The Drooping Tail

It seems like every person who talks about usability has to give their own assessment of Jakob Nielsen. I've never heard anyone who hasn't qualified their analysis of Nielsen with a comment along the lines of, "I don't always agree with what he says" (Steve Krug, Don't Make Me Think!), or  In fact, I think that pledge is a graduation requirement for potential designers and web developers. For those whose graduation predates Nielsen's popularity, they must make that pledge before being invited to any party thrown by web designers (who naturally throw the best parties).

So let me say it. Jakob Nielsen is a Dutch usability rock star. That statement alone should disqualify me from being invited to any parties in the near future.

But I don't agree with everything he says.

It's not that what he says isn't accurate. It's that it isn't always practical, and since context is important to everything we do on the Web, the context of which Nielsen writes isn't always clear in each column he writes.

His latest Alertbox column, "Data Visualization of Web Stats: Logarithmic Charts and the Drooping Tail," for instance, speaks to the inefficiency of going after "The Long Tail" in most websites. Nielsen surmises that unless companies can "con" users to generate their own content, then it would cost too much.

The exception to the context of which Nielsen writes is only "aggregators who get their products from others," and he goes on to say that, "companies who must develop their own [products] are usually served by staying away from the full tail." This context was not provided until the end of the article.

I can't argue with the stats or his conclusion, but I can argue with the word "con." You can't con someone into generating content they don't want to do generate on the Web. And if they are actually doing it, that means you're giving them something they want--providing a service. What they want may change over time as the Web I.Q. of users increases over time or as technology changes, but as long as you're providing them with something they want at that moment (provided it's moral, of course), then you're doing right by them, regardless of how the site makes its money.

Nielsen's negative view of the value of nonsearch advertising on the Web might explain his use of the word "con." In the same column, Nielsen wrote, "I believe Internet advertising is over-hyped and that advertisers are deluding themselves into overpaying." This undoubtedly is very limited in scope and lumps bad Internet advertising in with good Internet advertising, and is heavily influenced by Nielsen's repeated, habitual observation of the "banner blindness" phenomenon in user tests.

To be sure, how much companies should pay for Internet advertising will depend greatly on how effectively we deliver results within this constantly growing and effective medium. But with advertisers only recently getting the message that certain age groups are spending more time online than watching TV, it's far too early to say that it's "over-hyped." - Cam Beck

Comments

[Andjam]

Maybe the word "con" was a little tongue-in-cheek. Then again, Nick Carr talks about Sharecropping the Long Tail.