This doesn’t sound like DRM. It sounds like access control
This shows how the Mukurtu archive was not just about locking information away. It was more about deciding who should have access based on cultural responsibilities and community rules, instead of treating everything like public internet content.
“information wants to be free
This line points out a common belief that everything online should be open and accessible. The article uses it to question that idea, showing that it doesn’t really work when applied to Indigenous knowledge, since that kind of information isn’t meant for everyone to access or share freely.
we shouldn’t use LLMs to study uncertainty, because these models have already read many of the books we’re studying and will (presumably) already know the plot.
This part stood out to me because it pushes back against the main idea in a really honest way. The author is basically admitting that the method has flaws. If the model already knows the story, then it’s not really “predicting” anything like a human reader would. I think this is important because it shows the limits of AI in literary analysis. Just because it gives the right answer doesn’t mean it got there in the same way a person would.
topic modeling can be a powerful tool for discovery. But it would be a mistake to take this workflow as paradigmatic for text analysis. Usually researchers begin with specific research questions, and for that reason I suspect we’re often going to prefer supervised models.
Underwood is basically saying that just because topic modeling is useful doesn’t mean it should be the main way we analyze texts. I like how he points out that most research starts with a clear question, so we need tools that are more focused, like supervised models. It shows that digital tools aren’t one-size-fits-all.
Compiling resources on how to write within the form of a visual essay while omitting the discussion of storytelling, in both visual and written terms, as well as the importance of story structure, is of next to no help.
This quote shows that writing alone is not enough when working with data stories. The author is saying that if you only focus on writing and ignore storytelling, the result will not be very useful. It highlights that structure and how you guide the audience matters just as much as the words. It also suggests that data projects are not just technical or informational, they need to be shaped like a story to actually make sense to people.
Neither results seems particularly surprising, once we’ve constructed a just-so story to rationalize the answers.
This quote shows Lincoln’s main concern really clearly. He is saying that once people have a result in front of them, they are very good at inventing an explanation that makes it feel reasonable. The problem is not just bad interpretation. The problem is that the explanation can sound convincing no matter what the result is. That makes it harder to tell whether the data actually taught us something new, or whether we are just forcing a story onto it after the fact.
Knowing that there is a network and knowing what the network is are two different kinds of knowledge.
Mullen is explaining that general knowledge is not the same as detailed understanding. Someone might know that connections exist, but that does not mean they know how those connections actually look or function. This quote shows that visualizations help turn a vague idea into something specific. It is the difference between guessing something exists and actually seeing how it works.
Large, unregulated collections of behavioral data are a public hazard.
The author refuses to treat data collection as just a technical issue. By calling it a public hazard, he compares it to something that can harm society at a large scale, not just invade one person’s privacy. That wording shifts the conversation from individual responsibility to systemic danger. It suggests that the real problem is not only that companies collect data, but that they do so without meaningful limits or accountability.
If you go searching for patterns in the data, you'll find patterns in the data.
This points out that just finding patterns is not automatically meaningful or impressive. The author is basically reminding us that data can produce results, but that does not mean those results actually tell us something important.
because all of its records are derived from historical documents themselves, and so are eccentric, disparate, and historically and geographically uneven.
I like this quote because it shows why humanities data can be so hard to work with. The data is not neat or consistent since it comes from real historical documents, and that makes it more complicated but also more meaningful. It feels honest because it reminds us that human history does not fit perfectly into clean categories.
What became evident was that cleaning up or correcting values was a misleading—and even unproductive—way to think about how to make the data more useful for our own questions and for other scholars studying food.
I like this quote because it challenges the idea that data always needs to be “fixed.” Sometimes making data useful is more about asking better questions than forcing everything into one neat format. It feels honest because it shows that research is not always as simple as cleaning things up.
The term “cleaning” implies that a dataset begins as “messy.” “Messy” suggests an underlying order: it supposes things already have a rightful place, but they are not in it—like socks on the bedroom floor rather than in the bureau or the hamper.
This quote made me think about how the word cleaning already sounds biased. It makes the data seem wrong before anyone even works with it. I like this part because it shows that organizing data is not just technical, it also reflects human choices about what looks “right.”
Had we not taken the time to get to know our data with a data biography, we would never have realized how different all these data points were.
This example highlights how important it is to examine data carefully before drawing conclusions. The researchers initially thought the data showed a clear trend, but after investigating the data’s background, they realized the differences came from how the data was collected. This shows how easily data can be misunderstood without proper context.
For every piece of data you’re going to include in your story, you need to create a data biography — the background, or origin of your data. Just as you’d do a background check on a human source before publishing what they told you, you need to understand your data:
The author explains that journalists need to understand where the data comes from and the context behind it. Looking at the “biography” of the data helps determine whether it is trustworthy and appropriate to use in a story.
Despite the lavish production values, there's a strange sameness to everything. You're always in the same brown war zone.
Modern games and websites may have impressive visuals and high production budgets, but they often feel repetitive and lack originality. For example, COD has changed very little over the years. My main point is that high sales drive a focus on impressive visuals rather than creative innovation.
The OCLC’s definition of “literary greatness” is biased based on the libraries that OCLC represents, the list’s exclusive focus on physical books, and its emphasis on raw number of holdings, which is influenced by number of editions.
If the definition of “greatest” depends heavily on which libraries are represented and what counts as a holding, how much can we trust this list to reflect true global literary significance?
And yet we don’t know what they are actually consuming: the types of content that they are viewing, liking, sharing, or commenting on in the largest quantities.
If platforms refuse to share this information, how can researchers and the public really understand what shapes today’s culture?
In the process, I’ve had the occasion to learn this surprising fact: that it is both unexpectedly and especially difficult to find comprehensive data about the most popular social media content
McNulty notes that while platforms like TikTok reach people around the world, the data on their most-watched content is not available to the public. This lack of access makes it hard to do in-depth cultural analysis.
Born digital culture requires a born digital approach.
She says that our traditional ways of studying culture don’t work for TikTok-era media. Instead, we need methods designed for the internet.
Every week, the New York Times of course releases its famous list of “bestselling” books, but this list does not include individual sales numbers.
If well-known bestseller lists don’t actually show how many copies were sold, how much can we really trust them to reflect what people are reading?
Fortunately, there are a number of forward-thinking people who are already leading the charge for open book data.
Walsh notes that open data efforts, like public library checkout stats, are promising steps toward a more transparent view of how people actually engage with books.
I went looking for book sales data, only to find that most of it is proprietary and purposefully locked away.
The data that would help us understand reading trends and culture is hidden behind paywalls or restricted to industry insiders rather than being publicly accessible.
And data are never neutral; they are always the biased output of unequalsocial, historical, and economic conditions: this is the matrix of domination onceagain.
The authors explain that data shows what is happening in the world it comes from. So, if a society has inequality, its data will reflect that inequality as well.
This is the privilege hazard: the phenomenon that makes those who occupy the mostprivileged positions among us—those with good educations, respected credentials, andprofessional accolades—so poorly equipped to recognize instances of oppression in theworld
I like how this explains why biased systems continue to exist. It is not because people want to cause harm. Sometimes, they just do not notice the problem since the system benefits them. This makes bias more difficult to spot and address.
“What we choose to measure is a statement of what we value inhealth,” he explains.
If something is not being measured, does that mean society has decided it is not important?
Counting and classification can be powerful parts of the process of creating knowledge.But they’re also tools of power in themselves.
I think this is the main point of the chapter. Data is not neutral. Counting can help people get noticed, but it can also be used to control or hurt them. It all comes down to who is counting and for what reason.
Lurk-ing under the surface of so many classification systems are false binaries and impliedhierarchies, such as the artificial distinctions between men and women, reason andemotion, nature and culture, and body and world.
I like how this line makes it clear that categories aren’t just simple labels. They often hide power differences. When we treat something as a strict either/or, we ignore the complexity in between and usually end up privileging one group over another.
“What gets counted counts,” feminist geographer Joni Seager has asserted, andMunir is one person who understands that.
if being counted gives people visibility and resources, how do we balance that with the risk of making vulnerable groups more exposed?
“If everything is an archive, what meaning does the word have? What is so great about being an archive? Why have we given the word such power?”
What makes something truly worthy of being preserved and recognized as history?
In the face of this guilt-inducing backlog, special collections have turned towards digitization as a solution, prioritizing getting images of Black people online and hoping that will be enough.
Putting things online doesn’t automatically fix the problem. It might look like progress, but if there’s no real explanation or context, people still won’t understand what they’re looking at or why it matters. Just uploading images isn’t the same as making history truly accessible.
Academics continuously loosen the concept of the archives in vigorous debate and flowery speech, while hundreds of linear feet of Black history are stacked in secure shelving, unbeknownst and inaccessible to implicated communities.
This part highlights the disconnect between theory and reality. Scholars debate what an archive means, but real Black historical materials are sitting on shelves and not accessible to the communities they belong to. It shows how discussion does not always translate into action.
Most Unix systems are kind enough to allow just about anything in a file name.
This shows that even though systems allow a lot of flexibility, that doesn’t always mean users should take advantage of it without thinking ahead.
This merits a specific mention because detecting that a file name has one is not easy to do visually.
This explains why small details matter, since mistakes that are hard to see can still cause problems when working with files.
ibraries and archivesreally have to keep our roles whole and moving forward because we have a verydifferent point of view than the commercial guys. Wil
This idea directly connects to Besser’s argument that digital libraries should preserve traditional library values like access, ethics, and service, rather than becoming commercially driven platforms.
ickly. The current digi
How can we realistically preserve digital materials long-term if it requires constant attention, money, and people to keep updating them?
o preservation means "make copies."
Kahle is basically saying that the safest way to preserve digital material is to make multiple copies and share them, instead of trusting just one archive to protect everything.
nclusion, I argue for univeris within our grasp financially. It'sour gr
Kahle’s main argument is that universal access to knowledge is not just an ideal, it’s something we already have the tools, money, and systems to achieve if we choose to act on it.
So preservation means "make copies."
Kahle is basically saying that the safest way to preserve digital material is to make multiple copies and share them, instead of trusting just one archive to protect everything.
nclusion, I argue for univeris within our grasp financially. It'sour gra
Kahle’s main argument is that universal access to knowledge is not just an ideal, it’s something we already have the tools, money, and systems to achieve if we choose to act on it.
It is very possible that digital libraries will enable future humanities scholars to engage in new activities that we haven't yet envisioned.
What kinds of new activities might he be imagining, and are scholars actually ready for that shift?
Libraries (either digital or brick-and-mortar) have both services and ethical traditions that are a critical part of the functions they serve.
This highlights that libraries are defined not only by what they contain, but by how they serve users and uphold values like privacy and fairness.
Digital libraries will be critical to future humanities scholarship. Not only will they provide access to a host of source materials that humanists need in order to do their work, but these libraries will also enable new forms of research that were difficult or impossible to undertake before.
The author emphasizes that digital libraries aren’t just about easy access, they change how research can be done. It’s a shift from visiting a library to interacting with the material in new computational ways.
But it’s still not entirely natural, and he sometimes slips. About halfway through a recent nine-month research project, he’d built up so many files that he gave up on keeping them all structured.
I find this interesting and relatable because even people with years of experience can make mistakes. I sometimes give up and just put everything in one folder too. This shows how people often shift from structured folders to a “one bucket” approach when things become overwhelming.
Guarín-Zapata’s mental model is commonly known as directory structure, the hierarchical system of folders that modern computer operating systems use to arrange files.
This explains that computers organize files using folders inside other folders, similar to how people organize papers in real life. It connects back to the drawer analogy.
“I open a drawer, and inside that drawer, I have another cabinet with more drawers,”
I have experienced this when modifying game files on my PC, finding the directory, opening a file, then finding even more files, and even more when opening one of those.