Logic Data Modeling 4 - Normalization Example

[skillfullist]

this man is so good..i avnt completed the video yet ..bt hes kinda bringing out d joy in me

[nahsirah]

4:31 - the doctor has a skitzo episode, lol,,,,,,this is an awesome video. I am not in the states, but if you had a full detailed tutorial on data modelling, i woudl gladly pay for it.

[Riflemanm16a2]

Why are there so many names for the same thing? Entity=Relation=Table, Attribute=Column=Field, Instance=Row=Record. This really confused me at the beginning of learning about databases.

[gvimontslice]

@Riflemanm16a2 -- Think of these terms as falling into one of these two model categories: Logical and Physical.

Logical models have: Entities composed of Attributes. Physical models have: Tables composed of Columns.

Only the physical model supports the storage of records in rows.

[Riflemanm16a2]

@gvimontslice Thanks for the great explanation! Very clear and concise :)

[swu880]

Its difficult to see whats on the board. Should try to focus more on the board rather than on him all the time. Zoom in on where he writes.

Thank you for posting this vid. Aside from the above mentioned problem, its a wonderful lecture

[dustonly1986]

according to your opinion, it seems the foreign key only introduced in 3rd N.

the compound primary key introduced in 1st N.

[flutep]

i wish i had a lecturer as good as you:)

[heilla88]

Hey , i wanna thank you for the video series , i have an exam tomorrow and up to this point i haven't understood data normalization , this taught me allot.

[alihammadshah]

A table in a database is Entity and column's name in a table is Attribute. Am i getting it right ?

[gvimontslice]

Correct! A table in the physical db is equivalent to an Entity in the LDM. A column within that table is equivalent to an Attribute in the LDM.

[gvimontslice]

Don't confuse the definition of attributes with the instance of values within the materialized table once you've actually created a table based on the logical data model. Remember that Entities are to tables as Attributes are to Columns.

[keitaidenwaKern]

so I think you mean that you cant have attributes with more than one occurance. Like people cant have more than one name, but they can have more than one item that they are ordering. So then a seperate entity for items would need to be made with a relationship to the order?

[gvimontslice]

That's correct; the data elements that repeat -- those that potentially will have multiple instances -- are segregated into a separate entity. The relationship among these entities is driven by propagation of the foreign key from the main entity to the subordinate entity (from the parent entity to the child entity).

[chivasbross]

So, Elements are considered as rows or something else.

[gvimontslice]

When first learning about logical data modeling, it is common to confuse the definition of data structure with the instances of the data itself. Actually, the data elements that were referring to here are often called attributes. An attribute in the logical data model will become a column in the physical implementation of a table within a database schema.

[gvimontslice]

An instance of recorded data -- an actual record of information composed of one or more columns within a table -- becomes a row within the physical database schema.

[keitaidenwaKern]

so that means that a different entity instance cant have any of the same non key attributes. So if there was the customer name attribute, there could be two people with the same name. Then does that mean you would have to make the name its own entity? Like if multiple people had the name john, then would they make it a foreign key attribute liking it to a seperate entity with the primary key called john