My Pleasant Surprises - Qlik's Associative Difference

[Michael_Tarallo]

Hey Guys happy Tuesday! Today I will continue my series on some of my many pleasant surprises when first learning Qlik. My last pleasant surprise covered how Qlik creates data models without any complex data modeling skills and how the use of its metadata was a bit different from what I was use to - you can learn more about it here: My Pleasant Surprises Learning Qlik - a multi-part series. In this edition of the Qlik Design blog I cover one of our most important differentiators - the Qlik Associative Difference. By now you should know of my love for creating short succinct videos, and I personally feel they are more fun then reading walls of text - so in this entry I have embedded a video presentation and demonstration to show you Qlik's Associative Difference  - thanks for watching,

Try the app here:

https://webapps.qlik.com/associative-difference/index.html

NOTE: To increase resolution or size of the video, select the YouTube logo at the bottom right of the player. You will be brought directly to YouTube where you can increase the resolution and size of the player window. Look for the 'settings' gears icon in the lower right of the player once at YouTube. (video and sample files)

For more videos that can help you get started with Qlik Sense:

• New to Qlik Sense Videos
• Qlik Help - YouTube

2017_AssociaQlikSnippets.mp4

[robert99]

Hi Michael

Thanks for this

I don't even mention this aspect to a client as their eyes glazed over. But this is well explained. I might try this video on a client and see what they think

[Michael_Tarallo]

Thanks Robert - stay tuned - the app I used in this video will also be available for everyone to play with it.

[Anonymous]

Michael, excellent metaphor with the movies.  I'm going to use this to explain the associative difference.  It also has broad public interest, and is sophisticated-enough to relay the point (where the fruit is a bit too simple).  Can I get the movie app?  Does it also include somehow the "non-associative" visualization somehow, or did you splice that into the movie? 

[Michael_Tarallo]

Hey David - thanks - yes - this sample web app will be made available publicly shortly, I will send you a personal note.

[sirpod90]

Hey Michael,

nice summarized video about the difference of an associated.

The only thing I miss in the Qlik data preparation world is the possibility to handle associated loops. Because sometimes it's necessary to combine data with more than one table and if the tables are refereed to each other with a constrain or references, I have to rename the fields, join them or do some other stuff.

If Qlik could handle Loops in the future the discovery option would be much bigger and the preparation even simpler and faster.

Regards

Tobias

[Michael_Tarallo]

Hi Tobias - Thank you - I believe you are referring to Circular Reference? for those that may need it: https://www.youtube.com/watch?v=vPrsPXW2N2I

I will check with daz‌ to see if there are any plans to address this sort of thing automatically.

Dean - does visual data prep have a mechanism in place to correct a Circular Reference if found, I know that it informs you of the Circular Reference, but I believe you still need to manually rename the field this that correct?

Regards,

Mike T

Qlik

[sirpod90]

Hello Mike,

thanks for your fast response.

But I was more talking about Circular Reference on purpose. Because sometime in a little bit more complex data model you have the same data associated on more than one table.

For example in the video you posed you could add a table with more information to the city (eg. habitats, area size, leading party, economic value, ...).

At the moment you have the possibility to add the detailed city information twice to your model and link them separately to the fanclub city and the actor city. But now I can either search in the one city table or in the other but not both of them simultaneously.

I would wish that in future you could load one table to both of the information and handle the Circular Reference with a constrain or other techniques.

Thanks and regards

Tobias

[sspe]

I think the solution in that kind of scenarios, will be to create a Link table. While I understand your wish (and sometime have the same wish myself :-)), I think it will be difficult for Qlik to address. In the case where I need this functionality, I have build link table and/or composite keys to link several "City" keys to the same dimension.

/Steen

[Anders_Eriksson]

To allow circular references in the data model and handle it with constraints instead would break a fundamental cornerstone in how the QIX-engine works and it would no longer be the QIX-engine.

[Ian_Crosland]

Hi

The "magic wand" feature in the data manager has been designed to avoid circular references, the reason being you have ambiguous results from the logical inference - different results depending on whether you evaluate clockwise or counterclockwise.