Errata

The result I received from the query that you show in figure 1-12 is correct because there ARE relationships established in the source data.

Note from the Author or Editor:
In reality the problem is at page 8, you should load two tables (Sales Order Header and Sales Order Detail) together and then import the Product table afterwards, in a different import operation. In this way, only the relationship between the two Sales Order tables will be created and not the one between Sales Order Detail and Product.

Report CH02-01-FirstSample.xlsx in the companion content is not pivoted to match how you have it pivoted in figure 2-1 on page 17.

Also, OnlineOrderFlag returns 0 and -1 for me, not True or false.

I have just started this book and am disappointed so far. I hope it's something that I'm doing wrong.

Note from the Author or Editor:
The file CH02-01-FirstSample.xlsx contains the final workbook you obtain at the end of modifications described in Figure 2-16. You can easily create the same starting point by using the raw data you started with in Chapter 1. You can also use the CH01-02-PowerPivot.xlsx as a starting point for PowerPivot workbook, creating a new PivotTable following instructions in the paragraph preceding Figure 2-1.

The intellisense sidebar says: "When your write a formula in the PowerPivot window, the table name is not displayed by Intellisense when the database contains only one table."

However, if I import a single table (for example, FactInternetSales from AdventureWorksDW2008) and type "FactI" in the PowerPivot wondow's formula bar, the autocomplete dropdown contains the table name, "FactInternetSales."

Note from the Author or Editor:
You're right - this behavior changed a little bit since last beta we used during the book writing.
Actually IntelliSense works in this way:
- if you type an opening square bracket, IntelliSense shows only the columns of the current table (the one in which you are defining the calculated column or the measure)
- if you type the first letters of a table name, you will see both table name and column names in IntelliSense

In the book the formula says [Cost] = SUMX( Sales Sales[Quantity] * RELATED( Product[StandardCost] ) )
in the sample data the formula looks like this:

=SUMX( Sales_SalesOrderDetail, Sales_SalesOrderDetail[OrderQty] * RELATED( Production_Product[StandardCost] ) )

This is a small error, but variances between the information in the book and the actual implementation in the companion files occur a number of times throughout the book.

I would strongly encourage the authors to review the entire book for formula AND grammatical issues before printing a second edition. The book is full of excellent information, but the writing is awkward, making it difficult for at least this native English speakers to read the book.

Extensive use of the word obviously and starting sentences with of course are two examples of words that have no place on the written page. If something is obvious then why are you telling us about it? And of course is just filler text, not useful in a technical book.

Starting a sentence with however is bad, bad, bad. However is a modifier, if you must use it, make sure it comes after a comma, not a period. Nevertheless is another word that serves no purpose, it's just padding.

I am not an engineer and need books like yours to learn PowerPivot and advanced Excel functionality. Grammatical errors, sloppy language and formula errors make this learning process rather painful.

Yours,

Debbie Lefkowitz

Note from the Author or Editor:
the formula after the first paragraph references misspelled table names.
Instead of:
[Cost] = SUMX( Sales Sales[Quantity] * RELATED( Product[StandardCost] ) )

It should be:
[Cost]=SUMX( Sales_SalesOrderDetail, Sales_SalesOrderDetail[OrderQty] * RELATED( Production_Product[StandardCost] ) )

In the first paragraph, the last sentence, the book states, "This query can be constructed using the query designer..." However, when I construct the query using the query designer as shown in Figure 4-27, below the paragraph, I get a table with no rows, unlike if I use the hand-built query as shown on page 103 (which returns a table with 19,220 rows).

The reason is that, in the AdventureWorks database, no rows in the CustomerAddress.AddressTypeID field that have the value that corresponds to the value for "Primary" in the Person.AddressType table. In the first (non-query designer-generated) query, the restriction is on the right side of a left outer join and hence has no effect but when you design the query in the query designer it applies the restriction in the WHERE clause where it does have an effect.

So, I think that you can't generate the same query in the query designer, correct?

Note from the Author or Editor:
You're right, the SQL query shown at page 103 is not identical to the one generated by the wizard shown in Figure 4-27. The intention of the example was to show that you can use the query builder also to create a query with an high number of joins. Unfortunately, the specific SQL query shown at page 103 cannot be created with just the query designer for the use of the "Primary" filter in a LEFT JOIN condition. We cannot reproduce this particular join using the query designer neither by modifying the JOIN condition (that cannot accept comparison with a constant value) nor by adding an OR condition to the existing WHERE condition (which in the current form filter out all rows that would have been included in the first query). We're sorry for this mistake.
However, the purpose of the example was just to show a relatively complex query in the designer, even if we wrongly used a condition that cannot be created by using just the designer.

The formula listed on the page, which is the correct formula is not the same formula as is contained in the sample files. I rewrote the measure to the one in the book and everything came out fine.

Note from the Author or Editor:
The formula shown in the book for the AllContinents measure is correct. The sample workbook contains a wrong formula so if you put the AllContinents measure in the PivotTable as described in the picture you obtain a wrong result, but if you correct the formula with the one shown in the book then the result is correct.
The right formula is
CALCULATE( SUM( Orders[Amount] ), ALL( Cities[Continent] ) )

In the formulas for FiscalYear and FiscalQuarter the @MonthNumber column is referenced. In the demo file, the same column is named @Month. Thus, in these formulas you need to replace @MonthNumber with @Month to make them work.

The book shows the formula name for the measure as

"CalculatedOrderFreight ="

The workbook on the DVD shows this formula name as:

"CourierFreight ="

Note from the Author or Editor:
Correct, the measure should be called CourierFreight in order to replicate the screenshot in Figure 10-35