Re: OO and relation "impedance mismatch"
From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Mon, 04 Oct 2004 00:33:33 +0200
Message-ID: <41607e3c$0$25965$e4fe514c_at_news.xs4all.nl>
>>Fredrik Bertilsson wrote:
>>
>>>... OO purists are claiming that the object model
>>>should be design before the (or without respect to)
>>>the database schema.
>>
>>This is a development process issue. It may matter
>>wether (parts of) the object-models are designed
>>before, after or simultaneously with the database.
>>ISTM all three plans seem viable, as long as the
>>different designs get appropriate attention, and
>>appropriate cross-checking.
Date: Mon, 04 Oct 2004 00:33:33 +0200
Message-ID: <41607e3c$0$25965$e4fe514c_at_news.xs4all.nl>
Marshall Spight wrote:
> mAsterdam wrote: >
>>Fredrik Bertilsson wrote:
>>
>>>... OO purists are claiming that the object model
>>>should be design before the (or without respect to)
>>>the database schema.
>>
>>This is a development process issue. It may matter
>>wether (parts of) the object-models are designed
>>before, after or simultaneously with the database.
>>ISTM all three plans seem viable, as long as the
>>different designs get appropriate attention, and
>>appropriate cross-checking.
> > I dunno. I've done lots of modelling both ways, and I > can't imagine doing anything besides schema-design-first > ever again.
It's not always you who decides on the process. I am inclined to discuss tables - I like Downs' law. However, I've had some positive experience where use-cases (from earlier sessions) where thoroughly discussed ('objectified') in CRC-sessions before anybody mentioned 'table'. Schema design was downhill. Received on Mon Oct 04 2004 - 00:33:33 CEST