Perception of Colors of Dresses and Data Visualization

Finally, people are talking about color perception! Everybody is talking about the white/gold/blue/black dress that has gone viral as people discover that color perception is not absolute.

We cover this in our recent Oracle Press book on Data Visualization for Oracle Business 11g. From page 273: ‘We the authors often show a handful of optical illusions involving color perception/misperception during our data visualization workshops to drive home the point that we humans don’t see color as much as we “experience” it as our brains attempt to construct a color model for a given situation.’ In our presentation from BIWA Summit 2015 we showed a color spiral image.

Most people we talk to see a green and a blue spiral. In our presentations, we use a moving circle to demonstrate that these colors are actually the same value: RGB 0,255,151. Cick on the thumbnail of the spiral and take a print screen and use the eye dropper tool in Paint if you don’t believe me! Our brains perceive the colors differently because of the adjacency of other colors.

This becomes very important in constructing business intelligence dashboards. Color selection on dashboards affect how we perceive the data and the business insights obtained in ways we don’t even recognize. For sequential data, use sequential color schemes (e.g. increasing shades of blue for color-coding states based on increasing Revenue amount); for qualitative data use qualitative color schemes (e.g. various pastel colors of equal darkness to represent various brands).

For more information on the book that brother Tim Vlamis and I published in 2015, see our web page at


Omaha BI / DW User Group Meeting

Please join us at the Omana BI / DW User Group Meeting on Thursday, March 12th, 4 - 6:30pm at UNO's Mammel Hall.


Oracle Data Mining and Oracle R Enterprise Training

We are pleased to announce that Tim Vlamis will be facilitating a Live Virtual Class in Oracle Data Mining and Oracle R Enterprise.

  • Oracle Database 11g: Data Mining Techniques Ed 1 LVC - May 26 & 27, 2015
  • Oracle R Enterprise Essentials Ed 1 LVC - May 27 & 28, 2015

For more information, click here.


Our New Book is Out! Data Visualization for Oracle Business Intelligence 11g Webcast Feb 17

Techcast February 17, 1pm Eastern, 12 noon Central, 11am Mountain, 10am Pacific

Dan and I will kick off the introduction of our new book from Oracle Press titled “Data Visualization for Oracle Business Intelligence 11g” in a special IOUG BIWA Techcast on February 17, 2015 at noon Central.  Register here to attend .

We’ll cover some of the highlights from the book, take questions from the audience on their stickiest data visualization challenges, and offer advice on how to immediately improve your dashboards, visualizations, and analyses in OBIEE 11g.

Recognition of the importance of data visualization is growing every day in the business analytics community generally and in the Oracle business analytics audience specifically. We’ve seen the power of well-designed visualizations and how they deliver significant value to organizations through increased business insights, improved understanding of data relationships, and heightened engagement with business intelligence systems.


12c OBIEE is coming soon

My name is Arthur Dayton and I joined Vlamis Software Solutions in December this past year.  I just returned from the BIWA summit at Oracle’s headquarters in Redwood Shores, CA where I had the opportunity to hear directly from the Business Intelligence product development teams about their strategic objectives and what they will be working on in the next 18-24 months to achieve those objectives.

Having been a customer, and now working with customers, of Oracle it can feel at times like they aren’t hearing what you have to say about how the products can be improved to better satisfy customer demands and compete against other product sets that appear to be leapfrogging them with key functionality.  I’m happy to say that my perception of Oracle in this area and the reality are actually very different and that the development teams have been listening and are working very hard to improve on the end user experience inside of OBI and some of the pain and suffering involved with administering the development lifecycle across environments.

Visual Analyzer (VA) and Business Intelligence Cloud Services (BICS) were topics of emphasis as well as Oracle Transactional Business Intelligence (OTBI).  VA is a particularly exciting innovation coming in the 12c version of OBI and addresses many of the difficulties associated with end users ease of use, ease of creating visualizations and most excitingly for end users the ability to quickly and easily combine external data sources with existing OBI data!  This addresses a fundamental truth that a purely centralized data architecture, that often applies the 80/20 rule to ETL development, has ignored and, in my opinion, is a driving force behind slow adoption rates.  The heavy lifting that IT has done to give the end users 80% of what they need is only valuable in the context of the 100%.  So if we can’t enable the users to finish the work that IT will likely never get to, or can’t respond to in a timely manner, then OBIEE is reduced to just another way to barf data into Excel forcing the end users to do the rest of what they need offline.  This takes  all of the amazing capabilities of the BI Server out of the picture and tanks the ROI of the system because there is no way to bring the “finished” data back online without another long IT project that will begin promptly in the next 8-12 months.  The historical justification for this approach has been that we can’t possibly maintain a system that we let the end users add data to so hence we end up maintaining a system that the end users won’t use.

The end users have spoken.  What they do want from IT is vetted, reliable and easy to access enterprise information with the capability to deliver that data across distance, device and platform.  What they don’t want is a loss of agility and the ability to be responsive and innovative in order to get it.  In essence they require a “data development kit” that gives them a flexible means to satisfy their information needs using a common standard.  In the same way we in the development community have become accustomed to utilizing open source libraries that can extended to accommodate our particular needs the data consumption community has demanded the ability to do the same.

With 12c, end users of OBIEE will find it much easier to take advantage of the huge efforts that have been put into creating a curated source of the truth, and all of the wonderful visualization and enterprise delivery capabilities of OBI, without having to sacrifice the ability to add data that didn’t make the cut for the data warehouse team.  I know this will come with its own challenges and to some it may appear as a return to the old decentralized approach to BI that leads to maintenance nightmares and a divergence from organizational coherence.  But we must find a way for users of BI systems to have the best of both worlds that takes advantage of highly curated data in combination with user generated data sources.  Necessity is the mother of invention after all and if we are to survive as BI professionals we need to adapt to the reality of the market place.  Our end users live in a brave new world where the ability to capitalize on information is increasingly the differentiator between success and obsolescence.  Data development cycles that require months or years to apply simply will not be acceptable in this new world and I believe that Oracle has gotten that message and will deliver the tools to its customers to accommodate that reality.