl?from=PHP278.

The book selection includes many other valuable readings such as the classic "The Mythical Man-Month," high-quality Java and XML books and access to all technical books at fatbrain.com. Pineapplesoft Link receives a percentage on any book purchased by someone entering from our web site or the newsletter. This helps with the cost of running the newsletter.

Well Worth Reading

Pineapplesoft Link now reaches 2,650 subscribers. I'd like to see how fast it can grow to 3,000. Please forward this issue to a colleague and encourage him or her to subscribe.

This issue is atypical. It is not about technology, it is not about tools. It is about a book. Many technical books are published every month. Some of them are very good. Few however have a profound, lasting impact on our field. And fewer are relevant to both technicians and non-technicians alike.

The book I discuss this month is relevant to us all. Its impact on the field will be (should be?) profound. This is a book on software yet it is not technical. It aims to help the "business-savvy technologist or the technology-savvy businessperson" understand how to build effective and successful software.

I am talking of Alan Cooper's "The inmates are running the asylum."

Which Asylum

The software industry is booming. It is said to fuel the growth of the American industry, now in its longer period of prosperity ever. Around the world, software engineers are in high demand and software companies are appearing like mushrooms. That is a measure of success.

Another measure however is the rate of success, or rather for software the rate of failure. Let's face it, software projects that fail to meet expectations or that run into huge hidden costs or that are found too complex by users are commonplace.

The technical quality of software has improved thanks to object-oriented programming, new analysis methodologies, UML http://www.psol.be/old/1/newsletter/19990501_uml.html, better tools and programming languages. These techniques help control software development. When properly used, they have led to more robust software, that is easier to upgrade and maintain. However if the technical quality of the software has improved, its usability has not always followed.

User Acceptance

If you read this newsletter, chances are you are somewhat involved with computers. Look around you and, unless you live in Silicon Valley, you will find that there are more computer illiterate persons than power users.

This is a fact but most software fails to recognize it. Indeed many applications currently on the market are complex to use and challenging for the computer illiterate. This result in much frustration from would-be computer users.

When I left university, I thought people ought to train themselves and learn how to use a computer properly. Software ought to be easy to use, of course, but my definition of "ease of use" was influenced by years of formal training. After all, I reasoned, computers are powerful tools and if you want the power, you have to accept some pain.

With more experience, I realized this is no longer a sustainable position. When computers were rare and expensive, only trained operators could use them. Nowadays computers are cheap and widespread both in business and at home (the microwave oven, the video, the alarm, etc.). They should be usable by everybody.

I liken it to the phone service. In the beginning, people would not dial a number but ask an operator. As phones gained popularity, they became simpler to use and now we rarely need the operator.

Challenges

Today however software is challenging to use by all but trained people. Furthermore poorly though-out software is a huge time waster for everybody -- not to mention a major frustration. I will take two recent examples.

I tried to buy software at BuyDirect.com but their shop application would not let me enter my ZIP code. The application foolishly thought it knew where I am living better than I do, so it would not accept my order. When I entered an incorrect code, the shop accepted it but, of course, the credit card validation failed. I ended up shopping elsewhere.

How can a company build a shop so that (1) it treat its customers like idiots who do not know their address, (2) it does not even let them buy!

My bank, the Caisse Privee, is no better. In the beginning of the year, they rolled out their web banking solution and, much to my surprise, I found that I cannot pay taxes online. The reason? I am supposed to enter the address of the ministry. Of course I don't know the address and I don't really need it: I have been paying taxes for years with paper forms that have never required the address.

Is it not silly that a bank designs software that does not support common banking practices?

Alan Cooper would argue that none of these companies has designed its software to satisfy the user needs -- my needs, your needs. This creates frustration, wastes time and... cost them business. Says Alan Cooper "[when user are forgotten] Each new technology merely makes it possible to frustrate users with faster and more powerful systems."

Race for Features

The accepted wisdom in the software industry is that new releases should do more, not better. Pressured by time, vendors are trading quality for quantity.

Yet, except journalists (who rarely use the software for more than a few hours anyway), nobody cares how long the feature list is. We buy software because it solves a problem cheaply or, at least, we hope it does.

Luckily some companies have started listening. Again, I will take two examples.

I do all my writing with Sidekick from Starfish http://www.starfish.com. Sidekick has a minimalist word processor. It lacks fancy desktop publishing features (columns, etc.), it doesn't even have a spell-checker! What it does however, it does it well. First on my 266 MHz Pentium II, it loads in exactly 5 seconds. The tiny Wordpad can hardly beat that (4 seconds). Second it has its own document management: instead of files and folders, my documents are neatly organized in some sort of an index. I have never lost a document since I started using it.

Though Microsoft and Netscape dominate the market, Opera http://www.opera.com bet it could build a better browser and sell it. Compared with its super-loaded competitors, Opera is minimalist. Yet it renders pages faster and boasts an ultra-light download. Its users are happy to pay for the privilege -- there is no such thing as a free meal.

In his book, Alan Cooper shows that users rate a software as powerful if it does the job well, no matter how many features it has. He shows how to concentrate on core features and building them for success.

The Inmates Are Running The Asylum

Alan Cooper is famous as the father of Visual Basic. He now runs Cooper Interaction Design http://www.cooper.com, a design studio that helps companies build more effective software.

In "The Inmates Are Running The Asylum," Alan Cooper convincingly argues that software fails to meet user needs because it is rarely designed with the user, the real user in mind. He shows how this impacts on the business that creates the software (much like BuyDirect.com losing customers).

The book is not a technical book. It is not only about building better user interfaces. Ultimaly, it is about software project management. Alan Cooper does not argue about technology but on how to manage your company or team to build more effective software.

Anybody who has been in the software business long enough knows that erratic, capricious management has killed more projects than poor software engineering. This typically results in poor communication between engineering and management. Often management of software projects is reduced to planning but planning is a poor substitute for a vision and the willingness to execute it.

In his book, Alan Cooper introduces a methodology he has developed: he recommends introducing an interaction designer to the development team and design for "personas" (read the book!). The designer focuses on the user needs through "personas." This is an interesting approach to bridge the gap between developers and managers.

The team building the software has different goals and constraints than the user. In the words of Cooper, "While the *code* may succeed or fail on its ability to handle edge cases, the *product* will succeed or fail on its ability to handle daily use and necessary cases." The designer and "personas" make sure the user goals are represented.

There are other methodologies to address user interaction. However I am not aware of any other methodologies that have such a comprehensive and field-tested set of tools that so nicely complement software engineering with user interaction. Most methods in use are more like after-thoughts, trying to fix a software in the rush to production.

Get It

I hope I have shown you that "The inmates are running the asylum" approaches a problem that every software company faces. It brings interesting answers to common questions and it may well be one of these book that has a lasting impact on the software industry.

If you develop software or web sites for a living, if you manage a software development team or if you run a software company, then you ought to read this book. You will learn how to create more effective software and therefore build a more successful business.

Online Bookstore

Many bookshop carry "The inmates are running the asylum" but if you prefer online shopping, may I suggest you go through Pineapplesoft Link bookstore? We will receive a percentage on any book purchased by someone entering from our web site or the newsletter and it helps publish the newsletter.

Visit Amazon.co.uk http://www.amazon.co.uk/exec/obidos/ASIN/0672316498/1960.

Self-promotion department

I have no less than three XML/EDI presentations this month.

First there is a short presentation in Brussels, Belgium http://www.itworks.be/XMLinpractice.html.

Next there is a management briefing "e-Business and the Online Organisation" in Birmingham , England http://www.saaconsultants.com/saa_news/ebis_briefing.html.

Finally, there is an one day seminar on XML and electronic commerce in Paris, France http://www.technoforum.fr/Pages/Institut_XML_99/institut99.htm.

It appears I was too enthousiastic last month and I announced Jini, part 3 before it was published. It is now available at http://www.javacats.com/US/articles/ben/jini3.html. Part 4 will be available shortly.

About Pineapplesoft Link

Pineapplesoft Link is a free email magazine. Each month, it discusses technologies, trends and facts of interest to web developers.

The information and design of this issue of Pineapplesoft Link are owned by Benoit Marchal and Pineapplesoft. Permission to copy or forward it is hereby granted provided it is prefaced with the words: "As appeared in Pineapplesoft Link - http://www.pineapplesoft.com."

Editor: Benoit Marchal
Publisher: Pineapplesoft www.psol.be

Acknowledgments: thanks to Sean McLoughlin MBA for helping me with this issue.

Back issues are available at http://www.psol.be/old/1/newsletter/.

Although the editor and the publisher have used reasonable endeavors to ensure accuracy of the contents, they assume no responsibility for any error or omission that may appear in the document.

Mise à jour : juin 1999.
© 1999, Benoît Marchal. All rights reserved.
Design, programmation XSL & photo : PineappleSoft OnLine.