The Zotero data is removed in this step which, in the visualization, is reflected by animated the Zotero area band down to the 0-baseline. I think this could read/perceived incorrectly (e.g., as Zotero contributions shrinking to 0). Instead, you might consider just fading out the Zotero band when its corresponding data is filtered out.

(I'll be honest, at this step, I'm not exactly sure why the Zotero data is filtered out)

I might be being pedantic here but the epistemological claim long predates D3. I'd probably locate it in Bertin's Semiology of Graphics which developed the initial conceptual model of encoding-decoding. This conceptual model was then fleshed out by Leland Wilkinson's Grammar of Graphics, and Hadley Wickham's Layered Grammar (which was instantiated as R's ggplot2). So, the writing's argument holds but I would perhaps refine it to read more accurately.

(For what its worth, I'll also flag that visualization researchers have discussed this point as well; besides the Critical Reflections paper I linked to in an earlier comment, a pair of papers from Alex Bigelow and Miriah Meyer get into these issues too (AVI 2014 and IEEE TVCG 2016)).

As a tool builder who has used Playfair extensively in my tools' example galleries, I'm struggling with the analysis the writing is offering here because it doesn't accurately reflect why we chose to recreate Playfair charts. Rather than to "bolster [our] graphical authority" or "lend evidence to [our] technical mastery" (both of which, if I'm frank, read a little condescendingly to me), it was more to demonstrate the –expressivity of our tools. As charts from the historical cannon are hand-drawn, their designs often feature custom/ad hoc elements that are poorly supported by software tools which favor/require systematicity/consistency. For instance, the reason I particularly like the Playfair wheat & wages chart is because of the "Gantt Chart" of monarchs at the top—typically alternating, except for during the 17th century; Oliver Cromwell shown in white because he was Lord Protector, interesting things happening with the alignment of the textual labels, etc. Thus, recreating these charts is an evaluation/validation of the software/design approach—there's certainly a sense that this is a demonstration of "technical mastery" but that phrasing suggests a more ego-centric goal than what I think tool-builders are working through (i.e., what tradeoffs are the software abstractions making; see https://vis.csail.mit.edu/pubs/critical-reflections.pdf for more).

Here too, the writing's assertion does not, to me, accurately reflect current visualization research and design practice. For instance, in 2011, Jessica Hullman provided a pretty rich rhetorical analysis of visualization design (https://ieeexplore.ieee.org/abstract/document/6064988/), and there's been a real growth on focusing on these rhetorical implications over the last ~10 years (e.g., Hidy Kong's work on titles, alignment, and framing effects as another example). In design practice, see the discourse around Simon Carr's "Iraq's Bloody Toll"

I don't think this (particularly "begin to") is a very fair assessment of the state of visualization research and is likely to turn off a large set of readers whom, I imagine, the book is trying to persuade/convince. In particular, the high-profile BELIV workshop (https://beliv-workshop.github.io/) has been the site for work advocating exactly this point of view dating back to 2006 (https://beliv-workshop.github.io/2006/). And lots of visualization researchers have offered conceptual models about the "value of visualization" that go beyond the simplistic notions of ease/efficiency that this sentence focuses on. For instance, two of the most popular models are Jack van Wijk's (https://ieeexplore.ieee.org/abstract/document/1532781/) and Emily Wall's ICE-T (https://sites.cc.gatech.edu/gvu/ii/icet/).

Echo'ing my comment on the intro, this is hyperbolic language. I'd be very curious how many visualization practitioners/designers/researchers would agree with this statement (versus a more sober one that Tufte has played an important role). To me, and perhaps somewhat ironically for a book offering a counterhistory, statements like this erase the foundational contributions of other folks in the field (eg Jacques Bertin, Stu Card, Jock Mackinlay, Pat Hanrahan, etc).

The y-axis labels are rendering in a modern font for me. Can't tell if this is an error or intended.

I was a little confused about what comparison was being made here (i.e., "less precise" and "smoother" compared to what?). Initially, I thought it was some period of time in the depicted data—but couldn't tell any difference. But, then I realized it was continued context from the previous scrolly step about the third edition. Perhaps the writing could better preserve that context—e.g., "in this edition, the data-lines became less precise"?

Yes! The question of audience is something that has felt muddy throughout this chapter and, unfortunately, although I'm glad for this (and the subsequent) paragraphs, I'm not actually sure they're clarifying for me.

In particular, I'm still trying to work through a set of entangled issues:

(1) What is the chapter's stance on the Clarkson's Description? The initial writing made me think the chapter was positively inclined because it, rightly, was "viscerally affecting" in depicting the horror of the slave ships, that the use of abstraction and de-humanization was done purposefully.

(2) The later discussion—particular its use of the god trick—then read to me as a critique. And then visualization of resistance weaves seemed framed as a response to that critique: how do we adopt the view of the enslaved peoples, rather than the god view of the slave traders.

But, and maybe this is my misreading/misunderstanding the chapter, it seems strange to me to juxtapose the two visualizations together—or, even, to frame the latter as what "ought" to be visualized about this data because, following (1), even the chapter acknowledges how effective the original way. But, perhaps more importantly, would a visualization explicitly about resistance (and centered on the perspective of the enslaved) been as persuasive in its historical context?

I particularly appreciate that the writing seeks to distinguish what we, as "twenty-first century viewers" should be doing—but here too, I find the prescription a little muddy. Are we saying that, for data about oppressed people, there isn't value in the god trick and that such visualizations ought to always/primarily be centered on the perspective of the oppressed peoples? Maybe? I'm certainly open to being convinced; but I also wonder whether there remains space in modern day design practice for visualizations of historical data that can be equally viscerally affecting because they adopt the god trick—i.e., via lampshading.

I can't seem to leave a comment on the steps of the scrolly, but this comment should be associated with the first step about "Resistance."

I completely agree that the choice to focus on this variable is key. But, I wonder if the writing could spell out more explicitly why the choice to focus on Resistance is so important. This blockquote gets at the overall goal, but is articulated at a sufficiently high-level that it still felt difficult to operationalize to me versus the concrete example of Resistance. I think an explicit explanation would also help connect to the chapter's earlier points about "consider what is lost/gained" and "consider our responsibility" and make them read like trite.

I'm not sure that this is a definition that visualization designers/researchers would agree with? I think of an "infographic" as a representation that makes heavier use of illustrative elements (e.g., things that some might call "visual embellishments"). Eg from a quick google search: https://infographicworld.com/marketing-glossary/what-is-an-infographic-definition/

In contrast, if the focus is on "direct representation of data," I think either "isotype visualization" or "unit visualization" might be more apt? (I also find surprising the writing's assertion that the lack of abstraction was a reason for dismissing the Description from the visualization pantheon; which makes me doubly curious/hopeful for a citation to the dismissal)

I'd love a citation here because, from my perspective of the visualization pantheon, the Description hasn't been dismissed so much as never considered.

I love these last two paragraphs—I think they're conceptually rich and interesting, and I love that the writing begins to complicate the role of abstraction to illustrate the tradeoffs. My one quibble, though, is that this highlighted sentence (and particularly the call to "consider") feels like a somewhat weak way to end. For instance, I would consider these two paragraphs to be a pretty exemplary demonstration of what it would mean to "consider what is lost [...and...] what is gained" but then what? (I could well believe the "then what" might be out of this book's scope, but it nevertheless felt like the writing was building up to something only to leave things dangling and move on)

Minor, probably a result of churn, but this passage duplicates the first paragraph of this section/

I wonder if you want to reverse the order of the scrollytelling for the image? I feel like it would both better recall the original—and thus better recreate the visceral affect—if it started with the entire, original image and then used scrollytelling to highlight the individual sections with the associated explanation. To me, starting with an empty ship, and then incrementally revealing each section, robbed the image of its shocking/horror (and, thus, partly undercut the writing's argument).

There's a lot of really great content here. But, for readers like me (the technical/design/engineering/research side of the visualization community), I think the writing isn't landing with quite the impact that it could for a few reasons:

(1) In my interdisciplinary collaborations, I've noticed a difference in writing styles/norms between the humanities and the design/engineering disciplines. The latter tend to favor a top-down argument structure (e.g., a crisply articulated thesis that is then unpacked via clearly signposted topic sentences). I think that's because readers like me are trying to figure out how to operationalize the things we're reading/learning. So, right from the get go, we need a clearly articulate conceptual model so that, over the course of the rest of the writing, we can figure out how to integrate it with our existing mental models of practice/research.

In contrast, this piece takes a very bottom-up approach to the argument. For me, the experience of reading bottom-up writing is of assembling a mental model that feels more like a wobbly house of cards: ad hoc, duct taped together, and needing to constantly swap/rearrange it as more pieces of the conceptual contribution reveal themselves to me.

As a concrete example, for the first third of this chapter, I wasn't actually sure what I was supposed to be taking away. I almost wondered whether I should suggest titling the chapter "preface" instead of "introduction" because it opens by being focused inwardly (i.e., on the presentation of the homepage) in a way I'm more accustomed to with prefaces than introductions. Although that whole chunk of writing was very pleasant to read (which may also be a function of the fact that I had the pleasure of meeting y'all and learning about how the project came together!), I wasn't entirely sure what this chunk was hoping to do/communicate—or how it was hoping to influence my thinking.

(2) Related to the first point, while I personally find the exploration of a visualization counterhistory exciting and thought-provoking, I wonder if the writing could better motivate the goals of the counterhistory a bit more explicitly and clearly? That is, if someone isn't already bought into valuing the history of the field (or doesn't know how a counterhistory may/should affect their current practice today), how might the writing persuade them to care? Or, put another way, how can the writing speak and evangelize to an audience who is open-minded, but not yet "on side." To me, this feels like a particularly important thing for an introductory chapter to do, that seems missing in the current iteration.

(3) I'm on the fence about how central a role Tufte is given here. I think this depends on the audience you are trying to reach—I'm not sure that many (most?) visualization researchers/designers/practitioners (i.e., visualization "thought leaders") consider Tufte to play as influential a role as this chapter purports him to do. If this was the core audience, then I think the focus on Tufte could be watered down without losing much of the overall framing of "counterhistory"—because I think what this chapter describes is very much the history the field tells itself (relatively independently of Tufte, I think?).

On the other hand, if the intended audience are the folks one hop removed (i.e., people who produce/consume visualizations in their daily lives/jobs, but aren't necessarily plugged into conversations on the bleeding edge), I think Tufte serves as a useful foil. But, something about his treatment in this chapter feels a little caricatured to me (and I say this as someone relatively ambivalent about his role). I'm not quite able to put my finger on what specifically about the writing left me with that feeling, though.

(4) Starting with the "Two Stories of Data Visualization" (and particularly the subsequent chapter on "Every Datapoint is a Person"), I wondered whether the target of the book's critique is indeed visualization (i.e., the graphic representation of data) or whether it's more fundamental and broader practices of data (i.e., definition, collection, etc. more similar to the set of issues y'all discussed in Data Feminism). I really enjoyed all of the detail and discussion here—and I was convinced about the role that data played. But, I was perhaps less convinced about visualization's central/facilitating/empowering role in it. It's likely impossible to fully disentangle data from its representation (as the data table examples do a great job) but, if the book wants to maintain visualization as its target, I wonder if the writing could be refined a little to make its focus clearer/crisper?

(5) I wonder if the writing can be more explicit about its positionality? I think some of the early sections (and occasional passages throughout) set up an incorrect expectation for me of a much broader (i.e., more global) counterhistory. So, I was then surprised that this chapter maintains a relatively fixed focus on Western history. In fact, I might go further to say that the writing seems to be particularly fixated on an American point of view (e.g., I raised an eyebrow at the description of the United States as "the exemplary" colonial state; as a non-American and citizen of a former colonized nation, I would consider the British Empire to be the ultimate colonial power...). I think this focus is fine if the writing is explicit that it is primarily concerned with developing a counterhistory rooted in the West (and, at that, the United States).

I'm having a little bit of trouble with this argument because—although the subsequent text tries to explain it—I'm not exactly sure how (or persuaded that) a shuffle addresses the issues that this section opened with (empty space in timelines, teleology, etc.).

Although shuffled, the images are still arrayed out horizontally. And, given that the text is written in English, I read the images left-to-right in the manner of a linear timeline rather than the web of timelines in step 2 of this scrolly. (And I didn't realize until reading further, that I could actually interact with the images; and even then, it took me a very long time to realize I could drag the images around; perhaps switching the mouse cursor to be the "grabby hand" might make that clearer?)

I think I'd also be more persuadable if the writing was a bit more critical in its discussion of its design choices—eg, besides reading too self-assuredly like shuffle was the right call, I am very curious about what tradeoffs the shuffle is forced to make. For instance, doesn't the shuffled display make it hard to trace particular (alternate) paths through history? It seems to (perhaps necessarily, because it must display things on a 2D computer display) flatten things out.

On my browser, the scroll seems off sync with this image. The image first appears alongside this paragraph (which seems right), but then immediately zooms and then pans in between this paragraph and the next. So, as I'm reading the next paragraph, I've lost the overall image which makes it difficult to appreciate what the subsequent paragraph is describing.

And then image altogether disappears shortly after "lightographer's practiced hand" scrolls out of view.

In my browser, these paragraphs start getting occluded as I scroll because a timeline of visualizations appears over them. I'm using Arc (a Chromium browser).

It's not clear to me why a scrubbable timeline would address the start/end of the timeline? Was it because the timeline would be initialized randomly? Or because scrubbing necessarily means filtering -> so you're making more explicit that the display is only giving you a partial view into the full timeline/history? Or perhaps something else?

Is it worth describing what felt missing about this design as it pertains to the problem that is being setup about empty timelines?

Initially, I was really confused what the connection was to where the previous paragraph left off—aka why does foregrounding images help address the issues of an empty timeline. I got it once I understood the writing to be referencing the second image in the sidebar but there was some initial cognitive friction that you might want to address.

Also, perhaps consider adding explicit references to the figures (eg "Fig 1")?

I really love this point—as the writing suggest, it feels like one of the kernels of this project. But, depending on the audience, I wonder if the profoundness/implications might pass by too quickly as currently presented? I wonder if there's an opportunity to dwell on, and be explicit about, the significance of what "emptiness" in the context of a timeline suggests/could be read as/etc.

Think there's a name missing here?