Open BUK: Digital Labor, Media Investigation and the Downing of MH17

Abstract

This article considers the unique forms of digital labor that emerged in the wake of the downing of Malaysian Airlines flight MH17 over Donetsk, Ukraine in July of 2014. Whereas such investigations traditionally rely on expert analysis and strict information control, the Ukrainians took an unconventional, open-source approach to the case. By releasing key pieces of video evidence on social media, the Ukrainian government recruited a vast roster of skilled online analysts to work on its behalf without expending any financial resources. Placing this user activity in the context of scholarly studies of both fan labor and citizen surveillance, the paper argues that social and economic aspects of online culture enabled Ukraine to benefit significantly from the discourse produced by unpaid workers. Ultimately, the output of these laborers played a key role in counteracting Russia's use of global broadcasting and expensive online propaganda to dominate international debate surrounding MH17.

Co-Creating Safety and Security?: Analyzing the Multifaceted Field of Co-Creation in Finland

“…The third level of co-creation takes place "when individual[s] or organized groups of citizens provide input into the design of new tasks and solutions through crowdsourcing, focus-group interviews, written consultations, and public hearings that only allow a limited dialogue" (Torfing et al 2019:804-5). A well-known example of safety-and securityrelated crowdsourcing is the volunteering that took place to digitally aid the investigation of the downing of Malaysian Airlines flight MH17 in 2014 (Sienkiewicz 2015). As an example of a face-to-face event, in Chicago, consultative beat meetings are a mixture of focus-group interviews and public hearings.…”

Legal geography III: Evidence

“…Avakov's purpose for posting the video is not immediately clear, but presumably he would have been aware of the location and nature of the material he was putting online. Sienkiewicz's (2015) suggestion is that the posting was a deliberate act to enrol open source investigators who would perform the task of corroborating the evidence while broadcasting the subsequent findings to an international audience.…”

Recognizing military vehicles in social media images using deep learning

“…Figure 2 shows the learning curve for the best fold (accuracy: 0.8398). For experiment (2), the entire data set described in Section III-A was fed to ResNet50 pre-trained on ImageNet for feature extraction, recording the output of the final average pooling layer to obtain a 2048-dimensional feature vector for each image. These features were then split into training and validation sets and fed to a fully-connected neural network with 512 nodes, training for 75 epochs.…”

“…This is precisely the issue to which this paper contributes by presenting a system for recognizing military vehicles in social media photographs using computer vision and machine learning. 2 The paper builds on recent advances in the subfield of deep learning, which has brought about significant advances in 1 https://www.bellingcat.com 2 Available at https://github.com/DigitalGeographyLab/MilVehicles/ various computer vision tasks, such as object detection, classification and localisation [3]. Keeping the limited resources of non-and intergovernmental organizations in mind, the work leverages the Python machine learning ecosystem for openly available libraries, architectures and pre-trained models.…”

“…A prime example of a non-governmental organization is Bellingcat 1 , which conducts investigative journalism using open sources and social media, publishing meticulous reports on the ongoing conflicts in eastern Ukraine and Syria [2]. Intergovernmental organizations, in turn, may be exemplified by the Organization for Security and Co-operation in Europe (OSCE), which is currently tasked with monitoring the situation in the Donbass region in eastern Ukraine.…”