I really like how yor group geocoded each object and visualized the dataset in Tableau. I would like to see more interpretation as to the focus region for each institution because I am just really curious about how museums decide to purchase an object and build collections.

Another question I have is why you had to manually assign coordinates for the 86 museum objects instead of using Recognito like you did for the 12 archive objects.

I really like this analysis. Your group aptly points out both a change in language over time and the lack of tangible change on campus. I think it might be interesting to do a closer analysis of the documented sources and track if the "plans" have actually been implemented. Lastly, one overall feedback of mine if that it would have been better if your analysis had come right after the visualizations. Because they are separated into two sections, it is a bit difficult to see the direct relation between them.

It is interesting that the word "handicapped" was used most frequenty, instead of terms like "disabled" or "with disabilites." It seems like "handicapped" was considered the most appropriate language in the political climate of the 1970s. You might want to use the Trends tool on Voyant to see how the use of these different terms changed over time.

I have several questions on this part. Why did you choose to use data from 2020 in particular? What tool you used to create this box graph? Why does the size of colleges matter when we discuss co-education? What kind of correlation do you expect to see between the racial makeup and gender makeup of colleges?

I can definitely see how digital mapping is relevant to this project but cannot really imagine how Voyant will fit into the picture. What kinds of texts are you exactly planning to analyze, using Voyant? You could look at how student publications of Vassar and Hamilton have addressed concepts like accessibility and disability after ADA came into effect, but I am not sure if there will be enough documents to analyze because, as mentioned in this pitch, students with accessibility needs are highly underrepresented.

I think your group did a great job narrowing down the scope of the project to the Jasper Parrish Papers, and these are all very interesting research questions. One thing I am curious about is how many scholars have written about Parrish. How is your group's analysis of Parrish's life and legacy distinct from existing literature on the same topic? (besides the digital mapping component)

I think this may be precisely because these maps are scholarly maps. Members of academia tend to avoid making a "subjective" or "biased" argument, especially regarding historial matters. On the other hand, non-scholarly maps created bottom-up through community engagement (such as the Anti-Eviction Mapping Project referenced in Data Feminism) can more explicitly call out injustices. I want to learn more about the ways in which we can complement the limitations of scholarly mapping projects.

The Tech Bus Stop Eviction Map serves as powerful evidence to the case that tenant organizers are trying to make. I wonder what their goal was when they initiated this mapping project. Were they trying to show this to the decision-makers? To educate the public? Both? I think this is an important question because a compelling map based on accurate data does not always lead to solid policy changes.

I can see that you did a very traditional kind of literary analysis except that you used Voyant to guide you. It is interesting that the themes of your book was quite accurately reflected in your text analysis result because that was not the case for me. I am curious if you cleaned your data or changed any settings on Voyant in any ways to arrive at this result.

This is such an important point. I think textual analysis should be always be a starting point and/or a tool for an interesting research, not the ultimate destination. I want to know more about how digital humanists formulate their research questions and how they balance quantitative & qualitative aspects of their research.

I have not thought about this. It's a brilliant idea, considering that curators need to group artworks under a coherent theme/narrative in order to organize an exhibition. I wonder if museums like WCMA are already using tools like OpenRefine to keep track of and manage their collections of art in a structured manner.

I wonder why it says "information studies" instead of "Information Studies." Is it because the discipline in question here is interdisciplinary and flexible rather than having a series of set methods/issues? I am also curious about the relationship between information studies and data science. Do they overlap? Does one include the other?

This makes Hamilton's digital archive stand out among other archives that we examined for this week's blog. The plain text provided by the archive seemed pretty accurate. There were no errors or missing words. This is impressive because OCR technologies often have serious shortcomings when they are applied to historical documents/newspapers. I am curious what tool was used to harvest text from Woodhull and Claflin's Weekly.

As international student from Korea, I encountered race/ethnicity checkboxes for the first time while filling out my Common App. The Korean government (and higher education institutions) never collect such information because there is still a myth in Korean society that we are a "monoracial/monoethnic" country. Instead, people are usually classified as either a "Korean national" or a "foreigner." This perpetuate the myth I mentioned, erasing the lives of numerous people in Korea who do not neatly fit into those binary categories. Just like the title of this chapter, what gets counted counts (and also vice versa).

This statement makes me hopeful about the potential of data science. However, we need to remember the fact that conversting qualitative experience into data also has the risk of simplifying/objectifying that experience.

I also wrote my blog on "Mapping Memories of Africville," but I only wrote about the structural domain and why it is relevant in the case of Africville. Reading your blog post helped me realize that Mahon's website actually addresses all four domains through different mediums. My question for you and myself is, "How can DH projects tackle, not just expose the matrix of domination (in this specific case of Africville)?"

For some reason, the digital humanities projects I have seen till now (and the ones listed in Week 1 Blogging Instructions) were mostly creations. They were archives/tools used to present scholarly research to the public in an engaging, accessible, and interactive manner. So it is kind of difficult for me to imagine what it exactly is like to conduct "Digital Humanities research" (which is, at least in my opinion, different from conducting a research and then presenting them using tools of DH). I am excited to learn more through this course!

I am still a bit confused about the distinction between humanities computing and digital humanities. Are they fundmanetally different? Is it correct to say that Humanities Computing was focused on textual analysis whereas the scope and methods of Digital Humanities today are far more diverse? Or is the use of the term "Digital Humanities" simply a semantic change that was made to make the field of humanities computing sound more appealing for humanists?

I was conflicted between the two terms too a few months ago. It is fascinating to realize that it is not just a matter of grammar but also a matter of how we define the boundaries of DH.

The way data are being used against minoritized people's privacy and reproductive justice is deeply devastating. Target's pregnancy detection model reminded me of the way big tech companies monitor and restrict sex workers' online activity to avoid being held legally accountable for "facilitating" prostitution.

In this article, Olivia Snow warns that the same kind of algorithms will begin to target those seeking abortions after Dobbs. However, when I looked up what the situation was like at this point, I could not really find a research/article elucidating Dobb's impact on people's digital privacy. I am planning to look into the topic more, but it seems pretty clear that we need more research in this area.

While privilege hazard is very real, I also believe that increasing the representation of minoritized groups in CS/tech alone is just the bare minimum of what we can do to achieve data justice. The same is true in the political sphere: while increasing the number of elected representatives who have historically underrepresented is important, we should also acknowledge the limits of identity politics.

The term "imagined objectivity" reminded me of the South Asian history course I had taken last semester. We learned about how the British Empire used census data and its supposed objectivity in colonial India to pit different local groups against each other and establish hierarchies between them.

One remaining question for me is whether we can call colonial India's census data "data" because, nowadays, we tend to think of "data" as something that is digitally produced/processed/analyzed using computers. But how about data that are produced/processed/analyzed in non-digital ways?