Alternatives to the pixelated UI universe
Do you think the pixel is the only unit of measurement for building graphical displays? Come on, you can measure better than that! This month, the Cranky user offers tips for user-friendly HTML layout and interface design, and explains why pixels aren't always the best unit for the job.
I sometimes wonder whether graphics wouldn't be better off if graphics designers and developers had stayed with vector-based graphics. For all that pixels are wonderful (I have nearly two million of them in front of me right now), they sometimes make Web designers do really stupid things.
For example, most designers calculate the size of images in pixels, even though the size of a reasonably clear word (for example) is a function of multiple factors -- distance to monitor, the user's eyesight, display size, and more. Displays alone vary so much that specifying size in pixels leads to graphical images that are too big to be useful or too small to be easily read. In fact, in Web design, I would say that the one thing you can count on is that you cannot calculate the size of your image in pixels. So why do so many designers do it that way?
I think I can explain the drawbacks of over-reliance on pixels and suggest some design techniques for getting around them.
Variable text size
Different browsers offer different levels of control over graphical display. A few browsers allow the user to set a minimum font size, such that absolutely no text will be rendered below the set size. It's a great feature, although in a perfect world it wouldn't be necessary.
Apple's Safari browser was one of the first to add size control to its main-browser toolbar. Firefox lets users specify a display resolution and renders all text accordingly. On many pages, features like these yield beautiful, clear text that is easy on the eyes. Wonderful, but hardly a fix-all.
For example, consider what happens when a page's layout becomes entirely unusable given a different font size. If you're lucky the main text column will simply become too narrow to be easily read. If you're not so lucky the text might extend under or over another part of the display, resulting in overlapping or invisible text.
A browser that gives users more control over the display is a great idea, but it's no solution for poor design.
Graphics and text
|
Attempts that use graphical backdrops to make text render correctly are a major source of pixel problems. For instance, the Web site for Movable Type blog software (see Resources) uses a graphical backdrop for one of its menus. If you change the size of text, the menu no longer lines up as intended with the backdrop and some menu items become invisible.
Attempts to line up graphical elements with text, or even with each other, almost invariably fail at some point -- namely when the page's rendering changes.
Some designers use pictures of text as a way to control the layout of words. Don't do it: it's a bad idea. First, it introduces bloat. Second, it makes your page less accessible to people who can't see the pictures. And finally, it makes your page harder to read. Frankly, it's unlikely that you can do a better job of rendering text clearly for the user than a browser can; don't waste your time (or mine) trying.
View The cranky user: Oh, the pixel pickle Discussion
Page: 1 2 3 Next Page: Pixels, points, and ems