Keeping up with scientific papers on almost any subject is now a daunting task since the number of papers published is increasing exponentially. For example, within 12 months of Covid-19 being named in February 2020, there were over 300,000 citations in Google Scholar and about 50 of those papers had already been retracted. Today there are over a million citations on the topic. It is just not possible to read them all and Covid is not an isolated example.
In part, the problem stems from a change in publication practice from long established subscription journals run by societies for their members to the professional long-established publishers selling bundled journal titles to libraries - in which publication is slow but free to the author, to a new business model where authors pay thousands of dollars to publish quickly on-line and where peer review is basic or non-existent. Alongside that has been the rise of media managers promoting papers from their institute as novel and important together with a pressure to “publish or perish” to justify grant applications and tenure.
Academic publishing is now a US$12 billion industry with a return on capital of about 40%. This has led to the selling of fake papers, selling co-authorships, manipulation of h-scores, and the rise of the so-called “paper mills” that sell complete papers and authorships online and probably are involved in money laundering. Scientists too are cheating with plagiarism, AI graphics and text.
How then do you judge the scientific merits of a paper you read and cite?