In previous weeks we have looked at data in numerical or even alphabetical form, and I think we got a lot out of that information such as trends as well as what data is out there, where and how to get it. However a lot of data is very boring to look at and very long. This means, in particular for humanities data it looses a lot of its feeling and understanding. This week we looked at tools that can take in data and then churn it out in one of the visualisations, and how much customisation we can do like in Plotly and Raw. Examples like Mapping Police Violence and The Preservation of Favoured traces utilise movement and colour to very effectively and viscerally portray data. Mapping Police Violence had data points pop up like gun shots, and the rate at the changes of The Origin of the Species physically showed the theme of the progress of science. I thought these were really fascinating examples and summed up what we were trying to identify and explore in the class. Although I am more interested in mapping for my project I gained from the class that some visualisations match some data better, which can help me in selecting visual components for my project.
Here is my stacked bar graph using plotly rather than QueryPic
This class was very interesting to me, especially as I am taking literature as a major. What words we use and how we use them are vital in any field of study, as the communication of information, feelings and history is a salient part of humanity and society. Quantifying them into a manageable state or data set makes it easier to pick apart, reflect and study them was the purpose of this weeks class. It’s cool how you can start off with a question like when did we start saying World War I, or you don’t have to have a question at all and finding answers with in what you’re given is also a way of using the data. For example it was interesting in a sample from SameDif the comparison between Hilary Cliton’s speeches and Donald Trump’s. It showed what words they both used, and what words each only used with no cross overs. Clinton hadn’t used Mexico once, whereas it was one of Trump’s most used words. These programs offer such a simple way of finding windows into history by seeing how we write and talk about the world around us.
The management of digital heritage is definitely becoming more of a concern today then ever before. The problem I feel with digital heritage is that there is just too much. In the past where physical objects would get destroyed due to age, deterioration, handling etc., the collections of heritage available for preservation would get smaller per subject field as time goes on. When it comes to digital heritage as time goes on, the accumulation per subject field becomes more.
It is lovely to save everything but perhaps like trying to save everything in a fire we need to let some things go. As well as a system designed for improving data management we also need a system to dispose of data. Perhaps we could employ a bot for the job?
I had a go a creating a graph using Plot.ly this week. It absolutely looks no different to the graph from the tutorial however I did make this one all by myself, from scratch. #proudmoment Embedding the graph was easier said then done. Firstly, one must enable the private share link and secondly since I have coded this weeks blog in the text format, I needed to change the embedding code from an ‘iframe’ to ‘html’. I also decided we needed some style and colour on this blog page. Note: my coding skills are young and fresh with much to be learnt so basic coloured headings and some font change is all I could achieve for this week.
A Plot.ly Graph
This week we looked at the myriad of options concerning data and metadata, where to find it, what to do with it and what it is. I think that the Posner article really set the tone this week for me from doing the reading prior to the class, when he drew an analogy of what you would feel like if someone called your photo album a data set. I though that summarised his article well, illustrating the difference between humanities data and scientific data. Humanities data is most prominently linked intrinsically to people, culture, religion and real lives. This makes context and interpretation even trickier, but more interesting, to deal with, and I think that that was addressed well in acitivites and discussion during the class.
The class really opened up to me how much data there is, not how many records because obviously there’s lots of records of things especially in Western culture, but how many categories of ‘stuff’ is out there. Colonists bank records, water qualities, I even found how a dataset of average fruit intake for an Indigenous Australian primary school student. This class highlighted the importance of digital heritage, in particular in unlocking, fixing, and making sense of historic data to then churn it out in accessible format. What is the point of data otherwise? All of the things we found are super important and interesting to a plethora of people, cultures, organisations and have so many uses whether research, predictive or even personal. As I mentioned before because this is humanities data that stakes are so much higher as they are more actualised in the real world. However there’s no point if data is not in context, can not be categorised, understand or even read properly if there is mistake. The tools we used in the class fixed all these problems, plotly developed visual representations of data to be interacted and understood, openrefine to fix and categorise. Not only did we learn to use these tools, we understood in a broader sense what data is, how it can be used and finally how we can utilised in our own research contexts in cultural heritage.
One thing that I find really interesting was one of the graphs on plotly someone else had made, a pie chart illustrating the most used colours in Van Gogh’s artworks. What a cool use of data as well as graphing to portray something so organically creative and artistic and well non-mathmatical into a tangible piece of data – a good balance between the scientific data and humanities data Posner was discussing.
Here is my graph for male and females in 1901 – silly outlier Adelaide 🙂