search

JeremyTygh / Norse-Tales

A research project examining animal motifs in George Webbe Dasent’s translations of Popular Tales from the Norse.
1 stars 0 forks source link

Project-Update-8 #9

Open clk125 opened 3 years ago

clk125 commented 3 years ago

during our meeting this week we went over the pages that each of us populated as much as we could for this week. We also discussed the potential use of mapping for our project and have asked Dr. Birnbaum for some assistance on this front. For next week we are hoping to have more complete drafts of the actual text that will be populating our pages so we can work on finalizing our content as well as looking for images that could make our site more aesthetically pleasing. We are also thinking about how we can use mapping and JavaScript to help us on our project.

djbpitt commented 3 years ago

@clk125 Please let me know what you'd like to do with mapping.

JeremyTygh commented 3 years ago

We were thinking about mapping the characters’ perceptions of each other (e.g., positive or negative) in their interactions.

djbpitt commented 3 years ago

@JeremyTygh Do you have some ideas about the kind of representations you would want to create?

JeremyTygh commented 3 years ago

Possibly a graph with nodes representing characters/species connected by edges. I was thinking the edges could be made thicker to represent an overall positive relationship and thinner to represent an overall negative relationship (different colors could also be used to indicate the frequency of interaction). Do you think this would be a good utilization of networking technology?

djbpitt commented 3 years ago

@JeremyTygh Width may not be the clearest indicator of relative positive or negative relationship. Take a look at http://rap.obdurodon.org/sngraph.html (authenticate with userid "minas" [no quotes] and password "russianrap" [no quotes]). They used a color gradient to indicate degree of positive or negative value, and line thickness to indicate number of cooccurrences, which is sort of the opposite of what you describe. Frequency of interaction can be counted, so it seems to fit thickness more naturally than color would—not that color isn't numbers under the hood, but humans don't perceive those numbers proportionally.

Whether a network graph will reveal much of interest may depend on the volume of data. For example, if most of your characters and species occur or interact only a few times, the data may be too sparse for the graph to tell an interesting story. With more data, though, a graph may show the relationships more clearly than lists or tables or prose, which is sort of the reason we use visualizations in the first place.

Which of you is working with the network analysis software, and how far along are you in terms of being able to generate graphs of the type of information you care about, even if you haven't yet finalized the details?

JeremyTygh commented 3 years ago

We haven't necessarily assigned anyone to this task yet, but I would be willing to give it a try! We have all of our markup completed, so I believe we are far enough along to begin experimenting with visualizations of this type.

djbpitt commented 3 years ago

@JeremyTygh Check in with Elizabeth in the Inaugural Addresses project. She's overseeing their network development and she's met a few times with Elisa to discuss it, so she should be able to get you started with Cytoscape, if you think that would be helpful. I'd suggest working through Elisa's https://ebeshero.github.io/thalaba/cytosc.html, which is a good step-by-step tutorial.

There are other applications than Cytoscape that perform network analysis, so if you're already comfortable with one, you're welcome to use it. We generally recommend Cytoscape because it seemed to be the best of the lot when we reviewed the alternatives, and since Elisa uses it regularly, she's able to advise when we need help. To include visualizations in a web site we recommend exporting them from Cytoscape as SVG, since SVG has several advantages over vector formats for this type of data.

JeremyTygh commented 3 years ago

I will make sure to check in with Elizabeth and begin working through Elisa's introduction to Cytoscape. Thank you for the help!

peterbussch commented 3 years ago

Our team is also working with javascript and html/css for website design. We have a working navigation bar now, which is really helpful, but we we need to integrate the other work we're doing as a team. Looks like you all are doing great!

EdwinRosenberg commented 3 years ago

I just wanted to say your Website looks amazing! It seems like you guys are on top of everything & doing well. Idk how you would use Javascript for your project, but to reiterate what Professor Birnbaum said, you should talk to Elizabeth from our group, she's been killing it with the network analysis. I would say line thickness should refer to frequency and color should relate to positive/negative; I think that would be more intuitive to the user. I think it would be interesting to see the relationships between characters visualized this way, provided it would help you answer your research question.