Praise stealth

In this text, I hope to describe some of the issues of creating user-friendly sites. I will try not to affect any aesthetics, so this article is not intended for purely graphs. This article is about the interface.

So, we need to define what a good interface is. We can say that with a good interface, the user achieves his goal with minimal effort and time, receiving aesthetic pleasure from the beauty of design. But I like the other wording better: the interface is good when the user reaches his goal with minimal effort and time, without even noticing this interface. It seems to me that the information that the interface accesses is more important than this access, i.e. interface. Those. beautiful, but inconvenient interface worse than ugly, but convenient. So, I know two ways to make the interface more convenient.

Some problem is that the design should be pleasing to the eye. But after a while any picture begins to tire - if only it is not gray and imperceptible (in this case, the eye learns not to notice it). Accordingly, there is a dilemma - to make the interface beautiful, but tiring, or inconspicuous, but not tiring. An example is the Microsoft Encarta Encyclopedia. Her version of '97 was fabulously beautiful. And then came the year 1998, a new version came out - boring, gentlemen. Everything is brown, nothing decisively moves. But that is characteristic - with the 97th version I couldn’t work for more than an hour, I got tired, and with the new one I sat for five hours and then got up fresh and enlightened. That's it.

So you have to decide which interface to do — for the Web, this means choosing either an interface and a design that encourages a random user (reader) to linger, or an interface and design that do not prevent the user from receiving the necessary information. From my point of view, the dilemma is not too complicated. You just need to calculate the time that the average user will stay on your site, and if it is more than N minutes, you no doubt need to choose a modest option (making the first page pretty). The number N for me is about 15 minutes.
')
Thus, you will kill two birds with one stone - at the same time the surfer will be attracted to the first page, and the intelligent user will not be offended. A person who came to the link from the search engine, generally do not care about beauty - he is looking for information and nothing more.

There is another argument in favor of unobtrusiveness. It is clear that creating an absolutely beautiful design is impossible. By the phrase “absolutely beautiful design” I understand a design that everyone likes (aesthetically). Everyone has their own tastes. So gray (I emphasize - not ugly, but barely noticeable) design has one significant advantage - it is neutral and not disgusting to anyone.

I am not agitating for modesty at all. If you make a site for a design bureau, you need no modesty. But if you make a corporate website that is usually very tightly searched, it’s better to think about some unobtrusiveness.

But this is only part of the solution to the invisible interface problem. There is another part - associative clarity.

Personally, I do not like the dominant term "intuitive." If we proceed from the axiom, which says that the user is not just a fool, but a fool round, it is useless to expect from him the presence of intuition. You can only count on the appearance of any associations.

So I like the term “associative clarity” much more.

The problem is that for the user the site is just a window with constantly changing content. Accordingly, the WEB-designer has much less possibilities than, say, an interface designer of an operating system. He cannot, for example, control the opening and closing of dialog boxes, there is no drag-n-drop at all (in fact, all this is possible, but with a headache — his own and the user). And if there are fewer opportunities, then there is less choice of possible associations.

If you list the most frequently used associative models, it becomes clear that the choice is not too large. Operating systems most often use the desktop metaphor (Windows, Mac OS, etc.), most Unix interfaces use office cabinets. PageMaker carefully imitates the assembly workbench. But programs that do not have an associative interface model at all are much more. For example, neither Word nor Excel does not imitate anything more familiar. But Word is very similar to WordPerfect and WordPro. And vice versa. Sooner or later there comes a time when it is useless to count on a kettle. A person can hardly learn how to fully use WordPerfect - but an experienced WordPro user will master it without any problems.

A simple conclusion follows from the previous paragraph - do not reinvent the wheel. Count primarily on the prepared user. Find out what he already knows. Do the same.

I do not advocate plagiarism at all. Moreover, if you place the same button in the same place. No one will be offended. But the user will not get into the head a phrase from a children's movie - “where is his button”. And that's good. Here is another example. Many sites have a downloads directory. I know three people who in the relevant situation first of all check its presence, ignoring the WEB-interface. Why don't you have the same name for your catalog?

Actually it is like with operating systems. If the standard Ok button is used in the window of your program, no one will think that you are a plagiarist. You yourself do not think.