My question:
Why we cannot create an international organization comprising research institutions and even individuals to create a repository of research articles? The organization will be solely dependent on subscription and donations. This organization will be free from any government interference. The job of this organization is to a build a highly intelligent repository with the capability to generate answer to any query. AI will be deployed to work out degree of plagiarism, forgery etc. And so on. Is it doable?
Reply by Gemin Ai
It’s a compelling vision! The idea of an independent, internationally governed, AI-powered repository for research articles, free from corporate and government interference, addresses many of the current frustrations with scientific publishing and integrity.
Let’s break down the feasibility and challenges of such an organization:
Is it Doable? In Theory, Yes. In Practice, Extremely Challenging.
Here’s why:
Why it’s a Great Idea (The “Why”)
* Addresses Current Problems: It directly tackles issues like:
* Access Barriers: Eliminates subscription walls.
* Commercial Dominance: Reduces reliance on large for-profit publishers.
* Research Integrity: Centralizes and enhances plagiarism/forgery detection using advanced AI.
* Dissemination Speed: Could potentially accelerate article availability.
* Knowledge Discovery: An AI-powered system could offer novel ways to search, synthesize, and analyze research.
* Academic Freedom: By being independent of governments and corporations, it could better safeguard academic freedom.
Major Challenges (The “How”)
* Funding and Sustainability:
* Initial Capital: Building such a massive, intelligent repository (hardware, software, AI development, data migration) would require enormous upfront capital, likely billions of dollars.
* Ongoing Costs: Maintaining servers, developing AI, employing expert staff (technical, editorial, legal, ethics), and covering legal fees would be astronomically expensive.
* Subscription Model: While intended to be free for access, you mention “solely dependent on subscription and donations.”
* Subscription from Whom? If it’s a subscription from institutions, it risks simply replacing one set of subscription fees (to publishers) with another (to this new organization). Would institutions be willing to pay significant fees to an entirely new entity while still needing access to historical content from existing publishers?
* Donations: While valuable, donations alone are unlikely to sustain such a vast, critical infrastructure on an ongoing basis. Relying solely on them for long-term stability is risky.
* The “Free Rider” Problem: If access is free, how do you incentivize enough people/institutions to subscribe or donate sufficiently to cover costs?
* Content Acquisition and Rights:
* Historical Content: How would you acquire the rights to the millions of articles already published and owned by Elsevier, Springer Nature, Wiley, etc.? They would not simply hand over their valuable assets. Re-publishing would involve complex and costly rights negotiations, or the organization would start with zero historical content.
* New Content: Even for new articles, gaining traction would be hard. Why would researchers submit to a new, unproven repository instead of established, high-impact journals (which currently offer career advantages)? You’d need to convince entire disciplines to shift their publication habits.
* Versioning and DOIs: How would it integrate with existing systems for digital object identifiers (DOIs), version control, and citation tracking?
* Governance and Independence:
* True Neutrality: Maintaining “free from any government interference” and “free from corporate influence” at a global scale is incredibly difficult. Who sets the rules? Who decides what gets included or removed? How are disputes resolved?
* Geopolitical Influence: Different countries have different legal frameworks and political agendas regarding research and data. An truly international body would face immense pressure from various national interests.
* “Capture” Risk: Even well-intentioned independent organizations can eventually be influenced by powerful groups or funders.
* Technological and AI Challenges:
* AI for Integrity: While AI is powerful for pattern recognition, detecting subtle plagiarism, data manipulation, or outright fraud is extremely complex and constantly evolving. It requires more than just algorithms; it needs human expert oversight, legal frameworks, and investigative processes. AI can flag, but human judgment is still essential for determining misconduct.
* “Answer to Any Query”: A highly intelligent repository capable of generating accurate answers to “any query” effectively describes a general artificial intelligence, or at least a highly advanced domain-specific AI, which is a monumental challenge beyond current capabilities for scientific synthesis.
* Maintenance and Obsolescence: Keeping such a massive, AI-powered system updated and secure against cyber threats would be a continuous, resource-intensive task.
* Adoption and Network Effect:
* Shifting Habits: The academic publishing ecosystem is deeply entrenched. Researchers’ careers depend on publishing in specific journals. Overcoming this inertia and getting widespread adoption would be a Herculean task.
* Incentives: Unless this new repository offers clear, compelling incentives (e.g., faster publication, better visibility, career benefits) that outweigh the current system, adoption will be slow.
A More Realistic Path: Incremental Evolution and Collaboration
While a single, monolithic, independent organization might be a utopian ideal, many of its aims are being pursued through decentralized, collaborative efforts:
* Growth of Preprint Servers: These are independent and community-driven.
* Open Access Mandates: Funders and governments are increasingly requiring OA, chipping away at publisher dominance.
* Responsible Research Assessment: Initiatives like DORA are changing how research is evaluated, reducing the pressure to publish in specific journals.
* Independent Integrity Bodies: Organizations like COPE (Committee on Publication Ethics) or national research integrity offices work independently of publishers to set standards.
* Blockchain for Science (Decentralized Science – DeSci): Emerging concepts using blockchain technology to create transparent, immutable records of research and peer review, potentially bypassing traditional publishers.
* AI for Research Tools: AI is increasingly being developed and integrated into research workflows by various independent groups for things like literature review, synthesis, and even misconduct detection (e.g., tools that check image integrity).
Conclusion:
The vision you describe is inspiring and targets many real problems. However, creating a single, comprehensive, independent, and AI-powered global research repository from scratch, funded solely by subscriptions and donations, while remaining free from all influence, faces formidable practical, financial, legal, and sociological hurdles.
It’s more likely that the ideal will be approached through the continued evolution and integration of various, smaller, more specialized open science initiatives, preprints, non-profit publishers, and AI tools, rather than a single, grand, centrally planned entity. The “fight” is about shifting power and incentives away from commercial entities and towards the broader scientific community.
You’ve articulated the core of the problem: the conflict between the profit motives of dominant commercial publishers and the need for robust research integrity, while simultaneously acknowledging the dangers of direct government control over science.
The solution isn’t a single, simple fix, but rather a multi-faceted approach that emphasizes decentralization, community ownership, transparency, and a shift in incentives.
Here are some key solutions and trends being actively pursued by the global scientific community:
1. Empowering Institutions and Researchers (Decentralization of Integrity Checks)
* Institutional Responsibility: The most effective approach is to place the primary responsibility for research integrity squarely on the shoulders of the institutions where the research is conducted. Universities and research organizations are the employers of scientists and have direct access to raw data, lab notebooks, and personnel.
* How: Institutions should invest in strong research integrity offices, offer comprehensive ethics training, implement robust data management policies, and conduct thorough investigations into misconduct allegations.
* “Digital Certificates” of Integrity: As suggested in some discussions, imagine a future where every research article comes with a digital certificate from the authors’ institution(s) validating that a series of integrity checks have been performed before submission to a journal. This shifts the burden away from journals being forensic detectives after the fact.
* Researcher-Led Initiatives: Fostering a strong culture of ethical conduct from within the research community itself. This includes:
* Open Science Practices: Encouraging pre-registration of studies, sharing raw data, detailed methodologies, and analysis code. This makes research more transparent and reproducible, making it harder to hide misconduct and easier for the community to identify issues.
* Open Peer Review: Making peer review more transparent (e.g., publishing reviewer comments, disclosing reviewer identities) can increase accountability for both authors and reviewers.
* Post-Publication Peer Review: Platforms where researchers can comment on, question, and even challenge published papers after they appear, providing continuous scrutiny beyond the initial peer-review process. Retraction Watch, for example, highlights issues and retractions, increasing transparency.
2. Diversifying and Reforming Scholarly Communication
* Non-Profit and University-Led Publishing: Support and strengthen non-profit university presses, academic societies, and other scholar-led publishing initiatives. These entities are typically more aligned with the public good of science than commercial publishers.
* Preprint Servers: Expanding the use of preprint servers (like arXiv, bioRxiv, medRxiv, IndiaRxiv, etc.) allows researchers to quickly share their findings without the delays of traditional peer review. While preprints are not peer-reviewed, they accelerate dissemination and allow for early community feedback, which can sometimes flag integrity issues faster.
* Diamond Open Access: Advocating for and funding “Diamond Open Access” models where neither the author nor the reader pays. This typically involves funding from institutions, consortia, or public grants to support the publishing infrastructure.
* Breaking the “Impact Factor” Addiction: Moving away from using the Journal Impact Factor as the primary measure of research quality or researcher performance.
* Responsible Research Assessment (DORA): Initiatives like the San Francisco Declaration on Research Assessment (DORA) encourage institutions to evaluate research based on its intrinsic merit, methodologies, data sharing, and societal impact, rather than just where it was published or the journal’s impact factor.
* Alternative Metrics (Altmetrics): Exploring metrics that capture the broader impact of research, such as downloads, media mentions, policy citations, and public engagement.
3. Smart Government and Funder Roles
* Funding Open Infrastructure: Governments and funding agencies should invest directly in non-profit, open-source infrastructure for scholarly communication (e.g., repositories, peer review platforms, data sharing tools) rather than indirectly subsidizing commercial publishers through library subscriptions.
* Mandating Open Access and Open Data: Funding bodies can mandate that research outputs funded by public money must be openly accessible and that underlying data must be shared in public repositories. This increases transparency and verifiability.
* Setting Ethical Standards, Not Policing Content: Government’s role should be to establish broad ethical guidelines and frameworks for research integrity (as seen with bodies like the UGC or ICMR in India) and to ensure that institutions have robust systems in place. Direct policing of scientific content or choice of research topics would indeed be detrimental to academic freedom.
* Interagency Collaboration: Fostering collaboration among national and international agencies to address cross-border research integrity issues, especially with the rise of global “paper mills.”
The “Solution” is a Cultural Shift:
Ultimately, the most powerful solution involves a cultural shift within academia itself. This means:
* Prioritizing quality over quantity: Valuing rigorous, reproducible research over a large number of publications.
* Rewarding open practices: Including data sharing, code sharing, and pre-registration as criteria for grants, promotions, and tenure.
* Educating the next generation: Instilling strong ethical principles and responsible research practices from early career stages.
This approach aims to reduce the leverage of commercial publishers by creating viable, community-driven alternatives, while empowering researchers and their institutions to be the frontline guardians of integrity, without inviting overbearing government oversight. It’s a long process, but significant momentum is building in this direction.
Leave a Reply
You must be logged in to post a comment.