3. Footnote

3.1 Research proposal

The last couple of notes I’ve described my evolving thoughts about the thesis focus: I want to deepen my understanding of news ecosystems methods and the scope of the water issues that I’m looking at, particularly to incorporate the political context.

I got the all clear on this in my supervisory meeting, though I need to do the actual work of redrafting my research proposal. I’ll prioritise that next fortnight.

3.2 News ecosystems

Ahead of the supervisory meeting I reread a paper that has been a bit of a guiding light over the past few years – Stonbely et al.’s 2019 review of news ecosystems studies, and the typology that emerged from that work. It’s very clarifying and addresses with evidence what I discussed last time based on vibes: too much news ecosystem work is small sample, tightly-bounded case studies that aren’t generalisable, and there needs to be more work done developing scalable methods.

On exactly that note, Khanom et al. 2023 and Babel 2024 describe their work building a web crawler to scrape news articles in Missouri and natural language processing to assess coverage of community information needs. It’s very interesting work and seems to be able to do at a much larger scale what we were doing at PIJI. That includes some of the shortcomings of the Australian News Sampling Project, such as defining communities by political boundaries rather than organic ones, and limiting assessment to digitally-accessible professional news publications.

It’s amusing to find the authors write of the difficulty of finding or maintaining an up-to-date directory of news outlets that includes active URLs. I agree!

Babel also has a framework for assessing the health of local news ecosystems beyond the pretty ugly binary of presence / absence of a masthead. Continuing the ecological metaphor, he presents a version of the IUCN Red List (Thriving, Moderate, Critical, Endangered, Extinct) with markets allocated to a category based on the frequency and quantity of articles in that place that address CIN topics. It’s an interesting approach.

Related, I’ve put on my list for next week to review the Hess et al. Local News Proximity Index, which I remember being a more sophisticated version of this idea. I’ve read bits and pieces about it but I haven’t given it the close attention that it is due.

I don’t yet have a fixed idea of how I’ll take up Stonbely et al.’s call for deeper research methods. Deepening my assessment to incorporate both professional news outlets and other information sources, thereby mapping out a local information ecosystem beyond just its journalism inputs, looks more like the call they make for a rhizomatic approach.

That idea is pretty labour intensive though, since it involves me being in community in order to access information not available digitally, and so it is firmly on the case study side of the typology, not the scalable one. A compromise then might be to learn those scraping and NLP techniques to reduce the burden of that work instead – perhaps the work that enables the rhizomatic metaphor isn’t scalable, but I could improve my processes to conduct future content analyses at larger scale.

Another option might be to use those same tools to address the political v. organic communities issue. An earlier version of my research included a proposal to map coverage of different water assets – I could pick that idea up again and use a place tagging tool like spaCy to track the imagined ecological communities of particular newspapers. How much of the Murrumbidgee River is of interest to the Leeton-based newspaper The Irrigator? Which channels and wetlands?

That question might be of specific and narrow interest to me, a person obsessed with birds and swamps and wanting to include ecological frames in everything I do. But the method for assembling and assessing that data could be more broadly applicable for mapping coverage and, in so doing, getting past the imperfect use of local government areas as the organising structure for local news geographies.

3.3 Obsidian update

Speaking of new tools, I made a minor adjustment to my Obsidian workflow.

Like many, I use an adapted version of the Zotero > Obsidian workflow developed by Alexandra Phelan. I add to Zotero, highlight and annotate passages on the PDF reader there, then pull that data across to Obsidian. My annotation database then lives in Obsidian, where I do my writing, and Zotero remains my reference manager.

I don’t recall exactly how I’ve adapted her model, but I think that using each highlight colour in Zotero for a different purpose, labelling that purpose and then bringing both across into Obsidian was one of them. I know that bringing the page numbers across alongside each quote was my addition, because I remember what a pain it was to get that to work.

As well as the seven colours noted on the screenshot, I was also using grey to identify interesting references that I should track down. I found that this was not fitting with my work though – I’m hardly going to go trawling through a dozen different annotation notes in order to identify papers to add to my inbox for future assessment.

This past week I’ve fixed this with an intentionally basic approach: a new note that looks for grey in my Zotero annotated PDFs and on import into Obsidian, appends all of them to the bottom of a running ‘Reading Leads’ document.

This isn’t checking for duplicates, it’s not scanning for whether it already exists in my Zotero or Obsidian library, it’s not reformatting to a particular citation style – it’s just a very simple, ongoing, centralised list of everything that has stood out to me as at least worth finding out more about. Hugely helpful.

3.4 Recently

Babel A. 2024. Diagnosing news deserts: a path towards scalable assessment. Master of Arts Thesis. University of Missouri.

Campa P. 2018. Press and leaks: do newspapers reduce toxic emissions? Journal of Environmental Economics and Management 91(1): 184-202.

Khanom A., Kiesow D., Zdun M. and Shyu C. 2023. The news crawler: a big data approach to local information ecosystems. Media and Communication 11(3): 318-329.

Mathews N. 2022. Life in a news desert: the perceived impact of a newspaper closure on community members. Journalism 23(6): 1250–1265.

Neff T., Pawel P. and Pickard V. 2022. Philadelphia’s news media system: which audiences are underserved? Journal of Communication 72(4): 476-487.

Shaker L. 2014. Dead newspapers and citizens’ civic engagement. Political Communication 31(1): 131–148.

Smethers S., Bressers B., Willard A., Harvey L. and Freeland G. 2007. Kansas readers feel loss when town’s paper closes. Newspaper Research Journal 28(4): 6-21.

Stonbely S., Konieczna M. and Holcomb J. 2019. Mapping local news ecosystems, Phase 1: Meta-review of the literature, a typology of ecosystem studies, and a proposed method for the ideal ecosystem study. Montclair, NJ: Center for Cooperative Media, Montclair State University.