I'm currently reading Lindsay Poirier (@lindsaypoirier)'s dissertation—"Knowledge Representation in Scruffy Worlds: An Ethnography of Semiotic Infrastructure Design Work" (2018)—and just love the lead to Chapter 5—which I skipped to & had to share after reading the intro to it in the Intro:
Ch. 5. Catachrestically Designing Semiotic Infrastructures for the Human Services On June 9, 2016, Greg Bloom was the third speaker on a panel entitled "Solutions at Hand" at the Personal Democracy Forum - a conference that brings together technologists, hackers, government officials, academics, and journalists to discuss how technologies interact with government and society. At the time, Bloom was a Civic Imagination Fellow with Civic Hall Labs - a New York City-based collaboration center that designs technologies to advance the public good. Bloom’s talk was entitled "Building a Safety Net for the 21st Century," and in it, he described a project he had been working on for several years to enable an "Internet of help.
"By "help," Bloom was referring to the help offered by a variety of human services available in communities throughout the United States - homeless shelters, food pantries, domestic abuse shelters, and immigration services, for example. Bloom’s talk addressed how citizens find these services and how they determine whether they are eligible for these services - information, he argued, that is not easy to track down. During the talk, Bloom quoted his sister, a pubic defender that "helps poor people facing criminal charges in court," as saying, "When it comes to things that are really needed by people who are really in need Google is a ghost town" (emphasis original).
Here is the intro to Ch. 5, "Catachrestically Designing Semiotic Infrastructures for the Human Services", in the Introduction—which, note, isn't centrally about Google being a "ghost town" for those things "really needed by people who are really in need" but it is so fascinating:
Finally, Chapter 5 moves into a knowledge domain where the stakes for semantic stability and consensus are much higher. In March 2017, at the New York City School of Data workshop, I heard a presentation about an effort to design a data standard for information and referral - the branch of the human services domain that is concerned with helping people find public services for which they are eligible. For me, the effort embodied the political urgency of designing standard languages; developing common ways of talking about the human services is quite important for ensuring that people in need can quickly and accurately be directed towards services that can help them. Following this, I began interviewing designers of a variety of semiotic infrastructures for information and referral and analyzing design forums where the planning for these infrastructures took place. I recognized a lot of the same issues that had arisen in my study of Semantic Web infrastructure design; in the human services, due to bureaucracy, politics, and language habits, the meaning of words can be particularly challenging to pin down. In Chapter 5, I show how many semiotic technologists in the human services have had high hopes that semiotic infrastructures can help untangle muddled meanings in the human services, prodding the field towards adopting a shared language. However, these "standards" have never quite become standard. I argue that (mirroring trends that I describe throughout the dissertation) semiotic infrastructures in the human services have moved from very formal things to rather messy things. Semiotic technologists in this domain have come to understand standards as both necessary and impossible to fully achieve. They have experimented with opening these standards up, so that the meaning they represent can iterate and evolve. Newer, more experimental semiotic infrastructures in the human services are thus more about enabling collaboration and exchange than sorting out polysemy.
The close of the chapter, bringing in feminist scholarship re power relationships and @PopTechWorks's "Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor" (2018), makes one pause before thinking Google should focus on those who are really in need.
- A "ghost town" is different than a "search void" but similar in that it is "a byproduct of cultural prejudice" (Golebiewski & boyd, 2018). Perhaps in a ghost town we aren't so much exploited by what's online, but by the lack of infrastructure off. https://twitter.com/danielsgriffin/status/1016541092844466176