How people really browse

Links

Submitted by pedrito on January 8, 2001 - 08:56.

Here are some related links: original story

Primary navigation must die. (Bohmann)

Is navigation useful? (Nielsen)

Information foraging theory links.

What people really see when browsing.

Don't make me think! (by Steve Krug, book at Amazon.com, book at fatbrain.com) <= read this if you don't read anything else today.

login or register to post comments

Cite your sources!

Submitted by erika on January 8, 2001 - 15:32.

Pedrito, I liked the information in this article and the links you added at the end... but I found myself asking questions like this: How does Pedrito know how people really navigate? Has he tested them & interviewed them afterwards? Or is he just making an educated guess? Or his he lifting the information from somewhere else? Same sort of questions with the term "link scent" (where does this concept come from?) And the Yahoo thing... how many people do this, besides your mother? You should cite your sources in the body of the article. Make it clear when a concept is original with you, and make it clear when you are getting information from somewhere else. While the link list at the end is helpful, I don't want to have to go through all the links to figure out where you got your ideas from. It should be more clear than that, what originates from your research (and what kind of research you've done) and what you've borrowed from others (with the appropriate references).

login or register to post comments

I've seen people typing URLs into Yahoo

Submitted by themadman on January 8, 2001 - 22:25.

Erika, A lonnnng time ago, I used to frequent an Internet cafe after work. Connectivity here in India at that time was poor and unreliable, and this Internet cafe had a 64K leased line shared among ten computers. The sysadmin there was a good friend of mine and he'd let me use the facilities for free, in return for me helping out newbies who'd visit the cafe and wanted help finding something. This gave me a heck of a lot of insight into how people browsed. Over the course of two years, I stood there observing people and which links they followed, how they found the links etc. I watched people totally miss (as in "couldn't see") links that were, to me, very obvious. That's how I got into the user experience field. What's the point of my rambling? Just that I found that many novice users, who hadn't quite figured even the browser yet, would type the URL into Yahoo's search box (Yahoo would usually be the default home page on the machines)? Why, you ask? You really should read Don Norman's "Design of everyday things" for a good explanation. It's just that the Address bar was higher up on the screen, and most people's attention went directly somewhere to the centre of the screen, where of course they saw a text box, which looked like something they could type an Internet address into. So indeed that's exactly what they did. Of course, this behaviour is only of real novices. Once people figured out that there was such a thing as an address box, they used that. That said, I think a sweeping statement like "But they don't look at the navigation either!" is not accurate. It depends upon the person. There are many of us who'd actually look at a site's navigation and try and figure out the site structure from it, and I suspect most of the members of thelist are like that. :) Just don't create navigation like the one used on http://www.krysys.com. (this is a site designed by a person whose resume my HR department forwarded to me.) Madhu

login or register to post comments

Sources

Submitted by pedrito on January 9, 2001 - 06:49.

You're right ofcourse Erika, I should mention my sources. I'm not sure where I picked up some of the ideas, and to be honest I can't really be bothered to go and look up where for example the link scent idea comes from (it's not an academic paper). I wrote this article after reading a chapter of "don't make me think" (see list of links). It cristallised some of my ideas, as often happens when they're explained by someone else more clearly and more in depth. I did some research on sitemaps, and some general user testing over the past 6 months, and obviously I've read all Jakob's and related stuff. The article is really just opinion, a way of explaining some ideas that helped me think about how people really use the web.

login or register to post comments

more on the Yahoo thing....

Submitted by slyboots on January 9, 2001 - 12:21.

People typing URLs into a search engine is a genuine phenomenon. I was amazed too when I first started finding these things in the access logs for my site. And it's not just Yahoo! of course: it happens with Alta Vista and others. Folks also search search engines by company name. This is less astonishing, but if I'm looking for the site for The Foo Company, I usually try www.foo.com or www.thefoocompany.com first. I realize this doesn't always work, but it seems to work *most* of the time. Then there's the local site search. We worked so hard to provide effective, clear navigation on our site, yet the search is always one of the 20 most frequented pages. So I started paying closer attention to what people were searching for: sometimes it's gibberish that isn't on the site (and thus doesn't show up in the navigation), sometimes it's not. It's always interesting to look at, though. Perdito: nice article. "Link scent" will stick in my mind, for I've never come up with a good name for it.

login or register to post comments

Completely Anecdotal

Submitted by aardvark on January 9, 2001 - 13:47.

I'm sorry, but nothing in this article really proves or disproves anything. Adding a set of links and then a disclaimer that it's just opinion after it's been posted indicates that perhaps the article was submitted before it was fully thought out.

The only two examples you give are anecdotal. Yes, I see people type domain names into Yahoo all the time, and my own site logs support that. But how many people are doing that? By traffic to my site, less than 1%. By observation, well, the only people I really watch are my parents to make sure they don't buy something. So that's 50%, but from a highly skewed sample group. But you provide no numbers, studies, or real-world examples.

The example about visiting one site to get to another is common. But how common? I visit a couple Flash sites often because they maintain links that I otherwise wouldn't. They get updated to add new ones I haven't seen. They are link pages expressly built for that purpose. Otherwise, I rarely go to a site to find a link to another, assuming I saw no reason to bookmark it.

I'd rather see how your own observations perhaps support known rules, or where they may deviate. If it's an opinion piece, label it as such. Don't make assertions without citations or research to back them up, or at least state that it's anecdotal.

For example, I'd love to see how your assertions demonstrate a deviation from Fitts's Law. Microsoft has an article about the web and Fitts's Law (Fitts's UI Law Applied to the Web), and Tog has a good one that's structured like a quiz (A Quiz Designed to Give You Fitts). Obviously, this is a very specific example, but regardless, they are built around discussing a known element in a different way.

login or register to post comments

Submitted by bezalel on January 16, 2001 - 04:41.

I don't know about the others but I found your thoughts preety useful, Pedrito. I can't understand why there is a need of proof for all those things.. I mean you all seem so negative about it. We should be honest with ourselves. Don't wait for interviews and statistics to believe in facts! People often put URLs in search engines, people often push the browser's back button instead of the home link of the site, people don't consider more than one solutions and try the fastest one, people don't read the page they SCAN the page. All common truth I suppose. The meaning of these is to find a way for a better design concept according to the viewer's behaviour. To find a way to communicate. And you don't need statistics to do this.

login or register to post comments

Research has purpose...

Submitted by aardvark on January 16, 2001 - 09:07.

Bezalel, the reason you should make decisions based on data and research is because anecdotal reasoning is often incorrect. There are many cases where users have claimed that a feature of an interface speeds up their use of that interface, but when timed, it slows them down by as much as five times. In UI design and usability, decisions based on something other than facts are often wrong.

If you can take the quiz I mention above and get all the answers right on your first pass, then I'll defer to your judgment. However, that quiz only covers one miniscule facet of usability, Fitts's Law. As such, there's a lot more to learn about usability. To learn about the viewer's behavior, as you suggest, you very much need to observe and generate data before you can make assertions. Otherwise you are just guessing.

login or register to post comments

anecdotal reasoning versus stats

Submitted by pedrito on January 18, 2001 - 03:57.

Aardvark, I agree. Statistics rule (when they don't lie). Measuring rules, guessing sucks. (asking sucks as well).

However, anecdotal evidence can help us think outside of the stats. What I'm looking for is a better understanding of how people really use the web. I suspect it's a lot more complicated than we tend to think.

Another example that made me understand better how people browse is the fact that (yes, I said fact where it's only an observation. sue me) people seem to decide on the meaning and therefore usefulness of page elements by how they look. Typical example: a search box that doesn't look like a search box. (I did do a little bit of research on that.)

login or register to post comments

comment

Submitted by anarchymage on January 21, 2001 - 05:54.

hey all, i read this article and it seems to me that it wasn't a full-blown scientific study? people should appreciate that pedrito actually bothered to write this interesting article! i agree, i have seen many of my teachers (haha) type in urls in the search bar even though they are supposedly teaching us how to surf..... anarchymage

login or register to post comments

typing url's into yahoo

Submitted by paulnattress on January 22, 2001 - 03:06.

It's a very interesting point about people typing url's into the search box on yahoo and the like. Technically, they are following correct instructions. Most people are told that if they want to find something on the internet, they type it into yahoo and it will find it for them.
What we don't realise is that new users don't differentiate between url's and company names. Or maybe it could be that the whole web browser/web page screen is intimidating to them and that there's far too many options (buttons, links to click on, boxes to type stuff into) so they type their url into the most obvious text box (as pointed out above). Maybe a less complicated interface is what's called for.
Paul Nattress
Information Architect
Sage.com

login or register to post comments

More insights on how people search

Submitted by pedrito on January 26, 2001 - 13:33.

Here's an article with more insights on how people search: interview with a lady who analyses search queries for a living.

login or register to post comments

other article

Submitted by pedrito on February 15, 2001 - 08:29.

this article lists some of the research done on how people gather information, and lists implications for webdesign. I don't agree with all the recommendations, but most of them make a lot of sense, and are too often ignored. Seeing the recommnedations in context of the research really made things clear for me.

login or register to post comments

How People Really Browse...

Submitted by Durga on October 6, 2006 - 05:58.

hi I am Durga It's a very interesting point about people typing url's into the search box on yahoo, Google and the like. Technically, they are following correct instructions. Most people are told that if they want to find something on the internet, they type it into yahoo and it will find it for them.

login or register to post comments