Internet Technologies

The next speaker in this Social Media & Society 2018 panel is Lene Pettersen, who begins by highlighting the rise of the sharing economy. In this economy, the key stakeholders are service providers, users, and intermediaries, and these may not necessarily represent traditional commercial actors. Professionals and small firms are now emerging that use sharing economy platforms to provide professional services.

The final speaker in this session at ICA 2018 is Di Zhu, who explores the effects of personal news customisation on user enjoyment. Customisation here is seen as different from personalisation, which is algorithmically driven: customisation involves active, deliberate user choices, for example by choosing specific topics or indicating their interest in the stories they encounter.

I’m not seeing quite as many ICA 2018 sessions as I might like, because of other meetings, but this Sunday morning I’m in a session on ‘fake news’, whatever we mean by that term. Mo Jang is starting us off. He begins by noting that the knowledge production and publication system has diversified with the increasing role of online publication, and this has undermined gatekeeping processes. This has also led to the increasing spread of unverified information, rumours, hoaxes, and other forms of ‘fake news’.

The next speaker in this ICA 2018 session is Zane Cooper, whose interest is in the material constitution of big data. Big data make use of earth and labour that do not easily track with its digital manifestations: they generate a long supply chain of physical hardware that supports the big data cloud. There is therefore a need to distinguish between what big data infrastructures are (their constitutional logic) and what they do (their operational logic).

The next ICA 2018 session is on algorithmic culture and starts with Stina Bengtsson, whose focus is on the ethics of algorithmic culture. Apple’s AI assistant Siri is an example for this: it has been made to swear and say inappropriate things, and there are real questions about the ethics of subverting algorithmic culture in this way.

The final speaker in this ICA 2018 session is Elizabeth Dubois, who again highlights the moral panics about the effect of ‘the Internet’ on information flows. But there are many different media and platforms, where users exercise different media use choices. There is a need to better measure media habits, therefore, including their specific diversity, timing, and tactics.

The next speaker in this ICA 2018 session is Laleah Fernandez, who begins by highlighting the moral panics around echo chambers, filter bubbles, and ‘fake news’. There is limited evidence that these issues are major concerns, but to the extent that these are genuine problems, key users might be useful in addressing these problems, by nudging vulnerable users towards more sensible behaviours.

The next speaker in this ICA 2018 session is Bibi Reisdorf, whose focus is on the role of algorithms in shaping information flows, and on users’ understandings of the impact of such algorithms. Algorithmic literacy is not yet well researched; it extends beyond digital literacy and is specific to different platforms, too.

The second day at ICA 2018 starts for me with a panel on the personalisation of search, and the first presenter is Grant Blank. He begins by noting the importance of free-flowing information for society, but of course the media through which such information flows have changed over time, and this has affected media biases. Contemporary media now form a diverse media ecology.

Also in connection with the AoIR 2017 conference last week, I answered a few questions about the field of Internet research, and the conference, for the University of Tartu magazine. Here is what I had to say:

What are the major challenges in Internet research?

The central challenge is the object of research itself. The nature of the platforms, content, communities, and practices that constitute 'the' Internet is constantly and rapidly in flux – we are dealing with platforms like Snapchat that didn't exist ten years ago, and with practices like 'fake news' that were nowhere near as prominent even two years ago as they are now. This necessarily means that research methods, approaches, frameworks, and concepts must change with them, and that the toolkits we used to understand a particular phenomenon a few years ago may no longer produce meaningful results today. But at the same time we must beware a sense of ahistoricity: 'fake news', for example, does have precedents that reach back to way before the digital age, and we can certainly still learn a lot from the research that studied propaganda and misinformation in past decades and centuries.