Re: Is a Query langauge really necessary in an OODB?

In article <8b8016$9d7$1_at_nnrp1.deja.com>, Brian Price <brianleeprice_at_hotmail.com> wrote:
>
>> > In OODBMS, the situation would seem even more biased
 against query
>> > languages. Why do I, as a developer, need an interpreted
 query
>> > language, when I can simply iterate through a collection?
 If I need
>> > scripting capability, I can always implement part of the
 solution in
>> > Java. What would a query language be except an unneccessary
>> > intermediary?
>> > --
>> > Brian L. Price
 

>> One of the nicer benefits of a query language is that 3rd
 party tools
>> can be used.
>> For example, report writers like Crystal Reports.
>> If each OO application has it's own collection and accessor
 protocols,
>> then it is probably harder to use such a generic tool.
>> Just sompin' to think about, -tmind-
>>
>
>I know (from other posts/threads) you don't have much use for
 OO, but
>an overriding principle of good OO design is: Object shall
 implement
>their own behavior. The current crop of tools have it
 backwards.
>Rather than the report generator using some method (SQL) to
 reach
>inside my object (record), my object should create and pass a
 view of
>itself to the report generator (which is responsible for
 layout). Thus
>the 'generic' capability is not dependent upon a standard
 method (SQL)
>of violating encapsulation.
>

Well the OO approach violates *sharability* of data across multiple paragigms and languages. ("My method or no method")

Thus, we trade one evil for another. Joyous.

>--
>Brian Price
>
>Have compiler will travel.
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>
>

-tmind-

• Sent from RemarQ http://www.remarq.com The Internet's Discussion Network * The fastest and easiest way to search and participate in Usenet - Free!