Note: Why I will not use anymore

This is a short note on why I will not use Graph Commons for publishing data visualization in my research anymore. And also why I find it to have a misleading name. I have used it to map the early art trade at the Hôtel Drouot in the 1850s, and then again at the time of the German occupation of Paris in WWII, showing the networks of art dealers as experts, and the stately auctioneers. Also I used it in my research on anti-colonial exhibitions in Germany in the 1980s.

Detail of a network graph showing the cooperation between art dealers and art auctioneers in occupied Paris

I liked that readers could transform the graph themselves, zoom in, filter, etc. Why I cited it is because I liked the styling and grouping of nodes, which helped me to highlight research questions. That is why I have put Graphcommons URLs in my texts before. Now they changed their business model and my old links lead to a “not found” page, and at the moment to a small link to the “legacy” version of the graph. I could migrate the graphs, but if they are bigger than 500 nodes I would have to pay now. Which is honestly not worth it, as the platform has several other shortcomings, some of them I will mention below.

On the backend, there were always problems. The website was always buggy, I could only import through Google-sheets and not directly through the CSV option they had. Also they used Google Analytics. Which is both not what I want to do in research and teaching.

And then the name. I was careful to inform my students that I was using the tool for my research, but would not recommend it to them due to the connection to Google. I felt the need to do so, because it used this “commons” related marketing. But what did “commons” even mean for Graph Commons? Mostly that if you named a node, it would become a shared thing, you could find it in graphs by others. Not through any open data standard. Basically useless for computational analysis.

Is there a commons platform for linked open data? Of course there is. You can use Wikidata, which is a real digital #commons platform. Or to publish a dataset, you can use your academic repository. I use Depositonce of the TU Berlin, I also have used Github before for sharing data and code (Github is now owned by Microsoft, and of course also not a reliable place for research code and data) but I will now take better care that everything is on a proper repository with a citeable DOI. Or you can use open repositories, like Zenodo.

People can drop this data in their prefered dataviz tool. There is no citeable interactive data visualization that way though. I just hope academic data repositories will have more data visualization functionality in the future, in order to do the things that Graph Commons did for me, but in a sustainable and ethical way. I really would find citeable DOIs for interactive data visualization very helpful for digital humanities.

Das könnte dich auch interessieren …

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht.

Diese Website verwendet Akismet, um Spam zu reduzieren. Erfahre mehr darüber, wie deine Kommentardaten verarbeitet werden.

Suche in OpenEdition Search

Sie werden weitergeleitet zur OpenEdition Search