Re: atomic

#1

paul c wrote:

Quote:

I know that Codd wrote his first "big" db paper in 1969. At that time I
believe the understanding of physical atoms was simpler than it is
today but the word "atomic" in most people's minds inherited the physics
meaning.

I wonder if 1NF would seem clearer if it were expressed in terms of
"simplest" domains. I suppose there would still be people who would say
"but if I look at this way, it's not so simple", eg., when they are
talking about some compound key (versus composite key). But the rest of
us might not get drawn into the confusions they offer.

Even character strings have internal structure. Heck, one can even think
of integers as arrays of binary digits.

#2

"paul c" <toledobythesea (AT) ooyah (DOT) ac> wrote

Quote:

I know that Codd wrote his first "big" db paper in 1969. At that time I
believe the understanding of physical atoms was simpler than it is today
but the word "atomic" in most people's minds inherited the physics meaning.

I wonder if 1NF would seem clearer if it were expressed in terms of
"simplest" domains. I suppose there would still be people who would say
"but if I look at this way, it's not so simple", eg., when they are
talking about some compound key (versus composite key). But the rest of
us might not get drawn into the confusions they offer.

It might seem clearer expressed in that way, but it wouldn't express what he
needed to express. What he needed to express is the idea that none of his
arguments depend on the internal structure of the values, which they don't,
shouldn't, and can't.

1NF does not *require* that values be atomic. It asserts that values will
be *treated as* atomic. Big difference. Essential difference.

Roy

#3

paul c wrote:

Quote:

Bob Badour wrote:

paul c wrote:

I know that Codd wrote his first "big" db paper in 1969. At that
time I believe the understanding of physical atoms was simpler than
it is today but the word "atomic" in most people's minds inherited
the physics meaning.

I wonder if 1NF would seem clearer if it were expressed in terms of
"simplest" domains. I suppose there would still be people who would
say "but if I look at this way, it's not so simple", eg., when they
are talking about some compound key (versus composite key). But the
rest of us might not get drawn into the confusions they offer.

Even character strings have internal structure. Heck, one can even
think of integers as arrays of binary digits.

I don't know about that, when I'm thinking of a value in a db, say
"123", it seems enough to think of the whole thing as a symbol.

I agree. And because the value has an associated data type, the system
defines operations on it. Some of those operations may seem to reveal an
internal structure.

#4

On Oct 31, 4:31 pm, "Roy Hann" <specia... (AT) processed (DOT) almost.meat>
wrote:

Quote:

"paul c" <toledobythe... (AT) ooyah (DOT) ac> wrote in message

news:T6RVi.162603$th2.137089 (AT) pd7urf3no (DOT) ..

I know that Codd wrote his first "big" db paper in 1969. At that time I
believe the understanding of physical atoms was simpler than it is today
but the word "atomic" in most people's minds inherited the physics meaning.

I wonder if 1NF would seem clearer if it were expressed in terms of
"simplest" domains. I suppose there would still be people who would say
"but if I look at this way, it's not so simple", eg., when they are
talking about some compound key (versus composite key). But the rest of
us might not get drawn into the confusions they offer.

It might seem clearer expressed in that way, but it wouldn't express what he
needed to express. What he needed to express is the idea that none of his
arguments depend on the internal structure of the values, which they don't,
shouldn't, and can't.

1NF does not *require* that values be atomic. It asserts that values will
be *treated as* atomic. Big difference. Essential difference.

Roy

Can that be formalised? I agree with Bob that in general we have a
set of operators and they can allow us to see internal structure.
What does it mean for a value to be *treated* as atomic?

#5

On Oct 30, 6:45 pm, paul c <toledobythe... (AT) ooyah (DOT) ac> wrote:

Quote:

I know that Codd wrote his first "big" db paper in 1969. At that time I
believe the understanding of physical atoms was simpler than it is
today but the word "atomic" in most people's minds inherited the physics
meaning.

Etymology footnote: "atomic" literally means indivisible, from
the Greek "a" + "tomos": "not divided" or "uncut."

Both the physics and the math use of the word derive from the
original. Ironically, the physicists jumped the gun using this word
for the-thing-molecules-are-made-of. :-)

Marshall

#6

"Marshall" <marshall.spight (AT) gmail (DOT) com> wrote

Quote:

On Oct 30, 6:45 pm, paul c <toledobythe... (AT) ooyah (DOT) ac> wrote:
I know that Codd wrote his first "big" db paper in 1969. At that time I
believe the understanding of physical atoms was simpler than it is
today but the word "atomic" in most people's minds inherited the physics
meaning.

Etymology footnote: "atomic" literally means indivisible, from
the Greek "a" + "tomos": "not divided" or "uncut."

Both the physics and the math use of the word derive from the
original. Ironically, the physicists jumped the gun using this word
for the-thing-molecules-are-made-of. :-)

Science is always subject to revision. At the time the word "atom" was
chosen, it was thought that these atoms would never be divided.
Linguistically, "splitting the atom" is an oxymoron.

#7

On Oct 31, 11:02 am, paul c <toledobythe... (AT) ooyah (DOT) ac> wrote:

Quote:

David BL wrote:

Can that be formalised? I agree with Bob that in general we have a
set of operators and they can allow us to see internal structure.
What does it mean for a value to be *treated* as atomic?

I think it means that relational algebra operators are not allowed to
decompose it.

Nice!!!

Marshall

#8

paul c wrote:

Quote:

David BL wrote:

On Oct 31, 4:31 pm, "Roy Hann" <specia... (AT) processed (DOT) almost.meat

...

1NF does not *require* that values be atomic. It asserts that values
will
be *treated as* atomic. Big difference. Essential difference.

Roy

Can that be formalised? I agree with Bob that in general we have a
set of operators and they can allow us to see internal structure.
What does it mean for a value to be *treated* as atomic?

I think it means that relational algebra operators are not allowed to
decompose it.

Actually, the structure is illusory and representation-dependent.
Domains have operations that appear to reveal internal structure even
when that internal structure may not physically exist.

#9

On Oct 31, 11:46 am, Bob Badour <bbad... (AT) pei (DOT) sympatico.ca> wrote:

Quote:

paul c wrote:
David BL wrote:

On Oct 31, 4:31 pm, "Roy Hann" <specia... (AT) processed (DOT) almost.meat

...

1NF does not *require* that values be atomic. It asserts that values
will
be *treated as* atomic. Big difference. Essential difference.

Roy

Can that be formalised? I agree with Bob that in general we have a
set of operators and they can allow us to see internal structure.
What does it mean for a value to be *treated* as atomic?

I think it means that relational algebra operators are not allowed to
decompose it.

Actually, the structure is illusory and representation-dependent.
Domains have operations that appear to reveal internal structure even
when that internal structure may not physically exist.

Agreed.

One hundred and twenty three is an atomic value; a natural
number. The idea that there is a 1, a 2, and a 3 in there is illusory.
(Or rather, it is an artifact of the representation, not an
artifact of the value.)

You can write an expression that will give you those things,
but that shouldn't lead us to any conclusions.

X / 10 % 10 will give us the tens place. But what then
are we to make of X / 9 % 9? It gives us the nines place
in a base 9 number.

Marshall

#10

Marshall wrote:

Quote:

On Oct 31, 11:46 am, Bob Badour <bbad... (AT) pei (DOT) sympatico.ca> wrote:

paul c wrote:

David BL wrote:

On Oct 31, 4:31 pm, "Roy Hann" <specia... (AT) processed (DOT) almost.meat

...

1NF does not *require* that values be atomic. It asserts that values
will
be *treated as* atomic. Big difference. Essential difference.

Roy

Can that be formalised? I agree with Bob that in general we have a
set of operators and they can allow us to see internal structure.
What does it mean for a value to be *treated* as atomic?

I think it means that relational algebra operators are not allowed to
decompose it.

Actually, the structure is illusory and representation-dependent.
Domains have operations that appear to reveal internal structure even
when that internal structure may not physically exist.

Agreed.

One hundred and twenty three is an atomic value; a natural
number. The idea that there is a 1, a 2, and a 3 in there is illusory.
(Or rather, it is an artifact of the representation, not an
artifact of the value.)

You can write an expression that will give you those things,
but that shouldn't lead us to any conclusions.

X / 10 % 10 will give us the tens place. But what then
are we to make of X / 9 % 9? It gives us the nines place
in a base 9 number.

It's ironic you should mention base 9. It's a little-known fact that, if
a number is evenly divisible by 8, the sum of the digits of the number
in base 9 is also evenly divisible by 8. And if one repeats the process
with that sum until the resulting sum has only one digit and if the
original number was divisible by 8, the one remaining digit will be 8.

It's like magic. So it's really ironic you should mention base 9 in
particular.