Index/TOC Needed in HTML &/or Browser-Based Help? (take II)

[Beth Agnew]

I'm strongly with Geoff on this one. This is not the time to try to 
change user behavior and get them to adapt to a new way of finding 
information. People are increasingly overburdened, and especially when 
they are looking for help they need things to be easy and familiar. If 
you've used any of the "tag clouds" on sites such as Technorati.com/tags 
and Squidoo.com you know that keywords are not always applied in a way 
that the user finds intuitive. I can envision the day when the most 
sought for topics are indicated by prominence in the search list, but 
we're not quite there yet. I've always advocated redundancy in an index. 
Much better to have Print: documents and Documents: printing than to 
guess which one the user will go looking for. Indeed, if you want to 
help people adapt to a new way of handling information, provide both the 
old and new together so they can make the transition in a stress-less 
way. Our goal is still to design technology that adapts to the way 
people work, not force people to adapt to the way technology works.

Geoff Hart wrote:
> Taxonomies are also designed by the author, and thus have the 
> potential to be equally arbitrary. To work well, both require 
> audience-centered design: you can't impose an effective structure or 
> even a taxonomy if you don't think of the information from the user's 
> perspective. That insight into the user is what makes a TOC effective. 
> Its absence is what undermines a TOC, and what will undermine 
> taxonomies too.  ...
>
> That being the case, *why muck with what works*? Call it a table of 
> contents, and provide tools that facilitate the grouping of 
> information into appropriate categories ("taxonomies").
>
> <<And there wasn't an index tab. However, one of the MAML elements was 
> for keywords, and these keywords would be used to beef up the search. 
> There were also features like "Best Bet Keyword", which would push a 
> topic toward the top of a search list.>>
>
> Really. Dumb. Decision. Indexes work well because they organize 
> information conceptually and by context. A good index provides entries 
> such as "Print: troubleshooting font problems" beside "Print: 
> configuring a printer" so you can compare the two entries and guess 
> which one is most likely to answer questions. The keywords themselves 
> are merely the superficial evidence of the thought process that went 
> into defining and presenting these alternatives.
>
> Search engines cannot provide that context unless it's designed into 
> the text being searched; the keywords must not only be present, but 
> there must be meta-information that accompanies them to provide the 
> context for the keyword. Currently, that is the single biggest flaw in 
> all the search engines I've used. Defining better keywords won't by 
> itself help improve searches: the same time required to choose and 
> test keywords would be better spent creating an index.
>
> If you know anyone involved in the design aspect of this part of the 
> project, please pass that distinction along. It's not trivial, and 
> it's very important to users. 
-- 
Beth Agnew
http://www.301url.com/podcasting

Professor, Technical Communication
Seneca College of Applied Arts & Technology
Toronto, ON 416.491.5050 x3133
http://www.tinyurl.com/83u5u