When I was living in Korea, I was inspired by the multi-dimensional traditions surrounding poetry and literature there. Poetry readings were often accompanied by live musical performances. Visual artists would display paintings, photographs, or calligraphy interpretations alongside the poems, fusing textual and visual art. There were also playful poetry writing games and activities that engaged broader audiences.
This was in stark contrast to the often more solitary, cerebral reception of poetry in the US, which can come across as esoteric or inaccessible. I started thinking about how to make poetry more interactive, accessible, and enjoyable for others.
Later, while working for a literature festival, I realized poems themselves create spaces. For four years (until the pandemic), we would choose a poem to anchor the entire festival. Based on the themes in the poem, we would curate programs, speakers, and school activities. The conversations I overheard during the festivals were illuminating. Discussing a poem on the concept of hometown sparked exchanges about people’s personal histories. Another year, highlighting a poem on identity led to passionate discussions about purpose and meaning.
The poems acted as springboards into broader dialogues and shared experiences. Poetry became a co-created space for human connection.
This project was inspired in part by these experiences. Conversations with Alex opened up new possibilities for using technology to make poetry a more visceral, accessible medium.
The article on Minimal Computing suggests using only the technologies that are necessary and sufficient for developing digital humanities scholarship under constrained environments. While I agree with the rationale, I believe the issues are more nuanced.
In the article, the authors cite the example of WordPress versus a static site generator. They argue that WordPress was the right choice–while sacrificing security and maintenance–was easier to use. This example was presented as a binary but glossed over nuanced discussion of the tradeoffs. Rather than a binary choice between static sites or WordPress, creative solutions like automating file uploads to Jekyll sites can achieve simplicity for users while enabling flexibility. More than focusing on the specific technologies used, the focus should be on balancing complexity strategically.
Considerations beyond technical access are also crucial, including literacy, cultural norms, and geography. Framing minimalism solely as a reaction to technology maximalism overlooks contextual factors and imaginative space in between. Discussions should revolve around project objectives and community priorities.
Technology can organize information, reduce labor, and enhance access. A minimalist approach suits some scenarios but may miss opportunities. Thoughtfully applied technology can spotlight complexity and access as needed.
Dwelling on “what do we need? what do we have? what must we prioritize? and what are we willing to give up?” limits the discussion on what is possible. More open, future-minded conversations around possibilities better serve identifying genuine requirements. Neither maximalism nor minimalism is universally optimal and in digital spaces, a more open, considered approach might be more constructive.
I found the reading “Difficult Heritage and the Complexities of Indigenous Data” particularly interesting in light of articles both for and against the use of AI in preserving languages.
In an opinion piece on the Washington Post, Viorica Marian states that languages are being extinguished at a rate of about nine per year. And the stakes are high, she notes: “Crucially, this is about much more than language. If a majority of languages die in the space of a few generations, that will also bring about a collapse of ways of thinking and being. Because the interaction between language and the mind is bidirectional.”
Large language models like ChatGPT pose an existential dilemma for minority languages, promising both conservation while also posing tremendous risk. In an article for Popular Mechanics, Luke Ottenhof raises the question of whether these AI projects being done for minority language communities or simply on them?
What’s clear is that these initiatives require extreme care, consideration, and collaboration with native speakers. Data is power, and therefore how it’s collected, presented, and leveraged can either empower or further marginalize already vulnerable groups. Any use of AI to preserve languages must balance urgency with a commitment to minimizing potential biases and framing effects.
The “Difficult Heritage and the Complexities of Indigenous Data” reading points to some of the challenges of working with indigenous data including context for access or the ambiguity in the process of translation. The authors argue that the only path forward is “the only path forward is through slow, thoughtful, inclusive, and collaborative practices that recognize and privilege indigenous-centric research practices and ways of knowing.”
I’m left wondering how to strike that delicate balance. These systems embed the values and viewpoints of those building them. Invasive, rushed, or careless data collection could irreversibly eradicate the nuances that make each language unique. But acting too cautiously could mean losing the chance to record native speakers entirely.
During my time living in South Korea, I witnessed shifting perceptions towards dialects. Modernization had spurred the standardization of the Korean language decades prior, privileging the Seoul dialect. By the time I arrived, efforts were being made to revitalize regional dialects through classroom education. However, these moves felt too little, too late – the biases towards the standardized Seoul dialect were deeply ingrained after years of structural marginalization. Minority dialects still faced shame and stigma, their speakers viewed as unsophisticated. This example highlights the need to consider the long-term impacts of language preservation efforts, not just their urgency. Standardizing or digitally capturing languages can embed biases that might become impossible to reverse.
Data Galaxy, a startup seeking to tackle the issue of data governance in large companies, recently held a workshop on best practices in this emerging field. As organizations create, purchase, and make decisions based on more data than ever before, proper governance of that data is critical for success.
Data governance refers to the overall management and oversight of data in a company to ensure quality, security, and compliance with regulations. For modern data-driven organizations, a strong governance program can help with:
Increased reliability and quality: with reliable governance protocols in place, teams can trust that data is clean, accurate, and standardized across systems. Crucial for analytics and decision making.
Enhanced security and compliance: comprehensive governance helps safeguard sensitive data from unauthorized access or theft. Also ensures privacy regulations are met.
Better decision making: with governance maximizing data quality and access, leaders can feel more confident basing strategic choices on available data.
While the workshop was focused on the corporate context, I saw opportunities to apply these practices more broadly across sectors. The corporate priorities around efficiency and growth are distinct from the aims of academic digital humanities projects, i.e. Indigenous cultures in Difficult Heritage and the Complexities of Indigenous Data.” In the latter governance not only ensures data quality, but also upholds ethical research standards and community participation in decisions. There is divergence in both the goals and stakes. However, the workshop spurred consideration about how data governance programs could aid industries – both corporate and academic – if tailored appropriately to the context.
It also led me to consider how data best practices could produce reciprocal benefit if documented and shared across industries. Corporations could adopt more thoughtful community and ethical models for data collection. Meanwhile, academics could learn from efficient private sector data management to improve conservation and access. Overall, governance remains crucial wherever data informs important decisions – but its practice may differ across industries. By learning from each other, better standards can emerge sector-wide.
For the visualization assignment, I chose to work with NYC Citibike data. The data (so far) is presented as 123 separate csv files spanning 10 years. Some interesting observations while processing the data:
From June 2013 – January 2021, Citi Bike data is stored in the following format:
'tripduration','starttime','stoptime','start station id','start station name','start station latitude','start station longitude','end station id','end station name','end station latitude','end station longitude','bikeid','usertype','birth year','gender'
From February 2021 – July 2023, Citi Bike data is stored in the following format:
Post February 2021, gender, bikeid, and birthyear are no longer stored. Instead rideable_type and ride_id are introduced. The former stores the type of bicycle (electric, classic, docked) while the latter is a unique hash sequence. Unlike bikeid, which can be tracked across records, ride_id is a blackbox that is only useful as an index.
I used Plotly to create the visualizations and MapBox for the maps.
While some of the visualizations were interesting, the way the data was collected, structured, and presented limited what could be visualized.
Data then is an expression of those systems and worldviews and thus must be recognized as intrinsically tied to how the community defines itself.
Cultural Analytics Jennifer Guiliano and Carolyn Heitman
The data captured is more of a reflection of preferences and priorities of the Citibike system than a true representation of experience. Many of the factors we would consider part of the experience of using a Citibike (let alone biking in NYC) were unaccounted for.
Throughout the process of creating these visualizations, there were multiple levels of abstraction in play. The sheer volume of data, the structure of the dataset, and the resulting visualizations. Each layer was another step removed from experience.
In an effort to make the data more personal, I used the bikeid datapoint to track starting and ending stations over the course of a year. However, much like the bike itself, the data (and visualizations) are tied to the system within which it operates.
I started off with a simple enough idea: to programmatically map locations in a text.
I went to a store in London and then went to Lake Erie to go fishing.
The idea was to extract “London” from the text and then place it on a map. This proved to be more complex than I anticipated. Both SpaCy and NLTK offer entity recognition through pre-built models that do a decent job of extracting the location (in this case, London) from the text but aren’t equipped to add context.
After running through several possibilities, I came across Mordecai, a project authored by MSU professor Andy Halterman. It uses a SpaCy model to extract locations, it then runs the text through another model that predicts the context based on the other words in the text. In the case of our example, London would be correctly classified as London, the city in Ontario, Canada because of the contextual clue “Lake Erie.”
{ 'doc_text': 'I went to a store in London and then went to Lake Erie to '
'go fishing.',
'event_location_raw': '',
'geolocated_ents': [ { 'adm1_count': 1.0,
'admin1_code': '08',
'admin1_name': 'Ontario',
'admin1_parent_match': 0,
'admin2_code': '',
'admin2_name': '',
'alt_name_length': 3.2188758248682006,
'ascii_dist': 0.0,
'avg_dist': 0.11450381679389314,
'city_id': '6058560',
'city_name': 'London',
'country_code3': 'CAN',
'country_code_parent_match': 0,
'country_count': 1.0,
'end_char': 29,
'feature_class': 'P',
'feature_code': 'PPL',
'geonameid': '6058560',
'lat': 42.98339,
'lon': -81.23304,
'max_dist': 0.2692307692307692,
'min_dist': 0.0,
'name': 'London',
'score': 0.9999490976333618,
'search_name': 'London',
'start_char': 23},
I also tried with the following variations (there’s a Thames River close to both Londons):
A. I went to a store in London and then went to Thames River to go fishing.
B. I went to a store in London and then went to River Thames to go fishing.
The first example, returned two results London, England and London, Ontario with London England scoring slightly higher. The second example only returned London, England.
Given more time, I would have liked to try mapping the Black Mountain corpus. For this assignment, I chose to work with just one poem: “I, Maximus of Gloucester, to You” by Charles Olson.
After extracting the location data with Mordecai, I saved it as a CSV file, and I used Folium (an open-source mapping library) to map the datapoints
import pandas as pd
import folium
# read in the poem
df = pd.read_csv('poem.csv')
# map data to folium
m = folium.Map(location=[-6.1753924, 106.8271528], zoom_start=2)
# add markers to map
for i in range(0, len(df)):
folium.Marker(
location=[df.iloc[i]['lat'], df.iloc[i]['lon']],
popup=df.iloc[i]['text'],
icon=folium.Icon(color='green')
).add_to(m)
Mordecai extracted:
Oregon, Ohio
New England, Jamaica
Gloucester, England
While context was added to each of the locations found in the poem, only one of them was correct. I attempted a similar process with four other poems with similar results.
While this tool is useful when the cues in the text are obvious (or more likely common, e.g. Paris and the Eiffel Tower) the Mordecai model would need to be tuned differently for this type of text.
Some interesting questions that were raised in the process of completing this assignment were:
Could it be possible to extract location based on dialectical differences?
Mordecai was funded by the CIA as part of the Political Instability Task Force. What are the ethical implications of using a tool like this for alternative purposes?
Creative processes like poetry seem to be at odds with the statistical methods used in Natural Language Processing or machine learning. What kind of patterns are worth looking for?
In poetry, geographic references serve a variety of purposes. A map, while appearing multi-dimensional, can flatten literary analysis to just coordinates. Could a map reveal something new?
The Early Caribbean Digital Archive states its intent to refocus the archive’s spotlight on the pivotal contributions of enslaved and free African, Afro-creole, and Indigenous peoples in the Caribbean world. While this project appears to encompass many of the key elements associated with DH initiatives, it highlights both the potential and challenges inherent in DH scholarship.
In the introduction of “The Digital Black Atlantic,” the authors note several challenges in researching African diasporic communities:
a lack of a common body of scholarship for Black-centered digital studies creates challenges in framing or contextualizing discussions
the tendency to assume the epistemology of white, dominant, English-speaking cultures is universal
In light of these challenges, the construction of the ECDA prompts questions about both the materials used and the processes employed. It is immediately evident that there is a lack of diversity in terms of content and language. While the ECDA claims to employ digital tools to “remix” the archive, it does so from a perspective rooted in the Global North.
In the “Introduction” to “Global Debates in the Digital Humanities,” the authors acknowledge that scholars outside of the United States and the United Kingdom have deliberated over the “epistemic and political consequences of the English-speaking hegemony in DH.” When the Western model is imposed as the universal standard, local voices risk being overshadowed.
In “The Digital Black Atlantic,” the authors emphasize the necessity of engaging with local communities and paying attention to “what digital humanities looks like in particular African diasporic contexts.”
The ECDA, in its attempt to reconfigure the collection uniquely, raises a critical question: Is its endeavor to decolonize the archives inadvertently leading to their recolonization?
The Early Caribbean Digital Archive states that “the materials in the archive are primarily authored and published by Europeans.” If the ECDA is an act of “rememory” as described by Toni Morrison, we must consider whose memories are being recollected and by whom as we employ technology to bridge the past and the future.
Need help with the Commons?
Email us at [email protected] so we can respond to your questions and requests. Please email from your CUNY email address if possible. Or visit our help site for more information:
