Scan barcode
gretatimaite's review against another edition
I have published a long text on data feminism in my blog post which you can find here.
Also, you can find a much shorter version here. Just expect some spoilers (if you can spoil non-fiction) :)
One of the great advantages of this book is its accessibility without being simplistic or reductive. It might be different to someone who's not familiar with feminism but even then, I believe, over the course of reading, it would become clear what it's about.
Data feminism, then, is:
It tries to understand the inequalities produced and reinforced by data science and challenge them by data science itself. Moreover, its core belief is that of co-liberation - an idea that the basis of social issues is due to prevailing inequalities. Thus, drawing on bell hooks, D'Ignazio and Klein argue that data feminism is for everybody and it is about power.
The book is then divided into 7 chapters and each of them corresponds to one of the guiding principles of data feminism. I will briefly try to summarize what I have taken from each of it:
1. Examine power
a. Think of who has power and answer these three questions:
i. Data science by whom?
ii. Data science for whom?
iii. Data science with whose interests and goals?
2. Challenge power
i. They make an interesting distinction later on between being 'just' and 'ethical'. Being ethical, hence, is not enough to resist structural inequalities as it would assume there are some bad apples (people or tech) rather than that there's a structural issue. However, it's a good starting point to move towards justice.
3. Elevate emotion and embodiment
a. Focus on data visualisation (and visceralization)
i. It doesn't show 'objective' truth
ii. Emotion (or affect) is important to consider
b. Think of data visualization as rhetoric
4. Rethink binaries and hierarchies
a. Counts only what gets counted
b. 'when a community is counting for itself, about itself, there is the potential that data collection can not only be empowering but also healing.' (p. 120)
5. Embrace pluralism
a. Value the diversity of positions
b. 'There is a story about how every evidence-based argument comes into being, and it is often a story that involves money and institutions, as well as humans and tools. Revealing this story through transparency and reflexivity can be a feminist act.' (p. 137)
i. Hence, reflexivity is crucial
c. It's not enough just to use data for good, one should use it for co-liberation
6. Consider context
i. Context is also important when communicating and framing the data
ii. Thinking of 'raw' data production can also be a feminist strategy
7. Make labor visible
a. Often designing a data product, many contributors remain invisible and uncredited as we tend to credit only the work we see
i. Also think of all the emotional, affective, and care work that is done but usually goes unnoticed
b. Cultural data work - describes the work, often invisible, done by people who are responsible for moderating content online, for example by ensuring that FB feed is free of child pornography
c. The emergence of data production studies might help to uncover and challenge power dynamics by tracing down the roots of algorithms and visualizations back to the human and material sources
So here it is. I've done my best to keep it short as well as understandable. I don't really have much of a critique but that sometimes I did feel a bit bored. I think it's because I'm familiar with feminism and some points/ideas seemed common-sense. It doesn't undermine the importance of the book and I will come back to it, especially for the examples it uses. Some of them that I consider especially interesting are:
1. anatomyof.ai
2. womenstats
3. antieviction map
Also, you can find a much shorter version here. Just expect some spoilers (if you can spoil non-fiction) :)
Spoiler
First of all - a very VERY good book. I've never really thought about data science through the (intersectional) feminist lens. Not that I know much of data science but willing to change it :)
'a central aim of this book is to describe a form of intersectional feminism that takes the inequities of the present moment as its starting point and begins its own work by asking: How can we use data to remake the world?' (p. 5)
One of the great advantages of this book is its accessibility without being simplistic or reductive. It might be different to someone who's not familiar with feminism but even then, I believe, over the course of reading, it would become clear what it's about.
Data feminism, then, is:
'a way of thinking about data, both their uses and their limits, that is informed by direct experience, by a commitment to action, and by intersectional feminist thought.' (p. 8)
It tries to understand the inequalities produced and reinforced by data science and challenge them by data science itself. Moreover, its core belief is that of co-liberation - an idea that the basis of social issues is due to prevailing inequalities. Thus, drawing on bell hooks, D'Ignazio and Klein argue that data feminism is for everybody and it is about power.
The book is then divided into 7 chapters and each of them corresponds to one of the guiding principles of data feminism. I will briefly try to summarize what I have taken from each of it:
1. Examine power
a. Think of who has power and answer these three questions:
i. Data science by whom?
ii. Data science for whom?
iii. Data science with whose interests and goals?
2. Challenge power
it 'requires mobilizing data science to push back against existing and unequal power structures and to work toward more just and equitable futures.' (p. 53)
i. They make an interesting distinction later on between being 'just' and 'ethical'. Being ethical, hence, is not enough to resist structural inequalities as it would assume there are some bad apples (people or tech) rather than that there's a structural issue. However, it's a good starting point to move towards justice.
3. Elevate emotion and embodiment
a. Focus on data visualisation (and visceralization)
i. It doesn't show 'objective' truth
'The belief that universal objectivity should be our goal is harmful because it's always only partially put into practise.' (p. 83)
ii. Emotion (or affect) is important to consider
b. Think of data visualization as rhetoric
4. Rethink binaries and hierarchies
a. Counts only what gets counted
b. 'when a community is counting for itself, about itself, there is the potential that data collection can not only be empowering but also healing.' (p. 120)
5. Embrace pluralism
a. Value the diversity of positions
b. 'There is a story about how every evidence-based argument comes into being, and it is often a story that involves money and institutions, as well as humans and tools. Revealing this story through transparency and reflexivity can be a feminist act.' (p. 137)
i. Hence, reflexivity is crucial
c. It's not enough just to use data for good, one should use it for co-liberation
6. Consider context
'Big Dick Data is a formal, academic term that we, the authors, have coined to denote big data projects that are characterized by masculinist, totalizing fantasies of world domination as enacted through data capture and analysis. Big Dick Data projects ignore context, fetishize size, and inflate their technical and scientific capabilities.' (p. 151)
i. Context is also important when communicating and framing the data
ii. Thinking of 'raw' data production can also be a feminist strategy
7. Make labor visible
a. Often designing a data product, many contributors remain invisible and uncredited as we tend to credit only the work we see
i. Also think of all the emotional, affective, and care work that is done but usually goes unnoticed
b. Cultural data work - describes the work, often invisible, done by people who are responsible for moderating content online, for example by ensuring that FB feed is free of child pornography
c. The emergence of data production studies might help to uncover and challenge power dynamics by tracing down the roots of algorithms and visualizations back to the human and material sources
So here it is. I've done my best to keep it short as well as understandable. I don't really have much of a critique but that sometimes I did feel a bit bored. I think it's because I'm familiar with feminism and some points/ideas seemed common-sense. It doesn't undermine the importance of the book and I will come back to it, especially for the examples it uses. Some of them that I consider especially interesting are:
1. anatomyof.ai
2. womenstats
3. antieviction map
drj's review against another edition
adventurous
challenging
inspiring
reflective
slow-paced
5.0
Absolutely essential for anyone dealing with data, public policy, or marginalised communities.
madrosie's review against another edition
challenging
informative
inspiring
reflective
medium-paced
4.75