"DA Morgan" <damorgan_at_x.washington.edu> wrote in message
news:1109298953.738334_at_yasure...
> Haximus wrote:
>
>>> You completely missed the point of the demo.
>>>
>>> In one case two columns are returned ... in the other case three.
>>
>>
>> Yes I got that, I just thought it was pointless because you, as a
>> highly-paid IT professional, are paid to understand there are
>> repercussions when altering structures and schemas. If for some reason a
>> new column mysteriously pops up in table structure, as in your demo, it's
>> because someone else has control of your databases and is doing things
>> that you as the designer and modeler probably don't want them doing
>> (without your guidance). I could make a demo showing that deleting a
>> column can break everything, it would have about the same significance as
>> your demo because it's a 'no-brainer.'
>>
>>> It doesn't matter whether you put your clause into the ORDER BY, were
>>> that possible, it still breaks everything.
>
> Actually I would have thought that you as a supposedly competent
> developer would understand the development life-cycle and would
> appreciate the cost of going back over code: modifying and testing it,
> just because a schema was modified.
>
> Apparently not.
You seem to be convinced that because a feature is available you must then
embed it in everything. Wrong way of thinking, it's merely a shortcut,
ideally for use when developing queries but not intended for production
purposes. I'm of the mindset that it is always better to have a rich set of
features available and use them when the purpose fits, but otherwise don't
force yourself to use them just because they exist. Would you rather not
have the ability to do "SELECT *" whatsoever? If so, come clean and state
that you NEVER use it. Of course you do, we all do, but you know and I know
well enough not to embed such things in an actual application unless there
is a good reason to.
Received on Thu Feb 24 2005 - 22:23:17 CST
