30472 research outputs found
Sort by
Lessons from the Library: Extreme Minimalist Scaling at Pirate Ebook Platforms
Get PDFAt 33TB of data in its main collection, the highly illegal Library Genesis project is one of the largest repositories of copyright-violating educational ebooks ever created. Established over a decade ago in 2008 the goal of Library Genesis is nothing short of a modern Library of Alexandria, albeit without anyone’s legal sanction. As one of its administrators wrote: ‘within decades, generations of people everywhere in the world will grow up with access to the best scientific texts of all time. [...] [T]he quality and accessibility of education to the poor will grow dramatically too. Frankly, I see this as the only way to naturally improve mankind: we need to make all the information available to them at any time’ (Bodó 2018b). Rooted in its homeland’s Russian communist principles and particularly the Soviet isolationist copyright policies of the twentieth century, LibGen is a formidable resource and threat to conventional academic publishers.
The Library Genesis database had just short of 1.2m records (books) in 2014 (Bodó 2018a). As of January 2020, this capacity has doubled to 2.5m books. In this article, I examine the minimalist computational design choices taken by this maximal-in-intent, illicit archive of epistemological dissent and how such decisions have shaped the scalability and growth of the platform. This includes LibGen’s numerical subdivision of record identifiers into ‘buckets’ to work around directory file limitations in the GNU/Linux operating system; its use of md5 hashing of filenames within directories capped at 1,000 files to avoid future hashing collisions while allowing for on-disk integrity checking; and its use of the MySQL socket/network server as opposed to SQLite or similar disk-based database.
Beyond these computational details, though, the theoretical tension that this article highlights is the path dependencies that are set in (illegal) computational projects that have goals of absolute abundance and maximalist capacity, and the minimalist design principles that they must instigate at the outset to ensure a degree of scalability. I also query the ways in which the project’s contested mission statements target an economic (geographic) audience demographic with only minimalist access to high-capacity computing resources. I finally examine the limits on scalability of the distribution of the Library Genesis through its torrent archive and other distributed networking technologies such as IFS, which despite their promise of peer-to-peer redundancy fall down on an archive of this size
Does economic policy uncertainty matter for financial reporting quality? Evidence from the United States
Get PDFWe examine the effect of economic policy uncertainty (EPU) on the financial reporting quality of US firms over 1999-2015. We use accruals-based earnings management as a proxy for financial reporting quality and the index of Baker et al. (2016) as an EPU measure to show that they exhibit a positive and significant association. We also find a causal effect by employing three political polarization instruments for EPU. In a cross-sectional analysis, we further show that the positive relationship between EPU and earnings management strengthens for firms operating in politically sensitive industries, for firms in more financial distress, and during recessionary periods. We also provide evidence that increased financial constraints facilitate the positive relationship between EPU and earnings management. These findings are robust to the use of alternative measures of economic policy uncertainty and when we employ real earnings management as a dependent variable. These results indicate that managers aim to provide outsiders with an improved financial position of the company when EPU is high. Our findings suggest that investors, analysts, creditors, and regulators should be wary of firms’ financial reporting quality in periods of high economic policy uncertainty
Fairly assessing unfairness: an exploration of gender disparities in informal entrepreneurship among academics in business schools
Get PDFAssessing gender disparities in science commercialisation has been in the centre
of the unresolved debates on the inadequacies of the methods used to compare female and
male academics. Drawing from the literature on non–IP-based academic entrepreneurship
and gender disparities in science, this study used the “pair-matched” technique to isolate 406
female and male academics in business schools (203 of each gender from a sample of 729
academics) who share common characteristics regarding academic position, subdisciplinary
affiliation, and experience. The study confirms that a comparison of female and
noncomparable male academics could lead to an unfair judgement of female academics’
performance. However, the results show that even compared to comparable men, women are
less involved in remunerated consultations, generate a smaller proportion of their revenue
from consultations and are less engaged in the creation of consultancy companies. In
addition, the study allows us to quantify a leaky pipeline of both genders involved in informal
academic entrepreneurship and to identify four paths, from progressive to nonprogressive.
Most female academics follow a progressive entrepreneurial path but often struggle to move
from nonremunerated to remunerated entrepreneurial engagements. The study concludes
with implications for university administrators on knowledge transfer and gender inequality
From identity politics to the politics of power: men, masculinities and transnational patriarchies in marketing and consumer research
Get PDFDespite increased interest in men and masculinities in marketing and consumer research (MCR), mainstream research has neglected feminist perspectives that engage with issues of gender power relations. In particular, critical studies of men and masculinities (CSMM), including concepts such as the hegemony of men, patriarchies, transnational patriarchies, or simply transpatriarchies, are rarely theoretically or empirically developed. This chapter begins by highlighting the emergence of research on men and masculinities in MCR as based on images, representations and identities. This is followed by theoretical developments of CSMM, including the hegemony of men and transpatriarchies. We explain how these theories have appeared in recent research and how they can further impact change in relation to contemporary issues. This chapter then seeks to build on the momentum of recent feminist research efforts in MCR by calling out the more systematic gender power relations within the previously depoliticised research on men and masculinities
Unpleasant actuarial arithmetic: fair contribution rates for defined benefit pension schemes
Get PDFWe derive key properties of the actuarially fair contribution rate for defined benefit (DB) schemes, that equates scheme assets to liabilities for any given scheme member. The unpleasant actuarial arithmetic of both increased life expectancy and (especially) negative real yields has resulted in a massive rise in the fair contribution rate over recent decades. At present there appears to be little prospect of these rises being reversed. We analyse the implications for the viability of DB schemes, and consider the (potentially significant) impact of incorporating systematic risk into benefits
Collectives and epistemic rationality
No full textConsideration of collectives raises important questions about human
rationality. This has long been known for questions about preferences,
but it holds also with respect to beliefs. For one, there are
contexts (such as voting) where we might care as much, or more, about
the rationality of a collective than the rationality of the individuals it
comprises. Here, a given standard may yield competing assessments
at the individual and the collective level, thus giving rise to important
normative questions. At the same time, seemingly rational strategies
of individuals may have surprising consequences, or even fail, when
exercised by individuals within collectives. This paper will illustrate
these considerations with examples, provide an overview of different
formal frameworks for understanding and assessing the beliefs of collectives,
and it will illustrate how such frameworks can combine with
simulations in order to elucidate epistemic norms
Similar tactile distance anisotropy across segments of the arm
Get PDFA substantial literature has described anisotropy of tactile distance perception across many body parts. In general, the distance between two touches is felt as larger when the touches are oriented with the medio-lateral axis of the limbs than when oriented with the proximo-distal axis. In this study, we investigated tactile distance perception across the arm, measuring anisotropy on the upper arm, forearm, and hand dorsum. Participants made forced-choice judgments of which of two pairs of tactile distances felt larger and anisotropy was measured using the method of constant stimuli. Clear anisotropy was found on all three regions of the arm. There was no apparent difference in the magnitude of anisotropy across segments of the arm. We further measured the physical curvature of the arm and show that this cannot account of the perceptual anisotropy observed
Exploring FOI publicity patterns: the case of Italian municipalities
No full textThis paper aims to study FOI publicity implementation patterns by Italian municipalities' (>30,000 inhabitants). The analysis relies on data collected through an original survey of the municipalities' websites. Data allow inspecting the information that local governments disclose about the procedures for the presentation of FOI requests and how the FOI requests received in the past were managed. Cluster analyses reveal that the municipalities are rather heterogeneous in managing FOI publicity. Some of them provide full disclosure of relevant information; instead, others show non-compliance with the guidelines issued by the Italian central government. Regression analyses suggest that municipalities' size and income positively correlate with more elevated publicity. Civic capital is also positively connected with information about the procedures for presenting FOI requests. Factors affecting the demand for information about FOI, such as education and access to the internet, do not seem to be linked with publicity patterns
Monitoring COVID-19 on social media: development of an end-to-end natural language processing pipeline using a novel triage and diagnosis approach
Get PDFBackground: The COVID-19 pandemic has created a pressing need for integrating information from disparate sources in order to assist decision makers. Social media is important in this respect; however, to make sense of the textual information it provides and be able to automate the processing of large amounts of data, natural language processing methods are needed. Social media posts are often noisy, yet they may provide valuable insights regarding the severity and prevalence of the disease in the population. Here, we adopt a triage and diagnosis approach to analyzing social media posts using machine learning techniques for the purpose of disease detection and surveillance. We thus obtain useful prevalence and incidence statistics to identify disease symptoms and their severities, motivated by public health concerns.
Objective: This study aims to develop an end-to-end natural language processing pipeline for triage and diagnosis of COVID-19 from patient-authored social media posts in order to provide researchers and public health practitioners with additional information on the symptoms, severity, and prevalence of the disease rather than to provide an actionable decision at the individual level.
Methods: The text processing pipeline first extracted COVID-19 symptoms and related concepts, such as severity, duration, negations, and body parts, from patients’ posts using conditional random fields. An unsupervised rule-based algorithm was then applied to establish relations between concepts in the next step of the pipeline. The extracted concepts and relations were subsequently used to construct 2 different vector representations of each post. These vectors were separately applied to build support vector machine learning models to triage patients into 3 categories and diagnose them for COVID-19.
Results: We reported macro- and microaveraged F1 scores in the range of 71%-96% and 61%-87%, respectively, for the triage and diagnosis of COVID-19 when the models were trained on human-labeled data. Our experimental results indicated that similar performance can be achieved when the models are trained using predicted labels from concept extraction and rule-based classifiers, thus yielding end-to-end machine learning. In addition, we highlighted important features uncovered by our diagnostic machine learning models and compared them with the most frequent symptoms revealed in another COVID-19 data set. In particular, we found that the most important features are not always the most frequent ones.
Conclusions: Our preliminary results show that it is possible to automatically triage and diagnose patients for COVID-19 from social media natural language narratives, using a machine learning pipeline in order to provide information on the severity and prevalence of the disease for use within health surveillance systems
Data sources:
Access Repository Dashboard
Do you manage Birkbeck Institutional Research Online? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!