Ambient Findability

I found this book both thought-provoking and annoying. The author flits about various topics related to the issue of findability, discussing truly interesting things such as wayfinding in animals and humans, in nature, cities and the world wide web (Chapter 2). I also enjoyed his discussion of information retrieval in Chapter 3, where he explains the metrics of precision and recall and demonstrates how Zipf's Law means that precision and recall will both drop when the number of documents increases.

On the other hand he seems to like mentioning recent popular topics, with only the most oblique relevance to the theme of findability. For example, in the last chapter he discusses Gladwell's _The Tipping Point_ and _Freakonomics_ by Levitt and Dubner and their differing explanations for the fall in crime in NYC in the 1980s and 90s. All this to show that there are uncomfortable truths (e.g. Levitt's theory that abortion was a major factor in the fall in crime) and that some people will avoid or ignore this kind of information, and therefore "[t]he power of our culture and our surrounding information environment to mold us is nothing new". Eh? His discussion of the Semantic Web in Chapter 6 is similarly annoying.

Reading all this, I felt as if Peter Morville was trying to include all the most recent themes of interest in the "noosphere" and "intertwingle" (a Morville neologism) them in this book. But the different themes are never tied together nicely, at least not to my satisfaction. Perhaps this is because there are as yet no answers to the questions he is asking. That's probably the best way to think about this book: it's a book that asks interesting questions, but provides no answers. Therein lies its attractiveness and its enervating-ness.

*Contents*

*Chapter 1 - Lost and Found*: what is findability, why it's going to become ambient, how findability increases business value (National Cancer Institute website example)

*Chapter 2 - A Brief History of Wayfinding*: animal wayfinding (strategies: geocentric, egocentric, use of different senses, cognitive maps, collaborative wayfinding - honeybees), human wayfinding in natural habitats (similar to animals, using various subtle clues, plus use of tools including maps. Navigating the built environment - Kevin Lynch's _Image of the City_. Bad wayfinding design can lead to people dying. We apply wayfinding metaphors to the WWW:

bq. 'We us a mix of _trajectory_ metaphors (e.g., "I went to the IBM home page") and _container_ metaphors (e.g., "I found that inside Yahoo!"). We construct cognitive maps. We remember (and bookmark) landmarks and anchor points. We traverse paths or clickstreams in search of information objects. And we often become lost and disoriented.' (p. 38)

(cont'd) But it's difficult to map the WWW. "These spatial visualization approaches fail because there's no there there." (p. 38) All the same, spatial metaphors have real value. "Findability is a bridge that spans the physical and digital worlds." (p. 39) The Baldwin Effect:

bq. 'Baldwin asserted that organisms could survive ecological challenges by relying on acquired knowledge and skills, often learned from others, and that this may then channel antural selection to favor unlearned versions of the same behavior.' (p. 41) We can't rely on evolution: it takes too long. We have to rely on our intelligence, our languages, our culture, our ability to learn to navigate the noosphere.

*Chapter 3: Information Interaction* Mooers' Law: "An information retrieval system will tend _not_ to be used whenever it is more painful and troublesome for a customer to have information than for him not to have it." (p. 44) Information retrieval, precision and recall, why precision and recall fail dramatically when there is a large # of documents in the system. Metadata can help: Google combined popularity, content and metadata, which helped a lot. The people problem - relevance is highly subjective - how do we measure precision and recall when we don't know what's relevant? Evolutionary psychology - information sharing behaviour (i.e. gossip) is very important. Information innovators such as Amazon, Google, Flickr and delicious "tap into the gift of gossip and the power of popularity to inspire participation and improve information retrieval". Perhaps information interaction is a better term. Information seeking behaviour.

*Chapter 4: Intertwingled*: everyware, wayfinding 2.0 - GPS, geocoding, findable objects. Ethics of making everything - including humans - findable. RFID, satellites and surveillance cameras, exporting information to objects in our environment (e.g. the ambient orb).

*Chapter 5: Push and Pull*: The Web allows you to pull information on anything you want at any time without pushy salespeople. But push is important to keep people balanced, and to kick in inspiration for pull. Powerful forces today - marketing and design. Balancing push and pull in these areas, including homepages. Personalization and why it fails: the ambiguity of language, the paradox of the active user, the ambiguity of behavior, the matter of time, the evolution of need, the concerns of privacy.

*Chapter 6: The Sociosemantic Web*: the Semantic Web, making use of metadata, faceted classification, ontologies, taxonomies, folksonomies, social network analysis. Data becomes metadata (Amazon, Google). Popularity metadata is shifting power from the author to the reader.

*Chapter 7: Inspired Decisions*: Decision making traps, _Blink_, the web helps make better decisions by giving you more sources and more data. The pathology of information overload and how it effects the decisions we make. Digital libraries. (read less)