Re: Deep Normalization. Was: A Normalization Question

From: Jan Hidders <jan.hidders_at_pandora.be>
Date: 27 Jul 2004 01:50:08 -0700
Message-ID: <3c8cab4c.0407270050.293033c4_at_posting.google.com>


mikharakiri_nospaum_at_yahoo.com (Mikito Harakiri) wrote in message news:<8a529bb.0407261603.3a820055_at_posting.google.com>...
>
> Does "normalization" really stop there? One can decompose "b" into
> bits! But, that would be brand new super thingiest XDb3, of course.

The central issue is whether the character "B" (not its role as a unsystematic reference) is redundant. One way to know if something is redundant is to change one of its attributes and check for corruption. This is difficult to see with characters, because characters typically don't change and if they change we consider them to be a completely different character.

I realize the following is an extraordinary example, however a general data model can't have prejudices as to what data/changes it will accept. Suppose, the world is taken over by aliens that consist entirely of anti-matter. To make it easier for them to read our computers they want every bit in every computer to be inverted, thus 'B' needs to be updated to '=' (assuming 7-bits ASCII). Just updating one bit and not the others, corrupts the db. Below is approximately how XDb3 normalizes the three characters and inverting the bits does not corrupt the db.

GradeList DB101 Logic Calculus
1 ->2 ->2 ->2

Char Bit1 Bit2 Bit3 Bit4 Bit5 Bit6 Bit7 2 ->3, ->4, ->3, ->3, ->3 ->3 ->4

Symbol

3      0
4      1

Finally, let me add that I, for one, and the whole XDb3 development team welcome our new alien overlords.

  • Neo Hidders

PS. :-) Received on Tue Jul 27 2004 - 10:50:08 CEST

Original text of this message