I'll try to give an overview. Hopefully someone can correct me if I mess anything up.
Endeca Server is basically a type of database. The data in an Endeca Server domain is therefore made up of records. For example, a record might represent a single product, or a single sales transaction.
An attribute is a "column" in the "database". So for product data, the attributes for each product could be values such as Color, Size, Price, etc.
The record schema provides information about the structure of each record - the attributes to include.
For each attribute, the attribute schema provides information about the attribute value. For example, does the value have to be unique? Can the attribute have more than one value? What type of value is required (string, number, date, etc.)?
PDR and DDR are types of attribute schemas. Some attributes, called managed attributes, have a specific, limited set of values - for example, for a product, the Size attribute may have a specific set of available values (S, M, L, XL). For these attributes, the DDR is used for the attribute schema.
For attributes that do not have a specific limited set of values, the PDR is used for the attribute schema.
Hope that helps.
Can anyone help me out with the answers for
1.What's an attribute?
An attribute is the name of the column. In the example below, the following items are attributes:
Name, type, product ID, Size Range, Color, Number Sold, Price
As in here:
Name Bike Type ProductID Size Range Color Number Sold Price
2.Whats a Record?
Records are the fundamental units of data in the Oracle Endeca Server. Almost all information that is consumed by the Oracle Endeca Server, including raw data and the data schema, is represented by records. For example, here is a record:
Road-450 Road Bikes 4038 42-46 CM Red 171 1457.99
Its name is Road-450, its Type is Road Bikes, its Product ID is 4038, its Size Range is 42-46 CM, its Color is Red, its Number Sold is 171, its Price is $1457.99
3.Whats a record schema? A record schema is a collection of system records that defines the characteristics of the record's attributes. The schema consists of PDRs, DDRs, and other system records.
4.Whats an attribute schema? It is the same as the record schema.
5.What's PDR and DDR?
A PDR -- Property Description Record, it is a record that describes an attribute. For example, for an Attribute Color, its PDR will say that it can contain only values of type string, and can have only a single value per record. This essentially means that each record that has an attribute Color, can have color expressed in text (and not in HEX code, in this example ), and also, that each record that has a value for the Color attribute can only have one value, not many. In other words, if for example, a record is a bycicle, each bycicle can have only one color value -- be red, or blue, but it cannot be red and blue.
To continue an analagy, if a record is a shirt, for example, and an attribute is Size, then this attribute can have multiple values: S, M, L, for example.
A DDR is a Dimension Description Record, it is also a record that describes an attribute, but, comparing with a PDR, it describes other characteristics of an attriubute, beyond the ones included in a PDR. Each attribute has a PDR, only some of them have both PDR and DDR. DDRs are needed when you need to specify hierarchy, enumeration of values and other additional metadata on attributes.
Please dont give reference to the dev guide and admin guide as i was not able to understand from there.
I am one of the real people who created this documentation. I naturally care that it is useful. It has been reviewed and improved already. What would you suggest we do?
Please indicate if these additional explanations in my answer are helpful and if so, we will revise the doc again and include them. If they are not helpful, please indicate what would you like to improve.