Participation Inequality - Encouraging More Users to Contribute

Jakob Nielsen's Alertbox for October 9 is now online at:
  > http://www.useit.com/alertbox/participation_inequality.html
		
Summary:
In most online systems, 90% of users are lurkers who never contribute, 9%
of users contribute a little, and 1% of users account for almost all the
action.

------------

User Experience 2006 conference

  > Seattle, October 22-28 
  > London, November 5-11 

29 full-day tutorials, 3 keynote sessions open to all:

  > http://www.nngroup.com/events

Deadline to save $200 on Seattle registration: NEXT FRIDAY, Oct. 20

------------

A few people questioned my use of color-coding in the previous Alertbox
column: it contained a table of task performance, where successful tasks
were colored green whereas error-prone tasks were colored red.

Is this color-coding a violating of accessibility guidelines, considering
that 4% of users are red-green color-blind?

No, because the guideline for color-coding clearly states to use colors
(particularly red and green) as REDUNDANT cues, making sure that the
underlying information is also communicated with additional cues that
don't require color vision.

In my recent article, successful tasks were noted as, for example, 8
correct, whereas failed tasks were noted as, for example, 0 correct. Each
field states the number of users who performed the task correctly.
Somebody who can't see that the first field is green would still recognize
that 8 correct users are better than 0 correct users.

As another example, Yahoo! Finance shows stock quotes that are up in
green, whereas declining stock are red. If this had been the only cue,
then color-blind users would not have been able to tell the movement of
any given stock. But Yahoo! has a good user experience team that knows its
guidelines, so the design correctly uses color as a redundant cue: there's
also an arrow that indicates the direction of the move. (And, of course,
the graphic's ALT text says whether the arrow points "up" or "down", so
that blind users with screen readers get the information as well.)

Guidelines for improving usability for users with disabilities:

  > http://www.nngroup.com/reports/accessibility

------------

GERMAN translation of my new book "Prioritizing Web Usability"
  > http://www.amazon.de/gp/product/3827324483

------------

Natural Language Is Over-Hyped

I could have written the following myself, but Danny Sullivan beat me to
it:

Danny: "This is a rant. It's a rant from over 10 years of watching people
trot out natural language search as the 'killer' solution to the current
state of search."

  > http://blog.searchenginewatch.com/blog/061005-095006

Jakob: Natural language is hyped, because it seems, well, so *natural* as
an interaction technique. It just has two problems:

  1. It doesn't work
  2. Even if it worked, it wouldn't be the best UI for most purposes

Graphical user interfaces are almost always better, because it's faster to
click on something than having to describe it. Plus, of course, most
people aren't good writers.

AI has consumed mountains of research funding in the last forty years
without delivering the goods. I am not opposed to research, though I think
the money would be better spent on studying human behavior and discovering
better ways of fitting dumb tech to people's real needs. But it's naive to
believe that AI will produce something during the next few years that has
eluded armies of top scientists for so long.