When will scientists just self-publish? I mean seriously, nowadays there is nothing between a researcher and publishing their stuff on the web. Only thing would be peer-reviewing, if you want that, but then just organize it without Elsevier. Reviewers get paid jack shit so you can just do a peer-reviewing fediverse instance where only the mods know the people so it’s still double-blind.
This system is just to dangle carrots in front of young researchers chasing their PhD
Because of “impact score” the journal your work gets placed in has a huge impact on future funding. Its a very frustrating process and trying to go around it is like suicide for your lab so it has to be more of a top-down fix because the bottom up is never going to happen.
Thats why everyone uses sci hub. These publishers are terrible companies up there with EA in unpopularity.
It sounds like all it would take to destroy the predatory for-profit publication oligarchs is a majority of the top few hundred scientists, across major disciplines, rejecting it and switching to a completely decentralized peer-2-peer open-source system in protest… The publication companies seem to gate keep, and provide no value. It’s like Reddit. The site’s essentially worthless. All of the value is generated by the content creators.
Succesfully iniating this from the fediverse would be such a massive boost in public visibility and discoursive strength of the project of collectivization of information infrastructure (like lemmy).
Imagine we fluffin freed science from capital and basically all the scientists openly stated how useful this was
Those few top people are assholes who love the enormous power they wield over PhD students, postdocs and junior faculty, and they are usually editors on those big name journals. Unlike the people who actually do the work, they are getting paid from this system.
Ya that would be awesome and I think that movement would gain momentum really fast since most high profile labs have all had to deal with this nonsense.
That or legislation/open access rules to make these papers more accessible. One can dream.
the thing that they’re supposed to provide is peer review, solve that and we’re good to go. would be easier to do with some kind of central oversight and stable funding, we’re not talking about shitposting instance for 250 people that nobody will notice if it goes down
It’s commonplace in my field (nuclear physics) to share the preprint version of your article, typically on arxiv.org. You can update the article as you respond to peer reviewers too. The only difference between this and the paywalls publisher version is that version will have additional formatting edits by the journal.
If you search for articles on google scholar, it groups the preprint and published versions together so it’s easy to find the non-paywalled copy. The standard journals I publish in even sort of encourage this; you can submit the latex documents and figures by just putting the url to an arxiv manuscript.
The US Department of Energy now requires any research they fund be made publicly available. So any article I publish is also automatically posted to osti.gov 1 year after its initial publication. This version is also grouped into the google scholar search results.
It’s an imperfect system, but it’s getting much better than it was even just a decade ago.
Yeah I know about this, but personally in our field I don’t see anybody bothering with preprints sadly. Maybe we should though, sounds like the first step.
We (I’m a CS researcher) already kind of do, I upload almost everything to arxiv.org and researchgate. Some fields support this more than others, though.
Bone wars electric bugaloo. In the end you really do need a way to discern who is having an appreciable impact in a field in order to know who to fund. I have yet to hear a meaningful metric for that though.
Edit: I should clarify, the other option is strictly political through an academy of sciences and has historical awfulness associated with it as well.
Editors can act as filters, which is required when dealing with an excess of information streaming in. Just like you follow celebrities on social media or you follow pseudo-forums like this one, you get a service of information filtration which increases the concentration of useful knowledge.
When will scientists just self-publish? I mean seriously, nowadays there is nothing between a researcher and publishing their stuff on the web. Only thing would be peer-reviewing, if you want that, but then just organize it without Elsevier. Reviewers get paid jack shit so you can just do a peer-reviewing fediverse instance where only the mods know the people so it’s still double-blind.
This system is just to dangle carrots in front of young researchers chasing their PhD
Because of “impact score” the journal your work gets placed in has a huge impact on future funding. Its a very frustrating process and trying to go around it is like suicide for your lab so it has to be more of a top-down fix because the bottom up is never going to happen.
Thats why everyone uses sci hub. These publishers are terrible companies up there with EA in unpopularity.
It sounds like all it would take to destroy the predatory for-profit publication oligarchs is a majority of the top few hundred scientists, across major disciplines, rejecting it and switching to a completely decentralized peer-2-peer open-source system in protest… The publication companies seem to gate keep, and provide no value. It’s like Reddit. The site’s essentially worthless. All of the value is generated by the content creators.
Succesfully iniating this from the fediverse would be such a massive boost in public visibility and discoursive strength of the project of collectivization of information infrastructure (like lemmy).
Imagine we fluffin freed science from capital and basically all the scientists openly stated how useful this was
I can only get so erect, please stop.
Thank you, this justifies to introduce myself as campaign porn producer from now on
oh so this is the kind of stuff that turns on asexual people?
That would make them scisexual or politisociosexual I guess.
(What I’m trying to say is you have my bow)
So, shall we do it?
Those few top people are assholes who love the enormous power they wield over PhD students, postdocs and junior faculty, and they are usually editors on those big name journals. Unlike the people who actually do the work, they are getting paid from this system.
Ya that would be awesome and I think that movement would gain momentum really fast since most high profile labs have all had to deal with this nonsense.
That or legislation/open access rules to make these papers more accessible. One can dream.
It’s even worse for low profile labs because those publication fees eat up a greater proportion of our budget.
the thing that they’re supposed to provide is peer review, solve that and we’re good to go. would be easier to do with some kind of central oversight and stable funding, we’re not talking about shitposting instance for 250 people that nobody will notice if it goes down
I know about impact factor but still this system is shit and only works because people contribute to it.
Even Nature publishes shit articles now and then. Impact score is becoming a joke more and more.
It’s commonplace in my field (nuclear physics) to share the preprint version of your article, typically on arxiv.org. You can update the article as you respond to peer reviewers too. The only difference between this and the paywalls publisher version is that version will have additional formatting edits by the journal.
If you search for articles on google scholar, it groups the preprint and published versions together so it’s easy to find the non-paywalled copy. The standard journals I publish in even sort of encourage this; you can submit the latex documents and figures by just putting the url to an arxiv manuscript.
The US Department of Energy now requires any research they fund be made publicly available. So any article I publish is also automatically posted to osti.gov 1 year after its initial publication. This version is also grouped into the google scholar search results.
It’s an imperfect system, but it’s getting much better than it was even just a decade ago.
Yeah I know about this, but personally in our field I don’t see anybody bothering with preprints sadly. Maybe we should though, sounds like the first step.
We (I’m a CS researcher) already kind of do, I upload almost everything to arxiv.org and researchgate. Some fields support this more than others, though.
As if peer review weren’t massive fucking joke.
We should just self publish and then openly argue about it findings like the OG scientists. It didn’t stop them from discovering anything.
Bone wars electric bugaloo. In the end you really do need a way to discern who is having an appreciable impact in a field in order to know who to fund. I have yet to hear a meaningful metric for that though.
Edit: I should clarify, the other option is strictly political through an academy of sciences and has historical awfulness associated with it as well.
Editors can act as filters, which is required when dealing with an excess of information streaming in. Just like you follow celebrities on social media or you follow pseudo-forums like this one, you get a service of information filtration which increases the concentration of useful knowledge.
In the early days of modern science, the rate of publications was small, make it easier to “digest” entire fields even if there’s self-publishing. The number of published papers grows exponentially, as does the number of journals. https://www.researchgate.net/publication/333487946_Over-optimization_of_academic_publishing_metrics_Observing_Goodhart’s_Law_in_action/figures
Just like with these forums, the need for moderators (editors, reviewers) grows with the number of users who add content.
What’s the problem with peer-reviewed open access journals maintained by universities?
I agree but if it was that easy it would have been done already and there would already be another evil gatekeeper to hate.