by Charles Roe
In an effort to leverage the knowledge of several of the top minds in the Data Management industry, DATAVERSITY™ has been conducting a series of interviews on some of the most relevant topics in the field today. Recently, we interviewed Michael Scofield, an Assistant Professor at Loma Linda University.
Michael will be giving a presentation at the Enterprise Data World 2013 Conference in San Diego, CA from April 28-May 2, 2013. The presentation is titled “Graphic Techniques for Profiling & Understanding Production Data.”
The Speaker Spotlight Column (and its parallel venture the Sponsor Spotlight Column) is an ongoing project that focuses on highlighting several of the central issues represented at the many Data Management conferences produced by DATAVERSITY.
The primary emphasis of the interview was to question Michael Scofield on his work and history within the industry, with particular importance on his presentation at the upcoming conference:
DATAVERSITY (DV): Please tell us a little about yourself and your history in the industry e.g role at previous companies and/or universities (as opposed to job title), past experience and how you got started in the data profession and currently as a researcher?
Michael Scofield (MS): After getting my MBA degree at UCLA, I spent most of my career in the private sector doing decision-support, data architecture, and eventually evolving into data quality assessment. At Loma Linda University now, I teach part-time in Health Information Management (a field which used to be called “medical records”, but now is automated). I am also doing “institutional research” for another university. In higher education, that role is quite specific, providing the leadership with an understanding of student behavior coming from the enrollment database.
DV: What’s the focus of the research you are currently engaged in?
MS: We are wanting to mine our data for what factors may contribute to student success, or be predictors of the same. The Provost is also very interested in where the students will be coming from five years from now. I will eventually branch out into analysis of the cost of various academic programs, and determining how successful the students are at getting rewarding careers after graduation.
DV: What is the biggest change going in data architecture at this time?
MS: I think data architecture, as a discipline, has been a “forgotten child” in the rush to build new applications faster, especially with the popular “agile development”. I believe that “agile” runs quite contrary to the tradition and culture of thoughtful (but not paralyzing) analysis, and design of a stable logical data model which will serve the enterprise (or function) for a long time into the future. Fast development reflects a shortened attention span in the executive suites all over America. No so much outside the U.S.
DV: How does such a change affect your research?
MS: My immediate research is unrelated to data architecture.
DV: How do you think such a change will affect the industry over the next few years?
MS: Given my direct experience, and close observations (from a safe distance) of a number of for-profit companies which have had spectacular failures in custom-developed applications, I think that such failures will continue as long as CIO’s do not understand the nature of data as an asset, data architecture as an enterprise asset, and as long as CIO’s focus more on project completion without regard to quality and survivability of the applications being built.
DV: What are you going to discuss during your session at Enterprise Data World and what will the audience gain from attending your talk? (Please be specific about one or two issues you’ll be addressing, and the benefits the audience will obtain).
MS: My immediate topic at EDW is how graphic techniques can be employed to understand production data behavior, and spot quality problems in that data. In particular, over linear dimensions (the most common being time) a definitional discontinuity may undermine the usefulness of data being fed into a data warehouse. Understanding production data is much more than merely collecting physical structures and column definitions. It requires understanding the completeness, reasonability, and reliability of production data generated by business activity.
DV: How is Big Data going to affect data architecture in future?
MS: That is difficult to discern. Data architects may have to be more agile, and figure out how to integrate the architectures of multiple, disparate sources.
DV: What is something noteworthy about yourself that you would like to tell the conference attendees and our readers that they may not know?
MS: Something about me? I am interested in too many topics for my own good. I am an intellectual hedonist; just soaking up learning in history, science, and other topics. My latest lecture for general audiences is about Eleanor Roosevelt, and her activities during World War Two. Has nothing to do with data management, but it is enormous fun.
If you are interested in attending Michael’s presentation at EDW2013, please see the conference schedule at: http://edw2013.dataversity.net/agenda.cfm?confid=72&scheduleDay=PRINT
The presentation is on Wednesday, May 1, at 8.30am.
About Enterprise Data World:
Enterprise Data World is the business world’s most comprehensive educational event about data and information management. Over five days, EDW presents a diverse schedule of programming that addresses every level of proficiency, including keynotes, workshops, tutorials, case studies, and discussions.