further table design questions

#31

In article <is2p38$k4i$1 (AT) dont-email (DOT) me>, Jerry Stuckle <jstucklex (AT) attglobal (DOT) net> wrote:

Quote:

On 5/31/2011 8:38 AM, Doug Miller wrote:
In article<is1ma0$v4f$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:

I'm not changing the subject. That is a very valid question. After
all, how can you tell if a user has answered a question if you don't
even know who the users are?

Clearly you *are* changing the subject; it's *obviously* a different
question.
And I already told you how you can tell if a user has answered a question.
Perhaps you weren't paying attention.

Go back a few paragraphs and read it again. It's still there.

Yep, I did - although I trimmed a bunch of useless stuff.

You mean you trimmed the stuff that would show that I demonstrated you were
wrong when you claimed it wasn't possible to tell if a user had answered a
question, and the stuff that would show you trying to change the subject
after I showed you were wrong, and the stuff that would show you falsely
claiming not to have been changing the subject when I called you on it.

I can easily see why you would regard that as "useless stuff".

Quote:

And you again
show you have absolutely no idea how do design anything more complicated
than a flat file.

Perhaps he has such a need, perhaps he doesn't. But you're assuming facts
not
in evidence.

I am only assuming that, like with any database, you need some
indication of who your valid "users" are.

As I said -- assuming facts not in evidence. He has not indicated the need to
know that.

ROFLMAO! Please, tell us who the valid users are.

I never claimed it was possible to tell that -- only that the OP has not
indicated that he needs to.

Please do try to read more carefully in the future.

Quote:

The one table idea doesn't even meet 1NF criteria.

Absolute nonsense. Do you even understand what 1NF is??

A hell of a lot better than you do, obviously.

Clearly not: there is *obviously* no 1NF violation in what I suggested.

ROFLMAO! Clearly it is - if you understood anything about
normalization. But we've already seen you don't.

All right, then, show where it is.

Quote:

This is a database, not a flat file, and proper normalization is
important to resolve such issues.

What the OP has so far indicated that he needs does not demonstrate a
need
for
more than one table.

On the contrary, from what he has posted, a *proper design* would
contain several tables - as anyone who really understood database design
would understand.

More nonsense. It's becoming increasingly clear that you have a limited
understanding, at best, of database design.

You mean despite over 25 years of RDB design, doing thousands of
databases in that time, some with hundreds of tables?

Even if you have been doing it that long -- which I doubt -- that does not
mean that you know what you're doing.

I got started with relational database design with DB2 when working for
IBM in the mid 1980s. Maybe a few years before you were born.

No, actually, that would be about a decade *after* I started in the field. But
I won't hold your inexperience against you.

#32

On 5/31/2011 10:12 AM, Doug Miller wrote:

Quote:

In article<is2p38$k4i$1 (AT) dont-email (DOT) me>, Jerry Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:
On 5/31/2011 8:38 AM, Doug Miller wrote:
In article<is1ma0$v4f$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:

I'm not changing the subject. That is a very valid question. After
all, how can you tell if a user has answered a question if you don't
even know who the users are?

Clearly you *are* changing the subject; it's *obviously* a different
question.
And I already told you how you can tell if a user has answered a question.
Perhaps you weren't paying attention.

Go back a few paragraphs and read it again. It's still there.

Yep, I did - although I trimmed a bunch of useless stuff.

You mean you trimmed the stuff that would show that I demonstrated you were
wrong when you claimed it wasn't possible to tell if a user had answered a
question, and the stuff that would show you trying to change the subject
after I showed you were wrong, and the stuff that would show you falsely
claiming not to have been changing the subject when I called you on it.

I can easily see why you would regard that as "useless stuff".

Nope. You can't tell who hasn't answered a question if you don't know
who your users are. You are assuming you know the user's name. How do
you know that?

The answer is - YOU DON'T. Because you don't have a list of users to go by.

Quote:

And you again
show you have absolutely no idea how do design anything more complicated
than a flat file.

Perhaps he has such a need, perhaps he doesn't. But you're assuming facts
not
in evidence.

I am only assuming that, like with any database, you need some
indication of who your valid "users" are.

As I said -- assuming facts not in evidence. He has not indicated the need to
know that.

ROFLMAO! Please, tell us who the valid users are.

I never claimed it was possible to tell that -- only that the OP has not
indicated that he needs to.

Please do try to read more carefully in the future.

Then how do you know what that user id is? You just said you don't know
your users.

Quote:

The one table idea doesn't even meet 1NF criteria.

Absolute nonsense. Do you even understand what 1NF is??

A hell of a lot better than you do, obviously.

Clearly not: there is *obviously* no 1NF violation in what I suggested.

ROFLMAO! Clearly it is - if you understood anything about
normalization. But we've already seen you don't.

All right, then, show where it is.

This is a database, not a flat file, and proper normalization is
important to resolve such issues.

What the OP has so far indicated that he needs does not demonstrate a
need
for
more than one table.

On the contrary, from what he has posted, a *proper design* would
contain several tables - as anyone who really understood database design
would understand.

More nonsense. It's becoming increasingly clear that you have a limited
understanding, at best, of database design.

You mean despite over 25 years of RDB design, doing thousands of
databases in that time, some with hundreds of tables?

Even if you have been doing it that long -- which I doubt -- that does not
mean that you know what you're doing.

I got started with relational database design with DB2 when working for
IBM in the mid 1980s. Maybe a few years before you were born.

No, actually, that would be about a decade *after* I started in the field. But
I won't hold your inexperience against you.

I know you didn't start with relational databases in 1985. But if you
want to get into total experience, mine dates back to 1967 when I was
doing Fortran II on an IBM 1410. I had already been with IBM for
several years before I got into relational databases.

You're making a typical noob mistake. Looking at a small piece of the
puzzle without considering how it fits into the whole picture.

I doubt you have more than 2-3 years of RDB experience, if even that.

--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex (AT) attglobal (DOT) net
==================

#33

In article <is33eg$sb7$1 (AT) dont-email (DOT) me>, Jerry Stuckle <jstucklex (AT) attglobal (DOT) net> wrote:

Quote:

On 5/31/2011 10:12 AM, Doug Miller wrote:
In article<is2p38$k4i$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:
On 5/31/2011 8:38 AM, Doug Miller wrote:
In article<is1ma0$v4f$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:

I'm not changing the subject. That is a very valid question. After
all, how can you tell if a user has answered a question if you don't
even know who the users are?

Clearly you *are* changing the subject; it's *obviously* a different
question.
And I already told you how you can tell if a user has answered a question.
Perhaps you weren't paying attention.

Go back a few paragraphs and read it again. It's still there.

Yep, I did - although I trimmed a bunch of useless stuff.

You mean you trimmed the stuff that would show that I demonstrated you were
wrong when you claimed it wasn't possible to tell if a user had answered a
question, and the stuff that would show you trying to change the subject
after I showed you were wrong, and the stuff that would show you falsely
claiming not to have been changing the subject when I called you on it.

I can easily see why you would regard that as "useless stuff".

Nope. You can't tell who hasn't answered a question if you don't know
who your users are. You are assuming you know the user's name. How do
you know that?

That wasn't what you said originally. You're changing the subject again.

Originally, you said it wasn't possible to tell if a particular user had
answered any questions or not. I showed that to be false.

Now you've moved the goalposts again.

And you probably don't even see that the two situations are different.

Quote:

The answer is - YOU DON'T. Because you don't have a list of users to go by.

And you again
show you have absolutely no idea how do design anything more complicated
than a flat file.

Perhaps he has such a need, perhaps he doesn't. But you're assuming facts
not
in evidence.

I am only assuming that, like with any database, you need some
indication of who your valid "users" are.

As I said -- assuming facts not in evidence. He has not indicated the need
to
know that.

ROFLMAO! Please, tell us who the valid users are.

I never claimed it was possible to tell that -- only that the OP has not
indicated that he needs to.

Please do try to read more carefully in the future.

Then how do you know what that user id is? You just said you don't know
your users.

That makes no sense at all. I never claimed to be able to know what any
particular user id is.

Please try to read more carefully in the future.

Quote:

The one table idea doesn't even meet 1NF criteria.

Absolute nonsense. Do you even understand what 1NF is??

A hell of a lot better than you do, obviously.

Clearly not: there is *obviously* no 1NF violation in what I suggested.

ROFLMAO! Clearly it is - if you understood anything about
normalization. But we've already seen you don't.

All right, then, show where it is.

[Conspicuous lack of response is duly noted.]

Quote:

This is a database, not a flat file, and proper normalization is
important to resolve such issues.

What the OP has so far indicated that he needs does not demonstrate a
need
for
more than one table.

On the contrary, from what he has posted, a *proper design* would
contain several tables - as anyone who really understood database design
would understand.

More nonsense. It's becoming increasingly clear that you have a limited
understanding, at best, of database design.

You mean despite over 25 years of RDB design, doing thousands of
databases in that time, some with hundreds of tables?

Even if you have been doing it that long -- which I doubt -- that does not
mean that you know what you're doing.

I got started with relational database design with DB2 when working for
IBM in the mid 1980s. Maybe a few years before you were born.

No, actually, that would be about a decade *after* I started in the field.
But
I won't hold your inexperience against you.

I know you didn't start with relational databases in 1985.

That's true. It was actually several years before that.

Quote:

But if you
want to get into total experience, mine dates back to 1967 when I was
doing Fortran II on an IBM 1410.

Uh-huh. Sure. Whatever you say.

Quote:

I had already been with IBM for
several years before I got into relational databases.

Sweeping the floor in the mailroom doesn't count.

#34

On 5/31/2011 3:42 PM, Doug Miller wrote:

Quote:

In article<is33eg$sb7$1 (AT) dont-email (DOT) me>, Jerry Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:
On 5/31/2011 10:12 AM, Doug Miller wrote:
In article<is2p38$k4i$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:
On 5/31/2011 8:38 AM, Doug Miller wrote:
In article<is1ma0$v4f$1 (AT) dont-email (DOT) me>, Jerry
Stuckle<jstucklex (AT) attglobal (DOT) net> wrote:

I'm not changing the subject. That is a very valid question. After
all, how can you tell if a user has answered a question if you don't
even know who the users are?

Clearly you *are* changing the subject; it's *obviously* a different
question.
And I already told you how you can tell if a user has answered a question.
Perhaps you weren't paying attention.

Go back a few paragraphs and read it again. It's still there.

Yep, I did - although I trimmed a bunch of useless stuff.

You mean you trimmed the stuff that would show that I demonstrated you were
wrong when you claimed it wasn't possible to tell if a user had answered a
question, and the stuff that would show you trying to change the subject
after I showed you were wrong, and the stuff that would show you falsely
claiming not to have been changing the subject when I called you on it.

I can easily see why you would regard that as "useless stuff".

Nope. You can't tell who hasn't answered a question if you don't know
who your users are. You are assuming you know the user's name. How do
you know that?

That wasn't what you said originally. You're changing the subject again.

Originally, you said it wasn't possible to tell if a particular user had
answered any questions or not. I showed that to be false.

Nope, you didn't. What's my user id? How can you tell if I answered
the questions or not?

All you showed was if a particular userid - which was previously known
by some mystical means - answered a question. Not the user.

Quote:

Now you've moved the goalposts again.

And you probably don't even see that the two situations are different.

No, no movement of the goalposts. But you don't understand the
difference between a user and a userid.

Quote:

The answer is - YOU DON'T. Because you don't have a list of users to go by.

And you again
show you have absolutely no idea how do design anything more complicated
than a flat file.

Perhaps he has such a need, perhaps he doesn't. But you're assuming facts
not
in evidence.

I am only assuming that, like with any database, you need some
indication of who your valid "users" are.

As I said -- assuming facts not in evidence. He has not indicated the need
to
know that.

ROFLMAO! Please, tell us who the valid users are.

I never claimed it was possible to tell that -- only that the OP has not
indicated that he needs to.

Please do try to read more carefully in the future.

Then how do you know what that user id is? You just said you don't know
your users.

That makes no sense at all. I never claimed to be able to know what any
particular user id is.

Please try to read more carefully in the future.

But you did - you used it in your query. So pray tell, how did you
figure out what my user id is?

Quote:

The one table idea doesn't even meet 1NF criteria.

Absolute nonsense. Do you even understand what 1NF is??

A hell of a lot better than you do, obviously.

Clearly not: there is *obviously* no 1NF violation in what I suggested.

ROFLMAO! Clearly it is - if you understood anything about
normalization. But we've already seen you don't.

All right, then, show where it is.

[Conspicuous lack of response is duly noted.]

This is a database, not a flat file, and proper normalization is
important to resolve such issues.

What the OP has so far indicated that he needs does not demonstrate a
need
for
more than one table.

On the contrary, from what he has posted, a *proper design* would
contain several tables - as anyone who really understood database design
would understand.

More nonsense. It's becoming increasingly clear that you have a limited
understanding, at best, of database design.

You mean despite over 25 years of RDB design, doing thousands of
databases in that time, some with hundreds of tables?

Even if you have been doing it that long -- which I doubt -- that does not
mean that you know what you're doing.

I got started with relational database design with DB2 when working for
IBM in the mid 1980s. Maybe a few years before you were born.

No, actually, that would be about a decade *after* I started in the field.
But
I won't hold your inexperience against you.

I know you didn't start with relational databases in 1985.

That's true. It was actually several years before that.

ROFLMAO! And exactly which RDB was that? Visicalc?

Quote:

But if you
want to get into total experience, mine dates back to 1967 when I was
doing Fortran II on an IBM 1410.

Uh-huh. Sure. Whatever you say.

I had already been with IBM for
several years before I got into relational databases.

Sweeping the floor in the mailroom doesn't count.

Nope, but programming mainframes does.

You've already shown your ignorance. And you have to continue to
demonstrate your stupidity.

--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex (AT) attglobal (DOT) net
==================

#35

On Sun, 29 May 2011 08:31:52 +0100, geoff (AT) invalid (DOT) invalid wrote:

Doug and Jerry

Perhaps I should add a little more info re how people access the
site..

When a user first tries to join the site a username and password
(using .htaccess etc) are required. These are created by me and
allocated by the person running this research project to the people he
knows are participating in the project.

Next they select one of four areas to respond to.

Then they have to enter a usercode and password. These entries are
checked against a mysql table using php. Again these are created by me
and allocated by the person running the project.

Cheers

Geoff

PS I would be grateful if you can help me with my "odd display of
mysql table contents" posting...

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

further table design questions

comp.databases.mysql comp.databases.mysql

Re: further table design questions - 05-31-2011 , 09:12 AM

Re: further table design questions - 05-31-2011 , 11:02 AM

Re: further table design questions - 05-31-2011 , 02:42 PM

Re: further table design questions - 05-31-2011 , 06:46 PM

Re: further table design questions - 06-01-2011 , 12:45 AM