L to R: Kesia Davies, Ananya Dhandapani, Victoria Dey, and Liam MacLean.
At the end of the year, I like to take students to dinner. We also pay students to work on this project — and that involves time they may need to learn technologies and skills related to the digital humanities and history. But I think hard work and milestones both deserve celebration. Several members of the current project team are graduating and moving on; Victoria Dey joined us this year. There was a lot to fête. College and graduate school is a journey and I like to remind students that I appreciate them.
So (yet again), we are not a food blog, but this was a very delicious way of celebrating our labor. And no, we did not sit on the cheese.
Our student team works very hard on this project. We pay them for their time and provide them with skill training, but after the challenges of remotely working on this project over the summer, I was very pleased to be able to treat everyone to a well-deserved dinner at Eataly in Boston.
L to R: Liam MacLean, Jenia Browne, Kesia Davies, Ananya Dhandapani, Jessica Luo, Kira Torrieri
I promise that we’re not becoming a food blog, but the whole event seemed like something to commemorate after spending so much time apart from one another.
So now that we’ve gotten through two presentations, it’s time for some screenshots. The team presented at the Association for the Computers and the Humanities last week. Today we gave a presentation for folks at Northeastern. I am biased, but I think the students on the team are doing an incredible job navigating issues related to research and the technology, as well as presenting their experiences in clear and compelling ways. For the latter presentation, I turned VR on for a build-in-progress, and we pushed the Oculus view to Microsoft Teams. Trying to get VR headsets to play nice with various remote/pandemic contingencies has a particularly tricky part of this project. We’ve been building the project since March (including training!) and I think this aspect has slowed us down a little. I nonetheless think we’ve gotten a lot done despite the circumstances.
Anyway, it’s always weird showing a historical 3D reconstruction in progress, because the decisions we make today might not be the decisions we make tomorrow. We conducted an initial survey of source material at the beginning of the project, but it’s almost impossible to anticipate all the little details up front — sometimes we realize we have to go back to the archives when we’re building a texture or trying to figure out what is typical or appropriate at 8 Belknap Street.
3D render of proposed front parlor at Belknap Street c. 1829.
For example, none of us are satisfied with the wallpaper in this render despite extensive wallpaper research. We know that David Walker was renting a house erected in the mid 1820s, and we all think that wallpaper was probably up in the parlor, if up at all, and would probably date to that time period. We’ve been scouring digitized databases for examples and some archivists and librarians have been sending us samples of books bound in wallpaper to understand the range of what was available. So the students are still thinking about what this should look like — but what to do in the meantime? Not assigning wallpaper to the walls is also an interpretive choice. I think we opted for period-inspired paper as a placeholder to communicate that wallpaper should go here. This is why we think it’s so important to make our decisions public-facing. We’ll have some student-written blogs up soon that explain choices specific to certain objects. By the time the project is ready for public consumption, we hope to have a permanent feature on the site that links individual objects to the sources that underpinned their creation.
3D render of proposed front parlor at Belknap Street c. 1829.
Anyway, we have several more rooms to populate with objects that need to be revised in small ways. There are wonky reflection probes that are driving me bonkers even after spending hours trying to sort them out for these presentations. But most importantly we have to work through what sets of objects and spaces are going to tell the most meaningful stories about David Walker within the 8 Belknap Street build. This curatorial work is crucial for making this more than just a lovely model of a nineteenth-century home.
We’ve been busy working on the project these last few weeks — mostly with an eye to presenting at the Association for the Computers and the Humanities Conference, which will be held virtually, July 21-23. The conference program looks very exciting, and we encourage you to check out our panel presentation, hear about progress to date, and to meet our student team members.
Note: We realized yesterday that we lost some recent posts due to a server error. Over the next few days we’ll be working to restore those.
We’re now at the point in the project where the team has modeled and applied materials to several historical objects. When I teach students how to build models for virtual reality projects, I have them focus on wire-frame geometry and we map out places where we will apply materials to in the future. We build those materials in Unity and give said materials values that have them behave like glass, metal, wood, etc. They just don’t always look like these materials until we’ve added some environmental effects.
So sometimes, you spend a lot of time building objects like this and their appearance isn’t very satisfying yet:
Screenshot of Unity scene before we’ve applied reflection probes and post-processing effects. From a tutorial given by Jessica Linker and Liam MacLean on 6-17-21.
Yesterday we showed the student team how to add some layers of realism with reflection probes (Unity objects that determine what is reflected in reflective materials) and by building post-processing filters to augment what the main camera sees. The main camera usually determines what the user sees when wearing a virtual reality headset to view a project immersively.
Here’s what the same scene looks like with those objects installed and some of those effects applied:
Screenshot of the previous Unity scene with reflection probes and post-processing effects applied. From a tutorial by Jessica Linker and Liam MacLean given on 6-17-21.
I was aiming for a hazy look, as if someone had groggily walked into a room and encountered three mysterious cubes. The students were asked to imagine a way to style the scene using what they learned. We ended up with a range of different looks for the same objects, from an underwater look, to a retro video game look, to a spooky, shadowed look, and so on.
I often tell them that the trick to a lot of this is to let Unity do the heavy lifting. Also, post-processing can take the same scene and change the mood quickly. It’s much better to algorithmically update the values on a post-processing profile than it is to try to swap out textures for every object in a scene to create an atmospheric mood.
Some of these effects don’t always play well in VR, but that’s something we’ll fine tune when we have an assembled project we can playtest.
3D rendered models of the 3rd edition of David Walker’s Appeal (1830).
This is just a quick update to show off a model made and textured by Ananya Dhandapani, with a couple of tweaks by Jessica Linker, to test a close-up of the wrappers in the test scene. Ananya will probably want to tell you more about this process at some point in the future, especially as it has involved her learning about the structure of pamphlets and asking archives about wrapper colors. For now, please enjoy this screenshot!
l to r: Kira Torrieri, Jessica Luo, Kesia Davies, Ananya Dhandapani. Student project team members not pictured include Jenia Browne and Liam Maclean.
We took a trip to Beacon Hill to check out extant structures, take notes on materials, and to try out some 360 cameras. Here’s an unstitched spherical photo of members of the student team standing in Smith Court in front of the African Meeting House.
Here’s what image looks like stitched. You can click and drag to rotate the image.
It’s really hard to find visual evidence of what Belknap Street (as it bordered West End) looked like in the period before the Civil War. So we thought you might want to see two wood engravings that appear in Sketches of Boston, Past and Present (1851). These wood engravings seem to have been used first in 1840s editions of the Boston Almanac, and were reused for this publication. They give us a brief glimpse of what the neighborhood looked like about 10-15 years after Walker’s death.
The Abiel Smith School and Belknap Street, from Homans, Isaac Smith. Sketches of Boston, Past and Present (Boston: Phillips, Sampson, 1851).
This view looks up Belknap Street, toward Boston Commons, from approximately Walker’s front doorstep. The second structure from the right is the Abiel Smith School, which served the African American community. It is currently the site of the African American History Museum. The structure furthest to the right is no longer extant, but would have abutted a structure that William Cooper Nell once occupied. This may well have been what Walker could see from his front window.
The African Meeting House and First Independent Baptist Church, from Homans, Isaac Smith. Sketches of Boston, Past and Present (Boston: Phillips, Sampson, 1851).
Were we to take right at the Abiel Smith School instead of heading toward the Commons, we’d find the African Meeting House and the First Independent Baptist Church. The wall on the left is the back side of the Smith School. There are some houses on the far right which no longer exist, but much about this view is still recognizable today.
Much the way we navigated the twentieth century by using phonebooks, nineteenth-century Bostonians used printed directories to locate individual households and businesses within the city. Directory listings tend to reflect head of household and may list the dwelling place, place of business, or both. Boston directories in the early nineteenth century included segregated listings of “People of Color” that allow us to track individuals on Belknap Street (the site of Walker’s first residence and African American neighborhood) during non-census years. This is particularly useful in Walker’s case since he only appears in the federal census for 1830, around the time of his death. The state censuses from Walker’s time in Boston, which were taken every 10 years in years ending as “5,” have not survived.
David Walker first appears in the city directory for 1825. His last appearance is in 1830. His address appears differently in each directory, so we’d like to break down what you might find in a directory through his example.
Walker’s listing in the 1825 Boston Directory
It’s probable that Walker arrived in Boston sometime around 1824; city directories took time to prepare for publication, so Walker likely had to set up shop in the city prior to 1825 in order to appear in the city directory for that year. Frustratingly, there appears not to have been a city directory issued for 1824, so there’s no way to tell if Walker may have arrived in late 1823 or early 1824. He is not listed in the 1823 directory. But in 1825, Walker is operating (likely out of a stall) at the City Market, which refers to the area near Faneuil Hall. Quincy Market, which is the adjacent building, was being built around this time as well, to accommodate an expanding number of vendors. This directory doesn’t provide information about where Walker was living that year, though it’s not uncommon for Bostonians to just list a place of business.
Walker’s listing in the 1826 Boston Directory
By 1826, Walker has a store front on Brattle Street; several other Black merchants sold clothing on this block.
David Walker’s listing in the 1827 Boston Directory
In 1827 we get quite a different entry. Walker was still selling clothes on Brattle Street, but at a different address. This may indicate that he moved to a different store front, but it also could indicate construction and subsequent renumbering of buildings on the street. We’re trying to do more research to work this out. The “h.” stands for house, which later had the address “8 Belknap,” but for much of the 1820s Belknap Street may have had no formal numbering system, which was not unusual for the time. In addition to being a clothes dealer, Walker is an agent for Freedom’s Journal, the first African-American newspaper in the United States.
Walker’s listing in the 1828 Boston Directory
In 1828, his household is omitted from the directory.
Walker’s listing in the 1829 Boston Directory
And there’s no change to the listing in 1829.
Walker’s listing in the 1830 Boston Directory
In 1830, however, Walker has moved his family to Bridge Street, a site that is no longer extant but has been incorporated into the Massachusetts General Hospital campus. In August Walker dies. The census for 1830 tells us that the household was quite large; in addition to Walker, his wife Eliza, and their daughter Lydia Ann, there was another adult woman and several children. Because the 1830 federal census only lists household heads by name, we’re currently trying to figure out who these other occupants might be.
While our project relies upon building 3D models with a modeling program, such as Google SketchUp or Blender, and then assembling them in the Unity 3D engine, we’ll be trying out various 3D and immersive technologies so the student team has an idea of how each works. Anticipating a future field trip, Angel Nieves and I went out to Beacon Hill to try out some socially-distanced 360 photography with a GoPro Max. This is one of the captures we made; since the camera can be operated remotely with a cell phone, we’re hiding around a corner and using the view finder to capture Smith Court.
Stitched panorama of Smith Court. William Cooper Nell’s House, the African Meeting House, and the Abiel Smith School are visible in this image.
The camera has a stitching algorithm that merges the capture from two spherical, 180 degree lenses by identifying overlapping artifacts in each image it is piecing together. A similar technique is used to merge pictures in photogrammetry processes. Photogrammetry is a popular way of creating 3D models of physical objects through series of circuitously-captured photographs.
360 photographs require EXIF data that tell certain programs to interpret them as spherical captures rather than flat, distorted panoramas. We installed a WordPress plugin that lets us tell WordPress whether to interpret an image as a 360 image, or just a normal .jpg. The VR toggle in the lower right creates a stereoscopic view for your phone, which can be viewed with inexpensive VR devices like Google Cardboard. You can also view the image immersively in 360 by navigating to this page with a head mounted display, such as an Oculus Rift S or Quest, selecting VR mode, and looking around.
Normally, there’d be more people on the street, but because we took this picture during the pandemic, Smith Court was pretty empty. The 360 capture will be useful for thinking about spatial relationships between buildings in this area.