Turing’s Test 121
In using this phrase, I am not particularly referring to augmented reality journalism, a potentially interesting medium but one which, frankly,
still remains in a very embryonic stage at the time of writing. This is
instead the use of things such as big data and technologies to make journalism more effective and more efficient – a practice which is as old as the
deployment of the printing press for seventeenth-century news-sheets. For
Eric Eyre, the ability to analyse and track data is what helped him write
his story, organising the information that he gathered via very human
perseverance. In his 2012 analysis of machine-written news, van Dalen
observed that journalists did not simply reject automated assistance, but
instead – with regard to algorithmic sports stories – began to re-evaluate
their own core skills and consider ways in which they could make their
content “more human”, with a greater emphasis on interviews, commentary, and context than pure data.
73
Similarly, Thurman, Dörr, and Kunert observe that while journalists are sceptical (rightly in my opinion)
about the ability of robot journalists to source news stories, the rise of
algorithmic journalism will also expand the depth, breadth, and immediacy of information for them to work with on their own stories.
74
In less than a decade, algorithmic journalism has become a fixed element of news cycles particularly in the USA and the UK. Viewed from
the perspective of sheer volume, it would not be unreasonable to assume that automatically produced content is the future of journalism,
and yet during that time, the advances made by software have largely
been quantitative (more stories) than qualitative (moving into completely
different forms and genres of storytelling). This is by no means a blithe
assumption that such software is incapable of these developments, although the distinction referred to in Chapter 1, between what Dreyfus
calls “knowing-what” (which can be codified) versus “knowing-how”
(which cannot), posits a potential hard limit to what AI will be able to
achieve. For the foreseeable future, algorithmic journalism will work best
with information that is highly structured and in the public domain –
and the effect of such information on creating narratives from the huge
amounts of data which satisfy such conditions should not be underestimated. For more complex alternatives, however, the immediate changes
that are taking place are a greater use of algorithms and software to
help human journalists collate and understand data, what is called here
augmented journalism: the reporter today must be a cyborg, perhaps,
but the role of people in crafting stories remains as important as ever.
Notes
1 Jason Hall, Nineteenth Century Verse and Technology: Machines of Meter,
London: Palgrave, 2017, p. 113.
2 Automated Insights, “Rite Aid Posts 3Q Profits”, Yahoo! Finance, 3 January
2018, https://finance.yahoo.com/news/rite-aid-posts-3q-profit-212946446.
html?guccounter=1.
In using this phrase, I am not particularly referring to augmented reality journalism, a potentially interesting medium but one which, frankly,
still remains in a very embryonic stage at the time of writing. This is
instead the use of things such as big data and technologies to make journalism more effective and more efficient – a practice which is as old as the
deployment of the printing press for seventeenth-century news-sheets. For
Eric Eyre, the ability to analyse and track data is what helped him write
his story, organising the information that he gathered via very human
perseverance. In his 2012 analysis of machine-written news, van Dalen
observed that journalists did not simply reject automated assistance, but
instead – with regard to algorithmic sports stories – began to re-evaluate
their own core skills and consider ways in which they could make their
content “more human”, with a greater emphasis on interviews, commentary, and context than pure data.
73
Similarly, Thurman, Dörr, and Kunert observe that while journalists are sceptical (rightly in my opinion)
about the ability of robot journalists to source news stories, the rise of
algorithmic journalism will also expand the depth, breadth, and immediacy of information for them to work with on their own stories.
74
In less than a decade, algorithmic journalism has become a fixed element of news cycles particularly in the USA and the UK. Viewed from
the perspective of sheer volume, it would not be unreasonable to assume that automatically produced content is the future of journalism,
and yet during that time, the advances made by software have largely
been quantitative (more stories) than qualitative (moving into completely
different forms and genres of storytelling). This is by no means a blithe
assumption that such software is incapable of these developments, although the distinction referred to in Chapter 1, between what Dreyfus
calls “knowing-what” (which can be codified) versus “knowing-how”
(which cannot), posits a potential hard limit to what AI will be able to
achieve. For the foreseeable future, algorithmic journalism will work best
with information that is highly structured and in the public domain –
and the effect of such information on creating narratives from the huge
amounts of data which satisfy such conditions should not be underestimated. For more complex alternatives, however, the immediate changes
that are taking place are a greater use of algorithms and software to
help human journalists collate and understand data, what is called here
augmented journalism: the reporter today must be a cyborg, perhaps,
but the role of people in crafting stories remains as important as ever.
Notes
1 Jason Hall, Nineteenth Century Verse and Technology: Machines of Meter,
London: Palgrave, 2017, p. 113.
2 Automated Insights, “Rite Aid Posts 3Q Profits”, Yahoo! Finance, 3 January
2018, https://finance.yahoo.com/news/rite-aid-posts-3q-profit-212946446.
html?guccounter=1.