ImageMaps


ALT-tags are needed for each and every area. Some browsers, including
lynx, actually take you to a different screen when you enter an image
map. It gives you a list of links on the map. Now, if you make each of
those links have an ALT-tag, it makes for a much more readable list.

Plus, for eg, some sites can actually make very good use of this. Imagine
a site where it has maps of pretty much the entire world for download,
piecemeal. You put up an image of the entire earth, and make it an
imagemap. You can use ALT-tags to add much value here: You could, say,
give the name of the file in there [as, perhaps, e010n20.tar.gz, or
similar], and it's size. In image browsers, as you move the mouse over
that particular section of globe, it'll give that information somewhere
[eg, Galeon puts it on the status bar].

Something else to learn about imagemaps: They're not necessary a lot of
the time. Three points:

a) A lot of people use the fact that you can put mouseovers in
imagemaps. Don't get me wrong, mouseovers are a Good Thing (TM). Extra
UI feedback on what's about to happen is good. Lighting up text is double
double good. But have you considered that fact that IT's JUST
TEXT. Perhaps it'd be more professional, easier, and require less
bandwidth if you merely put some text in there?

b) If you're using a mouseover for each area, you need to download an
extra image for each and every area in the image. If you're precaching
the images [good], it'll take an extra {period of time} for someone to be
able to see your webpage [bad]. Say you have 5 areas in the imagemap. Each
of those has a mouseover. You need to download 6 images of the entire
map. Instead of 10 images that're 1/5 the size of the map. [each one
for highlighted and not highlighted].

Imagine each piece of image/area takes 3k:
5 [areas] * 3k [size] * 6 [number needed] = 90k
10 [separate images] * 3k [size] = 30k

You've just cut the size of your image requirements by a third. And
that's a geometric progression, kids. As the number of areas goes up,
you're /multiplying/ by the size of the image, not adding.

c) Browsers like Lynx require you to go into another screen to read
them. If they're site navigation, you've just made some go through an
extra click/keypress just to navigate your site. Think about it for
a moment. All you had to do was split the image up into the areas and
it would've worked just fine first time. If I can make multiple
images look like they're one big image, I'm pretty damn certain you, a
"professional web designer" can.

d) Some modern browsers don't support imagemaps. For example, the
only online browser for the Palm is Eudora. And it doesn't support
imagemaps. You know these new-generation phones that let you browse
webbily? Well, they can't use them either.