This chapter is the seminar deck S1 — What is Scientific Literature?. After Part V’s two platforms (Jade, Jason), the course turns to the practice of science itself: how results become literature, who makes that happen, and how the community accesses it. The deck is deliberately non-technical — it is about the process, not the machinery — and it prepares the ground for the next chapter, where the same author turns the survey into a method (the systematic literature review). Together they explain what the course’s own exam artefact is supposed to be.
Scientific literature is the result of a complex social process:
The process is articulated essentially around four stages:
In general, only when all four stages are well-developed does a scientific result become shared and successful. A result that is produced but never published is lost; published but not disseminated is invisible; disseminated but locked behind unaffordable access is unusable. The engineering mindset of the course applies here too: the chain is only as strong as its weakest stage.
Scientific literature is the result of the activity of many actors, who participate in the process with different aims and roles. Abstracting away from the motivations of actors in the research process could make it difficult to understand the process and its results — and becoming an active part of the process even more problematic.
The three hats overlap in practice: a researcher in a university is simultaneously a producer, part of a promoting institution, and a beneficiary of sponsors — with the corresponding obligations at every stage of the pipeline.
Scientific literature does not exist if not in a shareable form. An idea is not a scientific result per se — no matter how good it is or looks. A scientific result is something that:
Reproducibility and confutability are the main pillars of scientific methodology [Popper, 2002]. This is the course’s epistemological grounding: a result earns the word “scientific” not because it is clever, but because it is checkable — someone else could in principle redo it (reproduce) or show it wrong (confute). Everything the course has built so far — agents, logics, platforms — is subject to this test.
The primary sharable form for scientific results are articles (also called papers), collected and published:
When results are stable, they are often presented in extended form, in scientific monographies.
Scientific literature is subject to public control. Before it is published, an article is submitted in some form to a review process — for publication in a scientific journal, a book collection, a conference, a symposium, a workshop. Review is conducted by experts in the field, and concludes in a final evaluation:
The review is the social mechanism that implements the pillars of section 3: experts check that the article is reproducible and confutable as claimed, that the result is new, and that the form is shareable. The pipeline is a filter, not a formality.
Scientific literature is published (and disseminated) by publishers. Once ready, proceedings, journal issues, collections and monographies are scientific products — but not immediately typographic ones. Often, publishers intervene on the form of the scientific material before it is published: language, formatting, cover, illustrations — with the consent of authors and editors.
The role of publishers is twofold:
Publishers dictate the timing and pace for publication of volumes/issues and the total number of pages; they may also provide suggestions on the general goals of a published volume/journal. They handle organisational issues, and introduce and govern social & economical factors in the scientific process. They also take care of dissemination — and they rule access to scientific literature. The last point is the seed of the paywall problem of section 7.
Dissemination of scientific literature is a multi-faceted process:
Still, the main tool & measure of dissemination is citation:
Web resources for papers and citations: Google Scholar (scholar.google.com), Scopus (scopus.com), Web of Science (webofscience.com), DBLP (dblp.uni-trier.de). The course’s own resources: IRIS (cris.unibo.it) and the APICe group’s publications (pubs.apice.unibo.it).
Nowadays, access to scientific literature is mostly online. Traditional means are still widespread — participation in scientific events, access to printed materials in libraries and personal collections — but online access has gained more and more ground and is nowadays the most important means of access. Internet & Web technologies have obviously a key role: they allow an unprecedented flow of dissemination of and access to technical results — either published or yet unpublished — and might sometimes be the only way to access them, as it became more than evident during the COVID-19 pandemia. All international publishers have online publication sites; most of the relevant material is accessible online — some behind paywall.
Paywall vs. Open Access. Scientific literature can be either openly accessible or behind paywall. The paywall model is typically imposed by publishers, and is part of their business model: access costs are typically high — for private citizens, researchers, and institutions; public institutions mostly pay huge money to access some of it; the micropayment model was often proposed but never seriously considered.
Open access to scientific literature responds first of all to the fundamental need of science to be shared. The Budapest Open Access Initiative (BOAI, 2002) defines it precisely:
“By ‘open access’ to this literature, we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited.”
Also relevant: the Berlin Declaration on Open Access [Max Plank Society, 2003]. Two models implement open access: institutions can become publishers, and publishers can make it part of their business model by making authors and institutions pay for public access (the “gold” route).
The problem of paywall. Most of the scientific literature is produced by publicly-funded institutions — yet public institutions have to pay in order to access their own products behind paywall: paywall by publishers that did not pay for the public institutions’ work, which was instead mostly paid by public money. So the public pays twice for those research products.
In order to share results before any form of scientific publication, public digital repositories are used — e.g. arXiv (arxiv.org). They are also commonly used to make publicly shareable versions of scientific papers behind paywall, in order to provide some form of open access — e.g. pre-prints (before review) and post-prints (after review).
But public repositories are not the solution for open access: documents in public repositories are in general just technical reports — without validation by the scientific community, they are not scientific papers. The exception is when the paper has been published properly already, and the paper in the repo represents its open-access version.
The same holds for manuals and technical docs: whatever the sources, manuals and technical documentations are just that — no science there, just technical information, hopefully correct. The distinction matters for the exam artefact too: a survey is scientific literature when it follows the methodological discipline of the next chapter, not when it merely collects links.
The deck closes with a practical map of where the literature lives online.
Digital libraries from main international publishers: Springer (link.springer.com), Science Direct / Elsevier (sciencedirect.com), Taylor & Francis (tandfonline.com), World Scientific (worldscientific.com), Cambridge UP (cambridge.org/core/publications), Oxford Academic (academic.oup.com), InderScience (inderscience.com), SAGE (sagepub.com), Wiley (wiley.com), IGI Global (igi-global.com).
Digital libraries from main international associations: Association for Computing Machinery (ACM, acm.org) with the ACM Digital Library (dl.acm.org); IEEE Computer Society (computer.org) with IEEE Xplore (ieeexplore.ieee.org); International Federation for Information Processing (IFIP, ifip.org) with the IFIP Open Digital Library (dl.ifip.org).
Digital libraries by other players: DBLP (dblp.org), Google Scholar (scholar.google.com), Scopus (scopus.com), Web of Science (webofscience.com), ORCID (orcid.org), SciProfiles (sciprofiles.com), Academia.edu (academia.edu).
Two families matter for a student of this course. The publisher libraries hold the paywalled, reviewed literature; the association libraries (ACM, IEEE, IFIP) hold the venues where computer science is actually published — conferences first. The other players are indexes and aggregators: they do not publish, they make literature findable and citable.
Scientific literature is a social process in four stages — production, publication, dissemination, access — driven by researchers, universities, funding bodies and publishers. Its artefacts are shareable by construction: a result is scientific only if it is reproducible and confutable [Popper, 2002]. Publication is a public control (review by experts); dissemination is measured by citations; access is the battleground between paywalls and open access. For the exams and the course artefact: a survey is literature when it is structured, shareable, and checkable — not when it is merely a collection of links.
Scientific literature is the result of a complex social process involving thousands of skilled people world-wide, growing constantly, where individual, social, organisational, economical and political issues are often as important as scientific ones. The four stages are: production, publication, dissemination, access. Only when all four are well-developed does a scientific result become shared and successful.
Researchers (the producers, mainly involved in production but typically involved in all four stages: meetings, projects, books and journals, dissemination, access); universities & research centres (the promoters, whose aims include knowledge transmission, visibility and transfer to industry, competitive advantage); funding bodies (the sponsors, public or private, funding through projects or individual grants); publishers (who publish, disseminate, and rule access).
Because scientific literature does not exist if not in a shareable form. A scientific result must be presented and structured to be understood by non-authors, have a form shareable and accessible by the community, and give readers enough information to reproduce and possibly confute it. Reproducibility and confutability are the main pillars of scientific methodology [Popper, 2002].
Articles (or papers), collected and published in scientific journals, in proceedings of conferences, symposia and workshops, or as chapters of collections in the form of books. When results are stable, they are often presented in extended form in scientific monographies.
Before publication an article is submitted to a review process (journal, book collection, conference, symposium, workshop). Review is conducted by experts in the field and concludes in a final evaluation; if worth of publication, the article may be revised according to the reviewers’ indications and finally accepted. When published, it represents a piece of scientific literature. At conferences, public presentation and discussion are an essential part of dissemination.
Primarily earning money (at least not losing money); secondarily earning reputation through high-quality publications. They dictate timing, pace and number of pages, handle organisational issues, take care of dissemination, and rule access to scientific literature.
Citation. Paper A cites paper B in its bibliography; citations are a measure of scientific impact (even though citation is not necessarily a sign of approval, it is typically a good measure of relevance) and are typically used for evaluation of scientific production. Relevant web resources: Google Scholar, Scopus, Web of Science, DBLP, plus IRIS (cris.unibo.it) and APICe.
Most scientific literature is produced by publicly-funded institutions, yet public institutions have to pay to access their own products behind paywall — paywall by publishers that did not pay for the institutions’ work. The public pays twice for those research products.
By ‘open access’ we mean free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited [Budapest Open Access Initiative, 2002].
No. Documents in public repositories (e.g. arXiv) are in general just technical reports: without validation by the scientific community they are not scientific papers. They serve to share results before publication and to host pre-prints/post-prints of already-published papers — the latter being a legitimate open-access version. The same holds for manuals and technical docs: no science there, just technical information.
Publishers: Springer, Science Direct/Elsevier, Taylor & Francis, World Scientific, Cambridge UP, Oxford Academic, InderScience, SAGE, Wiley, IGI Global. Associations: ACM (ACM DL), IEEE Computer Society (IEEE Xplore), IFIP (IFIP Open DL). Others: DBLP, Google Scholar, Scopus, Web of Science, ORCID, SciProfiles, Academia.edu.