Article on Metadata Collection Strategies

The challenge in information environments is to get complete, accurate and consistent metadata applied to information and data resources. In this article co-authored with Maish Nichani we share strategies to collect metadata that lower the reliance on people in supplying metadata. We cannot completely remove people from the equation but we can prevent them from doing additional work, and focus the role of people on the value added metadata that machines and environment cannot automatically supply. View the article at PebbleRoad here. Download as a pdf here.

4 Comments so far

Stephen Bounds

Nice article Patrick and Maish!

Incidentally, although I think you are providing very good guidance on a problematic area ... I can’t help but wonder if the age of the “document” is nearing its end.

By that I mean the A4 WYSIWYG text editor style document.  Regardless of how much we set up templates and tag bundles, the ability to File -> Save As ... and then misapply all that diligently gathered metadata in an entirely different context is difficult to prevent.

I don’t pretend to know what the fix is, but I think we’re going to see much stronger ties between source systems and their document-style artefacts into the future.  It may require some work to convince people to let go of their pixel-perfect desktop publishing aspirations though!

Posted on May 15, 2013 at 08:54 PM | Comment permalink

Patrick Lambe

Lol… I learned a long time ago that people are truly ingenious at finding opportunities to “mis"-apply an intent. It’s the task of design I think to at least try to catch up…

Posted on May 16, 2013 at 09:53 AM | Comment permalink

I also thought that the article was a good discussion of metadata and the need to collect it. One thing that bothers me though was that I felt there was an underlying assumption that people couldn’t be trusted with creating or consistently applying metadata to their own work. You would think the creators of a document would be most familiar with it and therefore in the best position to decide on what metadata is applicable or not.

With that in mind, do you know if there’s been any research that looked at explaining to staff why metadata is important and how they should approach tagging? And if so, how effective has this approach been?


Posted on May 24, 2013 at 08:28 PM | Comment permalink

Patrick Lambe

Hi Justin - people tend to be okay (within the variances that human beings love to exhibit) at assigning tags/metadata for themselves. What we are not good at (even when trained) is assigning metadata/tags for others - there’s lots of research on this. Even trained librarians have variable consistency - just to give you a flavour of how difficult it is to govern consistency, the new Resource Description and Access guidelines for librarians runs to 940 pages. I’m not aware of any research on the effectiveness of training about the importance of metadata, but from my own experience, a great deal of metadata “for others” is extrinsic to the task within which the document is being created and stored. This means greater cognitive burden on the person creating the document - who has to think not merely about their task, but also about others who may need to use the document. Where such burdens exist, you will always have instances of omissions, mistakes, over-hasty or over-frugal decisions. This is where technology assistance and smart design comes in.

Let me give an over-simplistic example. I book a hotel online, and complete all my personal details in the registration form. I turn up at the hotel and they give me a new registration form - blank - to sign. I skate through the details and miss things out because I’m annoyed I have to give this information again, I just want to get to my room. Smart design lessens the load, lowers the overall cognitive burden so that when people do need to think about metadata, they are doing so for the stuff that counts. Make sense?

Posted on May 27, 2013 at 11:00 AM | Comment permalink

Page 1 of 1 pages

Commenting is not available in this weblog entry.

Comment Guidelines: Basic XHTML is allowed (<strong>, <em>, <a>) Line breaks and paragraphs are automatically generated. URLs are automatically converted into links.