IA Summit Program

17 February 2009

The IA Summit 2009 program looks to be stellar once again:

I’m particularly excited about the keynote speaker: Michael Wesch–a real live anthropologist and forerunner in the field of digital ethnography. In my talk on ethnography at the Euro IA conference in Amsterdam, I made the point that IAs need to be doing more ethnography. I wonder if he’ll make a similar point.

In this light, the “Evolve or Die” panel looks interesting to me too. So does “Strategies for Enabling UX to Play a More Strategic Role.” I can’t decide which presentation to go to for most of the other time slots–typical for the IA Summit.

Unfortunately, I have to miss Jesse James Garrett’s closing keynote speech. Arrgh. That may change, but I don’t think so.

Be there or be square.

I previously blogged on what I was calling rotating navigation–for lack of a better term. Mia Nothrop, from Razorfish, thankfully corrected me on the label. It’s widely referred to as a carousel or carousel navigation. See the Yahoo design pattern on carousel navigation.

Thanks, Mia.

The name of this blog–Experiencing Information–was inspired by two people.

  • First, Andrew Dillon is quoted as saying “data is stored, information is experienced.” He’s done a lot of work on the concept of information shape and document genre to show that how we encounter and interact with information is an important part of understanding information.
  • Second, the work of Professor Carol Kuhlthau has influenced my thinking on information seeking. In particular, she included emotions in her information seeking model. Actually, considering actions, thoughts, and emotions simultaneously–which Kuhlthau does–is a key part of my definition of “user experience” in general.

In 2004 I gave a presentation at the IA Summit in Austin, TX entitled Information Search Experience: Emotions in Information Seeking (ISX). This basically took the work of Kuhlthau and others and put forth a framework for a diagnostic tool that could be used in conceiving of information systems. A significant part of this model is that it takes user emotions into account. Parts of this framework have also appeared in other places, namely in an article in interactions and in my book.

Recently, Kuhlthau and company re-examined her framework for the Information Search Process (ISP) model. See “The ‘information search process’ revisited: is the model still useful?“. Not surprisingly, the authors (including Kuhlthau) find the ISP model to still be valid, even across different information seeking contexts:

The information search process model describes feelings, thoughts and actions in an information seeking task with a discreet beginning and end, where considerable construction of knowledge takes place. The description of the stages of affective, cognitive and physical experience of users continued to be found in this study. This indicates that the model continues to be a useful theoretical and explanatory framework for user studies in librarianship and information science.

In addition, the model continues to be instructive when designing user centred information services and systems. Its consecutive stages can form the basis for timed interventions in order to support users throughout the progress of a project. This research indicates that a crucial stage for interventions is the exploratory middle part where the formulation of focus is developed.

It’s this last part that has always interested me and that stands at the heart of my ISX model. But rather than necessarily taking Kuhlthau’s ISP stages as is, I suggest that you first need to uncover the seeking stages your particular user group actually goes through. This can be inspired from the ISP, but may have important variations. And of course the inclusion of actions, thoughts, and feelings in my ISX is directly inspired from Kuhlthau.

Not too long ago, I had the privilege of working with the good folks at the University College of London Interaction Centre (UCLIC). At the time, Stephann Makri was finishing up his doctoral work, and he visited LexisNexis to get feedback and valid a tool for designing information systems for legal researchers. See his forthcoming JASIST article on the method.

Stephann doesn’t include feelings in his tool, and it’s really more based on Ellis’ behavioral model of information seeking. But at a high level there are similarities between his approach, my ISX, and Kuhlthau’s ISP, I think. Basically, in conceiving of information systems, it’s helpful to understand the phases people go through in a systematic way. Then, you need to ensure that you match the features, functionality, and design of the system to support the user’s actions, though process, and feelings.

Sounds simple and obvious (which is good), but I’m not convinced project teams do this, and they certainly don’t do it explicitely. The tools (either mine or Stephann’s) provide important insight, I believe. But we’ve not used Stephann’s tool at LexisNexis, and I’ve never had success bringing the ISX model to a broader project team. (I’ve really only used it for myself.) So maybe as diagnostic tools the models are limited in their impact in practical settings.

Still, I think there is potential for theory to inform practice in this area. Maybe I need to revisit my ISX and formalize it (and package it) better.

Tagging, in general, leverages natural language. You tag as you would normally speak or use language in every-day life. And that’s a strength of tagging–one that makes it popular and scalable and usable and all that good stuff.

Why, then, don’t all tagging system use comma-separated tagging input mechanisms? There is nothing natural about these tags–either in entering them or using them:

Note that I’m complaining that there is no preferred term for these concepts, rather that the form that the systems require people enter tags in have caused people to find workarounds. Namely, entering compound tags (i.e., tags with 2 or more terms) isn’t how people naturally type. That you get multiple forms of tags for “New York City” is fine with me–a Good Thing even. But where there is a workaround, there is surely a fault in the system or workflow.

Consider this: “york” consistently appears in flickr’s top tags tag cloud. Really? Are so many people traveling to the lovely city of York in the UK and posting photos to flickr? Well, no. People are going to New York City and tagging their photos in a way that they would normally type: new <space> york <space> city, instead of the system-required newyorkcity or new_york_city or new-york-city. Here’s any example of a user doing that on flickr.

So, combined with the photos from the city of York that are tagged as such, “york” as a tag on flickr gets boosted up to a top tag.

WordPress does it right: you can simply type like you would an email or blogpost or whatever, using commas to separate compound tags.


Entering tags should always be comma-separated. Punkt.

Aufgrund der positiven Resonanz und zahlreicher Nachfragen veranstaltet NetFlow zwei Workshops mit mir auch in 2009. Die Workshops werden auf Deutsch gehalten.

Siehe auch den Flyer (PDF 64KB).

Die Teilnehmer Anzahl ist begrenzt. Du kannst dich online anmelden.

This isn’t exactly new, but I wanted to post on it anyway. Library Thing is trying to get an open-source-like project for classification for to replace Dewey going. It’s called the Open Shelves Classification.

Back in my library days, I might have joined in–but not now. There is a forum for the project on Library Thing, as well as a wiki.

While Ifind the initiative interesting and with some potential, I wonder how they’ll ever arrive at top-level categories or even facts. Good luck to them.


16 September 2008

My previous post pointed to activities at Google for automatically extracting quotes from books and linking them together. In his talk in this project, Google research Bill Schilit mentioned the Great Books of the Western World project and the Syntopicon. I had not heard of this, but it’s quite intriguing.

In a nutshell, Mortimer Adler, an American philosopher, and Robert Hutchins, president of the University of Chicago, headed up a project in the earlier 50s to index ideas across key books in the western canon. This index of 102 core ideas across 431 works is called the Syntopicon, or collection of topics, which spans 2 volumes. Here are the ideas that were indexed:

Volume I: Angel, Animal, Aristocracy, Art, Astronomy, Beauty, Being, Cause, Chance, Change, Citizen, Constitution, Courage, Custom and Convention, Definition, Democracy, Desire, Dialectic, Duty, Education, Element, Emotion, Eternity, Evolution, Experience, Family, Fate, Form, God, Good and Evil, Government, Habit, Happiness, History, Honor, Hypothesis, Idea, Immortality, Induction, Infinity, Judgment, Justice, Knowledge, Labor, Language, Law, Liberty, Life and Death, Logic, and Love

Volume II: Man, Mathematics, Matter, Mechanics, Medicine, Memory and Imagination, Metaphysics, Mind, Monarchy, Nature, Necessity and Contingency, Oligarchy, One and Many, Opinion, Opposition, Philosophy, Physics, Pleasure and Pain, Poetry, Principle, Progress, Prophecy, Prudence, Punishment, Quality, Quantity, Reasoning, Relation, Religion, Revolution, Rhetoric, Same and Other, Science, Sense, Sign and Symbol, Sin, Slavery, Soul, Space, State, Temperance, Theology, Time, Truth, Tyranny, Universal and Particular, Virtue and Vice, War and Peace, Wealth, Will, Wisdom, and World

It apparently took a massive team to read all the works, extract the ideas, and index them. And it cost a few million dollars. Apart from being completely impractical, there are obvious problems with the approach as well: a limited set of books, small scope of coverage (lots of philosophy), and bias from the people doing the indexing.

So, they are not my “great” books and not (necessarily) my key ideas. I didn’t see anything about music in the above list, for instance. And there’s a lot of overlap (where do Religion and Theology start and end?). Still, there are some interesting universal qualities to their collection of terms.

When you consider efforts like this–which started nearly 60 years ago–or things like the Science Citation Index from Eugene Garfield or the work of Paul Otlet, we see that a lot of the core ideas behind the new innovations in the digital era have predecessors in the offline world. I wonder what else we think we’re “inventing” anew that has already been thought of.


Get every new post delivered to your Inbox.

Join 73 other followers