The main goal in starting this blog is to describe how my projects have come about, and to outline my thought process as I continue to work. I’m sure many people out there could see my projects and know exactly how I did them, but several people have asked me to explain why I chose them, or how I came up with my idea.
I’ll be starting with Flowers in a Glass Vase. As I alluded to in my first post, this was my first DH project. After sitting in on three or four meetings with other GAs describing the projects they wanted to accomplish over the semester, I eventually decided I wanted to do something relating to Dutch flower paintings. (Most people assume this is because I love flowers and I’ve had lots of people since ask me about gardening . . . Actually, I decided to work on this topic because it’s an area of my field that I knew relatively little about, and wanted to expand my knowledge.)
I came up first with the idea of annotating a painting, before I was even aware of Omeka and Neatline. I wanted some way for people to learn what every flower was–and especially when they bloomed–simply by clicking on the image (requiring no knowledge of botany from the user!). A fellow GA initially suggested ThingLink, an online tool where you can annotate a still image with points containing written text and links. While ThingLink did what I was looking for, it’s not the most aesthetically pleasing and wouldn’t ever be more than just one annotated image. I next tried Prezi at the suggestion of another GA, but ultimately decided that it couldn’t do what I was envisioning.
Soon after, I was directed to Omeka, but more specifically Neatline. Quint Gregory, the director of the Collaboratory, set up a site for me and I spent several days just trying out different things (after I finally learned how to edit a Neatline page, which took me an embarrassingly long time to figure out). I was fortunate to be working in the Collaboratory the same semester as Nicole Riesenberger, Grace Yasumura, and Cecilia Wichmann, the three of whom were not only all working on their own Omeka projects, but were together building a best practices guide. I was about two weeks behind them in learning how to use Omeka and greatly benefitted from all their discoveries (and avoided their mistakes!).
Initially I was unsure about what Omeka–and more so Neatline–could do for me. Neatline is primarily used for mapping, plotting data “in space and time” (as their slogan explains). Time is something I was very much interested in, and while I was still hesitant about the mapping component, it was the possibility of a timeline (using the plugin SIMILE Timeline) that drew me in. (Eventually, I ended up not using it, but I’ll get to that later.) Further exploration of Neatline’s website and demos also made me realize that I didn’t need to rely on mapping–or rather, that I didn’t need to be doing traditional mapping. Instead of using a world map as the base for my virtual exhibition, I could use a still image, and “map” the flowers over it. Instead of plotting a journey or noting important sites, I could make each flower their own site, and use Neatline hotspots to provide information about them.
It was watching Grace testing out the “draw SVG” feature on Neatline that inspired me to take this a step further. The default for Neatline hotspots is just points. I could have done this–and it would have looked like a more polished version of my ThingLink. But as Neatline allows you to create custom shapes for your hotspots, I could make each flower into their own hotspot, effectively covering the whole vase. (Unfortunately for me, creating a custom shape requires an SVG code, which you can get from Illustrator . . . but not Photoshop. I had already meticulously “cut out” every single flower in Photoshop before I knew exactly what I was doing, and I wasn’t thrilled to find I had to do it again except in Illustrator. Soon enough though I had SVG codes to create a shaped hotspot for every flower, and a little trial-and-error moving had them in the correct place.)
Creating the shapes was only the start, however, as I had several decisions to make, which affected not only the aesthetic of my site, but also how visitors experienced and navigated the exhibit. First and foremost, I wanted to preserve the image, so that viewers could take it in, uninterrupted, if they so desired. This is where ThingLink was a problem, as well as regular point hotspots in Neatline. Bosschaert’s painting is a spectacular work of art, and I wanted my annotations to supplement it, rather than detract. So I decided to alter the transparency of the hotspots to have them be entirely invisible when they are not selected. I could only do this because I used shaped hotspots that covered the entire flower: if I had only used small points they couldn’t be invisible, as the user would never find them! As soon as the mouse hovers over the flower, the colored shape of the hotspot appears (and the flower’s name appears in the upper left corner) alerting you to the fact that there is something to click on. When you do click on it, the full record appears, giving you all the information about the flower and it’s blooming season, as well as three or four images, including photos and sixteenth- and seventeenth-century drawings. What I like most about this is how user-friendly it is: there are no instructions and you don’t need to be told how to use it; as soon as you move your mouse it becomes clear what the site does and how you use it.
Lastly, I decided to make the each their own color. I initially thinking of making the selection color uniform, and thought about using a plain greyscale. Eventually, however, I decided that grey just looked boring, and wasn’t celebrating the spectacular natural color and beauty of these flowers. So I made each selection color fit the flower (but not so close that a user wouldn’t realize it was highlighted–which proved rather difficult when it came time to do the white Alba Rose).
Another aesthetic choice I made here was using an Omeka plugin called “Hide Elements.” When adding items in Omeka, you fill out the metadata using Dublin Core, the standardized metadata vocabulary (title, creator, format, date, location, rights, etc.). While this is great for a work of art, for example, there is not a lot of metadata to keep track of for a flower–“blooming season,” for example, doesn’t fall under Dublin Core. So, to keep those unnecessary fields from showing up in each flower’s record, but to preserve the images (which were imported into Neatline as Omeka items), I used Hide Elements. This was one of the last things I did on the project, and something that felt a lot more important than it might seem: to me, it really felt like this site now had a customized appearance, and wasn’t simply filling in a standardized form.
The actual last thing I did was the timeline. This was one of my earliest ideas and something that felt very essential, but turned out to be the most vexing. Neatline does have its own timeline, but it’s not designed for 14 things happening over the course of only a few months, and got so crowded it could only show about four or five flowers at once. I decided to abandon Neatline and make my own:
I made it its own record and hotspot, and put it in the most logical place for a timeline: over the date, in the lower right hand corner.
Now of course, this isn’t everything I did. I could write about going through hundreds of pages of online versions of the Hortus Eystettensis (1613) and Hortus Floridus (1614) to identify plants which may or may not currently be extinct, creating a massive spreadsheet to keep track of all the flowers, spending several weeks tracing flowers in Photoshop and then Illustrator, spending several hours trying to place the hotspots correctly on the first try before I realized that you could drag them (not a great day), trying to find a font color that could work over all the flowers and then giving up and just putting the header text over the vase, or the dozens of background colors I tried before settling on a nice taupe (which looks great on all computers . . . except the screen in the Collaboratory, where it comes out a rather sickly mustard color). But reading about those things would probably be more tedious than actually doing them, and I don’t think they’re really necessary for understanding my process. But feel free to comment and ask if you want to know more!