Cristóbal Urbano
LightNews LightNews
xbal-urbano.bsky.social
Cristóbal Urbano
@xbal-urbano.bsky.social
170 followers 180 following 26 posts
Professor del Departament de Biblioteconomia, Documentació i Comunicació Audiovisual, a la FIMA @fima-ub.bsky.social de la Universitat de Barcelona @ub.edu
ORCID: https://orcid.org/0000-0003-0935-6436
Posts Replies Media Videos
xbal-urbano.bsky.social
Urbano, C., Rozemblum, C., & Banzato, G. (2025). Política de preprints en revistas académicas españolas: análisis de la claridad en las instrucciones a los autores. doi.org/10.31219/osf...

👉Preprint en revisión abierta en PREreview: prereview.org/preprints/do...

#preprints #OpenScience #journals
Análisis de la claridad con la que las revistas académicas españolas informan sobre sus políticas de aceptación de preprints en las instrucciones a los autores. Se busca determinar en qué medida estas revistas han adoptado políticas explícitas sobre preprints y cómo estas políticas se alinean con las recomendaciones de transparencia del Committee on Publication Ethics (COPE). Se realizó un análisis de las políticas de aceptación de preprints en 221 revistas académicas españolas clasificadas en los cuartiles Q1 y Q2 del Scimago Journal & Country Rank (SJ&CR) 2022. Se examinaron las instrucciones a los autores y las políticas de autoarchivo en los sitios web de las revistas, así como en los directorios Open Policy Finder (OPF) y Dulcinea. Las revistas se clasificaron en cuatro categorías: aceptación explícita, rechazo explícito, mención confusa y sin mención. Solo el 17,6% de las revistas analizadas exponen con claridad su política sobre preprints. El 15,4% acepta explícitamente los preprints, mientras que el 2.3% los rechaza explícitamente. El 40,3% presenta menciones confusas y el 42,1% no menciona los preprints. El estudio revela que la mayoría de las revistas académicas españolas no cumplen con la recomendación de claridad sobre preprints del COPE, por lo que existe un margen significativo de mejora en la forma como las revistas presentan sus políticas. Las revistas de ciencias experimentales y de la salud son más claras y proclives a aceptar preprints. Además, se observan inconsistencias entre las políticas de autoarchivo declaradas en los directorios de políticas de autoarchivo y las instrucciones en los sitios web de las revistas.
April 2, 2025 at 7:39 AM
xbal-urbano.bsky.social
#OA papers from OASPA @oaspa.bsky.social members concentrated in few big commercial players, with the top 5 publishers (MDPI, Elsevier, Springer Nature, Willey-Hindawi and Frontiers) accounting for the 70% of the output.

www.oaspa.org/news/fully-o...
OA output continues to be highly consolidated 3 publishers account for 54% of the total output covered by this OASPA dataset, the top 5 for 70% and the top 10 for 90% of it. This shows a slightly higher degree of consolidation compared with last year. We see greater levels of consolidation for CC BY licenses overall, and greater still for CC BY licenses published in fully-OA journals. The order of the top 5 publishers in terms of volume is the same as last year: MDPI remained the largest producer, Elsevier the second largest and Springer Nature the third (with the caveat that we had to estimate Springer Nature’s figures1 ). Wiley (including Hindawi) is fourth largest and Frontiers fifth.
February 6, 2025 at 7:57 AM
xbal-urbano.bsky.social
Ni blanco, ni negro... una gama de grises en el tema del #CVnarrativo. @profrachelgan.bsky.social pone un poco de "agua al vino" en el "hype" que se va dando sobre el #CVnarrativo aplicado de forma general a todo tipo de situaciones. Interesante su punto de vista
Postdoc wanted Investigating the impact of this approach on both funding and recruitment decisions is vital. Last year, I took part in one project aimed at doing just this, run by the University of Cambridge’s Action Research on Research Culture team. It looked at how narrative CVs affect shortlisting decisions in postdoc recruitment. When I recruit a postdoc to my research group, I first identify the criteria a candidate should meet. These criteria must be objective and act as a bulwark against cognitive biases. For example, if I am recruiting a postdoc to perform transmission electron microscopy, I am obviously looking for someone familiar with this very challenging technique. A standard CV in my field will include a list of a candidate’s skills, so I can quickly and efficiently shortlist those claiming experience in electron microscopy and test that claim at interview. Faced with a pile of narrative CVs, I found that information more difficult to track down. And many applicants—particularly international candidates, or those for whom English was not a first language—struggled to provide the information I needed. I became concerned that narrative CVs favour those skilled in English and self-promotion. More fundamentally, narrative CVs are best suited to answering whether a candidate is excellent in the broadest sense. This is not surprising, given they were pioneered by the Royal Society for its University Research Fellowship scheme for “the most promising early career postdoctoral scientists”. UKRI’s R4RI is adapted from the society’s Résumé for Researchers. Such fellowships demand a broad form of excellence. Vacancies such as mine, however, demand a specific set of competencies.
January 29, 2025 at 5:53 PM
xbal-urbano.bsky.social
A la biblioteca pública-escolar on jo treballava a finals dels 80 hi havia cua per agafar-los en préstec... El problema és que no "perduraven": l'enquadernació era molt dolenta i al tercer infant que volia llegir el llibre havíem de fer la respiració assistida al volum amb cola blanca ;-)
January 25, 2025 at 7:10 PM
xbal-urbano.bsky.social
Más sobre la relación sin equilibrio simbiótico entre herramientas IA y @wikipedia.bsky.social @viquipedia.bsky.social

▶️Wagner, C., & Jiang, L. Death by AI: Will large language models diminish Wikipedia?
doi.org/10.1002/asi....

Un artículo de #JASIST con evolución de las métricas de visitas.
Abstract: We argue that advances in large language models (LLMs) and generative Artificial Intelligence (AI) will diminish the value of Wikipedia, due to a withdrawal by human content producers, who will withhold their efforts, perceiving less need for their efforts and increased “AI competition.” We believe the greatest threat to Wikipedia stems from the fact that Wikipedia is a user-generated product, relying on the “selfish altruism” of its human contributors. Contributors who reduce their contribution efforts as AI pervades the platform, will thus leave Wikipedia increasingly dependent on additional AI activity. This, combined with a dynamic where readership creates authorship and readers being disintermediated, will inevitably cause a vicious cycle leading to a staling of the content and diminishing value of this venerable knowledge resource. Reviewing the Wikipedia readership from recent years, disintermediation by answer bots appears already prevalent and only to increase over time. Human readership peaked around 2016, at 106 billion page views, and has since dropped to around 90 billion views per year. Meanwhile, the number of automated (non-human) page views has doubled since 2017, from 14 billion to over 28 billion. Human page views are thus likely to continue their decline in favor of AI bot accesses. With that, Wikipedia's visibility will also diminish. In fact, with the widespread integration of AI tools into regular web browsers, we expect the disintermediation of Wikipedia to accelerate, leading to a more rapid decline and ultimate Wikipedia obsolescence.
January 6, 2025 at 8:20 PM
xbal-urbano.bsky.social
No menos documentado es el comentario a esa nota por parte de @lluisanglada.bsky.social‬ #ThinkEPI:

"¿Necesitamos un nuevo Unisist? Comentarios viejunos a un viaje de regreso al futuro desde el lejano año de 1971"
doi.org/10.3145/thin...
Resumen: Cristóbal Urbano publicó una documentada, inteligente y provocadora nota ThinkEPI recordando el programa Unisist de la Unesco que tenía por objetivo establecer un sistema mundial de información científica. La nota resume perfecta-mente el contenido e intenciones del programa y termina preguntándose si hoy si-guen necesarias políticas internacionales con respecto los productos documentales generados por la investigación. Mi nota no es ninguna réplica a aquella, pero sí la opinión de que está política internacional ya existe, solo que se está desarrollando bajo parámetros distintos a los de los años 70 y 80 del siglo pasado. Opino que el movimiento de la ciencia abierta parte de los mismos supuestos y tiene los mismos objetivos que Unisist, pero destaco tres grandes diferencias: la información científi-ca que hoy incluye los datos, la forma de usar la información ya que la abundancia de misma está dirigiendo los usos hacia lo más relevante, lo más fácil o conveniente de usar, y que el sistema creado para hacer esto posible no es centralizado, sino que se basa en la interoperabilidad entre nodos federados. Palabras clave: Unisist; EOSC; Documentación; Políticas de información
December 17, 2024 at 6:53 AM
xbal-urbano.bsky.social
"Generative AI chatbots as sources of web traffic"
> web analytics insights from RACO, an academic journal portal

👉 My contribución to _Anuario ThinkEPI 2024_ doi.org/10.3145/thin...

#AI #Chatbots #BibliographicSearching #WebAnalytics #SciComm #AcademicJournals
This paper is a reflection on the role of generative AI chatbots in the referred web traffic that reaches academic journals. Scenarios on informational search behaviour within the academic environment are explored, regarding the influence of these new tools based on Large Language Models (LLM) and on the generation of direct responses resulting from an iterative process of conversation with an AI chatbot. By exploring web analytics data obtained with Google Analytics for the RACO portal, a first approach is presented to the low volume of visits that, to date, it seems that academic journals are receiving from AI chatbots. Hypotheses are formulated to explain this low volume of traffic, assuming that the use of these AI tools is important and growing in the academic world. Finally, the role of Generative Engine Optimization (GEO) as a strategy for academic content optimization in generative AI-powered search tools is discussed, to decide whether it should be a suitable complement to the already established Academic Search Engine Optimization (ASEO) actions. Discussion is invited through a series of open questions on how new AI tools may be changing the landscape of bibliographic information search and use, and how that may affect scholarly journals in their visibility and optimization strategies for traffic acquisition.
December 6, 2024 at 6:27 PM