Boxes and Arrows: Comments by Jeff Seager

Jeff Seager — Thu, 09 Aug 2007 15:28:03 GMT

As you say, content is key. Readers will scroll if they find useful and relevant content in the portion of the page that is first revealed to them. That’s all. No magical formulas needed. This is basic human nature.
To extend the analogy with print media, though, the first-visible segment of the opening screen of any page should be where you place the strongest emphasis and put your best stuff (and establish or re-confirm corporate identity). Navigation and search elements (if used) should be present there and easily interpreted. It’s not that people WILL not scroll, but they MAY not if you haven’t prepared a suitably intuitive path for them.
Tracking data may be useful in aggregate, but we should be careful how we interpret it. On a site that doesn’t get much traffic, you may not have enough data to draw any reasonable conclusions. Some people move on simply because they’ve hit a page or a site that was irrelevant to their needs. That’s no fault of the design.
Finally, we’re not just designers. We’re also users or consumers of this communications medium. I think we’d all be better off if we’d set aside a lot of the “conventional wisdom” and honor our own direct experience of functional and dysfunctional design. As Milissa wisely suggests, some conventional thinking doesn’t translate that well to this medium.

Jeff Seager — Thu, 09 Aug 2007 13:53:37 GMT

Nice work, Joe. I think some people are desperate to see this as a literal toolkit, when in fact it goes far beyond that. What you’ve done here (and very well) establishes a way of thinking about information architecture that transcends all tools, but could integrate well with any tools. It’s a conceptual model, an extensible and flexible structural framework that would be useful in any programming environment.
In the end, our aim is always to create order out of chaos. This conceptual framework you suggest is very helpful for that.

Jeff Seager — Tue, 19 Aug 2008 20:54:47 GMT

“Companies should create rich and immersive website experiences, drawing from some of the techniques for game design to build brand affinity and differentiate their sites.”
I have trouble with “should” statements, unless the mandate is so clear as to be unarguable. This one is dubious.
When an immersive experience is needed, go for it! But in many cases, a company will benefit greatly from simply getting to the point. The point of this article supports an earlier model of the Web that insisted on measuring the success of a website by the amount of time spent there. More time may in fact mean the user is “stuck” and confused about what she needs to do to get what she wants.
A better measure is this: Did the user get what he wanted? Was the transaction profitable for all concerned? A game is not a good model for information access or commerce.
Also consider accessibility, which is a requirement for many government websites and a darned good idea for many others. Rich, immersive experiences are fine when progressive enhancement is practiced in their development, but many many people are unfamiliar with progressive enhancement techniques. That locks out a large number of people, when you figure the global impact, and in the long run it’s altogether a bad thing to encourage rich immersive experiences for their own sake.

Jeff Seager — Sat, 07 Aug 2010 04:20:17 GMT

For accessibility reasons, regardless of actual placement, it’s good to have the primary navigation close to the top of the HTML file because it’s one of the first things users will seek. Usually for me that means rendering it in the header or in a left-hand column (though I could use CSS to render it in the right-hand column, the logic of that escapes me).

When you disable scripts and CSS in your browser, or when you use a text-only browser like Lynx, the navigation should be among the first elements presented because it’s almost always going to be one of the most useful elements.

I’m an accessibility freak, though, so this is more important to me than it is to some. When blind users land on a page, they will generally need 1) An H1 heading that gives reassurance that they are where they want to be, 2) Navigation that is functional without javascript or other client-side assistance, and 3) Semantic structure. Because those elements benefit sighted users too, I would build this way for everyone.

Jeff Seager — Tue, 10 Aug 2010 11:30:42 GMT

I think it’s always useful when building a website to stop and ask “What can I subtract from this without losing anything important?”

I hope the trend moves toward fewer words that say more, and fewer useless elements on a page—not just more white space.