Re: Comparison of DB2 and Oracle?

From: Hans Forbrich <news.hans_at_telus.net>
Date: Wed, 20 Oct 2004 17:22:03 GMT
Message-ID: <%8xdd.22506$cr4.15445@edtnps84>

Mikito Harakiri wrote:

> Hans Forbrich <news.hans_at_telus.net> wrote in message
> news:<h3Scd.18205$cr4.15935_at_edtnps84>...

>> ...functionality that I see required in many
>> apps such as: workflow, message queueing, replication, subqueries, direct
>> http request/response capability, security, backup/recovery, admin &
>> management tools, job scheduler (akin to cron, but inside the DB), DB
>> initiated callouts to OS shared libraries, DB initiated mail & page, DB
>> initiated TCP calls, and so on.

>
> I alway wondered what is the true value of those bells and whistles.
> Let's not forget that RDBMS essentially is a SQL execution engine, and
> everything else should be judged from the perspective how well does it
> fit into that primary purpose. Therefore, let's go through your list
> itemized:

The value is simply in having a wheel around that doesn't need to be re-invented and maintained.

No matter how much one explains these away with "isn't it just ...", developers always seme to be reinventing these "justs". What you call "bells and whistles" seem to be a base requirement in 90% of the projects I've seen in the past 3 years - only the developer's don't realize the bells are already there so they either build or buy a completely new set.

If that wasn't true, JMS, MQ Series Queuing and Workflow (oh, sorry - it's WebSphere now), and the like would not have a reason for being.

Or are you saying - let's get back to commoditizing the SQL engine so we can recover some of the revenue from these capabilities? Or continue stretching project timelines to accomplish stuff that already exists? <g>

/Hans Received on Wed Oct 20 2004 - 12:22:03 CDT