179566 research outputs found
Sort by
Distant relations:The Palatine family, propaganda and print in early Stuart Britain
No full textThe marriage of Elizabeth Stuart to Frederick V, Elector Palatine, in February 1613, offered a lavish counter-attraction to the solemn mourning which had accompanied the death of her brother, Henry, Prince of Wales, just a few months before. Amidst dynastic rupture, the match fortuitously shifted focus onto the future of the line, rather than its recent loss. Indeed, following the birth of Prince Frederick Henry in 1614, and for the next sixteen years, until the birth of a healthy son to King Charles I and his consort, Queen Henrietta Maria, the ever-growing Palatine brood represented the next stage in the Stuart succession. Separated by distance from their potential subjects, it was important that the electoral couple continued to cultivate ties with the British people and, as their family grew, a familiarity with the prospective royal line was encouraged.A steady stream of printed images was enlisted in this campaign, designed to introduce the public back home to the Palatine family. Tracing the development of their visual portrayal in line with the shifting fortunes of the royal couple and their children, this article explores how these representations sought to naturalise a foreign royal family, forging cross-European dynastic loyalties. Whether portrayed as lineal standby or even alternative, images of Elizabeth’s family might endorse or, indeed, complicate dynastic rhetorics. As I shall argue, the affective bonds encouraged in word and image actually risked splitting Stuart allegiances
The artificial intelligence cooperative:READ-COOP, Transkribus, and the benefits of shared community infrastructure for automated text recognition
No full textBackgroundArtificial Intelligence (AI) and Machine Learning (ML) technologies are integral to developing sophisticated digital infrastructures. Ownership and stakeholder input are critical for creating AI systems that are both innovative and accountable. This paper examines READ-COOP (https://readcoop.eu), the first platform cooperative to develop and host its own AI and ML tools (https://transkribus.org). This case study demonstrates an alternative cooperative governance model for responsible AI infrastructure.MethodsWe employ Research In Action and qualitative questionnaires to analyse the development of READ-COOP, following its transition from funded European Commission (EC) project to cooperative business. We assess the cooperative’s structure, management, and community engagement from 2019 to 2024. Data was collected on membership dynamics, use of Transkribus, and feedback on the cooperative’s governance and operational efficacy.ResultsAs of October 2024, READ-COOP has 227 members from 30 countries, fostering a strong user base of over 235,000 registered individuals. Transkribus has transcribed approximately 90 million digital images of historical texts, demonstrating effective AI utilization in the cultural heritage sector, winning the European Union’s Horizon Impact Award 2020. The cooperative approach facilitates democratic decision-making, leading to sustainable growth, and significant stakeholder involvement. Qualitative feedback indicates high levels of satisfaction with the cooperative’s governance and the perceived integrity and utility of the AI infrastructure.ConclusionsREAD-COOP exemplifies that a cooperative business model can effectively sustain AI and ML infrastructures while promoting democratic participation and equitable ownership. This offers a viable blueprint for other sectors seeking to develop responsible and trustworthy AI solutions. We suggest that cooperative frameworks are particularly suitable for AI infrastructures initially funded through public grants, providing a sustainable transition from public development to long-term, sustainable community-ownership. We recommend wider application and exploration of cooperative models for hosting and developing AI and ML technologies to ensure their responsible creation, governance, and use
The future of research software is the future of research
No full textThe use of software is near ubiquitous in research, yet it is still under-recognized despite changes in policy and practice. Despite many successful initiatives to improve the culture around research software, the authors argue that it is essential that the development of research software anticipates changes in the research landscape and continues to support many different people that use it
Programming RISC-V accelerators via Fortran
No full textA range of RISC-V based accelerators are available and coming to market, and there is strong potential for these to be used for High Performance Computing (HPC) workloads. However, such accelerators tend to provide bespoke programming models and APIs that require codes to be rewritten. In scientific computing, where many of the simulation code are highly complex, extensive, and written in Fortran, this is not realistic. In this extended abstract we present an approach that enables driving such architectures via Fortran, avoiding code redevelopment
Comparing MOS, AB and BWS for speech synthesis evaluation
No full textText-to-speech (TTS) systems are often assessed using Mean Opinion Scores (MOS) tests. Despite their popularity, MOS tests have several limitations. In this study, we compare MOS to two forced-choice comparison tests: the AB test and Best-Worst Scaling (BWS). We examine participant consistency in each test and performance on a varying number of listeners and test duration. We are particularly interested in the evaluation of endangered languages where few native speakers are available for evaluation. We performed three separate listening tests each containing MOS, AB and BWS questions only. The tests involved stimuli from TTS systems of the Blizzard Challenge 2013 plus four additional modern TTS systems. To inspect participant consistency we calculated the inter and intra-participant correlation. The inter-participant correlation is the correlation between the systems scores obtained from the data of one participant to scores obtained with all participants. This measures consistency across participants. The intra-participant correlation is the correlation between the systems scores obtained from two randomly selected but balanced halves of a participant's data. This measures the internal consistency of participants, i.e. how consistent they are during the test. To analyse how responses change depending on the number of listeners and the test duration we created several versions of the listening test by systematically selecting via bootstrapping a subset of listeners from the pool of all listeners or a subset of stimuli from a pool of all stimuli. We performed 500 resamplings (with replacement) to create each subset and varied subset size (that is the number of listeners or stimuli selected). AB and BWS questions are longer than MOS because participants need to listen to more stimuli. However, we found that participants make quicker decisions in AB and BWS tests, likely due to the difficulty of using the MOS scale. We also found that participants are more consistent with each other and themselves in the BWS test. Finally, we found that the BWS test reveals more significant differences compared to both AB and MOS, especially with fewer participants or shorter tests
The protection of cultural property in times of armed conflict:Ethics, gender, and coloniality
No full textCultural heritage rests on imaginings of a shared humanity transcending national dividing lines. Yet, cultural heritage sites are frequently targeted in war. Thus, the politics of cultural protection is marked by tensions and contestations. Notably, the protection of cultural heritage in armed conflict is a higly militarised practice that rests on notions of protection that are broadly western-centred, masculinized and insensitive to the gendered and colonial power relations that undergird armed conflict and by extension the protection of cultural property. Informed by critical heritage studies, cosmopolitanism and feminist IR scholarship I elucidate the broad claims of this article through a feminist narrative analysis of the protection of heritage and cultural property. This involves identifying what is said and what is obscured and silenced in the narration of heritage protection. First, I zoom in on the wider storytelling that surrounds heritage protection, unpacking the ethical, gendered and colonial assumptions employed in that narrative. Second, I unpack the narration of military protection in the UNESCO military manual published in 2016, attending to its ethical underpinnings, its protection logics and its privileging of distinctively western military knowledge and best practices
Home working during the COVID-19 pandemic: The experience of drug and alcohol support workers
No full textBackground:Drug and alcohol support workers play a vital role in addressing the growing burden of substance-related harm and mortality. The COVID-19 pandemic led to an abrupt and significant shift towards home working for many in this workforce. This study explores these workers’ home working experiences, addressing a research gap and providing valuable insights for staff, organisations delivering public health services, and service users.Design and methods:This qualitative study explores home working experiences of 30 drug and alcohol support workers in northern England during the COVID-19 pandemic. Data collection included innovative digital methods: (1) digital timelines (n = 16); (2) in-depth interviews (n = 17); (3) five focus groups (n = 12). Timeline text was treated as qualitative text data. Interviews and focus groups were recorded, transcribed, and coded. Data were subject to Framework Analysis.Results:Seven themes were identified: (1) Difficulty balancing and separating work and home life; (2) Importance of setup, infrastructure and conducive work environment; (3) The move to remote/home working – a process; (4) Convenience and efficiency benefits; (5) Loss of the social – reductions in social connectedness and feelings of isolation; (6) The importance of the ‘office’ for connection, communication, socialising, and information sharing; (7) Managing remotely – the development and implementation of strategies and ways of coping.Conclusions:While home working offers some benefits for substance use support workers, providers, and service users, it also introduces significant challenges. Understanding these is critical for service optimisation. A hybrid (in-person/remote) delivery model, combining home and co-located, office-based working may be optimal
Nitrogen fixation in Arctic lichens and mosses:A survey across circumpolar subzones
Get PDFNitrogen bioavailability frequently constrains primary production in the Arctic with tundra communities vulnerable to ecological and metabolic disruption from climate variability. Diazotrophs associated with lichens and mosses are the primary source of new nitrogen (N) in the Arctic. We made 526 laboratory measurements of biological nitrogen fixation (BNF) in 272 lichens and 254 bryophytes representing 23 and 39 genera, respectively. These samples were collected from 49 tundra sites across the Arctic. We found 65 % of lichen and 44 % of bryophyte genera analyzed fixed N. We also identified potentially new cryptogam-diazotroph relationships in the lichen genera Asahinea, Nephromopsis and Thamnolia and the bryophyte genera Dicranoweisia and Amphidium. We found that while over 95 % of individual bryophyte samples fixed N within three months of storage, this dropped to less than 15 % after one year at room temperature. Individual lichen samples maintained a stable ~45 % fixation rate over nearly two years of frozen storage. Our experiments highlight the complexity in establishing robust BNF measurements required for model simulations.</p
The Draco Dwarf Spheroidal Galaxy in the First Year of DESI Data
Get PDFWe investigate the spatial distribution, kinematics, and metallicity of stars in the Draco dwarf spheroidal galaxy using data from the Dark Energy Spectroscopic Instrument (DESI). We identify 155 high probability members of Draco using line of sight velocity and metallicity information derived from DESI spectroscopy along with Gaia DR3 proper motions. We find a mean line of sight velocity of −290.62±0.80 km s−1 with dispersion = 9.57+0.66−0.62 km s−1 and mean metallicity [Fe/H] = −2.10±0.04,consistent with previous results. We also find that Draco has a steep metallicity gradient within the half-light radius, and a metallicity gradient that flattens beyond the half-light radius. We identify eight high probability members outside the King tidal radius, four of which we identify for the first time. These extra-tidal stars are not preferentially aligned along the orbit of Draco. We compute an averagesurface brightness of 34.02 mag arcsec−2 within an elliptical annulus from the King tidal radius of 48.1 arcmin to 81 arcmin
Rodent home cage monitoring for preclinical safety pharmacology assessment:results of a multi-company validation evaluating nonclinical and clinical data from three compounds
Get PDFThe presence of central nervous system (CNS) safety concerns during early clinical testing that were not picked up in standard preclinical assessment is a major cause of attrition in drug development. It is also very expensive, time consuming and potentially dangerous for clinical trial participants. Rodent home cage monitoring approaches have previously been shown to deliver significant animal welfare benefits through group/social housing, minimal interventions avoiding stress that can confound results, and in some cases also animal reduction benefits with the multiplex data acquisition requiring fewer total animals. Here we looked at the utility of home cage monitoring to uncover potential CNS effects not identified using standard safety pharmacology tests. We hypothesised that longitudinal behavioural assessment – by capturing non-evoked behaviour and reducing sampling artefacts – would be more sensitive to adverse reactions in preclinical animal models (i.e. rodents). To test this, we selected three compounds which previously passed standard safety tests but were failed later including two during clinical trials. We validated the general methodology for using home cage monitoring in safety assessment study designs from single doses to repeat dosing for up to four weeks. We then re-tested the three compounds in single dose studies. We showed that the methodology fits well with standard study Deleted: non-evoked, longitudinal behavioural assessment is both more relevant and suffers less from sampling effects, therefore designs. More importantly we uncovered significant findings in all three compounds that were not observed in the original classic safety pharmacology tests. The lack of such effects observed in standard preclinical assessment likely reflects functional differences between the limited observational snapshots characteristic of this approach and the more comprehensive temporal resolution enabled by continuous home cage monitoring
Access Repository Dashboard
Do you manage Edinburgh Research Explorer? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!