Statoil’s Lars Olav Grøvik wound up his keynote presentation at the Semantic Days conference (report on page 7) in Stavanger last month citing ‘Putt’s Law.’ This has it that, ‘technology is dominated by two types of people: those who understand what they do not manage, and those who manage what they do not understand.’ Having ploughed my way through the DAMA DMBOK Guide1 (review on page 3) I was left wondering, was this book written by the ‘understanders’ or the ‘managers.’
Travelling to conferences around the world I hear a lot of interesting and varied technical material. Unfortunately, not all presentations are created equal. Occasionally I think to myself, ‘what a lot of waffle; how can anyone get away with either a) such bland, motherhood and apple pie stuff or b) with such unsubstantiated claims?’ What is truly puzzling though is that, as I rant inwardly, or drift off into a jet-lag induced reverie, I may look around to see other people in the audience nodding enthusiastically in agreement!
I used to see this as evidence of the frailty of the human race. But it happens so often now that I am looking for alternative explanations. The most obvious one is that I have missed a crucial point, but I like to think that this is not always the case. I prefer to think that what is happening is that I am overly aligned with the ‘understanders’ side of the equation. My management credentials are more questionable. I wonder if the ‘nodders’ come from the other side of Putt’s duality, the non-understanding managers?
But why do these folks, wherever they are coming from, find stuff to agree with in vague talk? I think that this is due to how management speak language has developed. Those who study and write on management are constantly trying to derive rules and commonalities across different subject areas. Occasionally, as in accounting or law, this works, and a trans-industry discipline is codified, taught and practiced.
Elsewhere it is harder to find cross industry commonality. So the management gurus adopt a language that is deliberately vague. Words like ‘asset’ and ‘resource’ are preferred over old-style qualifications such as ‘drilling rig’ and ‘employee.’ The implication is that it doesn’t make any difference whether a ‘resource’ is a time serving domain specialist, a new hire, or an ‘outsourced’ individual working for a third party. This is a process of abstraction that hides such awkward granularity.
With a high level view, the debate, instead of getting harder, as you might expect, gets easier. At a suitable level of abstraction, when manager A makes a statement of a sufficiently abstract nature, then manager B can immediately agree with it—without there necessarily being any alignment between the two inner trains of thought.
Another observation is that at the intersection of business and IT (or another domain), words are invented, repurposed and shared. Technical terms used in one field are recycled in a different context. Their meaning evolves with time as does their currency. You wouldn’t get very far today using buzzwords from the 1990s. But if discourse and meaning ‘evolve,’ what is driving their natural selection?
Beyond the desire of the speaker to be up to date and ‘smart,’ the managerial tendency towards abstraction ‘selects’ words that cross domain boundaries. It is comforting to think that the same ideas apply in different places. Thus a term is used in a different context at the risk of meaning something rather different.
This process is a kind of linguistic entropy as words are re-purposed and used in a broader context, they move away from a precise meaning. Gaining nods, losing sense and fuelling games of buzzword bingo!
This process actually has quite a lot to do with IT, with master data, and, another great buzzword of today, ‘ontologies.’ I discussed some aspects of this in my ‘What is a turnip?’ editorial (February 2010). But I think that I missed a trick. The ‘solution’ to the turnip problem is not Wikipedia, it is Linnaeus! Instead of abstracting away, sharing words, sharing terminology someone needs to tie the thing down with some decent definitions.
My intent was to review the DM-BOK Guide in this context—to see if it nailed down the body of knowledge in a Linnaean sort of way onto which a science of data management could be built. But I was wrong-footed by the fact that the DM-BOK is actually volume II in a series of which volume I is the ‘Dictionary.’ So plan B is to order the dictionary and then to report back in a later review.
Preliminary findings from our reading of the DM-BOK (page 3) show that the Putt dilemma is everywhere and is at the heart of the data management issue. Is enterprise data management about ‘data’ or about ‘management?’
In our report from PNEC Volker Hirsinger offers more insight into these issues. Managing seismic data actually involves the opposite of abstraction. As Hirsinger shows, intimate knowledge of observers logs, navigation data, velocities, processing workflows and coordinate reference systems is required. There is more to it than ‘managing’ the large volumes of field data.
The risk of a DAMA-like approach is that, as specialist data managers join the generalists and dialog at a suitable abstract level, more warm feelings and possibly hot air will be generated than useful insights into the nitty gritty of technical data management.
I would think differently if a DAMA contained other users of technical data such meteorologists, nuclear and space scientists or microchip designers. Perhaps even some from the high performance computing brigade too—they manage seismic don’t they? That might make for a more relevant community. On which note it’s interesting that RESQML has adopted the NCSA/NASA HDF5 data standard for multi-dimensional data. I bet they didn’t learn that in the DM-BOK!
1 Data Management Association Guide to the Data Management Body of Knowledge. ISBN 978-1-935504-02-3, $74.95.
© Oil IT Journal - all rights reserved.