Ummm have you fellows ever worked on a data warehouse,
that is, worked with technology such as Cognos or Business
provide? tried to avoid a data warehouse by
using direct feeds from the heterogeneous databases?
The notion of near real time monitoring to feed dashboards
and using ontological layers to unite heterogeneous databases
ain't exactly new turf to be invented. What has changed is
the application of web services to reduce the protocols
on the wire, and the ever present seldom solved question
of warehousing vs using the real time source (which actually
is solved but not admitted by those trying to sell pure
web-service plays and justify some dubious patent applications).
In these systems, one is aware as the business objects are
competent. That is where the difficult, tedious, all-politics-is-local
challenges are. The reason for an upper-level abstract SA
followed by a layer of enterprise SAs is to homogenize these
as much as is possible and affordable. What is long term
is replacing the legagy. Thus, founder's effect.
len
Message
From: Klaus Backert [mailto:Klaus.Backert (AT) t-online (DOT) de]
Sent: Friday, January 27, 2006 2:31 PM
To: Nathan Young -X (natyoung - Artizen at Cisco)
To often we define expert as "someone who knows so much that they no
longer need to take input" what is needed is a definition:
"Someone who is constantly changing their criteria for what kind of
input to pay most attention to and can effectively evaluate those
criteria"
Physicists are a good example of experts, who are most of the time
consistently correct in their statements, as I think. There are
internationally installed, accepted and constant protocols for
becoming and being an expert in physics. Fraud is detected in the
long run at least. Its the same in technical disciplines based on
physics. Here we have areas, where there is only a small difference
between theory and practice - again: in the long run at least.
The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
initiative of ASIS <http://www.oasis-open.org>
The list archives are at
To subscribe or unsubscribe from this list use the subscription
manager: <>