Experiencing Information

A few months ago there was an interesting story in Smashing magazine that spotted some new trends in web navigation menus. By and large, the trends identified are seen from a visual design standpoint, including some style trends.

I’ve been noticing two other navigation mechanisms and styles that seem to gaining popularity. The first is what I call a section sitemap menu. This is basically a dynamic menu activitated on roll over or on click from a main navigation point. The layer that is revealed essentially shows a mini-sitemap for that section of the site. This is shallower structure for navigating and allows visitors to get an overview at a glance.

Here are three examples, from HP.com (which has a complete subnavigation in the menu), Philips.nl, and Otto.de (which allows browsing by different facets.

Both the HP.com and Philips.nl examples also integrate advertising and promoting into the navigation. Not sure how user-centered that approach is, but it’s surely a more seductable moment than a plain ad on the homepage. It probably overcomes banner blindness quite well, too.

The second trend I’ve notices is a double-column left-hand navigation area. Blogs sometimes have this. Here’s an example from Information Design Patterns. Or on Josh Porter’s Bokardo.com blog. I know I’ve seen more of this arrangement, but I don’t have more examples at this moment. They’re out there, though.

Let me know if you see any other trends out there.

I’d like to post some thoughts about presentations I saw at the Euro IA 2007 Conference. Already mentioned Are’s presentation.

Here’s a summary of mine, which is essentially the last slide in my presentation (available on SlideShare) that sums everything up:

• The cost of adding more information is noise. Don’t forget this when people talk about “unlimited shelf space” online.
• There are different types of sources of metadata to consider: user-generated metadata (e.g., tagging), technically generated metadata (e.g., entity extraction), and owner-created metadata (e.g., controlled vocabularies).
• There are also different types of structures of organization to give meaning and context to the metadata when you represent it: user-created structures (e.g., filtering tags for special interest groups), technically created structure (e.g., Google News page), and owner-created structures (e.g., a thesaurus).
• In the Long Tail, any and all types of metadata and types of structure are needed. Forget about the silly arguments that one will replace the other. Think of it as matrix with the types of metadata on the side and the types of structures on the top.
• Further, since niche markets fit the description of a bounded domain, and since traditional taxonomies and classification are often good strategies for organizing information in bounded domains, as Clay Shirky points out, AND as we move to a culture of niche markets, as Chris Anderson predicts, traditional IA and taxonomy will become more important.
• Additionally, niche markets are defined by the categories you create. Online, a “pile of information”–as David Weinberger says in Everything is Miscellaneous--begins and ends with the IA and organization you develop.
• IA in the Long Tail will be about second order design. You may not be able to customize each page or local navigation scheme. Instead, you need to provide people with the tools they need to make sense of information.
• This means a shift for IA to look at abstract, broader patterns of human information behavior and of information structures in a domain. Card sorting is great, but we need to go well beyond this. We need to look at users much more closely, as well as the inherent patterns of information in a domain.

Not the most practical talk I’ve given, but many people thanked for the talk and said it got them thinking. So it seemed to have been well-received.

I started using Daylife just after it first launched. Since then I dropped if from my “Daily Stuff” tabs in FireFox that I usually open simulatenously when I go online. I liked a lot of things about the service, but it just wasn’t something that I needed at the time.
Just having revisited the site, I noticed some major changes to the user experience. The huge, page-filling image that previously occupied the start page has been removed in favor of more genre-conforming elements for online news sites. And for the most part, the site is far less image-rich than before. In areas like “Celebrity” this is unfortunate, but overall it’s probably a good move. With more focus on text and links, Daylife should be able to better expose and leverage their algorithms and entity extraction. And I like the basic information design on the site, so it works well.

A next step for them might be to expose more user-generated content and metadata. Comments, blogging, tagging, etc., would set it apart from other similar news services. (Now, there’s a good quesiton: how can tagging be leveraged on a current awareness service, where articles come on and go off the radar in a matter of days? With no time to incubate a collection of tags, what do you do there?). But their API is already a huge step in the Web 2.0 direction, so I’m going to knock them.

Anyway, I’m going to give it another shot and add it back to my daily tabs.

The smart folks over at SlideShare came up with a powerful new service for the site: SlideCast. Haven’t used it yet, but it looks to be fairly simple and a good idea overall.

Personally, I never considered PowerPoint to be evil. It’s just another tool the communicate. Sure, it can be used wrong, and it has it’s own style of communication, but any medium does. SlideCasting looks like it will make posted slide decks much more powerful.

If you have experience with SlideCasting, let me know what you think.

Matt Hurst over at Data Mining: Text Mining, Visualization and Social Media points to this interactive map on the NY Times website to track political campaign funding. I guess I’m spoiled by the Trendalyzer mentioned in my previous post, but this level of interactivity is downright tame. It’s smooth and somewhat usable (although the banners and nav at the top of the screen cut of the date range controls so that I didn’t see them until I was done looking at it), but I craved for more interactivity and exposing relationships.

The thing I really wanted to somehow have the ability to overlay two or more candidates’ funding bubbles. Flicking between the two, Obama clearly gets more support from the Chicago area than Guiliani, for instance (which is no surprise). But what other interesting connections and relationships might also be revealed? How’s Barack stacking up to Rudy in NY? Or how about bubbles for Dems vs Reps? I don’t want to knock the NYT for doing this, but there just seems like so many other easy targets that could have made this so much better.

BTW, check out Matt’s blog for other neat things going on in the text mining and analytics realm.

There’s a interesting study in the February issue of JASIST about which elements are most important for determining credibility of news stories on automated news aggregator pages, like Google News. [1] Though the findings might be obvious (there’s nothing wrong with stating the obvious), the researchers point to three elements that are most important on such automatically created pages:

• The name of primary source from which the headline and lead were borrowed
• The time elapsed since the story broke
• The number of related articles written on the topic of the story

The researchers write: “…The findings from this study demonstrate that information scent is not simply restricted to the actual text of the news lead or headline in a news aggregating service. Automatically generated cues revealing the pedigree of the hyperlinked information carry their own information scent. Furthermore, these cues appear to be psychologically significant and therefore worthy of design attention. Systems that emphasize such cues in their interfaces are likely to aid information foraging, especially under situations where the user is unlikely to be highly task-motivated and therefore prone toward heuristically based judgments of information relevance. Navigational tools that highlight these cues are likely to be more effective in directing user traffic, as evidenced by early research on newspaper design (which highlighted the attention-getting potential of placement, layout, and color) and screen design (focusing primarily on typography and color…Finally, visualization efforts should focus on attracting user attention towards-and making explicit the value of-proximal cues instead of simply concentrating on visualizing the underlying information.”

This means to me that–even though the pages are automatically generated–there is still information architecture and information design that is critical to understanding and experience the information. Maybe machines won’t replace designers and there is a place for professions like IA in the future after all. Hmm…

[1] Sundar, S. Shyam, Silvia Knoblock-Westerwick, Matthias R. Hastall. News Cues: Information Scent and Cognitive Heuristics. JASIST 58(3): 366-378, 2007.

For decades, information science has developed and examined the notion of relevance in information retrieval (IR). By and large, the approach to measuring relevance has been rather technical. Recall and precision have been the two main measures:

• Recall looks at whether all of the documents relevant to a given query are returned.
• Precision measures whether only the relevant documents are returned.

To measure relevance, you first need to create a key. This is a list of matching documents in a given database to a given query. But this key is itself artificial and doesn’t take into account any of the significant contextual factors people employ when determining relevance in real-life situations. It’s made up ahead of time by group of people who themselves don’t have a real information need in a real IR situation.

Tefko Saracevic points to a broader model of relevance in his article Relevance Reconsidered [1]. This includes the notion of technical relevance, but takes a more holistic look at relevance accounting for information interaction in IR situations. In addtion to technical relevance, he adds other types to the mix:

• “Topical or subject relevance: relation between the subject or topic expressed in a query, and topic or subject covered by retrieved texts, or more broadly, by texts in the systems file, or even in existence. It is assumed that both queries and texts can be identified as being about a topic or subject. Aboutness is the criterion by which topicality is inferred.
• Cognitive relevance or pertinence: relation between the state of knowledge and cognitive information need of a user, and texts retrieved, or in the file of a system, or even in existence. Cognitive correspondence, informativeness, novelty, information quality, and the like are criteria by which cognitive relevance is inferred.
• Situational relevance or utility: relation between the situation, task, or problem at hand, and texts retrieved by a systems or in the file of a system, or even in existence. Usefulness in decision making, appropriateness of information in resolution of a problem, reduction of uncertainty, and the like are criteria by which situational relevance is inferred.
• Motivational or affective relevance: relation between the intents, goals, and motivations of a user, and texts retrieved by a system or in the file of a system, or even in existence. Satisfaction, success, accomplishment, and the like are criteria for inferring motivational relevance.”

A recent study in JASIST (July 2007) also shows that relevance is very situational and contextual [2]. The researchers looked at how people picked documents from random-ordered results lists from different search engines (Google, MSN Search, and Yahoo!).

“The findings show that the similarities between the users’ choices and the rankings of the search engines are low. We examined the effects of the presentation order of the results, and of the thinking styles of the participants. Presentation order influences the rankings, but overall the results indicate that there is no ‘average user,’ and even if the users have the same basic knowledge of a topic, they evaluate information in their own context, which is influenced by cognitive, affective, and physical factors.”

Cognitive, affective, and physical factors? Yikes. Recall and precision don’t look at any of these, yet these were found to be significant. So what does the traditional notion of relevance in IR really measure with recall and precision?

I believe there is a much broader context that needs to be considered–one that accounts for the entire information experience. Not sure what this is, but context and situation seem to trump recall and precision in real-world IR. Perhaps relevance isn’t even relevant any more in the online, ditigal world anyway. Perhaps we need a entirely new model for understanding how and when people select documents in IR situations.

[1] Tefko Saracevic (1996). Relevance reconsidered. Information science: Integration in perspectives. Proceedings of the Second Conference on Conceptions of Library and Information Science. Copenhagen (Denmark), 201-218.

[2] Judit Bar-Ilan, Kevin Keenoy, Eti Yaari, & Mark Levene (July 2007). User rankings of search engine results. JASIST (58, 9) 1254-1266.

Here’s something I’ve been thinking about for a long time and hope to work up into a presentation or story:

- With the advent of digital information available online, people pointed to how much more information there is than before. At first it was about the volume of information.

- But then others pointed out that it’s not the volume, it’s the access to information that changed. The information was previously available, we just couldn’t get to it.

- But really, you could get it if you had enough time. So my thought is that it’s not the amount of information or increased access to it, but the time it takes to find, use, understand, and experience information that has really changed.

This is an important aspect of Information Foraging Theory described by Peter Pirolli and Stuart Card: “We have argued that in an information-rich world, the real design problem to be solved is not so much how to collect more information, but rather, how to optimize the user’s time.” Foraging for information in the digital world is a trade-off between the perceived value of information and the time it takes to interact with and experience it.

Relevance, then, is also time dependent. Relevance guru Tefko Saracevic hints at this with the notion of Situational Relevance in a paper titled Relevance Reconsidered.

Perhaps the Time of Information needs more attention. Or is this so obvious that it doesn’t even need to be mentioned?

Spock is a new people-finding service available free on the Web. It is currently an invitation-only beta service, which means you must receive an invitation from Spock or a friend to sign up for the service. Apparently, their entity resolution technology is killer.

Any one get a login yet? I’ve requested one but am still waiting.

Scientists at the Walker Reading Technologies in Minnesota have an interesting new technique for improving online reading and comprehension. Basically, the human brain doesn’t deal with block text well. Instead, our eyes view text as if they’re peering through a straw. We only focus on a small area at once, the lines above and below can cause noise and distraction while reading.

Here’s the detailed study of the technique:
http://www.readingonline.org/articles/art_index.asp?HREF=/articles/r_walker/

Of course, we’re so used to reading block text, this might seem counter-intuitive. How could thousands of years of writing and printing be wrong? Well, physiologically block text is not the most conducive for humans to read.

Here’s an article about it with an example of the technique:
Live Ink offers better way to read text online
(Mark Coker, VentureBeat, May 10, 2007)

Be sure to check out the image of before and after formatting.

But do we really want all of our online texts looking like a haiku? For one, this would make pages many times longer. And printing would take reams of paper. So, solving one problem may cause others.

The interesting over-arching lesson from this, however, is that HOW text is presented affects how we read, understand, and interact with information. Information design is crucial to the user experience on many levels.

Maybe there’ll be a FireFox plugin to switch this kind of formatting on and off from your browser?