Is "Pervasive Nav" Dead?
Should Typical Tab/Menu Navigation Be Abandoned
in Favor of Trigger Words?
Henrik Olsen, on his www.guuui.com site
reports on a recent spool (Spool?)
that pervasive navigation—conventional top tab bars and/or multi-level
menus on the top or left of every page—is a proven failure whose time
must come to an end if we are to improve website search success rates. I think
the whole premise misses the point, throwing out the baby with the bathwater.
In a provocative article, on Navigation
Blindness, you can read that many pundits blame poor
web design on the weaknesses of "pervasive navigation. Pervasive
navigation refers to nav links that are on almost all
pages. They might be two-level tabs like Amazon, or Windows-like
dropdown/flyout menus, or semi-static, two-level menus on
the left like many sites use.
I believe this issue is perhaps the biggest one
in content-centric web
The recent sentiment stems from the many undisputed problems
of current pervasive nav (which though pervasive, is rarely
thorough). We are told that the solution is to cater to those
crazy, "goal-driven" visitors (is there some other kind?) with
trigger words or some other punditry flavor-of-the-month.
Here's my counter argument:
- I don't deny that many users skip the pervasive
nav. Much structured nav is a hodgepodge,
so why shouldn't they?
- The slow (but not invisible) emergence
of standards is partly a
deterrent. If all web sites used a fully RAM based collapsible
control from the early days, the whole mess would not
exist. (Modem speeds
prevented this, but just wait.) I recently updated another
of my sites, RSIRescue.com to such a design, based on the
popular techwriter's tool, RoboHelp.
- The inconsistent use of
RAM (not putting the full nav) in memory makes a
painful game of using persistent nav.
- The fact that body content changes
with any click might be the real demon
here. This "animation" draws the eye away from
what might have been a great
nav system. Users then get transfixed and seduced by prose
right-side feature callouts.
- The home page of http://www.analog.com is
offered up as proof that "trigger words" are the
solution and pervasive nav is dead. I say that nothing could
be further from the truth: it is proof that readers need
exactly what they've needed for 1500 years: a good table
of contents, in
close to a single glance as possible. The "expansion" of the level-two
headings right onto the page represents nothing more than readers' desperate
need for examples(!), clarifying the otherwise meaninglessness of the
level-one headings to the uninitiated. (For concepts, examples are king.)
This nonsense about pervasive nav being
- totally misses the issue,
- perpetuates a habit of throwing out the baby with the bathwater,
something the technologies love to do,
- and ignores the lessons of traditional information tools.
What will be
miraculous revelation/revolution? The epipany that web sites need intelligently-crafted
indexes... along with the obligatory chant, "Search
Must Die"? Duh.
What should you do on your site? Deploy a good
- table of contents,
- full-text search,
(yes they most certainly are useful)
These are the five fundamental "orders" of information
and they should be on all sites. Other orders include "popularity" (what's
hot), importance, and special orders for your own type of
data. For an intranet, the "revision history" is
a running log of content changes in date order, like my home
page and it should predominate the body of the page.
The fact that there is so much poor design out there and that
half of the current generation is
struggling to fathom the depth of
is no reason to adopt the sort of anecdotal browsing
implied as a direction. And the Analog site is far from
anecdotal. It is the essence of a RAM-based table of contents...
more than a good Site Map as a home page.