Groups | Search | Server Info | Login | Register

Groups > comp.databases > #95

Re: joining tables and the order changes

Show all headers | View raw

toledobythesea@yahoo.ca wrote:

> On Jun 23, 1:50 am, Arthur Ward <art.w...@noreply.xx> wrote:
> ...
>> SQL databases follow the Information Principle, which requires that all
>> the infromation content of the database is represented explicitly as
>> values of attributes of rows *and in NO OTHER WAY*. If one fact must
>> be distinguised from another in terms of the time (or sequence) at
>> which it became known, then the time (or sequence) must be part of the
>> fact. Once you have the sequence information you can use it to order
>> the facts when you finally report them.
> ...
>
> I'd rather say that relational db's depend on the Information
> Principle.

I am glad you picked up my careful reference to SQL databases rather
than relational databases, even if you didn't buy what I said about
them.

> It is widely ignored, such as when arguments against updating certain
> views are made.  These arguments remind of the pedestrian who looks in
> two directions then steps forward directly into the muck.  Even CJ
> Date opens the door in something he wrote recently giving a predicate
> for the usual Suppliers relation of something like "Supplier S# is
> under Contract".  No 'Contract' attribute in site.

(Typo? "Sight"?)  Why would there need to be a contract attribute?  I
see no violation of the IP (here).   Can you direct me to the relevant
article?

> I could agree that in this sense SQL databases 'follow the Information
> Principle', if only they didn't prevent certain view inserts and
> deletes!  

I will concede this point only because the distinction between an SQL
database and an SQL DBMS is hazy and incomplete.  The manipulations the
engine supports may force certain compromises on the database design,
but some SQL implementations are worse than others.  Maybe one day there
could be one that doesn't have these limitations on view updating.

> But I gather they do, so they aren't 'following' anything
> except arbitrariness, which is what the IP was intended to counteract.

Art

Back to comp.databases | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

joining tables and the order changes shawrie <tourerukcom@googlemail.com> - 2011-06-22 07:44 -0700
  Re: joining tables and the order changes Ben Finney <bignose+hates-spam@benfinney.id.au> - 2011-06-23 07:47 +1000
  Re: joining tables and the order changes Arthur Ward <art.ward@noreply.xx> - 2011-06-23 08:50 +0000
    Re: joining tables and the order changes David Kerber <dkerber@WarrenRogersAssociates.invalid> - 2011-06-23 09:09 -0400
      Re: joining tables and the order changes Arthur Ward <art.ward@noreply.xx> - 2011-06-23 13:23 +0000
        Re: joining tables and the order changes David Kerber <dkerber@WarrenRogersAssociates.invalid> - 2011-06-23 14:35 -0400
    Re: joining tables and the order changes toledobythesea@yahoo.ca - 2011-06-24 10:40 -0700
      Re: joining tables and the order changes Arthur Ward <art.ward@noreply.xx> - 2011-06-25 12:33 +0000
        Re: joining tables and the order changes paul c <toledobythesea@yahoo.ca> - 2011-06-25 14:44 -0700
        Re: joining tables and the order changes paul c <toledobythesea@yahoo.ca> - 2011-06-25 14:53 -0700
          Re: joining tables and the order changes Lee Fesperman <firstsql@gmail.com> - 2011-07-04 00:39 -0700

csiph-web