| author | Sambasivan, Nithya et al. |
| title | ‘Privacy is not for me, it's for those rich women’: Performative Privacy Practices on Mobile Phones by Women in South Asia |
| howpublished | (SOUPS 2018). Baltimore, MD: USENIX Association |
| year | 2018 |
Much design around security and privacy assumes (and advocates or attempts to enforce) a policy of “one user per account” (or device). This paper considers a sociological context (working-class women in India, Pakistan, and Bangladesh) in which that assumption is inappropriate. The authors conducted qualitative focus groups with ~200 women in 2017, attempting to understand how women in South-West Asia perceive and control their privacy on shared mobile phones, and how the designs of the devices do or do not fulfil the women’s expectations.
A bit over half the participants reported sharing mobile devices in their household. Devices-sharing seems to be more of a gender issue than a class issue: “When women had mobile phones, their devices were often viewed as ‘family’ devices.” In some cases the situation was worse than that; a woman’s use of technology was “mediated” by someone else or their phone was routinely checked by someone as a kind of surveillance. Participants generally did not view these situations as a violation of their privacy; instead they viewed “privacy” as a foreign cultural value that was not relevant to them. (Privacy as a cultural value was perceived as an artifact of the upper class.) In contrast to these espoused cultural norms, all respondents reported taking some actions to control privacy on their devices.
The subject-matter of this paper is really imporant. It’s intuitive that privacy as a cultural value is less common in low-SES contexts; validating and unpacking that intuition is important as our own culture’s relationship with privacy is evolving (and competing memetically with other cultures). I was particularly interested to read about how privacy-protecting behaviors and technology designs play out in contexts where privacy is explicitly not valued. Also, from a professional perspective, n=199 is impressive.