In october 2000, as IT specialist of DW in a multinational company, I
had the opportunity to hear Dr Hackathorn speaking about Web Farming.
I thought it was a little bit early to already speak about THE new
big thing but now... I must conceed that it is coming rapidly!

The WebFarming group is growing. Messages posted from all over the
world. I believe this won't take the 2 or 3 years i thought it would
take.

As I decided to specialize in these new Information Systems (DW,
Knowledge Management, WebFarming), I perhaps can help you by giving
my personal opinion.

Your questions where:
1. The validity and Reliability of Web Content
2. Web Searching method (which is productive)
3. Integration the WF system with the business intelligent system.
Are there any challenge out of this 3 ?

I think the challenge of WebFarming is not different of the one of
any "InfoQuest"...

This challenge has nothing to do with technology. Did you ever tried
to transfer datas from HTML to DataBases? very easy, in fact. I tried
with Yahoo Meteo Services to study soft-drinks sales related to the
temperature. The VB program is only 100 lines long !! And with XML,
it becomes more easier. Some companies already offers daily XML-files
on their websites or via email.

As you study Knowledge Management, you discover a very interesting
concept : the 3 phases of knowledge. that is something that can help
us to answer the questions of WebFarming.

1st phase: You learn from a person. You have to build a network of
relationships with persons or "social entities" that you know you can
trust. Pay attention: these trustable persons are not always the same
depending of the subject. For every subject, you have to build an
expertise map. This map will help you to trust the info you find on
the internet, analysing the authors & references. There is nothing
here that can be automated.

2nd phase: You learn from a document. Internet content is very
volatile, I agree but, based upon your expertise map, some documents
can very rapidly be considered as "trustable". Here, you cand find
some tools on the market that can help you. I didn't studied the
subject in deep yet but I would suggest to base the selection on the
authors & references, at least in a first phase. Dr Hackathorn
explains that very brillantly in his book, so I won't speak of what I
don't know.

3rd phase: You learn from a program. Here we speak about full
automatization. In this case, it would be a real challenge to build a
program that automatically load the DataWarehouse with all document
datas. But if you think about very specific questions, it becomes
really easy (as I did for Yahoo Meteo Services).

The big challenge is for me to trust your sources and this work has
to be done manually by specialists of the subjects. For the moment,
we are at the beginning of the utilization of Internet Knowledge.
I would just say : don't trust documents, trust sources : people &
organizations.

Bye
Fredt