I understand that browser loyalty is fierce. But alienating
part of your audience because you've written code that is optimized
for one browser or another is bad all around - both from a business
and usability standpoint. Worse yet, people have fallen into
the trap of watching browser statistics websites, taking a look
at which browser is in the lead, and designing their site for
it. The problem with this is that no one set of statistics is
going to tell you exactly what the entire web is using. People
love to quote statistics and swear by them... here's why you
shouldn't.
Take a look at the April
2005 statistics from TheCounter.com. This is an unscientific
sampling of websites which use a counter from TheCounter.com.
When the page loads a counter, it looks at what browser the
user has and records it.
Now, according to this set of statistics, Internet Explorer
has a strong lead in the browser war and Netscape/Mozilla has
about a 5% browser share. This isn't true for the entire world,
but some people will embrace it as so. What I found even more
interesting though were the statistics toward the bottom. Last
month, over 6,000 people were recorded using Netscape 3.0, and
about 700 were using Netscape 2.0 or lower! People think that
they don't need to design sites that can be used in these older
browsers because "nobody's using them." Really? What
if one of those 6,000 (or more) people using an old browser
is a potential customer of yours?
The "top" browser could change at any time, but people
are going to keep using what's installed on their machines.
As developers, we care about having the latest and greatest
browser technology -- but most people don't. You're only kidding
yourself if you believe that they do.
Here's a real-life example for you. A few years ago I designed
a website for a nonprofit organization, a snowmobile club. Since
it was a pet project site for a family member, I decided to
have some fun with it and create a "snowing" effect
with what was at the time bleeding-edge browser layer technology.
It looked great in the browsers I tested it with, and to the
best of my knowledge I coded it in a way that people with older
browsers would just see the site without the snow.
As soon as the site went live, everyone in the organization
was complaining. For some people, the snowing effect evidently
was causing their browsers to reload the page over and over,
so that it would load a few lines, reload completely, load a
few lines, reload completely. What a mess. Other people saw
the snow fine, but as the snow fell it caused their pages to
scroll down with the snow moving down the page. What browsers
were these people using? Some were using old versions of AOL.
Some were using versions of Netscape and IE even older than
the ones I tested the code on. While the snow worked fine in
most 4.0-or-better versions of Netscape, the 4.08 version evidently
handled things differently and caused the incessant page reloading.
IE 5 made the page scroll down automatically to follow the snow.
Needless to say, we quickly removed this effect. While it looked
great and passed our in-house usability testing, it simply did
not work for the people who were using it.
In our efforts to be cutting-edge, we often forget the feelings
of the audience that we're trying to reach. I've often said
that most people in this business are much more talented than
their work will show. We simply can't utilize all of the new
technology we possess until the users catch up and become ready
for it.
So what does this mean for us? We have to adhere to a standard
that works in both browsers. I used to be a die-hard when it
came to adhering only to the printed standards. Lately I've
become a bit more flexible. As long as an HTML tag degrades
gracefully in the opposing browser, I'll use it. IE supports
a tag for fixing the background image as a watermark -- this
is a nice effect, and because Netscape passes over the tag without
any problems, so it is "safe" to use. HTML purists
will disagree, I know, but if it doesn't cause problems for
the opposing browser, I will use it. If, however, a tag causes
problems for any other browser, I don't use it.
Unfortunately, with the ease and availability of WYSIWYG editors,
the problem of noncompliant code is becoming even more prevalent.
People who scratch their head at the sight of bare-bones HTML
code are building web pages. They don't know, or don't care,
that the pages they're building are not going to work for everyone.
Ideally, everyone who wanted to design sites would spend time
actually learning to code (novel, I know.) When their sites
break, they would know how to fix them. They could spot noncompliant
code and remove it before it ever hit the web.If you aren't
familiar enough with HTML code to troubleshoot it yourself,
the best solution for this is to validate your code. Run your
code through a good validator, or use an editor with a validator
built in (Dreamweaver's is very good.) Iif people are familiar
with the properties of the tags they're using, they'll know
whether or not a piece of nonstandard code is actually going
to cause a problem.
Summary:
* Never design your site for one particular browser.
* Test your site in many different browsers and versions.
* Validate your code, and become familiar with the properties
of nonstandard tags.
Webmaster's Corner is written biweekly by Jill
Cataldo, Senior Web Developer for NewRamp.com. She has been
working in web development since 1996. Articles are copyright
©2005 NewRamp and may not be reproduced without permission.
Visit the
Webmaster's Corner Articles Archive
NewRamp.com offers web
hosting and technical support that is second to none. We
guarantee 99.99% uptime for your site!
