Re: Just one more anecdote

From: Gene Wirchenko <genew_at_ucantrade.com.NOTHERE>
Date: Thu, 04 Aug 2005 10:03:05 -0700
Message-ID: <bbi4f1p2hgj5h786kjq120vk4vqtqj7fvd_at_4ax.com>


On Thu, 04 Aug 2005 09:04:29 GMT, "David Cressey" <david.cressey_at_earthlink.net> wrote:

>"Gene Wirchenko" <genew_at_ucantrade.com.NOTHERE> wrote in message
>news:a9r1f1h7h47sk13foh61n6f4dbklhnmvfj_at_4ax.com...
>
>
>> I like ERDs, too. It does not take much to read them. I like
>> having notations that are fairly easy for non-IT people to follow.
>
>Agreed.
>
>I disagree with "Down's Law" ("Everybody understands tables just fine")
>just a little bit.
>
>I think everybody THINKS they understand tables just fine, but they miss
>some of the finer points.

     Yes, and then when someone who knows insists that rows are not records, the one who thinks he knows gets his back up.

>In particular, the fact that it takes a linkage table to specify a
>many-to-many relationship is outside of most people's active thought
>process. They'll recognize it if you force them to, but they won't do it
>on their own. They will accept a line with two crow's feet on the end far
>more readily than a linkage table.
>
>And this means that an ERD works better than a table diagram for discussing
>the requirements with non IT people.
>
>IMO, of course, and I surmise IYO, too.

     Quite.

     "What is that crowfoot there?" "What?" Points. "Oh. That means that there can be more than one of them."

     Not a horribly complex explanation, is it? UML fails badly on this score.

Sincerely,

Gene Wirchenko Received on Thu Aug 04 2005 - 19:03:05 CEST

Original text of this message