Monday, April 23, 2007
On being "pro-browsing"
In a recent Harper’s article discusses Rick and Megan Prelinger's library in San Francisco. Gideon Lewis-Kraus wrote,
“For the most part, it is easy these days to find what you’re looking for;
one hallmark of digital efficiency is that the more specific the query, the
most efficient the query-based search. Megan and Rick, however, would
like to help you find what you are not looking for.” (p. 48)
Instead of organizing by dewey decimal system, they opt for a system inspired partly by Aby Warburg, who had a burning desire to present a collection of objects such as books as a guided experience. Each book is thematically connected to its immediate neighbors in a section, and each section would lead to the next to encourage discovery and browsing. As Fritz Saxl, a student of Warburg, put it,
“…the books together – each containing its larger or smaller bit of
information and being supplemented by its neighbors – should by their
titles guide the student to perceive the essential forces of the human mind
and its history.” (p. 54)
I couldn’t help but apply Warburg’s “law of the good neighbor” to online music. Beyond bringing to mind jokes such as Cusack’s character in High Fidelity organizing his records in the order he bought them, this idea perfectly captures what a good collection should be: personal. Because personality is what is lacking in online browsing.
By contrast, a website like Amazon.com is eerily accurate in its recommendations, but not good at transitions and contextual relationships. It points out books that should be on my bookshelf based on what I have already bought, but probably not those that would best appeal to interests not easily expressed or quantified. This is the dark side of the “long tail” phenomena - we might be trapped by our own choices in a world that is forever of our making. If there is a failure in the online music shopping model, it because it works too well.
This is a different argument from the “paradox of choice” argument presented by Barry Schwartz, which states that people are confused by and cannot appropriately process choice. I claim that instead of being overwhelming to the average person, the technology that guides people to potential products (thus eliminating the “paradox”) borders on being too clinical. In a way, we have overshot the goal. Automated recommendations are accurate in guiding our tastes, but lacking in contextual information or the leaps of logic that a good librarian will routinely make. By comparison, online recommendation engines have us stewing in our own tastes.
When I need to be rescued from entirely-too-efficient music recommendations I go to a local record store. There, you’re never quite sure of what the result of your musical foraging will be. The only certainty is that you will find music there; the selection, organization, and presentation is up to the owner. If the owner is particularly savvy s/he will distinguish between “funk” and “soul” but more frequently they would both be funneled into “R&B.” Reggae and latin genres might all get lumped into “World.” What may be contiginous sections in one store might be fractured and re-arranged in another. Plus, you’ll feel like you’re looking through the dustbins of history, where the almost-could-have-beens and non-charting albums live. It’s a breath of fresh air from the paint-by-the-numbers world of online browsing.
Contextually there’s little connecting The Cosmic Jokers on Ohm, a German rock band, and Sandy Bull on Vanguard, a spacey folk artist. However when set side by side they have an uncanny common love for excessive effects, jamming, and side-long tracks. A fan of one would no doubt be interested in the other, but the likelihood of a that recommendation being made is slim. I would enjoy bossa nova, despite the fact that I have literally no knowledge about the genre. Browsing online record stores by genre only brings out the most popular in each category, not the most relevant to your tastes. A good record store will present these moments of surprise by its selection and sorting. It’s Warburg’s “law of the good neighbor” all over again. The mere fact that there’s so much knowledge online to sort through makes retrieving it more challenging. Similarity measuring algorithms are still, even in 2007, no match for the instinct and depth of knowledge of a talented librarian.
There is also a limit to what Amazon.com knows about me. It doesn’t know I enjoy funky covers by steel bands. I’ve never searched for them, nor would they have any in stock if I did. I picked up one of my favorite records at a dusty Florida flea market for a dollar for the interesting cover and track selection. The LP featured a notable Isaac Hayes cover that was simultaneously funky and strangely mournful, and started me searching for the cream of the steel band records. After intense searching through several hundred limp covers of “Yellow Bird,” I found some gems: Amral’s Trinidad Cavaliers Steel Orchestra (phew) doing Gwen McRae’s “90% of me is you,” The Dutch Rhythm Steel & Show Band covering “Down by the River,” and St. Peter & Paul C.H.S. Shooting Stars Steel Band (is there some rule that steel bands must have long names?) doing tragically bad versions of “What’s Goin’ On” and “Kool is Back.”
To be able to find these records for me an automated system would have to incorporate semantic tagging or be an exceptionally intuitive neural network, to say nothing about the scarcity of these titles and how they would be input. The commercialization of the online space also presents certain restrictions on what is available. A few months ago I heard an album by the Lijadu Sisters. My reaction to the album couldn’t be predicted. Although I enjoy Fela Kuti, it would be difficult to extrapolate this preference to an out-of-print album by identical twins on a different label, if for no other reason than it is not commercially available. No sale, no price, no reason for a music website to tell you about it.
Searching and browsing in the real world is still a necessary and worthwhile pursuit for the active mind. The presence of infinite answers online may lead to a lack of patience people have for looking and browsing. When a world of knowledge is at your fingertips, it’s easy to forget it wasn’t always this way, and for the most part, still isn’t; for every record or book that has been digitized, there are dozens, if not hundreds, still on shelves. Guy Kawasaki calls increasing your breadth of information intake “perching in other trees.” The human mind has an amazing ability to draw connections. But if we just sit in the same tree, analog or digital, we’ll never get the opportunity.
“For the most part, it is easy these days to find what you’re looking for;
one hallmark of digital efficiency is that the more specific the query, the
most efficient the query-based search. Megan and Rick, however, would
like to help you find what you are not looking for.” (p. 48)
Instead of organizing by dewey decimal system, they opt for a system inspired partly by Aby Warburg, who had a burning desire to present a collection of objects such as books as a guided experience. Each book is thematically connected to its immediate neighbors in a section, and each section would lead to the next to encourage discovery and browsing. As Fritz Saxl, a student of Warburg, put it,
“…the books together – each containing its larger or smaller bit of
information and being supplemented by its neighbors – should by their
titles guide the student to perceive the essential forces of the human mind
and its history.” (p. 54)
I couldn’t help but apply Warburg’s “law of the good neighbor” to online music. Beyond bringing to mind jokes such as Cusack’s character in High Fidelity organizing his records in the order he bought them, this idea perfectly captures what a good collection should be: personal. Because personality is what is lacking in online browsing.
By contrast, a website like Amazon.com is eerily accurate in its recommendations, but not good at transitions and contextual relationships. It points out books that should be on my bookshelf based on what I have already bought, but probably not those that would best appeal to interests not easily expressed or quantified. This is the dark side of the “long tail” phenomena - we might be trapped by our own choices in a world that is forever of our making. If there is a failure in the online music shopping model, it because it works too well.
This is a different argument from the “paradox of choice” argument presented by Barry Schwartz, which states that people are confused by and cannot appropriately process choice. I claim that instead of being overwhelming to the average person, the technology that guides people to potential products (thus eliminating the “paradox”) borders on being too clinical. In a way, we have overshot the goal. Automated recommendations are accurate in guiding our tastes, but lacking in contextual information or the leaps of logic that a good librarian will routinely make. By comparison, online recommendation engines have us stewing in our own tastes.
When I need to be rescued from entirely-too-efficient music recommendations I go to a local record store. There, you’re never quite sure of what the result of your musical foraging will be. The only certainty is that you will find music there; the selection, organization, and presentation is up to the owner. If the owner is particularly savvy s/he will distinguish between “funk” and “soul” but more frequently they would both be funneled into “R&B.” Reggae and latin genres might all get lumped into “World.” What may be contiginous sections in one store might be fractured and re-arranged in another. Plus, you’ll feel like you’re looking through the dustbins of history, where the almost-could-have-beens and non-charting albums live. It’s a breath of fresh air from the paint-by-the-numbers world of online browsing.
Contextually there’s little connecting The Cosmic Jokers on Ohm, a German rock band, and Sandy Bull on Vanguard, a spacey folk artist. However when set side by side they have an uncanny common love for excessive effects, jamming, and side-long tracks. A fan of one would no doubt be interested in the other, but the likelihood of a that recommendation being made is slim. I would enjoy bossa nova, despite the fact that I have literally no knowledge about the genre. Browsing online record stores by genre only brings out the most popular in each category, not the most relevant to your tastes. A good record store will present these moments of surprise by its selection and sorting. It’s Warburg’s “law of the good neighbor” all over again. The mere fact that there’s so much knowledge online to sort through makes retrieving it more challenging. Similarity measuring algorithms are still, even in 2007, no match for the instinct and depth of knowledge of a talented librarian.
There is also a limit to what Amazon.com knows about me. It doesn’t know I enjoy funky covers by steel bands. I’ve never searched for them, nor would they have any in stock if I did. I picked up one of my favorite records at a dusty Florida flea market for a dollar for the interesting cover and track selection. The LP featured a notable Isaac Hayes cover that was simultaneously funky and strangely mournful, and started me searching for the cream of the steel band records. After intense searching through several hundred limp covers of “Yellow Bird,” I found some gems: Amral’s Trinidad Cavaliers Steel Orchestra (phew) doing Gwen McRae’s “90% of me is you,” The Dutch Rhythm Steel & Show Band covering “Down by the River,” and St. Peter & Paul C.H.S. Shooting Stars Steel Band (is there some rule that steel bands must have long names?) doing tragically bad versions of “What’s Goin’ On” and “Kool is Back.”
To be able to find these records for me an automated system would have to incorporate semantic tagging or be an exceptionally intuitive neural network, to say nothing about the scarcity of these titles and how they would be input. The commercialization of the online space also presents certain restrictions on what is available. A few months ago I heard an album by the Lijadu Sisters. My reaction to the album couldn’t be predicted. Although I enjoy Fela Kuti, it would be difficult to extrapolate this preference to an out-of-print album by identical twins on a different label, if for no other reason than it is not commercially available. No sale, no price, no reason for a music website to tell you about it.
Searching and browsing in the real world is still a necessary and worthwhile pursuit for the active mind. The presence of infinite answers online may lead to a lack of patience people have for looking and browsing. When a world of knowledge is at your fingertips, it’s easy to forget it wasn’t always this way, and for the most part, still isn’t; for every record or book that has been digitized, there are dozens, if not hundreds, still on shelves. Guy Kawasaki calls increasing your breadth of information intake “perching in other trees.” The human mind has an amazing ability to draw connections. But if we just sit in the same tree, analog or digital, we’ll never get the opportunity.
Labels: browsing harpers online internet amazon
Thursday, April 05, 2007
Obligatory Twitter post
I give up – you’ve got me, blogosphere echo chamber. With all the discussion about Twitter, I feel obliged to at least offer an opinion on it. So here goes:
I’m currently in the middle of a collaborative, qualitative study of undergraduates. We are using journals and focus groups to examine how younger individuals use multiple mediums or modes of interaction. A response that came up time and again is that taking the time to talk with another person is less common than texting, emailing, and using other less (graphically, contextually, information) rich modes. The causes are anybody’s guess: narcissism, lack of time, or too many acquaintances to take the time to interact in a meaningful way with all of them. But the trend is clear: individuals seem to prefer modes of interaction and technologies that allow them a large degree of control paired with convenience. Twitter tickles that sweet spot perfectly.
That said, despite other opinions, I do not believe Twitter will impact the majority of individuals’ social relations. Certain people with addictive personalities indeed might have difficulty using a service such as Twitter because it offers a near-instantaneous stream of information. However, Twitter isn’t any more responsive than, an Internet connection, which can facilitate all manner or near-instantaneous interactions, or text messaging. Perhaps it engages people who may otherwise not feel this style of technology is “for them,” but I am not sold on mobile technologies being the next major online movement.
When a new technology is introduced, critics are ready with proclamations of how it will drastically impact our lives. These claims include but are not limited to: television disassociates you from reality, Nintendo makes you dumb, and the Internet diminishes your capacity to connect with others. Rarely these warning come to bear in any significant fashion. It has been repeatedly shown that most people do not have difficulty processing large amounts of information; the excess gets ignored, or the information source avoided. In this and almost every case of a new commodity, the technology can be easily avoided. Cell phones, for all their connected-ness, still have an “off” button.
I’m currently in the middle of a collaborative, qualitative study of undergraduates. We are using journals and focus groups to examine how younger individuals use multiple mediums or modes of interaction. A response that came up time and again is that taking the time to talk with another person is less common than texting, emailing, and using other less (graphically, contextually, information) rich modes. The causes are anybody’s guess: narcissism, lack of time, or too many acquaintances to take the time to interact in a meaningful way with all of them. But the trend is clear: individuals seem to prefer modes of interaction and technologies that allow them a large degree of control paired with convenience. Twitter tickles that sweet spot perfectly.
That said, despite other opinions, I do not believe Twitter will impact the majority of individuals’ social relations. Certain people with addictive personalities indeed might have difficulty using a service such as Twitter because it offers a near-instantaneous stream of information. However, Twitter isn’t any more responsive than, an Internet connection, which can facilitate all manner or near-instantaneous interactions, or text messaging. Perhaps it engages people who may otherwise not feel this style of technology is “for them,” but I am not sold on mobile technologies being the next major online movement.
When a new technology is introduced, critics are ready with proclamations of how it will drastically impact our lives. These claims include but are not limited to: television disassociates you from reality, Nintendo makes you dumb, and the Internet diminishes your capacity to connect with others. Rarely these warning come to bear in any significant fashion. It has been repeatedly shown that most people do not have difficulty processing large amounts of information; the excess gets ignored, or the information source avoided. In this and almost every case of a new commodity, the technology can be easily avoided. Cell phones, for all their connected-ness, still have an “off” button.
