This was an ugly instance of the line monster. Here's attempt 2 if it don't
work I'll put the thing on a web page:
1. INTERVIEW: Louis Rosenfeld and Peter Morville on Information
Architecture
We interview the two authors of "Information Architecture for the
World Wide Web, 2nd. Ed." by O'Reilly. The polar bear strikes back.
WebReference: Tell us what's new in the second edition?
Peter Morville: Well, for one thing, the second edition is
heavier than the first. It's more than twice as long (almost
500 pages). We didn't intend to write a longer book, but the
truth is we've learned so much in the past four years, we simply
had more to say. New topics in the second edition include
enterprise information architecture, business strategy and
thesaurus design. We've also added two in-depth case studies (a
corporate intranet and an online community) that show how
information architecture design happens in the real world.
WR: How has information architecture changed in the last four
years?
PM: We've become much better as a community at sharing ideas and
best practices. The SIGIA-L mailing list and the ASIS&T IA Summits
have really brought information architects together. The emphasis
has shifted from designing new sites (pre-1998) to redesigning
existing sites. This has provided the opportunity to develop a
bottom-up methodology that incorporates content analysis, user
testing and controlled vocabulary development.
As information architects, we continue to push beyond traditional
disciplinary boundaries. While Lou and I both have educational
backgrounds in library and information science, we are constantly
drawing lessons from other areas of expertise. Business strategy,
knowledge management, and social network analysis are just some
of the new frontiers.
WR: Yes, in your book you talk about not being afraid to cross
over to the design realm. We all end up wearing a number of hats.
How do you do that, not ruffle too many feathers, and keep the
creative juices flowing?
Lou Rosenfeld: It's not easy, and the best way to deal with it
is for the members of an interdisciplinary design team to accept
that feathers will be ruffled and move on. Really, there's no way
that creative collaboration can take place without friction.
Perhaps the biggest source of tension is language. Programmers
may have a very different definition of "functional requirements"
than designers do. Usability engineers might not have a clue what
"brand equity" means, and marketers might have never heard of "task
analysis." Suddenly, all of these folks get thrown together;
naturally, it's hard for them to communicate. Information
architecture, user experience design, and many of the other
recently minted design fields are really organic efforts to bridge
these communication gaps and develop shared vocabularies that help
us understand each other and work together.
WR: How do you sell information architecture in today's tough
economy?
PM: There's no question the economic downturn has hit the IA
community hard. Given that IA is a relatively new, abstract
discipline this is not surprising. Steve Krug (author of "Don't
Make Me Think") summed up the challenge as that of selling luxury
services in a franks-and-beans economy. My contention is that
while IA may currently be perceived as a luxury service by many
managers and decision makers, IA can have a big impact on both
top and bottom lines. A well-organized e-commerce site will
generate more revenue. A well-structured intranet will save the
company time and money. Our real challenge is to help the broader
business world understand the real economic impact of IA.
My latest approach is to talk about "findability." Thanks to
successful evangelists like Jakob Nielsen, more and more managers
understand that usability is important. We need to take them the
next step, showing that with large web sites and intranets a user's
ability to find what they need is one of the largest factors in
ensuring usability. In short, findability precedes usability.
You can't use what you can't find. Hopefully the second edition
of our book will help make this case to a broader audience.
WR: Can you bottom-line the benefits of IA for us? Nielsen talks
about hundreds of percent savings with usability makeovers. What
can good IA do for ROI? Do you have any figures you can give
before and after from a rearchitecting you did on e-commerce sites?
LR: Some of my colleagues will disagree with me, but it's very
hard, if not impossible, to bottom-line IA. An architecture is a
holistic, integrated system of smaller components. Some of those
components can be measured, but I don't think you can measure the
whole. And just measuring the components can really skew reality.
For example, it's hard to measure how well a search system performs
without also looking at the site's taxonomy; in fact, they're often
difficult to separate (Yahoo! provides a great illustration of
this). The point of IA is to help people find information, and it
takes a whole system to do that, the sum of many parts.
It's extremely hard to measure the cost of finding information;
it's even harder to measure the cost of not finding information.
When we are looking for information, we really don't know what's
out there, and what we're missing. Sometimes ignorance is bliss,
but if your efforts to find medical information to help a sick
parent are unsuccessful, you won't feel that way.
There are just a few decent case studies out there that express
the monetary benefits of doing IA work (for example, read "You
Think Tomaytoes, I Think Tomahtoes" which appeared in the April
1999 issue of CIO magazine
(http://www.cio.com/archive/webbusiness/040199_nort.html)), but
take a close look and you'll see that the numbers are predictive.
They estimate future savings or increased revenue; they don't
actually prove it. So realize that when you see ROI numbers for
IA, you're really looking at soft numbers.
WR: You talk about the importance of metadata, thesauri, and
controlled vocabularies for organization of content. How
important are these content tagging devices? Seen any out of
control vocabularies?
LR: Most vocabularies, spoken and otherwise, are out of control,
despite the best efforts of dictionary editors around the world.
This is also true of the sets of terms used for labeling
navigation options on most large sites. For example, what I call
"home," you call "main page." And while my intranet portal has
a link called "home" that gets back to the portal's main page,
sub-sites within that portal may also use "home" to link to their
main pages. This all gets confusing, and illustrates why it's
useful to standardize and control vocabularies in certain situations.
Tagging content with terms from indexing vocabularies and thesauri
can be really helpful for searchers. For example, if I search a
recipe site for "snacks," tagging would allow me to retrieve recipes
for chicken wings and nachos, even if those recipes' files don't
include the word "snack." Just keep in mind that such vocabularies
can be prohibitively expensive to develop, maintain, and apply to
content.
WR: Explain how indexing thesauri enable browsable indexes?
PM: By defining preferred terms (or acceptable metadata values)
and tagging documents with those terms, you lay the foundation for
automatically generating useful browsable indexes. If you simply
let indexers apply uncontrolled keyword metadata tags to documents,
you can still technically create a browsable index, but similar
documents will be "dispersed at the point of indexing" because
different indexers used different keywords to describe the same
thing.
For example, a product index on Microsoft.com might list some key
documents under A for Access and others under M for Microsoft
Access. A controlled vocabulary will specify that "Microsoft Access"
is the preferred term, and a good thesaurus will provide for a "see"
reference in the A section of the browsable index (Access, see
Microsoft Access).
For a good example of a browsable A-Z product index, see
http://peoplesoft.com/corp/en/indices/prod_index.asp
WR: Reading your book, it sounds a bit like doing database
normalization for a web site. Break everything down and put it
back together again in a logical way. A fair analogy to IA?
LR: Sure, that works. But let's just use this as an analogy;
data is much less complex than the semi-structured text that
constitutes much of our sites' content. That's the stuff
information architects are typically grappling with.
Speaking of analogies, you could also say that IA is a lot like
information retrieval, or designing an application's interface,
or developing a corporate identity, or modeling a corporate
organizational structure, or like many other things that involve
design and organization. IA is like all of these things at one
point or another, and it shouldn't be surprising: it's an
interdisciplinary field, and besides, just about everything we
create has a structure and contains information.
WR: What are the top ten IA mistakes?
LR: I'll limit it to seven mistakes (my lucky number):
1. Letting software drive design.
Too often, purchasing a new search engine or some other IA-related
tool is the excuse for taking another look at a site's information
architecture. This is completely backwards; the architecture
should be informed by content and the needs of both users and the
sponsoring organization; the architecture's functional requirements
should in turn drive technology purchasing decisions.
2. Providing too few ways of navigating.
Not everyone wants to search your site, so invest some effort in
enabling browsing. Not everyone's a regular user of your site, so
invest some effort in developing a table of contents, or a guide
to the site's content. And so on. Users' information needs and
seeking behaviors often vary enough to require alternative means
of navigating.
3. Providing too many ways to navigate.
Conversely, it's tempting to be all things to all people. But
it's also expensive to develop and maintain all of these
alternative schemes. They can really clutter up the architecture
and actually make it harder to find information. Follow the 80/20
Rule: choose the few (20%) best architectural components that will
serve the major audiences (80% of all users). Better than going
broke trying to serve those last 20%.
4. Trying to do everything at once.
Large, complex sites-corporate intranets or portals for example-
involve huge amounts of content, diverse user audiences, and enough
politics to turn a congressman's stomach. You can't re-architect
the whole thing-all two hundred departmental sub-site-in six months
like your boss asked. Tell him he's wrong, or start looking for
a new job.
5. Believing that there's a "right" information architecture.
An information system is always changing. Content grows, gets
stale, while users become more sophisticated. Their interests
change, as do the business goals and practices of whatever entity
is footing the bill for the site. With so many moving targets,
it's impossible to perfect an information architecture; instead
your goal should be continuous improvement over the long haul.
6. Practicing imbalanced information architecture.
Your IA research and design must be informed by three areas of
investigation: content, users, and business context. If you
leave one out, this three-legged stool collapses. Try to use
research and testing techniques that address each area, and staff
your team with expertise in each area.
7. Following gurus.
Our field is too new and too unformed for anyone to know
everything. And it will continue to change; I doubt that our
knowledge of IA will ever keep pace with the problem space. If
you're getting IA advice, avoid people who claim to have all the
answers. The first words out of a smart information architect's
mouth should always be "it depends."
Information Architecture for the World Wide Web, 2nd. Ed.
By Louis Rosenfeld and Peter Morville
O'Reilly & Associates, $39.95
ISBN: 0596000359
http://books.internet.com/books/0596000359
# # #
About the authors: Peter Morville and Louis Rosenfeld both have
library of science degrees from the University of Michigan, and
collaborated on the first edition of "Information Architecture
for the World Wide Web," aka the polar bear book. Peter is
president of Semantic Studios, an information architecture and
strategy consultancy at http://www.semanticstudios.com . Lou is
principal of Louis Rosenfeld LLC, an Information Architecture
consulting firm at http://www.louisrosenfeld.com. Lou is
currently part of the Nielsen Norman Group's world usability
tour.
--
Larry C. Lyons
ColdFusion/Web Developer
Certified Advanced ColdFusion 5 Developer
EBStor.com
8870 Rixlew Lane, Suite 204
Manassas, Virginia 20109-3795
tel: (703) 393-7930
fax: (703) 393-2659
Web: http://www.ebstor.com
email: [EMAIL PROTECTED]
Chaos, panic, and disorder - my work here is done.
--
> -----Original Message-----
> From: Larry Lyons
> Sent: Thursday, August 29, 2002 2:07 PM
> To: CF-Community
> Subject: FW: WebReference Update: Interview: Lou Rosenfeld and Peter
> Morvi lle on Information Architecture
>
>
> Interesting interview with the writers of "Information
> Architecture for the
> World Wide Web"
>
> larry
>
> --
> Larry C. Lyons
> ColdFusion/Web Developer
> Certified Advanced ColdFusion 5 Developer
> EBStor.com
> 8870 Rixlew Lane, Suite 204
> Manassas, Virginia 20109-3795
> tel: (703) 393-7930
> fax: (703) 393-2659
> Web: http://www.ebstor.com
> email: [EMAIL PROTECTED]
> Chaos, panic, and disorder - my work here is done.
> --
>
> -----Original Message-----
> From: Webreference Update [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, August 29, 2002 12:00 PM
> To: Larry Lyons
> Subject: WebReference Update: Interview: Lou Rosenfeld and Peter
> Morville on Information Architecture
>
>
> ((((((((((((((((( WEBREFERENCE UPDATE NEWSLETTER )))))))))))))))))
> August 29, 2002
>
> ___________________________ Sponsors ________________________________
>
> This newsletter sponsored by:
> Instant Messaging Planet Fall 2002 Conference & Expo
> VoiceXML Planet Fall 2002 Conference & Expo-9/26-27, Boston
>
>
______________________________________________________________________
Structure your ColdFusion code with Fusebox. Get the official book at
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
