How fragmentation happens


Revision as of 19:13, 1 April 2011 by Dbalzer (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

From the TC 372 Workshop Compendium

So many databases ...

... and still more to come?

Over the past two or three decades, databases have sprouted in every corner. With little or no guidance from standards, we now often have dozens of film-related databases in a single institution.


Consolidating legacy databases is a time-consuming task.

Authentic statements from a past century:

"Why should creating a database be different from creating a filmography for a book?"

"If I do not share scholarly interests with my colleague, why should I share a database with him?"

"How can I stay in control of my data if anybody around here can change it?"

"My database should reflect my interests, not those of a committee."


HRH Princess Anne marvelling at a 1980s desktop computer.

Those were the days:

Personal desktop computers encouraged the creation of personal databases. Many of them still exist (and continue to grow).

Some of these databases are real gems of scholarship, reflecting thousands of hours of diligent research. However, integrating them with modern knowledge bases can require further hundreds of hours of manual work.

<mediaplayer image=''></mediaplayer>
From: Introduction to Biomedical Ontologies. A training course in eight lectures by Barry Smith. University of Buffalo, 2009 (License: "Free for use in any way")

Barry Smith is a philosopher working on the foundations for representing human knowledge in information systems.

In his introductory lecture to students of biomedical informatics, he addresses the issue of integrating hundreds, if not thousands, of databases created by biomedical researchers in the past decades.

• Previous: How EN 15744 and EN 15907 came into being • Up: Contents • Next: The case for reference models