This CAVAL-funded project – after a long period of gestation – has finally come to fruition. Written by Indigenous librarians for use by undergraduate students and the librarians who advise them, the toolkit is designed to help people to be think through relevant issues when citing Indigenous knowledges in academic writing.
This is not a particularly surprising finding, but it's good to see someone's collected some data. Ironically the study suggests that non-native English writers may be more likely to avoid their writing being flagged as written by SALAMI if they use GPT to re-write their original.
In this study, we evaluate the performance of several widely-used GPT detectors using writing samples from native and non-native English writers. Our findings reveal that these detectors consistently misclassify non-native English writing samples as AI-generated, whereas native writing samples are accurately identified. Furthermore, we demonstrate that simple prompting strategies can not only mitigate this bias but also effectively bypass GPT detectors, suggesting that GPT detectors may unintentionally penalize writers with constrained linguistic expressions.
What this says about the state of university education is left as an exercise for the reader.
Regular readers will perhaps by unsurprised that I'm somewhat pessimistic about the likelihood of success for this initiative, but I wish them all the best. Anything that includes corporate publishers and ignores the impact of ranking tables and the pressure to publish journal papers is unlikely to resolve the toxic incentives at the heart of academia.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
Alissa McCulloch outlines exactly why the recent statement on WLIC 2024 from the IFLA President and President-elect is a slimy attempt to justify IFLA's decision to hold their next conference in a theocratic dictatorship that imprisons people for heinous crimes such as “swearing on Facebook”, “being part of an out-of-favour religious tradition” and “engaging in consensual intimacy with an adult you are not in a heterosexual marriage with”.
Queer people deserve to exist in public venues. Queer people deserve to be safe in public venues. And while IFLA WLIC is a private event, it also has grand aspirations of being a conference where people from all corners of the Earth can meet and learn from each other. ‘Change happens through dialogue & certainly cannot happen if we are not present,’ reads the IFLA statement. Yes, it certainly cannot happen if a whole class of attendees cannot safely be present in the country at all.
I bet the conference dinner is going to be a real hoot, with UAE's ban on public dancing and all.
A blast from an NLA past, check out this great short talk. As best I can tell, Flinders did indeed cancel their Libguides and have moved to a combination of LMS modules, clear website content, and reading lists to convey key information to students.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
SUPARS and other largely forgotten systems were the forerunners of the contemporary search engines we have today. While the popular history of the internet valorises Silicon Valley coders – or, sometimes, the former US vice president Al Gore – many of the original concepts for search emerged from library scientists focused on the accessibility of documents in time and space. Working with research and development funding from the military and industry, their advances can be seen everywhere in the current online information landscape – from general approaches to ingesting and indexing full-text documents, to free-text searching and a sophisticated algorithm utilising previous saved searches of others, a foundational building block for contemporary query expansion and autocomplete.
The Australasian Open Educational Practice Special Interest Group are looking for a Co-Convenor (Research) – could it be you?
This position is primarily responsible for identifying and coordinating SIG research activities that encourage the membership to become involved in collaborative publication and research presentation opportunities related to OER and OEP. The position will support at least one SIG conference paper at ASCILITE annually, with the intent to develop opportunities for AJET publications (or other OA journals). Additionally, the position will liaise with ASCILITE to determine support for research activities (for example, organising guest speakers), and identify (with the membership) opportunities for funding.
For a long time, the basic paradigm has been what we have termed the “read-write web.” We not only consumed content but could also produce it, participating in the creation of the web through edits, comments, and uploads. We are now on the verge of something much more like a “write-write web”: the web writing and rewriting itself, and maybe even rewiring itself in the process. (ChatGPT and its kindred can write code as easily as they can write prose, after all.)
We face, in essence, a crisis of never-ending spam, a debilitating amalgamation of human and machine authorship.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
The ASCILITE conference will be in Christchurch, Aotearoa this year and the Call for Papers is open. Draft your ten-page paper or 500 word abstract about “People, Partnerships and Pedagogy” now!
On May 23, the Council of the EU adopted a set of conclusions on scholarly publishing that, if followed through, would spell the end for academic publishers and scholarly journals as we know them. On the same day, the adoption was followed by a joint statement of support by the largest and most influential research organizations in Europe. At the heart of the goals spelled out in the conclusions and the statement of support is the creation of a “publicly owned and not-for-profit” infrastructure for scholarly publications.
I am generally very skeptical about statements like this because academia has so far shown very little appetite for fixing its self-inflicted system of transferring truckloads of public money to a small number of monopolies in exchange for an enormous amount of bureaucracy and access to its own research. But it's kind of true – there does seem to be a bit of movement from funding bodies like the EU which seem to finally be losing patience with how their money is wasted.
This headline really should read something like “Italy withdraws from Berne Convention, declares war on creativity” given the ruling flies the face of every international agreement on the extent to which states can arbitrarily restrict creative use of public domain art. In this case, the artist has been dead for five centuries, yet the state of Italy effectively claims copyright over representations of David. The legal precedent here is catastrophic for common sense and for anyone who wants more art and artists in their life.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
ALIA's Phase 2 consultation paper is out, as well as a report on the Phase 1 consultation. I found the updated consultation paper quite disappointing, as ALIA seems to be moving backwards from their stance in Phase 1, and now appears to be suggesting adding even more credentialism to the profession. If you're thinking of writing a submission, be sure to also remind them that when it comes to Work placements: paying people is the fair thing to do.
This is a nice outlines of “minute papers” to help students understand how much they actually understood from a class, and help educators understand how effective their teaching is.
Yes, it's a media release from Clarivate. Why am I sharing this? Because if you're in negotiations with Clarivate or one of their many subsidiaries this year, keep this article at the front of your mind. Clarivate have so much cash sloshing around in their back accounts that they are using it to boost their own stock price and make shareholders richer. They can afford to give you a reasonable price.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
Emily Bender helps us to get to grips with how Large Language Models “understand” language and why they can't think, in this intriguing thought experiment.
In order to support the development of OA in Australia, and in particular understand how university OA Policies can support OA performance, we are interested in understanding in more detail why and how formal OA policies are developed, maintained and monitored at Australian universities.
Australia's Federal Minister for Education has asked for a House committee to look at generative AI use in education. So if you were hoping all the news stories would end, I'm sorry to be the bearer of bad tidings.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
From the Copyright Clearance Center (CCC), this is an interactive report/infographic highlighting key points in the scholarly publishing life-cycle where there are particular problems related to PIDs and metadata. Worth a look for research liaison librarians and metadata wizards.
The ARDC is calling for expressions of interest to work with them establishing an 'Open Source Program Office'. The idea seems to be a little like research data management support, except for open source software used in research – there is a significant problem with replicability, sustainability/maintenance, and preservation when it comes to the use of open source software (or really, any software) within academic research.
This was launched a couple of years ago but I only just heard about it. Disregard the horrendous backronym, because the idea is good. It's a recognised problem that academic journals disappear, and open access journals without a sustainable funding model are particularly prone to this. Project Jasper is a collaboration between CLOCKSS, DOAJ, Internet Archive, Keepers Registry and PKP to create kind of backup archive for open access journals. If your library has some defunct journals running on OJS in a cupboard somewhere, why not check it out.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
Aaron Tay has been doing some deep dives into how SALAMI systems are being integrated with search and discovery systems. This post is another interesting one analysing two recent research papers on the subject. A key point from the second paper highlights the conundrum for those trying to provide dynamic “answers” that are factually correct and in fluent natural language:
The paper makes the point that engines that try to closely paraphrase or directly copy from citations are likely to have high citation precision recall, but the answer is likely perceived to be less fluent or useful because it may not answer the question. The more the system tries to paraphrase, the more likely the answer will be rated more fluent or useful but the chances increase it makes a mistake doing so.
Degrowth journal is organised as a free, academic, open-access, international, transdisciplinary, and peer-reviewed journal that focuses on advancing the goals of degrowth. It will be published online including open issues and special-issues, and later, rolling submission.
This is not a typical academic journal:
First, we're looking for writers: researchers sharing their latest scientific discoveries or their perspectives about specific issues; thinkers wishing to share their exploration of a topic in an essay; bookworms eager to write a review of their latest read; students excited to summarise their theses; doers who have a story to tell about a conference, a meeting, or anything else they feel might be relevant to the readers of this journal. This is a safe place to grow wild ideas; don't hesitate to submit.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
Some very interesting news about draft legislation being negotiated in the EU Council. The path of EU legislation is long, winding, and full of opportunities for corporate lobbying at every stage, but this does sound promising. It seems member states may finally be waking up to the fact that replacing usurious pay-to-read fees with extortionate pay-to-publish fees isn't a very good deal.
Aaron Tay has a really approachable explanation of the range of approaches being taken by web search and question-answering tools when it comes to integrating large language model-based SALAMI. Tay's exploration shows how different approaches to combining large web indexes and large language models can result in vastly different results.
Ed Summers created pymarc 25 years ago, and a major new version has recently been released. Here he muses on MARC, web archiving, ephemerality, and why MARC won't die:
It’s a bit of a paradox how web resources can be so fleeting and impermanent, while data formats (like MARC) that circulate by virtue of the web can turn out to be so resilient.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.
Alex Barthel runs the numbers on academic language and learning advisors in Australian universities: there's not many of them, and the ratio is getting worse as student numbers increase.
An open online book containing short scholarly API code examples for interacting with various “scholarly web service APIs” including Scopus, CrossRef, PubMed, the World Bank, and more.
Libraries and Learning Links of the Week is published every Monday by Hugh Rundle.
Subscribe by following @fedi@lllotw.hugh.run on the fediverse or sign up for email below.