Using a real-world example of a particularly challenging categorization task, we'll look at data-layer solutions for several different approaches to categorizing, cataloging, organizing, systematizing information. Data-modeling geeks, bring your thinking caps! Once we define the problem space and some terms we'll use for discussion, I'd like this session to be very participatory.
I'll try to keep our discussion grounded in concrete reality—using examples from research, medical records, standards compliance, and product formulation—but the issues involved apply to many domains of information. I'll also illustrate approaches that fail, using examples from multi-national corporations and local/county/federal government.
Our holy grail will be a single, elegant data model that is equally well suited to several major types of categorization tasks, but at worst maybe we'll agree on a best model for each of several different categorization systems.
Host: Allen Poole
cdsnw.com
Comments (3)
L. Allen Poole said
at 11:04 am on Jan 22, 2010
Intro slides available here:
http://pauseonerror.pbworks.com/f/TaggingVsLinnneus750.mov
L. Allen Poole said
at 2:24 am on Jan 15, 2010
Great! There are some interesting UI challenges to explore as well. For example, a hierarchical taxonomy can sometimes be very elegantly modeled in a recursive structure, but Filemaker's UI features don't make it easy to help a user navigate/query/report recursive data.
John Sindelar said
at 11:28 pm on Jan 14, 2010
Very much looking forward to this Allen. I'm especially interested in the 'costs' of different modeling approaches: what does a given approach do to importing records, to searching, etc.
You don't have permission to comment on this page.