|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | RE: GISList: Quality model?? |
| Date: |
12/07/2001 10:34:42 AM |
| From: |
Anthony Quartararo |
|
|
Interesting site and certainly an effort I would applaud. However, on the
surface, there seems to be little unique and groundbreaking information
here, just proven processes and methodologies applied to a new discipline
and industry segment. That is valuable work in and of itself, however it
still has the same fatal flaws and traps as every other acronym-laced
"quality" initiative.
Having ISO 9000 registration actually means very little in the big
industries, and it means even less in the GIS-related industries. All that
the ISO certification conveys is that a) an organization has spent alot of
time documenting their business processes, b) they communicate those
processes to employees, clients and consultants, c) they maintain records to
prove it. Oh yeah, and it also means that some ISO quality consultant makes
alot of money, and the newly registered company gets to stick an "ISO" label
on their marketing materials.
I have yet to meet a qualified ISO registrar who knows beans about GIS,
digital product quality or anything related to geospatial services. Sure,
there's alot that know how to audit a QMS for software product, that's not
unique. Microsoft could be ISO registered and it would not require them to
decrease ANY of there thousands of known software bugs.
The SEI CMM, ISO 14000, Baldridge Awards, etc.etc.etc. are all in the same
boat, some work better than others, but none works 100% of the time for 100%
of the users. There can and will not be a truly valuable standard of
quality in GIS or related industries until we have standards of what it is
that we are trying to make "better" by all these quality initiatives. What
I mean is, that we do not even have compliance to a standard of
interoperability for a "pole" from all the major GIS software vendors, let
alone how a telecommunications model works, or and electric model works,
etc. etc. etc. So how can a quality standard validate a service or product
against an unknown? It can't, and that's fatal flaw #1.
The second fatal flaw is that the "independent" registrar auditors are paid
by the companies seeking certification, whether ISO, CMM, etc. I doubt I
need to elaborate on the details of this flaw.
How long has ISO TC 176 been working on a technical standard that could
apply, among other things, to GIS? The OGC and countless other academics
have spent PhD's man-years tackling this problem in the nth degree of detail
with still no comprehensive and collective standard model that fits just
right.
A third flaw is, well unavoidable and is not going to be diminished.
Interpretation of the standard. Since all of these models and standards are
written in such a way as to be less "prescriptive" and more "descriptive",
there is ample wiggle room for what they actually mean in a given industry
setting. This is both good and bad and has certainly led to a lot of
companies becoming registered to ISO and others that haven't the slightest
clue what its all about. But you can't have a standard so focused that it
really only applies to a single industry and/or segment, right? That's what
happened to the US automotive industry. Oops.
Lastly, (I know I have vented long enough), even companies, CEO's included,
that are ISO certified don't get it. How many times have we all heard this,
"Better quality means higher costs" If you have heard that then you can
rest assured that the people selling that are selling "junk bonds". What I
mean is, they have missed the entire purpose and effort of the quality
management industry. As Motorola and others have demonstrated with Six
Sigma, higher quality actually means more production efficiencies, less
waste and rework, and that should all translate into more competitive
pricing for consumers (ie. utility or telecom GIS users, etc.) rather than
cost increases. Just because offshore labor may be cheap, and it costs
nothing to throw another dozen set of eyes on something to make sure
"quality is improved", it does not follow that the end product is going to
be even .5% better because of all that checking, rechecking, rework, etc.
because, product still gets delivered that is not acceptable and I am sure
many of us have experienced this over and over again.
Another quality model is not what we need, we need to make the ones we
already have work. Until then, I'll wait for the movie version.
Anthony
here are the URLs.
model :http://itsqc.srv.cs.cmu.edu/escm/index.html
documents: http://itsqc.srv.cs.cmu.edu/escm/escm_model.html
News release : http://news.cs.cmu.edu/Releases/demo/85.html
>Hello listers,
>
>I just saw a news release on a quality model released by Carnegie mellon
>university recently from Pittsburgh, which they call is for IT enabled
>services. The published documents in their website says that
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
