How is this a thing? How has nobody just started hosting their own papers? What does it need except a fairly basic website and some storage for the papers themselves? Forgive the ignorance, I’m an IT guy not a scientist…
Some have, but even that depends on licensing options. Many universities have institutional repositories with folks working hard to get affiliate papers uploaded for open access, but they still have to follow the publisher’s license. Some publishers allow OA upload in an IR after an embargo period. Some do only if you pay for general OA publication (extra cost on top of the regular publishing costs, although subscribe to open or read & publish deals sometimes take care of the fees). Some allow it as a matter of course. Some allow it if the author requests it at some point. Some just don’t care and never allow it.
There are also university presses or nonprofit publishers, but their models often aren’t that different. It should be treated as a public service, especially for research given public monies to be completed, but it’s currently just business as usual.
Because the journals provide a quality gate. To be published in, say, Nature is a career peak for most scientists. While counting references to a paper can tell you some things about its relative merit, it’s not as clear an indicator as having a PNAS, Cell or similar on your resume.
They have created a market for their name so it self-perpetuates.
It’s a bit of a circular problem. Certain journals have a reputation of publishing higher quality work, so if you see where it’s published, you’re more likely to read it. Since it draws in readers, it leads to more citations. More citations means more people want to publish there, meaning that the journal gets to be more selective and gets to choose the cream of the crop. Thus maintaining their reputation of publishing higher quality work.
In certain fields, at least, there are important steps these papers provide such as screening and review that are simply not feasible through as self-hosted. People who understand what the paper is about and can sniff out bullshit - be it cooked numbers, incorrect figures, improper citations, etc. are an important part of the process. Heck, even among academic papers out there, some are much lower ‘quality’ than others in that they are frequently bought off or have poor review processes allowing fluff and bad science to get through.
With all that being said, scihub is a thing and even paid journals are often easily pirated.
Peer review is false security, so much bad and fraudulent science gets through, but due to the stamp of authority people are less skeptical. Additionally it’s harder to publish good science.
Don’t get me wrong, it’s not a way to solve everything. But an authoritative body can build credibility and hold onto it. People should still be skeptical and still review, but that’s a normal part of the scientific process. Knowing what’s more and less credible is a normal process of research, and learning to assess credibility is important too. Peer review doesn’t need to be torn down as a concept, it just needs to be taken with a healthy grain of salt, like all processes. This is part of why I mentioned how some journals are more reputable than others - it’s a reflection of how often their peer review misses important things, not a reflection of how bullet-proof their science is. Everyone makes mistakes, the goal should always be to make less.
Also, to be clear, I’m talking about the post-research and pre-publish step, not the pre-research proposal step - that form of peer review can fuck right off.
Also of great importance which I should have probably highlighted in my initial post - this is really dependent on the field itself. In medicine people put in effort for that kind of review. I’ve peer reviewed quite a few papers and I’ve received really good advice from peer reviewers on some of the papers I’m on. Certainly this can happen in environments where this kind of review isn’t necessary, but the institutions that exist do make it a lot easier. An open source self-hosted model would make it really hard to get an idea of how many eyes were on a particular paper, and would make keeping up with continuing education difficult… of course unless groups of people made their career reviewing everything that emerges and putting together summaries or otherwise helping to sift through the noise.
So I actually spent a few seconds thinking about it and I think the main problem would come down to moderation? Ultimately if someone who wrote a paper wanted to distribute it they could do so using existing sites like pastebin or GitHub. The concern with running one myself would be “how do I know this doesn’t contain child abuse images or something”.
How is this a thing? How has nobody just started hosting their own papers? What does it need except a fairly basic website and some storage for the papers themselves? Forgive the ignorance, I’m an IT guy not a scientist…
Some have, but even that depends on licensing options. Many universities have institutional repositories with folks working hard to get affiliate papers uploaded for open access, but they still have to follow the publisher’s license. Some publishers allow OA upload in an IR after an embargo period. Some do only if you pay for general OA publication (extra cost on top of the regular publishing costs, although subscribe to open or read & publish deals sometimes take care of the fees). Some allow it as a matter of course. Some allow it if the author requests it at some point. Some just don’t care and never allow it.
There are also university presses or nonprofit publishers, but their models often aren’t that different. It should be treated as a public service, especially for research given public monies to be completed, but it’s currently just business as usual.
Because the journals provide a quality gate. To be published in, say, Nature is a career peak for most scientists. While counting references to a paper can tell you some things about its relative merit, it’s not as clear an indicator as having a PNAS, Cell or similar on your resume.
They have created a market for their name so it self-perpetuates.
That quality gate hasn’t been doing it’s job for a long time.
That may be the case but a Science article on the resume is still something every working scientist covets.
It’s a bit of a circular problem. Certain journals have a reputation of publishing higher quality work, so if you see where it’s published, you’re more likely to read it. Since it draws in readers, it leads to more citations. More citations means more people want to publish there, meaning that the journal gets to be more selective and gets to choose the cream of the crop. Thus maintaining their reputation of publishing higher quality work.
In certain fields, at least, there are important steps these papers provide such as screening and review that are simply not feasible through as self-hosted. People who understand what the paper is about and can sniff out bullshit - be it cooked numbers, incorrect figures, improper citations, etc. are an important part of the process. Heck, even among academic papers out there, some are much lower ‘quality’ than others in that they are frequently bought off or have poor review processes allowing fluff and bad science to get through.
With all that being said, scihub is a thing and even paid journals are often easily pirated.
Peer review is false security, so much bad and fraudulent science gets through, but due to the stamp of authority people are less skeptical. Additionally it’s harder to publish good science.
There’s a lot of people who understand this better than me who can explain it. Here’s one starting point. https://www.experimental-history.com/p/the-rise-and-fall-of-peer-review
Don’t get me wrong, it’s not a way to solve everything. But an authoritative body can build credibility and hold onto it. People should still be skeptical and still review, but that’s a normal part of the scientific process. Knowing what’s more and less credible is a normal process of research, and learning to assess credibility is important too. Peer review doesn’t need to be torn down as a concept, it just needs to be taken with a healthy grain of salt, like all processes. This is part of why I mentioned how some journals are more reputable than others - it’s a reflection of how often their peer review misses important things, not a reflection of how bullet-proof their science is. Everyone makes mistakes, the goal should always be to make less.
Also, to be clear, I’m talking about the post-research and pre-publish step, not the pre-research proposal step - that form of peer review can fuck right off.
Also of great importance which I should have probably highlighted in my initial post - this is really dependent on the field itself. In medicine people put in effort for that kind of review. I’ve peer reviewed quite a few papers and I’ve received really good advice from peer reviewers on some of the papers I’m on. Certainly this can happen in environments where this kind of review isn’t necessary, but the institutions that exist do make it a lot easier. An open source self-hosted model would make it really hard to get an idea of how many eyes were on a particular paper, and would make keeping up with continuing education difficult… of course unless groups of people made their career reviewing everything that emerges and putting together summaries or otherwise helping to sift through the noise.
So I actually spent a few seconds thinking about it and I think the main problem would come down to moderation? Ultimately if someone who wrote a paper wanted to distribute it they could do so using existing sites like pastebin or GitHub. The concern with running one myself would be “how do I know this doesn’t contain child abuse images or something”.