I am not quite sure, but it's nice to end a newsgroup thread peacefully :-)
Drago Ganic
P.S.: I saw that there was a bug in my message because I used >> characters.
Therefore I repeat it (for other newsgroup readers)
Frank,
You and I are entities (something, object .. whatever term You prefer). As
such, we have properties (features, attributes, fields, ...) which
"describe" us . The whole set of the properties "identify" us from other
entities in the Universe. Identification is equivalent to uniqueness,
because You (and I) believe that we are unique and have at least ONE
property which value is different. If we are described with the "same" set
of properties (but different values of those properties) we belong to the
same Entity Type (or Class ... name it). This is also a shortcut, because
it's easier to group the individual entities in sets/types. You and I are in
the same set :-)
Yes, the ID numbers are a TECHNICAL shortcut because the number of
properties is infinite and we never know them all explicitly .. but we do
know (or believe to know) that we are unique. Therefore we need an
identification vehicle. "Names" of people were a nice try to solve the
problem. If we lived in a small village with 200 inhabitants and with no
communication that would be OK. Today, numbers (integers) are used. In 10 or
so years fingerprints or eyescans will be used (for entities which have
them, of course :-).
So, YES the IDs are a technical thing in so far as NAMES ("Drago Ganic", or
"Frank van Bortel") are technical because we do not now the values of all
the attributes. In that case (If we would know everything about anybody) the
"natural" key would be it's attribute-value set. In the absence of that
knowledge, we add (or invent or use) an attribute which we call Name,
Surname, Name+ Surname, SSN etc.
In the discourse above I have not mention Microsoft, Oracle, IBM, SQL, Java,
Basic, pointers, indexes, hash tables etc. So it is and it is not Technology
:-)
Greetings from Croatia
Drago Ganic
"Frank van Bortel" <fvanbortel_at_netscape.net> wrote in message
news:ctt37g$51p$1_at_news3.zwoll1.ov.home.nl...
> Drago Ganic wrote:
>> Frank,
>> You and I are entities (something, object .. whatever term You prefer).
>> As such, we have properties (features, attributes, fields, ...) which
>>
>>>>describe<< us . The whole set of the properties >>identify<< us from
>>>>other entities in the Universe. Identification is equivalent to
>>>>uniqueness, because You (and I) believe that we are unique and have at
>>>>least ONE property which value is different. In we are described with
>>>>the same set of properties (but different values) we belong to the same
>>>>Entity Type (or Class ... name it). This is also a shortcut, because
>>>>it's easier to group the individual entities in sets/types. You and I
>>>>are in the same set :-)
>>
>>
>> Yes, the ID numbers are a TECHNICAL shortcut because the number of
>> properties is infinite and we never know them all explicitly .. but we
>> do know (or believe to know) that we are unique. Therefore we need an
>> identification vehicle. People "names" were a nice try to solve the
>> problem. If we lived in a small village with 200 people and with no
>> communication that would be OK. Today, numbers (integers) are used. In 10
>> or so years fingerprints or eyescans will be used (for entities which
>> have them, of course :-).
>>
>> So, YES the IDs are a technical thing in so far as NAMES ("Drago Ganic",
>> or "Frank van Bortel") are a technical because we do not now the values
>> of all the attributes. In that case (If we would know everything about
>> anybody) the "natural" key would be it's attribute-value set. In the
>> absence of that knowledge, we add (or invent) an attribute which we call
>> Name, Surname, Name+ Surname, SSN etc.
>>
>> In the discourse above I have not mention Microsoft, Oracle, IBM, SQL,
>> Java, Basic, pointers, indexes, hash tables etc. So it is and it is not
>> Technology :-)
>>
>> Greetings from Croatia
>> Drago Ganic
>>
>> "Frank van Bortel" <fvanbortel_at_netscape.net> wrote in message
>> news:ctrh1m$f6s$1_at_news5.zwoll1.ov.home.nl...
>>
>>>Mark A wrote:
>>>
>>>>"Frank van Bortel" <fvanbortel_at_netscape.net> wrote in message
>>>>news:ctqf0h$2ln$1_at_news1.zwoll1.ov.home.nl...
>>>>
>>>>
>>>>>SSN is a prime example is a technical key. Why is
>>>>>it people do not have a problem recognizing me by name,
>>>>>but computers do?
>>>>>--
>>>>>Regards,
>>>>>Frank van Bortel
>>>>
>>>>
>>>>That is not true. Names are not unique (maybe yours is). For most
>>>>people,
>>>>there is another person somewhere with the same exact name. Not true
>>>>with
>>>>social security name or employee number (at least it is intended that
>>>>these
>>>>are unique).
>>>>
>>>>
>>>
>>>Even when I was called John Doe (which is a common name
>>>in the US of A, I believe), people know me. The name
>>>may not be enough, but the features of my face, along
>>>with the name certainly do.
>>>And if that's too hard to understand/program for computers
>>>and uniquely identify me, how about my birth date, my father's
>>>and mother's names, and their birth dates? Or the fact I write
>>>to this ng, and have been doing that for about 10 years now?
>>>
>>>Chances are smaller than the number of people on this
>>>planet, that is not unique.
>>>
>>>But - it is not compact; in the times memory was scarce
>>>it surely would not have been acceptable. So, take a number.
>>>Nice and easy, small, and guaranteed to be unique on top of
>>>it all. And we call it: number.
>>>And as this particular number was invented by the (US) Social
>>>Service, let's call it a Social Security Number.
>>>
>>>I don't have one, but I live in a country with a civilized
>>>social security system :^) I've other numbers attached to
>>>me...
>>>
>>>But that is what a technical key is: a meaningless
>>>number (BTW: is SSN really meaningless, or is it coded?),
>>>instead of (mostly many) fields, together making up a
>>>unique identifier.
>>>
>>>--
>>>Regards,
>>>Frank van Bortel
>>
>>
>>
> Guess we are in agreement, then.
>
> --
> Regards,
> Frank van Bortel
Received on Thu Feb 03 2005 - 08:07:24 CST
