[Week 3] 🗺️ Find the edge of a map 💬📈 Visualize chat logs

[brooklynnorthcross]

Do

• Go to the edge of a map you interact with regularly and document it in some way (experience report, Polycam app to 3D/LiDAR scan it -- a free iPhone, Android app).

This could be the map of a electric scooter app being tested in a neighborhood as in this example, the edge of a subway/transit map, the limits of Google Maps' search (e.g. try searching for tacos versus taqueria and see if you can find the rough edges of the resulting maps).
Your choice to interpret a "map" as you see fit.
As you document, reflect on who gets to make it, whether algorithms and data are in play, whether you trust the map reflects reality, and whether and how you get to have a say in making the map (if, for example, you disagree with how it reflects reality!).
Include a description of all the steps you'd need to go through to influence the edge of the map being redrawn and by whom.

Read

• Read data thinking class notes on the ethics of exporting and sharing Zulip data -- https://github.com/onefact/datathinking.org-codespace/blob/acb3f0e2a6665d71ba3f7a973cb430ac1b682ab9/notebooks/university-of-tartu/230316-embeddings-text-data-like-chat-logs-and-networks.ipynb

We decided to keep this a public chat to enable future students to benefit from previous students' experience, and to be able to make the data we generate easier to analyze and teach from.
(datathinking.zulipchat.com is a web-public stream, described here: https://zulip.com/help/public-access-option). Feel free to share any reflections about this on Zulip or over email! This is a constantly evolving open source bootcamp that depends on your feedback and experience.

• https://www.publicbooks.org/cooking-monasteries-arithmetic-lorraine-daston-on-the-history-of-rules/
• https://www.theverge.com/23753963/google-seo-shopify-small-business-ai
• https://subpixel.space/entries/life-after-lifestyle/ (or listen to this being read: https://youtu.be/iK3oK50AFlg)
• Read this zine: https://www.jennyodell.com/free-watch.html
• Read about time, temporality, and the history of capitalism: https://sci-hub.se/https://doi.org/10.1093/pastj/gtz014
• Read about numbers and nutrition, mental health, diet, and exercise: https://www.thenourishmentproject.com/ - what is lost when we regain control using numbers? Who decides what "good" numbers mean or are as we use them to make sense of ourselves?

Watch

• Watch Berger's Ways of Seeing episode on the Language of Advertising: https://archive.org/details/WaysofSeeing/Ways.of.Seeing.S01E04.Language.of.Advertising.1974.VHSRip.AAC2.0.x264-rattera.mp4

Code, Visualize, Install

• Use the Jupyter notebook from lecture to visualize the Zulip chat data to answer a question you might have about it, and post a screenshot on Zulip -- Jupyter notebook: https://github.com/onefact/datathinking.org-codespace/blob/main/notebooks/princeton-university/week-3-zulip-chat-data-analysis.ipynb
• Use homebrew to install rectangle on Mac (https://formulae.brew.sh/cask/rectangle) or use Chocolatey to install powertoys on Windows (https://community.chocolatey.org/packages/powertoys, documentation here) and try using either to rearrange windows with keyboard shortcuts
• Test using maccy on Mac or copyq on Windows for clipboard management (try using the keyboard shortcut to, for example, copy and paste some experience reports from datathinking.zulipchat.com into a large language model)

[brooklynnorthcross]

For the edge of the map activity, I tried to find a beauty supply store in the Los Angeles area. For context, I am staying in the Westwood area of Los Angeles for the summer. If you don't know, a beauty supply is a store where many Black people source their beauty products and they are also referred to as the "hair store" in Black culture. At these stores, Black people can find items that they typically wouldn't be able to find at typical stores like Target, and usually at more affordable prices. Google maps was not able to provide me with helpful results. They were showing me cosmetics stores like Ulta and Sephora, stores that sold products for professional [white] salons, and hair salons. After scrolling through a long list of results, I was only able to find a couple true beauty supply shops, or at least how I would define a true beauty supply. This could be because I'm in a predominately white area, but I didn't believe that there were only a couple of them in this big city, so I looked to Tik Tok to prove this true.

While doing this activity, I was thinking about the conversation that Tik Tok is replacing Google as a search engine. When I cannot find what I'm looking for on Google search--when I find an edge of the map-- I often go to Tik Tok. So, I went to Tik Tok to search for beauty supply shops in Los Angeles and I was able to find infinite options. When I search "beauty supply los angeles" in the Tik Tok search bar, the app is able to deduce my desired meaning of beauty supply: the staple store in Black culture and not a chain cosmetic store and not a salon. I'm unsure if the search results are tailored based on the algorithm that recognizes the content that I interact with, or if the search results would be the same for every user.

When thinking about data thinking at the edge, a big question is who gets agency in removing themselves from maps. But in this case, there is a counter question of how does Google remove people from maps. How does google's definitions, and misunderstandings of non-white cultural language disappear people, places, and businesses from maps?