StaRKeY
starkey at wanadoo.nl
Sun May 16 17:27:52 CDT 2004
Hi, long read no write:-) Interesting Susan/Martin, ofcourse the combination first/last -name is a repeating group since more than one human object can carry the same name property. This 'problem' is probably the reason why social security started using unique identifiers/numbers:-) How do we do this in real life on the streets? No number is used luckily but we do use unique identifiers without thinking about this since we're so grown into it. The whole human object is taken into account when trying to recognise a 'known' identity but what if 99.9% is recognized and this 0.1% not.... we have doubts and need to check by either ask this identity direct or find out indirect. Ofcourse the source has to be 100% sure about the statement given otherwise gossip or misunderstanding is born:-) Returning to Susan's remark, for database efficiency reasons we should probably have one name table containing all known name combinations on earth and an autonumber for relational matters and one unique idetifying human object table containing a global unique identifier and the FK reference to a name combination (since this is a repeating group:-)) This would get rid of duplicate names and thus be more efficient data management. Makes me think of Assembly versus Visual Basic:-) Hihihi getting inspired now... In fact like the use of DNS the internet has for identifying the unique IP addresses we global inhabitants should have a HNS (Human Name System) used the other way around (number = name combination instead of name = ip-address) where name combinations are kept globally so we would only have to reference this domain system LOL... More inspired...by using a unique identifier (read social securitynumber) added with a unique country identifier added with the known name combination (at birth registration) we'd have a unique global human identifier:-) Assuming the ID of name combination Eric Hans Starkenburg would be 123456789 my global unique human identifier would be something like this; 104501533-NL-123456789 So Susan I agree, one name table should get rid of these ugly duplicate values:-) Nighty night folks -----Original Message----- From: accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com]On Behalf Of Susan Harkins Sent: zondag 16 mei 2004 22:58 To: 'Access Developers discussion and problem solving' Subject: RE: [AccessD] normalization question You would think so Martin -- and you're most likely right. But, what if????? Even with an AutoNumber value as the primary key, there's no way to distinguish one from the other -- you have to depend on the relationships to get it right. In fact, and this is what I'm really getting to -- isn't a duplicate name really just a repeated value? I'm bordering on ridiculous chaos here I know, but well... it's a Sunday and I'm working, so might as well mess with everyone's heads, right? ;) Susan H. Why would you have a table with only names? In my experience there is usually another qualifier? But I am sure someone here will come up with an interesting approach to this one. -- _______________________________________________ AccessD mailing list AccessD at databaseadvisors.com http://databaseadvisors.com/mailman/listinfo/accessd Website: http://www.databaseadvisors.com --- avast! Antivirus: Uitgaande bericht is niet besmet. Virus Gegevensbestand (VPS): 0420-4, 14-05-2004 Getest op: 17-5-2004 0:27:52 avast! auteursrecht (c) 2000-2004 ALWIL Software. http://www.avast.com