Re: foundations of relational theory?

Jonathan Leffler wrote:

> Lauri Pietarinen wrote:
>
>> Jonathan Leffler wrote:
>>
>>
>> The problem with constraints like these is that, yes, you safeguard
>> your db from corrupt data, but getting the information back into the
>> application is not that easy: you have to parse the error message,
>> get the name of the constraint and translate it into a meaningfull
>> error message for the user. I don't know if anybody does this, so,
>> in effect you end up coding the same checks twice: in the app and in
>> the db.
>>
>> This is where we run into the frustrating "mismatch" between app
>> language and db environment.
>
>
> Hmmmm, well yours is probably a slightly clearer exposition than mine,
> but way back an aeon ago in this thread, I commented on exactly this
> phenomenon.
>
> Let's see...ah, there it is...groups.google.com...Oct 13...

Ups, I must have missed that one ... ;-)

Lauri Received on Wed Oct 29 2003 - 09:32:12 CET