KnowledgeBase software / search engine software?

Subject: KnowledgeBase software / search engine software?
From: "Hart, Geoff" <Geoff-H -at- MTL -dot- FERIC -dot- CA>
To: "TECHWR-L" <techwr-l -at- lists -dot- raycomm -dot- com>
Date: Wed, 23 Apr 2003 08:56:40 -0400

Nadine Underwood reports: <<We are in the processing of publishing a
KnowledgeBase for our customers to use over the internet. The idea being
that they log on to the internet and search the KnowledgeBase for a
particular problem/query before logging a call with the Support Desk.
However, we are having problems with our KnowledgeBase not returning any
appropriate knowledge. We have basically constructed our knowledge to be
separate documents under a folder structure on our server as opposed to
holding all knowledge in a standard (NFO) database.>>

Two things are probably going wrong here. First, the folder structure may
not represent the actual hierarchy of the information as users perceive it.
If that's the case, you must reorder the folders and subfolders to match the
manner in which people expect to find them; for example, you may have
separate folders at the same level such as "printing" and "problems", when
in fact browsers expect to find "problems" under "printing" (i.e., one
problem-solving directory per topic).

One advantage of collections of online documents is that even when you only
have a single Web page for "solving printer problems", you can point to this
page from as many locations as seems appropriate. So you could have links to
this page under "problems", "printing", "troubleshooting", "stop bothering
tech support and read the fine Web page", and whatever else you expect
readers to use as search terms.

Second, the most common problem with knowledgebases is either the lack of an
index or the creation of an inadequate index. A good index trumps a search
engine every time, because in creating an index, the indexer provides
context (through the choice of headings, subheadings, and cross-references)
and synonyms (alternative ways to express the same concept) that search
engines cannot provide. Moreover, the indexer creates these terms based on
an understanding of how your audience thinks of the concepts expressed in
each topic; software can't yet do that, and probably won't be able to do so
for a good number of years.

Any good knowledgebase software should let you include several index
keywords and cross-references for each topic. (That's a tautologous
criterion, by the way: if the software doesn't do this, I don't consider it
a knowledgebase. Your definition may vary.) If you're using straight HTML
authoring tools, consider third-party indexing tools such as those from
DevaTools ( or HTML Indexer (

<<Has anyone implemented a similar solution with just a search engine?>>

Search engines rarely work as well as their designers hope. Each one
requires users to master a whole new idiosyncratic approach to searching,
and the vast majority can only find words that are physically present in a
document; thus, they don't correct for typos while entering the search term,
fail to account for users who misunderstand abstruse concepts such as
Boolean search definitions, and provide no means of compensating for the
user's lack of knowledge of your jargon (missing synonyms). Moreover, few
accept stemming (e.g., finding topics containing "print" if you enter
"printing" as a keyword). If you do use a search engine, make sure it's not
the only tool you provide for finding relevant topics.

--Geoff Hart, geoff-h -at- mtl -dot- feric -dot- ca
Forest Engineering Research Institute of Canada
580 boul. St-Jean
Pointe-Claire, Que., H9R 3J9 Canada

"If you were a member of Jesse James's band and people asked you what you
were, you wouldn't say, 'Well, I'm a desperado.' You'd say something like,
'I work in banks,' or 'I've done some railroad work.' It took me a long time
just to say, 'I'm a writer.' It's really embarrassing."--Roy Blount, Jr.

Last chance to order RoboHelp X3 and receive a $100 mail-in rebate,
PLUS free RoboScreenCapture and WebHelp Merge Module. Offer expires
4/30/03! Order here:

Help celebrate TECHWR-L's 10th Anniversary starting this month!
Check out the contests at
Happy birthday to you, happy birthday to you, happy birthday TECHWR-L....

You are currently subscribed to techwr-l as:
archive -at- raycomm -dot- com
To unsubscribe send a blank email to leave-techwr-l-obscured -at- lists -dot- raycomm -dot- com
Send administrative questions to ejray -at- raycomm -dot- com -dot- Visit for more resources and info.

Previous by Author: Alternative to the word module?
Next by Author: RE: Hostility towards STC
Previous by Thread: Re: Newbie Question: DBMS & IBM Mainframe Apps
Next by Thread: Re: Writers as SME's

What this post helpful? Share it with friends and colleagues:

Sponsored Ads