Faceted Search Design – Ordering Facets

I’ve been lurking on the IAI’s mailing list for some time now, but recently someone posted a question that I just couldn’t resist answering:

We are getting ready to roll out a new faceted search option and I’ve been asked to make recommendations regarding the order of the facets and their characteristics.  I am having a hard time finding specific information about standards or best practices.  I repeatedly come across Stephanie Lemieux’s recent article, Designing for Faceted Search, stating that both facets and values should be based on importance.  While this is great, can anyone point me toward supporting information or is this something that is just understood?  Are there general guidelines for when to list characteristics alphabetically versus when to list them in descending order?

Ok, good call on the poster’s part – I had been vague in my article (Designing for Faceted Search, originally published in KM World), mainly because I had a broad audience and a word count limit. But I supposed that I should clarify a bit…  Here’s my response:

Academy of Motion Picture Arts and Sciences Metadata Symposium

I just finished moderating the Digital Motion Picture Metadata Symposium at AMPAS.  The day covered all aspects of metadata from pre-production through production, post, distribution and archiving.

We had presenters from Pixar, Sony Pictures, the Academy of Motion Picture Arts and Sciences, Marvel Studios,  Warner Brothers, CNRI, Gracenote and the Library of Congress.

We saw examples from productions including The Incredibles, Wall-e, The Curious Case of Benjamin Button, Syriana, Ocean’s Eleven and others.

The day was packed with presentations that addressed all aspects of the metadata lifecycle for Digital Motion Pictures.

Digital Motion Picture Metadata Lifecycle AMPAS Symposium

Podcast and session summaries coming soon.

Podcast on Folksonomy & Taxonomy in the Enterprise

I had the great pleasure of doing a podcast a few weeks ago with Paul Miller, podcaster for Nodalities (magazine & blog), on hybrid approaches to folksonomy and taxonomy and their role in the enterprise.

We discussed the now tired debate of folksonomy vs. taxonomy, and focused on the strengths and applications of each approach. I covered how organizations are leveraging social tagging and what some of the pitfalls are in the enterprise context.

I also talk a lot a few of the hybrid approaches to taxonomy & folksonomy:

• Co-existence
• Tag-influenced taxonomy
• Taxonomy-influenced tagging
• Tag hierarchies

I cover some interesting examples and tools (ZigTag, Flickr & Library of Congress), as well as the new directions in “intelligent tags”, like MOAT.

You can hear more about these approaches at the Semantic Technologies conference next week, where my colleagues Paul Wlodarczyk and Richard Beatch will be presenting on the topic on my behalf. Listen to the podcast…

Follow me on Twitter: @stephlemieux, @earleytaxonomy

Forest for the Trees: How Taxonomy Design Is Like Systems Engineering

Thanks to my wife, I’ve been learning a little bit about systems engineering, a form of engineering that addresses the complex interactions of multiple systems. As she says, you need to consider systems engineering when the interrelationships between systems are as complicated as the systems themselves. For example, to reduce automotive traffic you need to research social behavior, road design, business, and the environment. To study ergonomics you need to study the human body, computer design, application design, and user efficiency needs. And don’t get me started on the U.S. healthcare system.

The very first step of systems engineering is to understand the full scope. Ground transportation isn’t about cars and trains, for example, but about the entire surface of the earth: population clusters, topography, climate, and distances. Taxonomy starts this way too, with facets like people, documentation types, product lines, and access levels.

Or at least it should. Sometimes it’s easier to look at a single component of your world and tabulate it in isolation — department offices, account records, HR data — and imagine someday expanding to something larger. But this doesn’t work, not in the long run.

Can’t it just be like Google?

I often get frustrated by those who think Google is the greatest search engine that ever parsed. Don’t get me wrong – I like Google, I use Google, I employ it as a verb. But if I hit the search button and get wonky results, I recognize that they are wonky and am not afraid to blame Google. (Full disclosure: I have a library science background which I’d like to think has made me into a pretty good searcher, so I will usually try a few different queries before I point the finger at the machine.)

On most of our consulting engagements, at least one person will say “I want our search to be more like Google.” I have a few problems with this kind of statement. Partly it’s that most folks aren’t terribly critical when it comes to evaluating the relevance of Google results. It’s what we know, it’s what we’re used to. We don’t mind that Wikipedia is almost always the first result on any query – many might find that a “feature”.  We’re generally happy to take whatever shows up in those top 10 results and roll with it regardless of what it is, mostly because we can’t know everything that is out there so we trust Google to filter it for us. We satisfice (yes, it’s Wikipedia, joke intended)

Take these same folks and plunk them in front of their enterprise search and ask whether the top ten results are usually “good enough”… Most will answer no. Why? Because we have a better idea of what information is out there and what would make for a good result.