Update references to be more explicit?

[AoifeHughes]

References are difficult to link in text, this could become confusing for readers. For example in module 1.

rds-course/coursebook/modules/m1/1.1-WhatIsDataScience.ipynb

Line 59 in 53c121a

"Together with data scarcity, the other aspect that we should never underestimate when we speak about data is whether they are representative of the phenomenon we aim to study. In a recent article by Anna Rogers, the author considers the following argument: “the size of the data is so large that, in fact, our training sets are not a sample at all, they are the entire data universe”. Rogers replies to it by saying that this argument would stand if the “data universe\" that we use for training for instance a speech regognition system was the same as “the totality of human speech/writing\". It is not, and will hopefully never be, because collecting all speech is problematic for ethical, legal, and practical reasons. Anything less than that is a sample. Given the existing social structures, no matter how big that sample is, it is not representative due to (at least) unequal access to technology, unequal possibility to defend one’s privacy and copyright, and limited access to the huge volumes of speech produced in the “walled garden\" platforms like Facebook. \n",

The use of Roger's name does not make it clear that they are being referenced formally.

rds-course/coursebook/modules/m1/1.1-WhatIsDataScience.ipynb

Line 265 in 53c121a

"Rogers, A. (2021). Changing the World by Changing the Data. arXiv preprint arXiv:2105.13947.\n",

I can see two possible solutions to making references more explicit :

1. Hyperlink to the existing references section when appropriate
2. Use BibTeX https://github.com/takluyver/cite2c

• This would be my preference, and I volunteer to experiment with this.

[AoifeHughes]

Did some digging - this seems more promising https://jupyter.brynmawr.edu/services/public/dblank/jupyter.cs/Examples/Calico%20Document%20Tools%20and%20Bibtex.ipynb