« More Is Metadata? | Main | Remember the Formalists? If you do, you're probably wrong. »

March 23, 2009


Steve Tuck, Datanomic

The obvious lesson to be learned from this analogy is that DQ (be they Don Quixote or Data Quality consultants) sometimes tilt their lances at Windmills; they put their efforts into fighting false enemies, often to the detriment of addressing the real challenges. If only they listened to the Sancho Panzas of the world they might understand the context of the data quality issues and focus on fighting the battles that matter - the ones that deliver real value to the business.

Jim Harris

I thought that this article was fantastic for two reasons.

First, although I started and ended my academic pursuits with the study of computer science, in between I was a literature major and Don Quixote was one of my favorites.

Second, I must admit that earlier in my career I occasionally failed the Sancho Panza test. As an “expert data quality consultant” enamored with my own knowledge and experience, I ignored the client’s Sancho Panzas and went tilting at windmills and my beautifully architected, wonderfully coded, elegantly implemented technical solution resulted in...complete and utter failure.

As my career advanced, I developed a useful Quixote/Panza split-personality that not only improved my perspective but also focused my efforts on understanding and collaborating with my client.

Daragh O Brien

An excellent post. It ties in well with a series of articles I wrote back in 2006 for the IAIDQ, and the theme of a lot of my conference presentations since then, about putting the Information Quality agenda back in the language of the business (ie in terms they'll understand) and also linking it to real objectives so that you can make the case for change.

Back in the day, unfortunately, all our articles were pdf'd, but the archives are being brought into HTML at the moment, so I'll post a link to them when they are available.

As Steve says, we need to ensure that we are putting data quality issues in context and only fighting the battles that really matter to the business - particularly in the current economic climate

Bob Lambert

Great post Paul. It seems to me the reason this is a problem for data more than process, software, and infrastructure is that data isn't as far along in the general scheme of things. While influential constituencies have built up around the other three, we're just now beginning to understand what it means to manage data as a key resource. Traditionally data is seen as integral to the software, and not considered on its own merits.

Data consumers typically haven't seen the sausage factory from which their data emerges, and until the field is further developed SP and DQ will continue their quest. Part of the answer will be embedding in the culture data concepts, just as the culture overall has begun to understand concepts like software modularity.

The Sancho Panza/Don Quixote is something that we can apply in so many aspects of our lives - thanks for your post!

The comments to this entry are closed.