The DataFlux Community of Experts is a forum for industry thought leaders to provide perspective and insight and engage in discussions on issues surrounding data governance, data quality, data integration and master data management. Our regular contributors include:
- David Loshin, president of Knowledge Integrity, Inc. and author of Master Data Management and the soon-to-be-released The Practitioner's Guide to Data Quality Improvement
- Joyce Norris-Montanari, consultant and president of DBTech Solutions
- Jim Harris, blogger-in-chief at Obsessive-Compulsive Data Quality
- Dylan Jones, founder of Data Quality Pro and Data Migration Pro
- Charles Blyth, MDM blogger from CharlesBlyth.co.uk
- Phil Simon, consultant, author of two books including his most recent title, The Next Wave of Technologies: Opportunities from Chaos , and blogger at Phil Simon Systems
- Rich Murnane, data architect and enterprise data operations manager, iJET Intelligent Risk Systems
- Robert S. Seiner, president and principal consultant of KIK Consulting & Educational Services, LLC, and publisher of The Data Administration Newsletter (TDAN).
Application Designers Need to Bake Data Quality In
Posted by Dylan Jones in data quality on July 30th, 2010
Take a look at the vast majority of article inches, conference presentations and blog posts devoted to data quality and you’ll find an interesting trend – they’re mostly focused on managing data quality after the systems, business processes and data have been created.
As an industry we tend to ignore the most critical aspects of data quality, which is designing quality measures into applications, so that we make it far more difficult for defects to arise in the first place.
Talking About Data, Part I
Posted by Phil Simon in data quality on July 29th, 2010
In this first in a series of posts, I”ll cover the art of speaking about data-oriented matters. I’ll try to debunk the myth that all of us who deal with data on a daily basis are, by definition, poor speakers.
The stereotype exists for a reason, though. Some of us “data folk” just don’t speak very well. I have seen really smart people fail when presenting data-oriented findings. They’ve had great ideas, but have fumbled as they tried to find the right words or medium for their message.
Through the Data Gazing Glass
Posted by Jim Harris in data quality on July 28th, 2010
Through the Looking-Glass was Lewis Carroll’s sequel to Alice’s Adventures in Wonderland. Both books, as well as most of his other writing, are examples of the genre of literary nonsense, where both sensical and nonsensical elements are used to challenge the standard conventions of either language or logic, or both.
Although I am definitely a fan of literary nonsense, I am not a fan of nonsense when it comes to data quality (DQ) best practices.
Multidomain MDM, Entities, and Roles
Posted by David Loshin in data governance, master data management on July 27th, 2010
I have been continuing my thoughts about some inherent issues with the concept of multi-domain master data management. On one side the concept of “master” data implies only one instance of each entity within the master data environment. From the other side, the concept of multiple master data domains suggests a master data organization centered on representing specific data entity types such as customer, product, employee, part, vendor, member, etc.
Assessing the Enterprise
Posted by Joyce Norris-Montanari in data integration, data profiling, data quality on July 26th, 2010
Good luck getting an entire enterprise’s data needs and usage assessed in the near future! Isn’t that always the issue? The data…the data…the data – it is always about the data! In my last blog, I talked about how corporations want reporting and analytics quickly. In fact, some companies are implementing agile methodologies to speed things up!
If you are one of the companies who have not created an enterprise inventory, then continue reading this blog. You have to understand the data and computer systems for the corporation, but how can we do that quickly? My suggestions follow:
Some Practical Tips for Making the Invisible Visible
Posted by Dylan Jones in data governance, data quality on July 23rd, 2010
In some of my most recent expert Q&A sessions on Data Quality Pro, the benefit of “Making the Invisible Visible” has surfaced. What these interviewees were referring to was the effect that takes place when defective data, invisible to the naked eye, is finally brought onto the centre stage and laid bare for all to see.
Although these insights were from two totally different viewpoints, one from a data analytics perspective, the other from a data governance angle, both had the same result – jolting people into action.
Read the rest of this entry »
Continuous Partial Attention
Posted by Phil Simon in data quality on July 22nd, 2010
A few months ago on this site, I ran a series about some of the causes of data quality problems in many organizations. To shamelessly plug my own work, the three pieces are:
- What Clients Don’t Tell Each Other
- What Consultants Don’t Tell Clients
- What Clients Don’t Tell Consultants
While each post addressed a separate topic, there was an underlying theme to the series: DQ problems on data migration projects stem from the differing agendas and priorities of different groups. What’s more, sometimes DQ is overlooked altogether. The question tends to be less about whether DQ suffers, but to what extent.
Data Transcendentalism
Posted by Jim Harris in data quality, metadata on July 21st, 2010
Extending the topic of my blog post Predictably Poor Data Quality to include metadata, Beth Breidenbach wrote the excellent blog post Predictably Poor MetaData Quality, where she explained how “similar human motivations come into play for metadata quality as with data quality.”
“Human behavior,” according to Beth, “is both the root cause and the solution. Technology doesn’t cause or solve the data quality challenge. Rather, it’s a tool that exacerbates or aids human behavior in either direction.”
I definitely agree with Beth. I have always found it puzzling when organizations try to resolve people issues by applying more technology or simply better technology.
Only people can resolve people issues.
Conflicting Thoughts: Centralization and Modeling
Posted by David Loshin in data governance, data migration, data quality, master data management, metadata on July 20th, 2010
During the early days of computing, two issues that plague us today were not of any specific concern. The first concern is standardized data models and semantics, and the second is the question of centralization vs. distribution of data assets.
Let’s start with modeling and semantics: most early computing was targeted at batch processing of transactions, with significant encapsulation of a complete set of data elements within each record. Earliest approaches did not have much need for a model, rather, only a list of the data values required to execute each transaction and their maximum sizes. It would not be surprising to see similar data element lists in the files used for different applications, but when there was little overlap across line-of-business transaction processing, there would be no need for aligning those nascent models.
Speed to Market!
Posted by Joyce Norris-Montanari in data quality, master data management on July 19th, 2010
I don’t know about you, but my clients are demanding their analytical and reporting environments set up very quickly these days! There does not seem to be time to complete the proper design and documentation prior to programming. Are we truly back here AGAIN?