iaidq.org

Sitting in Bonnie Neill’s session I was interested to hear her discussion of the importance of semantics and the problems that arise due to people using terms in the way they understand them (or not) compared to the way others understand them.

In the first 10 minutes she made me think about pornography. Bonnie talked about how in our organisations we have people talking about ‘dirty data’ but it is not clear what the definition of not dirty data is. And this reminds me of how an American judge, Potter Stewart once defined pornography…

“I can’t define it, but I know it when I see it”.

Bonnie identified miscommunication as a key issue.. we are not precise communicators and this leads to a universal problem of miscommunication.

A root cause is acronyms, geographic regions (local idioms and also how organisations manage across geography), slang, mergers of companies (different cultures and different short-hand). Also, terminology gets frozen in a system which causes the business process to freeze at the level of thinking and understanding that persisted when the system was first built.

What interested me about this part of Bonnie’s presentation was how it echoed and reinforced much of the themes of the presentation I will be doing with Arnt-Erik Hansen later in the week as well as many of the themes i’ve raised in articles in the IAIDQ newsletter over the past few years.

It is nice to know that I’ve not been alone in my experience of this point of pain, but equally it is interesting to note that the EXACT same awareness of the challenges of language in information quality exist both at the level of selling the idea to senior management to get them supporting the effort as well as at the level of actually doing the work.

Business rules are also an issue – they are often not articulated but are locked in people’s heads… which leads to private sets of business rules… this leads to a communications barrier due to the assumptions made by individuals. This reminded me of much of the theme of Walid el Abed’s tutorial yesterday.

Bonnie also pointed out that meaning often gets buried in the system. Because each system has its own system of meanings, it makes it harder (if not impossible) to share information across systems … or across organisations… or between organisations.

Exposing the terminology helps solve the problem. Good definitions in a data dictionary spell out the problem.

There was lots of sharing of war stories about data dictionaries and our experiences of defining information… I think the summary of that part of the session is “grrrhhh”, because that’s a phrase we all used.

As I thought about the topics raised by Bonnie, I couldn’t stop thinking about a particular marx brother’s skit. I dug it up on Youtube and might (technology and time permitting) make reference to it in my own presentation… but here it is for now..

Bonnie went on to talk about derivations (derived terms and derived values)… unfortunately single definitions are difficult if not impossible achieve, hence synonyms etc. need to be documented.

Bonnie also stressed the need to exercise stewardship… in otherwords getting the name of the business person(s) who defined a term and getting them to own it. This is consistent with good governance practices and sound principles of quality management.

What is an address? Grant Robinson had a great example about his experiences in water management.

Term confusion is also a key issue.

Unfortunately I had to step out of Bonnie’s session early (the first coffee break) to help give the kiss of life to a colleague’s laptop so I missed the rest of her session. I hope others will fill in the gaps in what I missed.

This entry was posted on Tuesday, September 23rd, 2008 at 6:35 pm and is filed under Blogging the Conference, IDQ2008, Using Dictionaries and Glossaries as the Basis of Data Quality. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.