If the TimesMachine is the bookish offspring of the Grey Lady, then Madison is her quirky niece.
Awkward familial metaphors aside, the New York Times' new archive—named after Madison Avenue, once the embodiment of the U.S. advertising industry—is a tool built on a crowdsourcing platform that allows readers to peruse ads that appeared in print in the 1960s. And, if the peruser should so chose, she could, one at a time, label, tag, and transcribe those ads like so:
Created by the Times' Research and Development Laboratory's eight-person team and launched last week, the archive works kind of like Captcha: Because digitized advertisements can't be transcribed by computers, they need the help of the human eye. The point of Madison is to have readers be those eyes.
"As we were looking through our old issues, it really struck me that advertising provided this really unique view into our cultural history, and that was a different view from the one provided by the news content," Alexis Lloyd, creative director of the R&D lab, tells me. "Our crowdsourcing approach is a way of both exposing people to this part of our archives and then also having them contribute some of that metadata."
Beyond showing off the adds, and using readers to archive them, Lloyd says the simplicity of the tool can be expanded to allow for four main goals.
- To satisfy historians (and Mad Men fans).
- To build a potential data set based on the number of users and the identification of ads.
- To help advertisers understand their brands.
- To demonstrate Hive, the "modular, flexible, open-source platform" as Lloyd put it, the team developed for creating crowdsourced applications.
The last one, demonstrate Hive, is arguably the most intriguing—and forward-thinking. Madison could simply be a fun tool for history buffs and advertisers to check out, but it's the platform that Lloyd and her team seem most excited about, and hope can be applied outside of the Times.
"You could easily see things like financial data or satellite imagery that you might want people to help with in a citizen journalism effort," Lloyd says. "There are journalistic applications, but there are also applications for different organizations that have their own amazing archives that can't be searched in any way."
For now, Hive is simply a framework for the Times. And it's a nifty one, thanks to a small gamification element: As readers "find" more and more ads, they're rewarded with positive reinforcement in the form of arbitrary titles, from "Reader" at the beginning, to "Fledgling Finder" after 10 ads, and so on. "We wanted to give people some recognition for the things they've contributed," says Jane Friedhoff, a creative technologist at the Lab.
And in true crowdsourcing spirit, the team plans to release the code for Hive as an open source project, letting anybody use it for their own work.
What people use Hive for will depend on what content they need readers to sift through. The most obvious way for Hive to be applied outside of the Times may be in advertising, in which certain brands can explore their own archives and engage their consumer or audience base. In an interview last week with Ad Age, Lloyd hinted at the idea, saying, "Brands are clearly interested in their own history."
On the other hand, Hive could be applied straight to journalism. Users could live translate and verify news from foreign papers, or put their heads together to scan through images searching for people, places or objects. For another, a platform like Hive could open up avenues for publications to delve into archives—recent or farther back—alongside their readers, offering both a chance to better understand their titles.
Of course, that's all extrapolating from a tool the Times only just rolled out, but as Lloyd puts it, applying crowdsourcing was always the "initial goal" of the project. Where Madison engages users, TimesMachine exists in stasis. Madison creates an interactive cycle (users help out, then share their findings, attracting more users); TimesMachine let users search and peruse without including the next step.
But if Hive doesn't penetrate the advertising and journalism fields, that's fine too. According to Lloyd, Madison will eventually host ads from other decades as more users engage with the project. In the meantime, it exists as a treasure trove for classic ads, like this one from August 1969:
We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.
