Sunday, April 19, 2009

The Data Model of Dorian Gray

Mrs. AdvisoryBored and I spent the week at the Enterprise Data World conference talking data with friends both new and old. It is a joy to get together with others who want to discuss metadata, data quality, data governance and Twitter names beginning with the word data. Yes, the last one did occur in the wee hours of the morning and yes, it is possible that some consumption of alcohol was involved. Did I mention that it was a data management conference?

A more serious topic of discussion was the concern that data management has become a profession (job, role or set of activities) of older people. I tried to point out that young people such as myself were entering the profession. They were quick to point out that 50 didn't count as young. Who knew!

While I share that concern, I think that there are aspects of the profession that will always skew the age of practitioners toward the "experienced" end of the scale. That would include a focus on:
  • data, not on the computers and applications
  • an enterprise view of the organization, not an operational view
  • semantics, categorizations, meanings and definitions
  • specialized jobs (traditionally found only in larger organizations and consulting firms)
  • planning and coordination
I lost you at semantics didn't I?

Look, experience need not be synonymous with age, so if we want to bring younger practitioners into the field then we have got to do a few things different.

First we must teach the core competencies that serve as the foundation for data management training later. It parallels the effort to teach computational thinking skills like numbering systems or sorting algorithms without using computers at a young age so they are prepared to learn computer programming later, if they wish. These competencies - things like classification, abstract thinking, information literacy, computer literacy and set theory - ensures students are prepared to learn data management at a more appropriate time. Wax on, Wax off.

Second we must build awareness of data management as an area of study, as a set of skills and as a profession. Most of you are probably thinking database administration when I speak of data management. Fair enough, database administration is certainly part of data management, but so are data modeling, data warehousing, data security and any number of other jobs or roles. Our biggest challenge in this area, as Karen pointed out on numerous occasions at the conference, is to be visible to younger professionals by participating in the communication channels they use. This means we need to be talking data management in blogs (or this one), on Twitter, on LinkedIn, in discussion forums and on wikis.

Finally we must develop, or at least help to develop, curriculum to support the needs of new entrants into the field as well as the ongoing professional development of those who have already chosen data management as a career path. DAMA International made a huge leap in this area last week with the release of The DAMA Guide to the Data Management Body of Knowledge (DM-BOK). The challenge now is for data management professionals to carry this information to local colleges and universities. Additionally, we need to push beyond the basic database training to teach a broader range of data management activities. As Peter pointed out on several occasions, we teach students how to build new databases, but how often do we really do that in our professional lives.

Let me leave you with one final thought. Data management is both a set of activities and a profession. Many people who do the former don't consider themselves the latter. They are, nevertheless, contributing to the body of knowledge, for better or worse, and we need to connect with them just as much as an up-and-coming metadata analyst. Consider the outcomes that Mrs. AdvisoryBored identified in her original proposal for the Business Information Management class (MGMT 215) at Edmonds Community College (see the end of page two). There is no guarantee that the students, particularly those from the Business department, will embrace those data management principles and/or the profession, but she has at least had the opportunity to introduce the concept of data as a managed enterprise resource.

Tags: , , ,


Anonymous said...

What's your twitter handle? Are you Corey_smith? I'd like to follow you.
-Bonnie O'Neil

Rob Drysdale said...

Excellent discussion about the need to attract and develop "newbies" to the field of data management. I really like the point about giving them the fundamentals to move into the profession later even if they don't go there right away.

I think we also need to do a better job of marketing the profession to the business to drive up the demand for more data professionals. I don't think that most businesses understand the value of their data and therefore the need for data professionals especially when they are out buying and implementing ERP's. If we can get more people saying "we really need a data architect" rather than saying "who's that?" "what do they do for us?" we can really help get more people thinking about it as a career.

Karen Lopez - said...

You've touched on the real issues:

- we aren't growing many new data management professionals
- what few new ones we have are learning about it from random websites or authors who know nothing about databases and data management
- we aren't asking colleges and universities to cover these topics in order to produce professional-level knowledge and skills.

I've also seen that companies try to hire people who can cover all the rows in the Zachman Framework, down the data column. I think that is a huge mistake. We should be developing professionals along the same rows to span multiple columns. So we shouldn't be looking for the non-existent "Data Professional" who can and want to work only with data, from strategic planning to performance tuning scripts and schemas.

I don't believe that people are naturally drawn to work up and down a single column, but are normally draw to tasks that span columns and focus on a few rows (Plan versus Architect versus Build or Maintain).

Great post.

Corey Smith said...


I am indeed on Twitter. I am @heycorey on Twitter and would be happy to have you and anyone else interested in this topic to follow me there.

Fair warning, however, you may also hear about a favorite stout, worm ranching or other general observation on life that don't tie directly to IT and IT education.


Hula Betty said...

so I've thought about this... A LOT... Did a young copy repair man plan to shift into data management and meta theory? Was it the enticement of unlimited training possibilities, career advancement or free vendor T-shirts?

When name dropping Zakman or Holcman will get you past the bouncer at the club or a salient data descriptor ensures a date with a hot super model, than you'll have them lining up.

Until than, maybe the way to bring followers to the doorstep of data theory is to convince business of the data architect.

Project managers went through this and appear to have won... ITT Tech Institute now runs ads telling everyone the path to a good career is in project management. Of course now the market is flooded with paper lions who wave their PMM certs and take jobs from more capable practitioners for a 1/3 less in salary...

Ok so maybe that is not the way to lift Data Management above the roll of geeky theory spouted by old analysts around the table over a stout...

Maybe just figuring out how to make data management execution oriented with practical delivery to the bottom line will allow the theory and intellectual growth to ignite and consume those who can see past today's limitations.

Or not.