Re: Oracle NULL vs '' revisited

Serge Rielau wrote:
> DA Morgan wrote:

>> Wait ... isn't that DB2 (Viper) including new "compatibility" features?
>> Why yes I think it is. <g>

Seems reasonable. Perhaps you could teach that trick to Microsoft.

> What's mathematically right (which is the debate started by
> _Oracle_users_ here) has little to do with what's popular (in the same
> sense as one-party elections popular).

The conversation was about strings ... not math. A zero byte string may or may not be NULL depending on the product. But there are no differences when it comes to math and dates.

> Further note that I at no point stated:
> "It's wrong because the SQL standard says so".

No need to be defensive ... at least with respect to this thread. Anytime you want to bring DB2 to a single set of specifications for basics like the number of bytes in an object name though you can be as defensive as you wish. <g>

> W.r.t. the arrow and the target the underlying math is a tad older that
> Larry I dare say. The standard fixed his mathematical error.
> Sometimes Oracle chooses to follow (like on outer join syntax),
> sometimes not.
> You could say Oracle has shot their arrow, barely hit the target and
> over time redefined the bulls eye....

Math may be older than Larry ... he may have a different opinion. <g> But string concatenation and string length is not. It is a problem that is quite recent.

> Lastly:
> I speak three languages, so does DB2
> My national pride has not been hurt by learning English.

But I wonder if it could survive my French? <g>

> My professional pride has not been hurt by teaching DB2 CONNECT BY and
> (+). In fact I have enjoyed it very much.
> I chuckle on both fronts...
>
> Cheers
> Serge

Are you coming to OpenWorld? <g>

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org