Curt Sampson wrote:
>>Because in fact you have advocated removing the OO stuff.
> Actually, I'd suggested thinking about removing the OO stuff.
Man, aren't we into splitting hairs?
You actually stated your case quite strongly and indeed if you hadn't,
the thread would've died long ago.
Whatever. You're just dick-waving.
Enjoy your life :)
> Starting
> a discussion about the concept is far from "advocating" it. And in fact
> I'd backed off the idea of removing it. However, now that it appears to
> me that table inheritance actually breaks the relational portion of the
> database, I'm considering advocating its removal. (This requires more
> discussion, of course.)
Except apparently you have no life, oh well, not my problem.
>>Writing extra code, no matter how trivial, increases the odds that a
>>mistake will be made.
>
>
> Yeah. But using a broken table inheritance model is far more likely to
> cause bugs and errors. It certainly did when I tried to figure out what
> you were doing using inheritance. Not only did I get it wrong, but I'm
> not at all convinced that what you were doing was what you really wanted
> to do.
I wasn't using inheritance. I didn't post an example. And all agree
that PG's model is broken and eventually needs to be fixed.
Three strawmen in one paragraph.
Again, you're dick-waving and further discussion is not useful.
>>You mean you accidently supported the argument that this approach is,
>>perhaps, more error prone?
> No, supported the argument that table inheritance is either
> ill-defined, broken, or both.
Then what you're saying is you've been arguing all this time against it
without understanding how it works?
Because either
1. If you understood how it worked then you screwed up your more complex
view-based analogue, therefore supporting the argument that you've shown
that the mapping is more error prone.
2. Or you screwed up your code because you've been dick-waving without
bothering to learn the semantics of the PG OO extensions, which doesn't
really enhance your credibility.
Which is it? The idiot behind door number one or the pendantic boor
behind door number two?
>>We don't need the binary "integer" type, either. We could just use
>>"number". Yes, operations on "number" are a bit slower and they often
>>take more space, but ...
>>
>>Shall we take a vote :)
>
>
> If you like. I vote we keep the integer type. Any other questions?
Sure ... why the inconsistency without explanation?
--
Don Baccus
Portland, OR
http://donb.photo.net, http://birdnotes.net, http://openacs.org
