BKIAFWWW.RVW   980327

"Information Architecture", Louis Rosenfeld/Peter Morville, 1998,
1-56592-282-4, U$24.95/C$35.95
%A   Louis Rosenfeld lou@argus-inc.com
%A   Peter Morville morville@argus-inc.com
%C   103 Morris Street, Suite A, Sebastopol, CA   95472
%D   1998
%G   1-56592-282-4
%I   O'Reilly & Associates, Inc.
%O   U$24.95/C$35.95 800-998-9938 fax: 707-829-0104 nuts@ora.com
%P   226 p.
%T   "Information Architecture: For the World Wide Web"

I would have no argument with the authors on their initial point.  The
vast majority of Web sites; so many that one can legitimately say
almost all; are badly designed, difficult to navigate, and much less
useful than they could be.  Most of those who build Web sites do so
without a definite vision or a consistent plan.

Further, I would agree that librarians, trained and experienced in
organizing access to information, are the one group that have been
underutilized in the explosive growth of the Web itself.  Techies,
graphicists, managers, and marketroids have all had input to the
proliferating pages, but precious few have seriously considered how
users (surfers, clients, browsers, customers: however you want to name
them) were going to use a given site.

Unfortunately, the book still has problems.  I wasn't even sure I
wanted to start it, with a title like "Information Architecture."  Was
this to be another theoretical tome on data warehousing, or the bland
platitudes about management information systems that business
consultants produce with tedious regularity?  Once into the text, it
was a relief to find that there was a specific topic to be dealt with. 
However, that relief was short-lived given the immediate, and abiding,
tone of pomposity that pervades this volume.  That the second sentence
in the book refers to information handed down from on high, engraved
on stone tablets, is ironically apt.  What is involved here, after
all, is Web site design, not rocket science.

While chapter one is supposed to address what makes a Web site work,
the bulk of the text belabours the obvious fact that most sites don't
work.  The points to be considered in determining what people like and
don't like in a site are a start, but are not exceptionally helpful
and definitely aren't exhaustive.  Given that the concept of the
information architect is so central to the book, it is surprising that
the term is not defined until the second last page of chapter two,
following a rather pedestrian review of the various skills needed for
Web site creation and maintenance.  Chapter three is both useful,
dealing with organizational schemes and structures, and good, in its
explanations of them.  The tips on navigation design, in chapter four,
don't present any problems, but don't present much help either. 
Chapter five's discussion of labeling is overly long, and not very
clear.  It is the only discussion of the topic that I have seen, but
the concept is not presented very clearly, and those who need it most
will probably be the ones to give up on it first.  Search engines are
probably going to be obtained off the shelf, so the theory in chapter
six will have limited use.  Chapters seven and eight look at the
process of designing a site, given all the parties that must be
involved.  The material tends to the "motherhood" type of statements
that consultant's books fall prey to: good advice if you can get them
to work, but not terribly helpful for actually making anything work in
the real world.  The advice on production in chapter nine is a bit
more realistic, but fairly common.  A case study concludes the book in
chapter ten.

Now, I will definitely grant the authors yet one more point.  They
deal with the design of the site, rather than merely the page.  All
too many books on the topic (and I have reviewed a depressing number)
are concerned with fonts, backgrounds, text colour, imagemaps, and
other factors that have only the slightest bearing on accessing
information.  Rosenfeld and Morville, to their credit, do not get
bogged down in this morass of minutiae.  Still, while the points they
make are valid, and apposite to the current Web environment, the
advice in the book is of limited utility to those who must actually
produce pages.  It *is* worth getting the text to avoid a $20,000
consult.  And maybe you can get the CEO to read it and realize that
the job can't be done in a single afternoon.

Having said all that, I do think the book should be read.  It is quite
singular in its approach, and it has many unique points to make.  In
response to the initial draft, one author pointed out an important
point: the book is based on material that was used in live
presentations, and the statements that give the book its pontifical
tone in print are effective and humorous on stage.  The authors'
intentions are good.  It is unfortunate that they have paved the way
to a book where the most important aspects are lost because the
content requires so much of the reader.

copyright Robert M. Slade, 1998   BKIAFWWW.RVW   980327