viz. - data collection

The Image of the City, Revisited: MIT’s Place Pulse Project

Lisa Gulesserian — Sun, 06 May 2012 14:00:00 +0000

Image Credit: MIT's Macro Connections Group

Last week, as my students in my Rhetoric of Suburbs & Slums class presented their final movie projects, I was reminded of how we often judge a place after only a cursory glance. One group project especially got me thinking: “The Divide,” a student-made film that explored the differences between East and West Austin, included many images from East and West Austin along with candid interviews of residents from both sides of the divide. My students’ video reminded me of MIT’s Place Pulse project, which in turn reminded me of Kevin Lynch’s seminal urban planning book from 1960, The Image of the City. As a culmination of my time blogging about cities the last few months on viz., I’m going to talk about “imageability” and intimacy in Austin (and beyond).

As coined by Kevin Lynch, “imageability” is “that quality in a physical object which gives it a high probability of evoking a strong image in any given observer. It is that shape, color, or arrangement which facilitates the making of vividly identified, powerfully structured, highly useful mental images of the environment.” An “imageable” city is one that is readily identifiable by its landmarks and landscape. Connecting “imageability” to our daily lives, we make decisions on where to go if we’re unfamiliar with a city by judging what we see in the moment we see it.

East Austin Fence — Image Credit: karenjeanette's flickr

In their video, my students interviewed West Campus residents about their views on East Austin. Many of the interview subjects mentioned “chain-link fences” and “refuse” as they were describing East Austin. My students then asked their interviewees if they’d spent much time in East Austin—their answers were often phrased as “No, because it’s unsafe.” Seeing chain-link fences and trash was a deterrent for these West Campus students to venture across I-35 (no matter that the city’s most dangerous areas—in terms of the highest occurrence of murders, aggravated assaults, and rapes—aren’t even on the east side of town!).

Image Credit: Screenshot of Place Pulse

But are these visual cues universal? A group of researchers at MIT’s Macro Connections Group has made it their goal to find out with Place Pulse. The group describes the project as “an attempt to generate quantitative data on aspects of cities that are hard to quantify, such as the effect that urban looks have on our perception of a city’s safety or our own perceived level of prosperity. To answer these questions we crowdsource the comparison of pairs of images that show randomly chosen urban landscapes.” When you visit the site, you see two images side by side, then are asked questions like “Which place looks more safe?” or “Which place looks more touristy?” or “Which place looks more livable?” The site is meant to emulate our experiences in unfamiliar places. A chain-link fence on an unknown city street might make us vote for the other place as “more safe.” Or an outcropping of flowers in someone’s front lawn might make us deem it “more livable” than its partner picture. An initial visual cue affects our opinion of a place, and Place Pulse helps track what kinds of cues stimulate specific reactions in urban environments.

Image Credit: Screenshot of Place Pulse

The researchers at Place Pulse have been collecting visual cue data for a little less than a year now. They’ve even started to open up their data set (and their site’s visitors) to independent researchers around the world. Now, you can set up a question, along with the types and locations of Google Maps images, to get answered by anyone on the internet.

Image Credit: Corbo E. on yelp

I’m curious to see what would happen if I were to ask “Which place looks more safe?” for two images of Austin—one of the Drag in West Campus, one of East Sixth Street on the East Side. To those not as intimately familiar with the wonderful trailers, bars, and artists’ studios on Austin’s East Side, the chain-link fence might signal “danger.” To me, that same chain-link fence signals “a perfect place to lock my bike (if the bike racks are already full) while I eat beet fries at the Liberty Bar.” The images of a city are key, but so are our intimate experiences of a place. As Kevin Lynch says: “We are not simply observers of this spectacle [of the city], but are ourselves a part of it, on the stage with the other participants. Most often, our perception of the city is not sustained, but rather partial, fragmentary, mixed with other concerns. Nearly every sense is in operation, and the image is the composite of them all.” At first glance, “image is everything.” But with a closer look, it’s not the whole story.

We're Watching You - Google Analytics and viz. Readership

Cate Blouke — Thu, 05 May 2011 23:34:32 +0000

Screen Shot of Google Analytics dashboard view for viz.

Earlier this semester, we here at viz. decided it might be nice to get a better sense of who we were talking to. As good rhetoricians, we felt it might be pertinent to know more about our audience. So we installed the handy-dandy (free) Google Analytics tool for tracking all sorts of information about our readership. The results have been surprising at times, reassuring on the whole, and ultimately quite useful for thinking about how better to serve you, our readers. The following is a discussion of both Google Analytics itself and the trends we've noticed on our site over the past few months.

Screen shot of daily readership graph

First and foremost, the Analytics tool will give you cold, hard numbers. You can track the number of visits, the time spent on the site, and whether the vistitors are unique or repeat customers. We have between 300 and 800 visits per day here at viz., and over 20,000 views each month. While we're pretty happy with these numbers, no self-respecting academic would object to increased readership. So we've been thinking about ways to broaden our audience and make the site more accessible/searchable - even though we've been pleased to discover that our readership stretches well beyond the continental United States.

Screen shot of the analytic map of viz. readership's geographic distribution

That's another thing Analytics can do - track from where in the world visitors are accessing the site. And, of course, being Google, it gets pretty specific...

Screen shot of the analytic map of viz. readership in the United Kingdom

Analytics will also tell us how people are making their way to the site. One thing we discovered is that our post titles could use some work - Google's search function picks up first on page titles, so the more specific we can be, the more likely it is unique visitors will discover our work. One of our most visited pages, is a post from last February that comes up when people search Google for "Diane Arbus." Analytics will tell us which pages on the site are getting the most hits, and how people are getting there - both in terms of the referring site and the keywords they use to find the page. While it will give us this information for the site as a whole, we can also narrow it down individual pages.

Screen shot of analytics source break-down for Ashley's post on athletes and body images

Another one of our top-trafficking pages, Ashley's post on athletes and body images, is garnering attention from a wide variety of sources. While we plan to work on increasing our Facebook and Twitter presence, our three most popular pages (at present) are an interesting indication of the ways in which we can better shape our content.

You may (or may not) have noticed some tweaks we've made to the page this semester, and we've got plans in the works for a more significant overhaul next year. One thing we did was move our tag cloud to the upper right hand corner of the page. As a result, one of our top three most visited sites is the page hosting blog posts tagged as "political cartoons."

We've become much more conscientious of tagging this semester, especially given Noel and Megan's work with the Steve-in-Action project, investigating the connections between composition and social tagging.

We also made it easier for readers to re-post articles by adding this widgit at the bottom of every page. We've seen some great results in terms of traffic from Facebook, Twitter, and other social networking sites where people are sharing our work.

While perhaps this shouldn't have been as surprising as we found it, visitors are accessing the static content of the site more than we had realized. Our teaching and visual theory pages are apparently being used fairly regularly, so we plan to continue to update those resources and make them even more useful.

In the four months since we began tracking this information, the launch page for visual theory has receieved nearly 1,000 visits - not bad, but it could be better. Did you even know we had a theory page available? It's in the menu bar at the top of the page, along with the teaching link that will take you to lesson plans our contributors have generated over the years, and the views link that hosts various interviews we've conducted with photographers, game designers, and other visual rhetoricians. We've got plans to revamp these pages next year, starting with the teaching page. Aside from our daily visual analyses, we plan to bring you more (specifically) pedagogically applicable content.

It's been a good year here at viz., and we're glad to know a bit more about who's out there. Thanks for reading, and we'll see you again soon!