Re: Ever met a good data modeler?

From: Daniel Morgan <damorgan_at_x.washington.edu>
Date: Sun, 30 Nov 2003 21:47:28 -0800
Message-ID: <1070257680.897437@yasure>

John wrote:
> Daniel Morgan <damorgan_at_x.washington.edu> wrote in message news:<1070218159.300927_at_yasure>...
>

>>John wrote:
>>
>>
>>>Hi group,
>>>
>>>Have you ever met a good data modeler? Why do you think he or she is
>>>good? Thanks for sharing!
>>>
>>>
>>>John
>>
>>Yes.
>>
>>Because they don't create unresolved many-to-many relationships. Because 
>>their table structures are somewhere between third and fourth normal 
>>form. Because they understand how to reasonably denormalize when moving 
>>from OLTP to OLAP. Because they understand the difference between a star 
>>schema and a snowflake. Because they know when, and how, to leverage 
>>object-relational features.
>>
>>And because they know enough about the database for which they are 
>>designing to take advantage of its strengths and avoid its weaknesses.

>
>
> Thanks all. Is there such thing as data modeling consultant? Have you
> ever met a data modeling contractor? My impression is that you need to
> know a lot about the business process to be able to model the data
> proerply. Assuming that you have financial data modeling background.
> Can you really come to a manufacturing environment and start doing
> data modeing? Some of the entities can have hundreds of attributes.
> There can be hundreds of entities. If you don't know them well enough,
> can you really model them? I guess the answer is probably "well, you
> study them. Given enough time, you will be able to model them
> correctly". But I just like to hear other people's opinions.
>
> John

I have. I have performed that function myself and have worked with others that do it too.

You don't so much need to understand a specific business as you need to understand how to ask questions. Good questioning is something taught in law schools and in the best medical schools. Certainly it does take a good solid knowledge of business, of legal principles, of accounting practices. But more to formulate the questions than to have some specific or special tribal knowledge.

All accounting packages, for example have the same core elements. They have assets, they have liabilities, they have money coming in and money going out. But not all make refunds. The good questioner always inquiries about the unusual.

If I had a choice, as a modeler, between someone that knew how a specific business or industry worked but was weak in data modeling or someone that was a data modeling expert but understood basic business principles ... I go with the later every time.

-- 
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)

Received on Sun Nov 30 2003 - 23:47:28 CST