Every web site should be innovative in design and content, but, when it comes to usability, a slightly conservative mindset is the best option.

Most visitors are used to particular ways of navigating sites and entering data in forms. If you keep web design simple by sticking to those well-known ways, you're reasonably certain that most of your users will understand how your site works.

In web development practice, this may mean copying usability concepts from other web sites. If thousands of well-known web sites have a navigation bar on the left, your users will immediately understand what the off-color bar to the left with the many underlined bits of text is for.

Copying well-known web development concepts can be dull, even though it may be simple. As the first column explained, simple web sites are often mistaken for dull sites. 

All web designers, who have fallen in love with the endless empty spaces of the web, think that ‘simple’ is equivalent to ‘dull’. Web design is attractive because designers want to leave their mark on these spaces and fill them with new ideas. Yet, on the web, there are few fundamental rules.

For example, if you have a great usability idea for web development and try it in a few web sites other web developers might pick it up and copy it or write about it. It's quite possible that, only one year after you originally conceived it, your idea could have grown into a basic law of usability.

Web developers and web designers are, almost by definition, concerned with new ideas and with trying them in their sites. New ideas usually start as quite complicated affairs, mainly because the inventor doesn't yet know which parts of his or her idea will be important in the long run and which parts can be safely removed.
If web developers want to avoid complicated web sites at all costs, they should avoid new ideas. Understandably, they don't like that idea, and simplicity and dullness in web development have become one and the same.

A good way to strike a balance that will satisfy both innovative web developers and conservative end users is to use only one new idea in any web site. The users can concentrate on understanding this single new idea, while the rest of the web site is comfortably familiar.

Suppose you create a web site which contains lots of forms. An extensive navigation provides access to these web forms, and the visitors need some help entering the data. Finally, the data has to be validated and the results of this validation presented to the user.

Of these three usability problems, only one should be solved by a new idea. If you opt for a totally new way of navigating the forms, the help and validation functions should be based on old, well-known web design concepts. If you want to create an extensive and innovative help function, the navigation and validation should be as simple as possible.

Scientific approach in web development has three distinct advantages. First, it reduces user confusion, which is always a good idea.

Secondly, this web design approach forces you to think and rethink the web site and your ideas on a fundamental level. Like which site feature is the most important: navigation, help function, or validation, or which of your new ideas is the best.

Finally, the goal of incorporating new ideas in your web sites is to find out if they work, if they make a certain task easier and more intuitive, or if they provide a better interface for it. Therefore, the web site you use them in is an experiment to find out if your idea is a good one. This experiment should be as scientifically correct as possible.

In any scientific experiment, if you want to find out if something works you isolate it. You make sure that all other factors that may influence the outcome of the experiment are as predictable as possible. Thus any reaction to the experiment is a reaction to the one unknown factor you allow.

In web design terms, any user feedback on or statistics of the web site as a whole should ideally reflect only on your new web development idea. All items other than your new idea should be well known, even a bit boring, so that your visitors will use them without thinking and won't comment on them.

But web development is not an exact science and truly scientific experiments cannot be performed on web sites, since even the simplest web site contains far too many unknown factors.

Nonetheless, it's a good idea to be as exact as you can, to isolate new ideas and test them, without a clutter of other new ideas that may distort the outcome of the experiment. Even if your data won't be scientific evidence, you'll have a better chance to discover how your idea works in practice and how you can improve on it.