After a great, big brainstorming session (documented here), I have a nice range of possible projects. I also can’t seem to stop thinking about the research into Amazon that I began in our Critical Experiences class in the Fall Semester. There is a lot left to be explored and uncovered there. I think I need to decide if I am all in on pursuing that for my thesis, or if I simply continue that work along the side, while exploring one of the brainstorm topics instead.

For the sake of this exercise, I’m going to explore what it would look like to build a thesis around the Amazon research that I have pursued so far. While researching user reviews on Amazon, I started to also investigate political products on the platform. These were typically flags or apparel. In this research, I came across a company which holds about 10 patents for brands that are almost exclusively sold on Amazon. (CENTUKE is one of the brands that I’ve been observing). These brands appear to sell products which are print-on-demand with a wide range of graphics and messaging. Many of these brands specifically feature messages which cater towards bipartisan American politics, with a predominance of right-wing content. Watching these product lines grow and change has only further piqued my overall curiosity into the intentions, motivations, logistics, and overall impact of this business model.

I've also stumbled upon a heavily repeated photoshopped image of a woman holding a flag at a Trump rally. This image has become a go-to product photo for flags across several of these fast print brands. So far, I've noted 32 different brands who frequently use this image on their flag product pages on Amazon. The range of flag messages that this woman has held is extraordinary. She has "held" every imaginable flag including"Santa Trump", "Anyone but Trump", "Gay Pride", "Heterosexual Pride" and so many things in-between. I have so many questions about and for this woman, and the usage of her for these products.

‍

‍

Conceptual Questions:

1. What could be the possible business strategy behind the product offerings that I have observed? If these brands/products are not about an effective sales strategy, then what is their purpose in existing (even if only existing digitally)? Who benefits from these products being online? What is the difference between an imagined (but purchasable) product, an imagined product, and a fully-realized product? What is the life story of these photoshopped products? Who conceives of them? Where do they live (in ads? in places outside of Amazon?)
2. What is the purpose/function/capability of research in art? Is art the correct response to uncovering certain types of research? What does declaring that impact and purpose are your justification for an artistic experience mean about aesthetic art? Balance between impact, purpose and user experience/aesthetic pleasure? What role does viewer comfort have in a work of art?
3. Looking more into the reused image of the Trump rally flag woman. Can I find out who she is? To what purpose/reason am I so curious about this image? How did it come to be used so much? Does the woman know? What does it mean for any of us if our image and personhood can be used in this way?
4. What does truth in advertising mean when it’s done online? What is the influence of productization on vaccine hesitancy?
5. What role does the platform have in policing products and sellers? Is Amazon knowingly using deceitful practices to permit this type of productization? (Misrepresentation of quantity amounts; failure to monitor content; confusing information architecture (brand name above or below title has a different meaning and different search capabilities); inability to easily jump to last page when browsing by brand).

Technical Questions:

1. Web Scraping on Amazon
2. Data Collection and Analysis
3. Machine Learning for Sentiment Analysis
4. Tracing and tracking images, products and advertisments across platforms.
5. Is there a way to programmatically find, follow, and document products on amazon to their potential advertisments on the web? Why not?
6. How to build, read, and utilize large datasets
7. How to make inferences/build knowledge about the world through data

Possible Projects

Iteration 1: Writing + Political products on amazon

• Find stories here:
• The story of the woman with the flag +Lin Chunhai Investigation
• What stories are told simple by seeing a group of images sold together.
• I’m looking for a hat. I’m looking for a hat and I believe X. → These hats
• I’m looking for a hat and I believe Y. → Those hats
• The story of the life cycle of a product that may never exist and compare it to the life cycle of a product that does exist. What does “exist” mean when it’s a digital representation of a potential product.
• How many different conspiracy stories can I tell with the same dataset?
• Is there a way to walk people through the observations and products and invite them to draw conclusions? Demonstrate how presuppositions and suspicions can end up reinforcing strange ideas. How we start to form stories to make sense of the world.

Iteration 2: Web development + Political products on amazon

• Straight facts/truth telling. Interactive data representations of the data that I’m starting to collect. Look at The Markup and Red vs. Blue. What does “red” advertising look like vs “blue” advertising? But also, it all being underneath one umbrella. Look at NYTimes interactive pieces for reporting and storytelling.
• https://www.nytimes.com/interactive/2021/02/04/parenting/working-moms-coronavirus.html
• Maybe it’s an amazon seller account that is either managed by AI or invites any user to upload an image and then auto-generates the information to add that as a product on Amazon marketplace. (Problem here is I don’t actually want to help Amazon create revenue.) Is there a way to do this where it ends up interfering with Amazon algorithms in some way? Maybe it’s about creating really bad listings and flooding the search tags? Could I set a very specific target, of anti-vax messaging, and crowd out the marketplace?

Iteration 3: Digital design/Communication Design

• Creating an actual storefront with these items in it. A walk-thru shopping experience of one of these vendors? Or a t-shirt/ice cream truck in NYC/DC? A vending machine? Bringing an element of physicality to these theoretical objects. The temporality of these items vs the permanency of their existence.
• Or maybe it’s set up to look as though it sells the products, but they are actually just posters or screens, to better articulate the digital existence of these items. Elaborate packaging?
• Vending machine that holds 20,000 “products.” Could do in blender. A vending machine that just keeps stretching wider and wider. Purchasing them is very easy, but then you have to see the mechanical steps that go into the product. Print on demand, shipping from China, American distribution, etc.
• Storefront…What story are you telling? If you are focused on the impact of having this messaging being more pervasive, then a storefront display is a more appropriate physical articulation. Maybe if you have the mannequins wearing white shirts, you could project the shirt graphics on to the models. This would allow you to change the “targeted ad” based upon the user/viewer. Maybe it’s done randomly based upon any product from just one of these brands, or maybe its based upon data from the user (taken either voluntarily or surreptitiously).
• https://nerdist.com/article/projection-mapping-on-clothes-is-now-possible-great-for-making-memes/
• Or maybe the audience has to be the ones “wearing the shirts.” You have 6 people enter at a time, and 3 are facing the other 3. You can see them wearing certain shirts and end up making inferences about those people. Maybe one group was asked to supply data which leads to the “targeted product” and the other group is “wearing” random products. So some of the users think the product selection is a reflection of the users and the other group thinks that it’s random. (More of an experiment setup, then an interactive art piece?)
• What about models/runway fashion shows?
• Hanging clothes to dry on a line / these absurd hats on a hatstand
• There’s something unusual about pulling these items and placing them in the context of a home. Three hats that say “I pee in pools” “Let’s go brandon” and “Free Masons” describe a person and possibly a worldview. Maybe it’s just a series of hat racks featuring these products? Or a hatrack along the wall with blank hats facing out, so we could utilize data and project the graphics dynamically.
• A performance piece where you build the set, and hire actors to set up that photo of the Trump flag. Users then have to stand in the image holding a green sheet, and we can print out an image that randomly selects a flag image.

Iteration 4: Clicker Game

• In the style of Universal Paperclips, Kitten Games, or CandyBox!, this could be a clicker game where the player is an Amazon seller. “Buy-A-Hat”
• Number of:
• Products
• Products with reviews (sell faster)
• Pages of products
• Brands
• Unassociated Brands (brands below)
• Sellers
• Milestones/Events
• Reaching enough products to have the brand page lie about total # of products
• Political events, start selling items that foster bipartisanship, then more and more extremist views…white supremacy

‍