Recently Tableau has published a white paper on LOD or level of details computation. In the paper, there is a chart on the precedence of filters which is very nice.

Especially this shows by what precedence various filters work in succession. It is essential for any Viz developer to understand this precedence and use them accordingly.

I found that the chart above ignored sets, which are also a class of filters by its own.

Sets as Filter

Sets are a special class of filters which narrow the scope of the data set. When you create a set and put it on the filter shelf, all the following calculations will take place in the context of the set. The filter conditions for the set can be configured in this interface:

The definition of the filter can be based on dimension or measure or both.

This is meant to be a general incremental extraction scheme. It is derived from an example of our own work with Salesforce data.

We use Alteryx to extract data from Salesforce and then share the data source in yxdb/tde format. The initial extraction was simple and brute: extract data fully everyday. Now the data volume is huge and the extraction is taking hours.

[Update - a couple of related posts have been published recently:

Counting Active Orders via Pivot

Counting Active Customers ]

Introduction

Things come and go, including yours truly. C'est la vie. Life started when I checked into Earth on birthday. When will it end? Will see. In between, I am considered as an active person in the national census.

We often need to count moving parts of a limited lifetime at a specific time. This is called stocktaking sometimes.
5

So far I attended twice the San Francisco Bay Area Tableau User Group (SFBATUG) meetups. First time was at Twitter HQ in March. This time was at Charles Schwab on July 29, 2015. In between, seems I missed quite a few.

Each time, it was fantastic. The events were well organized. The speakers have prepared and delivered great presentations. The venues are among the best in town with nice food and wine. I am so grateful.

The following of my articles have been selected as the Best of Tableau Web recently:

May 2015

Histogram via Size()

June 2015

Histogram via LOD

July 2015

Coding Case Statement Made Easy Replacing Data Source the Fast Way Very grateful to the Tableau community for the inspirations and encouragements!
1
Blog Archive
Loading
Dynamic Views theme. Powered by Blogger. Report Abuse.