#pubmet2023 — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #pubmet2023, aggregated by home.social.
-
What do publishing platforms and editors need? Look at the presentation from @eliopellin and @diver they delivered at #PUBMET2023 in September of this year and learn more about what CRAFT-OA does to improve the landscape of #DiamondOA publishing.
-
Two cool things from @pubmet, which ended today.
1. A paper name badge. I appreciate reducing plastic waste.
2. A lynx badge. I love both cats and badges.
-
Predrag Pale's ending keynote on AI is provocative. #PUBMET2023 #AIethics
-
What to do about AI in academia?
1. Nothing. Industry will welcome this.
2. Ban. Academia has been shocked and the use of tools goes against the principle that the work must be yours. Regard use as equivalent to plagiarism.
3. Moderate. Allowed if the instructor permits it. Bans don't handle the issues or educate about risks of using AI tools. Can we instead persuade users why these tools won't help them learn or produce good content?
-
UNESCO, EU, etc., say a human perspective must be central to the regulation of AI. Tools do not have personalities or personhood. Few countries have AI regulation or laws; the EU has been held up by lobbying.
UNESCO has just released guidance on AI in education and research: https://www.unesco.org/en/articles/guidance-generative-ai-education-and-research
-
LLMs may be biased due to biased training materials, but inexperienced users may not be aware of these biases, errors, or 'hallucinations'. The output may only present one view, but science needs to assess multiple views and varied information.
We don't know the source of the concepts outputted by LLMs, but in academia we need to attribute ideas and use citations.
-
This is problematic for scientific research and writing, because uncritical use is dangerous. Creates fake science, from which there may be no return. Human faking already exists, but at a less scale.
The black box problem is that we can't explain exactly how LLMs work because they use billions of parameters, nor can we reproduce outputs.
-
To regulate the use of AI, Ivana Kunda says new laws and regulations will lag behind. We can instead apply old principles to the new technology. 🧵
AI 'intelligence' in large language models is very different to human intelligence. It shuffles words around to create new sentences by guesswork, without knowledge of the meaning (a 'stochastic parrot'). It may sound sophisticated, but be totally wrong or invented.
-
AI in academia: Jelena Madunić researched use in open educational resources. Must give the model a role, context, standards, and formats for output. Followed the CLEAR Framework for Prompt Engineering: https://doi.org/10.1016/j.acalib.2023.102720. Got back a full course plan with lesson plan, etc., but the level of complexity/technicality is often too simple. The plans are basic and look reasonable, but they lack creativity and experts can see the flaws and gaps.
-
Structured #PeerReview at Elsevier helped reviewers cover more topics and increased inter-rater agreement, says Mario Malicki. Read more: https://pubmet2023.unizd.hr/malicki-abstract/
https://www.elsevier.com/reviewers/how-to-review#structuring
#PUBMET2023 -
Keynote talk at #PUBMET2023 from our Research Integrity Manager @mattjhodgkinson: don't waste a good crisis in publishing ethics and peer review. Crises can be useful to effect change, as Saul Alinsky observed. Do we need - and can we achieve - a Kuhnian paradigm shift from competition to collaboration, and from 'quality' to transparency? #PublicationEthics #PeerReview
-
🚀 Tiberius Ignat takes us on an optimistic trip to 2065, when researchers have rejected 'dual use' of research and embraced a radical openness that sees universal open data, deep scrutiny of data aided by AI, and widespread popular research participation. #PUBMET2023 #Futurism #OpenScience
-
Are "data" relevant to AHSS? Yes, say Ljiljana Poljak Bilić and Kristina Posavec.
"DARIAH-DE understands all those sources / materials and results collected, written, described and / or evaluated in the context of a research and research question in the field of human and cultural sciences, and in machine-readable form for the purpose of archiving, citation and for further processing as research data."
https://de.dariah.eu/en/forschungsdaten -
How can #OpenScience principles apply to anthropology? Olga Orlić presents example of the mixed methods SOLIDARan project on research students.
Some data can't be open due to sensitivities, so FAIR (findable, accessible, interoperable, reusable) interacts with CARE (Collective Benefit, Authority to Control, Responsibility, Ethics): https://www.gida-global.org/care
-
Jane Mahony presents publishoa.ie, a plan for publicly-owned, diamond #OpenAccess journal and book publishing infrastructure in the Republic of Ireland and Northern Ireland. They're getting advice from Jisc, Finland, and the Netherlands.
First stage is a digital directory of Irish scholarly publishers - many links to the UK. Next, exploring open platforms like Coko's and OJS.
-
So, focus on:
- Interoperability
- Small wins
- Quality UI/UX
- Co-creation and ambassadors
- Skills
- Balancing marketing and community building -
However, there are lots of silos and disagreement. For example, #OpenAccess is fragmented into APCs, repositories, diamond OA. What infrastructure should be public vs private, what are the value-added services, what are Open Science business models?
Researchers are often not involved in infrastructure design: they don't care about technical details; they just want as little hassle as possible.
-
Everything is digital
Everything is big
Everything is connectedNatalia Manola of #OpenAIRE lays out plans to move to an open scholarly ecosystem: #OpenScience is the vehicle for reproducibility, ethics, and more, with positive feedbacks in an interconnected system.
-
CW: Oops
In my sleep-deprived state, I spilled coffee over the nice white conference tablecloth at #PUBMET2023. On the plus side, I got to use one of the few Croatian words I know: pardon (a cheat really, as a loan word).
They have a very Italian approach to coffee here - tiny cups. #Coffee
-
Ukraine has reached >80% of scholarly publications #OpenAccess, says Nataliia Kaliuzhna (@KaliuzhnaN).
Unusually, this is highest in the humanities and mainly in national journals (diamond OA or low APC). However, Ukrainian institutional repositories are lagging behind.
#PUBMET2023 #Ukraine -
First #PUBMET2023 speaker is Kathleen Shearer of COAR, who highlights recent recommendations for equitable #OpenAccess:
https://on-merrit.eu/
https://www.unesco.org/en/open-scienceCOAR has a vision for non-profit, global, interoperable infrastructure for publishing and #OpenScience as a public good, including repositories.
-
Our Research Integrity Manager @mattjhodgkinson didn't allow a missed connection to stop him reaching PUBMET 2023 at the University of Zadar this morning. https://pubmet2023.unizd.hr/programme/ #PUBMET2023
-
UKRIO is at PUBMET 2023!
Tomorrow at 9:45 am, Matt Hodgkinson will be moderating PUBMET 2023's first session on "The brave ‘new’ world of institutional publishing".
On Friday at 9:30 am, Matt will be speaking on "Not wasting a good crisis in publication ethics" in session 5 of PUBMET 2023.
To view the full programme: https://pubmet2023.unizd.hr/programme/
#PUBMET2023 #PublicationEthics #ResearchIntegrity #SocialImpact