Huh? You have to scroll loads on that site at 640x480 and their home page has text rivalling k10k for size.

They're right about 800x600 (users of higher resolutions tend not to maximise their windows as often anyway), but their site isn't a good advert for being cautious with technology.
Apart from anything else, it's considerably slower than it could be. The 75K of HTML for the cluttered home page could be reduced by at least 15-20K by using CSS for text formatting and box borders, and would render quicker. At 5 million page views per month that adds up to a frightening amount of wasted time and money for everyone involved.

Why not use newer technologies to produce a better, faster site? Where's the print style sheet? Why not improve accessibility for search engines and non-graphical browsers? Why not use valid markup? Why isn't it built better than a site from 1997?

If anything, Economist.com demonstrates the perils of being a late adopter.

Posted by: Matt Round on April 23, 2002 04:16 PM

 

OK, I'll bite. They don't use newer technologies because, as the preface states, they are "not willing to ignore any technology that accounts for more than three percent of total site traffic."

Economist.com looks similar in my Netscape 4.7 and IE 5.5. Among sighted users, a site built with legacy code is accessible to a wider audience. Pull out the legacy code and you have an ugly presentation for legacy users. And my unscientific feedback is that visually impaired users have as much trouble using an accessible site as they do one built with legacy code. For the time being, Economist.com and many other companies simply have no reason to abandon legacy users in favor of an ideal.

Posted by: Jack Schonchin on April 23, 2002 05:06 PM

 

"my unscientific feedback is that visually impaired users have as much trouble using an accessible site as they do one built with legacy code"
Having fewer tables helps those whose readers do point them out and with CSS you can often put some unimportant elements at the bottom of the source to avoid them having to go through them first (also helps with search engines). And what do FONT tags do for the visually impaired?

I agree they shouldn't be ignoring 3% of users, hence me suggesting using limited CSS. Netscape 4 users would have a perfectly usable site. They've already got one cosmetic touch in there that doesn't appear in NS4 - a link hover colour.

Perhaps they feel it's worth all the extra HTML to make sure NS4 users get their box borders and v3 browser users are comfortable, but it seems people rarely consider these trade-offs. How many potential visits/page views are lost by the site being slower than it could be?

Posted by: Matt Round on April 24, 2002 04:35 AM

 

Page weight is a large (pun intended) concern at Economist.com, and one we're working to rectify.

When we launched the site in October 2000, the total home page load was around 90K, equivalent to other news sites at the time. Since then our design has grown up and out, and we've become a bit heavier than we expected.

A main design project this spring is to alter our code and shave 10 to 20 percent off our total page weight without altering the design. We are also beginning to integrate style sheets and cleaning up our HTML to improve overall accessibility without lessening legacy system access.

As you are aware, many factors affect a design, and our focus has wandered to other, equally important tasks--like adding and improving content--in recent months. But increasing usability is important and not going unnoticed.

-David Wertheimer
Design director, Economist.com

Posted by: David Wertheimer on April 24, 2002 09:29 AM