Question
Are scholarly journals outdated? If so, what should replace them?
Answer
Journal publishing issues (no pun intended) are merely a symptom. The real problem is the professionalization of research. We need to re-amateurize inquiry.
Research publishing started as a way to share new knowledge. Now it has become a medium for defensive IP based business model insurance for companies, job-independent career documentation for individuals, a way to lay a paper trail for grant funding for University-based scholars, and finally a way for disciplinary societies (the new guilds) to achieve an IP balance of power with the competitive part of the information marketplace (business trade secrets, patents, and university publishing).
This even applies to the liberal arts, though more weakly. Through complicated money flows that involve (in the US) the NAE, Pew, political think tanks etc., even an apparently non sequitur paper about Shakespeare ends up becoming as much a part of the scholarly-industrial complex as some technical engineering paper.
That's really what this is: a "scholarly industrial complex" that's a relic of the industrial age. I am not saying industrial modes of scarcity-driven production are bad. But in the age of the Internet, there is no justification for applying them to anything that does not involve physical atoms. Research output is bits. It's not like making steel.
In the scholarly-industrial complex, the idea of actually sharing information for the purpose of stimulating intellectual exchanges has gone from being 100% of the purpose in the old days of Abbe Mersenne's living room or the original Royal Society, to maybe 10% (optimistically). The other 90% of the purpose of journal publishing has nothing to do with the original purpose. It's about perpetuating the institutions that live off the publishing. This explains why 90% of the papers published in most disciplines is mostly junk DNA in humanity's collective knowledge genome.
I am not saying the work that goes into them is useless, but the value most of the time is not in the paper, but in the person who does the work, and their immediate social research context. Good research mostly diffuses through word of mouth, not journals. I don't think any of my papers belongs in the 10% non-junk (there, I admitted it). The 90% does NOT need to be "published." It should merely be written up and posted online, and informally discussed. I mean come on, everybody has LaTeX for FREE today, and can get a website for free as well.
The other 10%, yeah, sure. Put it in journals if you want. But by definition that stuff is so great it will diffuse with extraordinary rapidity with or without journals, and get extreme scrutiny. If someone publishes a credible proof of the [math]P\neq NP[/math] conjecture, trust me, you don't need journals to have that thing spread like wildfire and a dozen smart people staying up nights to verify the proof. The journal as an intermediary can be completely bypassed.
Really, for the 10% that deserves publication, the journals should serve more as commemorative "gold record/platinum record" status. Maybe every preprint that gets forwarded on email more than 100,000 times should be in the "Nature: Gold" journal.
But that doesn't mean the WORK behind the other 90% is junk. Only the papers themselves. The work in my case did have impact: it trained me to be smarter, and led to more consequential work. The students I worked with benefited. A few peers got some inspiration out of the work (as I did out of theirs). I got a couple of jobs on the strength of the work. But ALL of this impact was through talks, personal conversations etc. NOBODY ever got wiser or benefited from actually reading my papers AFAIK. At best, they looked at one of them, got interested, and emailed me to talk in person and have me explain some proof or code to them. Even the jobs I got were based on people actually getting to know me, and trusting my abilities. The CV and publication list were merely a passport and visa that got me the interviews, and this is a terrible state of affairs. I've seen very talented people denied opportunities because they lacked the passport/visa. And I've seen morons get great jobs they didn't deserve on the strength of junk publication lists. These are the alpha and beta errors of the scholarly-industrial complex, and the error rates are unforgivably high (and rising higher every year).
So if most (90%) of the value is being created in ways OTHER than journal publishing, what's the point of an elaborately bureaucratic system to control access to "published" status? Some say it's about peer review, but that doesn't actually work well enough these days to justify the system. See my answer to:
How viable is the academic peer review system today?
Maybe it is about broader recognition? I think that's actually set up in toxic ways. I think the healthier "recognition hunger" behaviors can be supported without the publishing system. See my answer to:
How much does ego and the desire to be recognized as an individual impede scientific progress?
And finally, maybe it's about the sheer scale of knowledge production that demands the professionalization? Nope. See my answer to:
How might we build an education system that is centered on creating rather than replicating knowledge?
Research publishing started as a way to share new knowledge. Now it has become a medium for defensive IP based business model insurance for companies, job-independent career documentation for individuals, a way to lay a paper trail for grant funding for University-based scholars, and finally a way for disciplinary societies (the new guilds) to achieve an IP balance of power with the competitive part of the information marketplace (business trade secrets, patents, and university publishing).
This even applies to the liberal arts, though more weakly. Through complicated money flows that involve (in the US) the NAE, Pew, political think tanks etc., even an apparently non sequitur paper about Shakespeare ends up becoming as much a part of the scholarly-industrial complex as some technical engineering paper.
That's really what this is: a "scholarly industrial complex" that's a relic of the industrial age. I am not saying industrial modes of scarcity-driven production are bad. But in the age of the Internet, there is no justification for applying them to anything that does not involve physical atoms. Research output is bits. It's not like making steel.
In the scholarly-industrial complex, the idea of actually sharing information for the purpose of stimulating intellectual exchanges has gone from being 100% of the purpose in the old days of Abbe Mersenne's living room or the original Royal Society, to maybe 10% (optimistically). The other 90% of the purpose of journal publishing has nothing to do with the original purpose. It's about perpetuating the institutions that live off the publishing. This explains why 90% of the papers published in most disciplines is mostly junk DNA in humanity's collective knowledge genome.
I am not saying the work that goes into them is useless, but the value most of the time is not in the paper, but in the person who does the work, and their immediate social research context. Good research mostly diffuses through word of mouth, not journals. I don't think any of my papers belongs in the 10% non-junk (there, I admitted it). The 90% does NOT need to be "published." It should merely be written up and posted online, and informally discussed. I mean come on, everybody has LaTeX for FREE today, and can get a website for free as well.
The other 10%, yeah, sure. Put it in journals if you want. But by definition that stuff is so great it will diffuse with extraordinary rapidity with or without journals, and get extreme scrutiny. If someone publishes a credible proof of the [math]P\neq NP[/math] conjecture, trust me, you don't need journals to have that thing spread like wildfire and a dozen smart people staying up nights to verify the proof. The journal as an intermediary can be completely bypassed.
Really, for the 10% that deserves publication, the journals should serve more as commemorative "gold record/platinum record" status. Maybe every preprint that gets forwarded on email more than 100,000 times should be in the "Nature: Gold" journal.
But that doesn't mean the WORK behind the other 90% is junk. Only the papers themselves. The work in my case did have impact: it trained me to be smarter, and led to more consequential work. The students I worked with benefited. A few peers got some inspiration out of the work (as I did out of theirs). I got a couple of jobs on the strength of the work. But ALL of this impact was through talks, personal conversations etc. NOBODY ever got wiser or benefited from actually reading my papers AFAIK. At best, they looked at one of them, got interested, and emailed me to talk in person and have me explain some proof or code to them. Even the jobs I got were based on people actually getting to know me, and trusting my abilities. The CV and publication list were merely a passport and visa that got me the interviews, and this is a terrible state of affairs. I've seen very talented people denied opportunities because they lacked the passport/visa. And I've seen morons get great jobs they didn't deserve on the strength of junk publication lists. These are the alpha and beta errors of the scholarly-industrial complex, and the error rates are unforgivably high (and rising higher every year).
So if most (90%) of the value is being created in ways OTHER than journal publishing, what's the point of an elaborately bureaucratic system to control access to "published" status? Some say it's about peer review, but that doesn't actually work well enough these days to justify the system. See my answer to:
How viable is the academic peer review system today?
Maybe it is about broader recognition? I think that's actually set up in toxic ways. I think the healthier "recognition hunger" behaviors can be supported without the publishing system. See my answer to:
How much does ego and the desire to be recognized as an individual impede scientific progress?
And finally, maybe it's about the sheer scale of knowledge production that demands the professionalization? Nope. See my answer to:
How might we build an education system that is centered on creating rather than replicating knowledge?