Welcome to OSINTEditor Sunday Briefing, a new weekly OSINT newsletter. As this is the inaugural edition of the newsletter, a brief introduction is warranted. The goal of OSINTEditor is to deliver cutting edge OSINT information and methodology in an open-access format. Moreover, there is an aim to combine Open-Source Intelligence with the power of geography and story-telling. This newsletter is part of that goal. Each Sunday morning this newsletter will deliver: a new OSINT techniques and tools, recap an insightful article from the global community of OSINT practitioners, debunk a piece of misinformation or disinformation, and then share a unique piece of cartography. Feedback on the content herein is most valuable as this is for you, the reader.

As far as who the author of this newsletter is… My name is Alexander Haynes. I am currently working on a Ph.D. somewhere in Taiwan, focusing on disinformation research. My location may bias some of the content in here towards the Asia-Pacific Region, but I’ll try to keep a global perspective. Before pursuing a Ph.D. I worked as a journalist (which I still dabble in) with a focus on historical perspectives. Geography and cartography research is very neat to me, so I cannot wait to dig out obscure maps and images from those fields at the end of each of these newsletters. And hopefully, this is the last you will read about me. Onto the content –

OSINTEditor Sunday Briefing: Identifying and dissecting propaganda

This week is about identifying and dissecting propaganda: a quick Twitter analysis tool, the Taliban’s Twitter presence via Line of Actual Control, and self-congratulatory People’s Republic of China propaganda on ‘correcting the minds’ of Hong Kongers.

MakeAdverbsGreatAgain (Allegedly)

The tool Make Adverbs Great Again has an arduous name that does not sound serious nor can it be abbreviated to MAGA for obvious reasons. I will just refer to it as Allegedly, the name in their homepage URL. Naming conventions aside, Allegedly is a fantastic tool for quickly identifying tweeting trends, including: repetition, repeated hashtags, time of day tweeted, service tweeted from, languages tweeted in, and type of tweets. The tool is powerful in providing a quick overview to help determine if someone might be a bot, what the bot’s behaviour could be pushing, and how to further investigate it.

Take for example Apple Daily Taiwan, a news service that can be a bit repetitive in their social media push. The ‘Repeated Tweets’ section at the bottom of the page is probably the most helpful part of Allegedly. As can be seen, Apple Daily Taiwan has been repetitive posting about the Tokyo Olympics, DogeCoin (狗狗幣), a poll regarding the preferred way to receive monetary vouchers, the Taiwan Golden Melody Awards, and Afghanistan. All of these tweets are perfectly reasonable and show the varied, authentic behaviour of a news outlet tweeting.

However, that does also bring up the downside of Allegedly – it can only analyse so many tweets. As shown, the tweets only go back to 22 July. This is likely not to do with the tool itself, but limits Twitter has put on the number of tweets allowed to be aggregated at once. Regardless, showing behaviour across time can help to identify what region a bot may be targeting. The quick overview that Allegedly provides makes it a great tool that anyone can bookmark and use to spot possible inauthentic behaviour.

Kandahar’s Don Drapers from Line of Actual Control

Sticking with the Twitter analysis, Line of Actual Control wrote a concise and well-constructed article investigating the rise of the Taliban’s Twitter presence and propaganda. LOAC delivers analysis using Twint, another more extensive Twitter analysis tool. (Because it is more extensive, investigating an account on Allegedly may be good practice to determine if it is worth the time to investigate using Twint).

The Taliban have an extensive social media campaign that has been effective due to targeted campaigns in multiple languages. The Taliban have been able to effectively engage users in English, Arabic, Pastho, Dari, and Urdu, each language having some specific propaganda associated with it. Their stark implementation of social media and an effective use of propaganda mean the Taliban will be lingering around with targeted campaigns for quite some time. Read Kandahar’s Don Drapers for the full analysis.  

State Media Influence into Hong Kong – is it really changing attitudes?

China Review News Network (CRNTT, 中國評論新聞網) released a report this week detailing the influence of Central Chinese State News Agencies, such as Xinhua News Agency and People’s Daily News Agency, on Hong Kongers. CRNTT concluded that these central news agencies have played a critical role in swaying public opinion in the safekeeping of the “One Country, Two Systems” agreement, pressuring the dissolution of pan-democratic civil institutions, and in the future will ensure that there is a, “Hong Kong ruled by Patriots”. The review further concludes that, “This [trending approval to states news agencies on social media] is an objective fact and what the people want!” (這隻客觀事實,也是民心所向!).

The report is self-congratulatory in the most pandering of manners. While the assessment is correct about some influence, their methodology to infer what the data means is questionable and misrepresents what is happening in Hong Kong.

The first page of the assessment is the most accurate part of the report, for all the wrong reasons. The chart shows significant developments in Hong Kong’s political situation that correspond to articles from a central news agency. These include such op-eds as the July 2021 Xinhua and People’s Daily articles which referred to the Professional Teachers’ Union (PTU) as a “malignant tumour” that needed to be eradicated and a July 2020 article that called for an investigation into a pan-democratic primary. In the weeks following those respective articles, the PTU disbanded and 53 pan-democrats were arrested under National Security Law charges. The analysis claims that these only happened because the central media was able to turn the eyes of Hong Kong towards these issues.

From the inside of the propaganda machine, the above tale sounds congratulatory and paints the central media as one that can ‘correct’ the ‘ills’ of Hong Kong. From the outside, however, it is evident these are state outlets creating public pressure out of internal pressure that was experienced before the op-eds communicated such to the public. More bluntly: state media is part of the machine to pressure, troll, and penalise opposition in Hong Kong, Taiwan, and abroad.  

The second page of the assessment reaches a level of absurdity there is a relevant XKCD comic. The first graph represents Hong Kong netizen reactions to central news articles with “approval,” “sarcasm/mockery,” and “dissent”. This data comes from ‘social media’ and does not specify which social media – just local Hong Kong social media. There are realistic correlations to events, such that ‘likes’ would have been expected to go down through the end of 2019 as pro-democracy protests continued, and reached a low during 2020-Q2 due to the spread of COVID-19. Sarcasm and dissent decline, however, and CRNTT relates this to the welcoming of “True Patriots ruling Hong Kong” after the implementation of the NSL.

Even if these charts are sourced from authentic data, there was a wave of Hong Kongers deleting social media accounts with the implementation of the NSL, which would have pulled most of the ‘opposition’ from social media. Furthermore, an ‘opposition’ member is quoted as saying, “Times have changed,” (時代變了) with regards to social media becoming more supportive of the central news agency; there is no source for this quote. Times have certainly changed in Hong Kong, just not exactly in the way this report assumes.

The third page represents the topics of different central news agency articles, and how much interaction certain articles have received from Hong Kong netizens. The top three are: National Security (國家安全), COVID-19 (新冠疫情), national defence and diplomacy (國防外交). There is not necessarily refutation of this, although once again, no data source is present. However, the article positions these topics as representative of Hong Kongers the resolution of “One country, two systems,” in their minds. Such conclusions cannot be drawn from a pie chart. Furthermore, how this pie chart even represents ‘interaction’ (was it based on comments, likes, views?) is inconclusive.

The article above is fairly blatant propaganda with statistics taken from unlabelled sources that Hong Kongers have expressed their desire for such change. The reality on the ground is much darker with a growing number of citizens and politicians self-censoring. Regardless, this report and these graphs may be touted around as evidence of the changing hearts and minds of Hong Kongers. The narrative will be cycled around in op-eds and such papers as Global Times. By December, this misinformation may even make it into mainstream western newsrooms. This is how such data was baked.

A Sober Map – Next Stop: Nothing Happens

(Original source for the comic is Hong Kong Worker (職人阿港) on Instagram).

The above is technically a comic and not a map. But there is a map on the third strip depicting the Kowloon side of the Hong Kong MTR, where “Nothing Happens” between Mong Kok and Sham Shui Po. That Kowloon red stop is supposed to be Prince Edward station, where on 31 August 2019 police responded to a man wielding a hammer on a tram, only to pepper spray, beat, and tear gas an entire tram filled with mostly protestors returning from the island.

The following Twitter thread from citizen investigator Lazyyyyyyy details what happened on 831, now two years ago. 香港人加油.