Re: Mixing OO and DB

From: Bob Badour <bbadour_at_pei.sympatico.ca>
Date: Thu, 06 Mar 2008 17:44:19 -0400
Message-ID: <47d065b3$0$4033$9a566e8b_at_news.aliant.net>


topmind wrote:

>
> Robert Martin wrote:
>

>>On 2008-03-04 17:06:57 -0600, topmind <topmind_at_technologist.com> said:
>>
>>
>>>
>>>Robert Martin wrote:
>>>
>>>>What you are missing is that there is
>>>>no restriction that other objects use the data from the employee_table.
>>>>So while a class certainly defines the methods that can be used to
>>>>access and manipulate the data within that class, there can be many
>>>>different classes that use the same data in very different ways.
>>>
>>>In other words, each class becomes a little roll-your-own database
>>>with a very custom interface.
>>
>>Each small group of classes becomes a little roll-your-own data access
>>and manipulation scheme that is perfectly tuned for it's very specific
>>purpose.

>
>
> Which is over-kill for the task-level. Often one ends up only
> instantiating stuff once, for example. If you do that, you might as
> well use procedural to avoid bloated code.
>
>
>>>RDBMS instead factor common collection-
>>>handling and attribute-management idioms into a central tool.
>>
>>Granted.  Very useful.

>
>
> Then why wouldn't you want that in OOP also? Answer: because you'd
> have to rethink encapsulation and also face the same problems that
> prompted Dr. Codd to formulate relational.
>
>
>>>OO does
>>>not, making it primitive and tedious.
>>
>>Balderdash. Programmers organize the data they extract from the
>>database into forms that are more convenient for the particular
>>processing they need to do.

>
> Our techniques for measuring "more convenient" apparently are very
> different. Either one of us is way off (or both?), or they are
> subjective personal preferences.

Neither. Martin is just talking out of his ass. Yet again.

[snip] Received on Thu Mar 06 2008 - 22:44:19 CET

Original text of this message