While open access seems great, and the reasoning behind it is inline with my ideals, I still have problems with "gold" OA, which seems to be what is being referred to within this article:
>Put another way: Publishers are still going to get paid. Open access just means the paychecks come at the front end.
Firstly, the fees imposed by journals are thousands of dollars, which is far too much for many researchers to pay. It would seemingly largely prevent the publication of independent research within such journals.
This was mentioned in the article:
>In fact, many academics still don’t publish in open access journals. One big reason: Some feel they’re less prestigious and lower quality, and that they push the publishing costs on the scientists.
However, the article seemingly (and contradictorily) earlier implies that Gold OA is a solution to pushing the cost onto the researchers:
>Academics are not paid for their article contributions to journals. They often have to pay fees to submit articles to journals and to publish.
However, under Gold OA this is only exacerbated, with large fees being everywhere on the publication-end. The readers don't have to pay, but now the authors do.
Lastly, the article mentions predatory publishing, however fails to note that this phenomena is caused by Gold OA in the first place. In fact, it is sometimes specifically called "predatory open-access publishing". The idea behind predatory publishers is that Gold OA incentivizes publication (as they now get paid per-paper), leading them to seek out and accept as many papers as possible regardless of quality.
While open science certainly is in-line with my views, I'm not convinced that Gold OA is a good solution here.
Agreed. Throwing gobs of money at Elsevier, Springer, etc., regardless of whether it's front-end or back-end, doesn't seem like a winning proposition to me.
I'm not going to say that they don't add any value, but the existence of overlay journals like Discrete Analysis, and other high-quality "pure play" open-access journals like JMLR, and JAIR, etc., suggest to me that it's possible to create a journal system where very little money is required on either end. I think our aspiration should be to see most scientific publishing move to such a model.
I tried to negotiate with one journal regarding their OA fees, and I didn't get anywhere. Their journal, which I preferred because it was more specialized (it's not particularly prestigious because few people know about it) charges about 40% more than the cheapest OA journal in my field. The journal staff didn't seem to care how much cheaper their competition is. Not surprisingly, almost no one publishes OA in that journal.
Still, the cheapest in my field is AIAA journal at $2000 per article. That's still outrageous. Given my understanding of what happens behind the scenes I find it hard to believe any journal actually needs that much money to stay profitable.
I'd be happy to pay an order of magnitude less, say $200, if that could ensure the article stays online for a long period of time. I don't know anything about starting new journals, but I hope competition helps this situation. I might start a journal someday...
Unfortunately the term Gold Open Access has been co-opted by the publishers. When the OA movement was using the term, they meant "published open access by the publisher, rather than posting alternative editions elsewhere (Green OA)". The publishers then started using it to mean author-pays Open Access.
Thus, when OA advocates use the term Gold OA, that gets interpreted the way you do above - whereas they usually intend for the fees to be low or non-existent, for authors. Some have started to use Diamond or Platinum OA for that, but it's hard to get that to stick now.
The point is: there are definitely Open Access models possible where publishing does not entail thousands of dollars of publication costs. This has been proven by many quality journals already.
Sure, but unfortunately this is what it's become. I generally use the Green/Gold/Diamond split, where my own preferences go along the lines of diamond > green > gold. 'Diamond' journals exist, but they are rare.
> One big reason: pirates, including Kazakh neuroscientist Alexandra Elbakyan. Her (illegal) website Sci-Hub sees more than 500,000 visitors daily, and hosts more than 50 million academic papers.
Definitely citation needed on the “(illegal)”.
FYI, this site is very useful and keeps track of the ever-changing sci hub links: https://sci-hub.now.sh/
Question - what makes sci-hub.now.sh, which is essentially acting as a kind of manual DNS, more resilient to censorship than the actual DNS system that necessitates all these ever-changing links? And can we automate it?
So you can't censor sci-hub.now.sh (at least in the US, where linking is typically protected speech). So that's one thing.
Also, the actual DNS system doesn't have a way of redirecting you to another site when the record has been censored, so it's not very resilient in the first place.
I'd guess it's the same as what makes Wikipedia resilient: they're just links. As long as Wikipedia and Wikidata can link to the current location of Sci-Hub, why can't other sites?
Ha, that's interesting - the concept and execution seems almost identical a the project I shared [1] here more than a year ago: https://whereisscihub.now.sh/
An example is Vox's current tirade against YouTube with the goal of censoring certain creators in the name of one-sided "inclusivity."
What happens when Vox aims their influence at scientific publications publishing research with uncomfortable implications? For example, the genetic factor in IQ, violence, and development.
That's all I'm saying. I don't think their proclivity for censorship will merely stop at social media platforms.
This is stuff that's severely misrepresented by the alt-right. Yes, there are statistical variations in IQ and criminality across social groups. No, there's no consensus whatsoever that all or even most of that is due to any sort of "genetic factor", and the plausible contribution of something like that is well within the "too small to matter" range!
Besides, these distributions are such that they can only ever matter in the large, as in "explain other large-scale social phenomena"; they don't tell you ANYTHING about any individual person that you couldn't garner way more easily and more reliably within a few seconds of interaction or even observation.
Most people (as in 99.999%) who bring up "race-IQ" connections as if they were some sort of forbidden truth are either white supremacists or clueless memebots, spreading a memetic infection of white-supremacist origin without even realizing that this is what they're doing. It's pure race baiting and white supremacy, not a serious attempt at social debate.
Despite your well thought-out response, you didn't address my point: Vox has shown a strong proclivity for censorship, and there's no reason to believe they will stop at social media platforms when arguing for inclusivity.
I'm happy to address your comment though, because I'd like to relieve you of the mental gymnastics you're engaged in:
> there's no consensus whatsoever that all or even most of that is due to any sort of "genetic factor", and the plausible contribution of something like that is well within the "too small to matter" range!
Where did you get this information? A quick glance at the "Heritability of IQ" Wikipedia page [1] clearly states that "the heritability of IQ for adults is between 57% and 73% with some more-recent estimates as high as 80% and 86%."
There are probably hundreds of twin/adoption/family studies on the heritability of IQ within and between races. This is simply one of the most researched topic in this field, and claiming that it's "too small too mater" and "there's no consensus whatsoever" seems disingenuous to me.
For example, some scientists have claimed that the heritability of blacks is significantly lower than that of whites [2], and this is due to poverty, the history of slavery, etc. This was addressed by Osborne [3] with his Georgia Twin Study comparing 123 black and 304 white pairs of twins (12-18 years old). Osborne used three different aptitude tests and found an overall heritability of 50% for both blacks and whites. Another study done by Rowe [4] studied the correlations of background variables and outcome measurements using cross-sectional correlational matrices with both independent variables (homes, peers) and outcomes (achievements, criminality). The black and white matrices were almost identical.
The scientific consensus is that IQ (specifically g) is both environmental and genetic.
> Besides, these distributions are such that they can only ever matter in the large, as in "explain other large-scale social phenomena"; they don't tell you ANYTHING about any individual person that you couldn't garner way more easily and more reliably within a few seconds of interaction or even observation.
I totally agree with you here. You can't judge an individual by group averages.
I should also add, I'm not saying people are subject to a kind of race/IQ determinism. But in large-scale, yes these things matter.
> Most people (as in 99.999%) who bring up "race-IQ" connections as if they were some sort of forbidden truth are either white supremacists or clueless memebots, spreading a memetic infection of white-supremacist origin without even realizing that this is what they're doing. It's pure race baiting and white supremacy, not a serious attempt at social debate.
This appears to me as an attempt to discredit those with uncomfortable arguments. If the truth is uncomfortable, we still have to address it and pursue it. There are people who dedicate their entire academic career to studying these questions, with the goal of people discussing and arguing over their work.
The science shows that there are racial differences in IQ. In fact, it also shows that east Asians and Ashkenazi Jews on average have much higher IQ scores than whites. Am I a Jewish-supremacist for bringing that up?
Not a serious attempt at social debate? Yes, calling 99.9999% of the people attempting a serious social debate "white supremacist" is why we can't have a serious debate.
People like you are why we can't have a serious discussion about anything nowadays. You instantly try to shut down any discussion by "alt-right" or "white supremacist" attacks.
Unfortunately for you, I'm not white or a white supremacist, so your ad hominems don't work on me so you'd better find another tactic.
For all the nonsense about "alt-right", it's the "alt-left" extremists that I fear as the alt-left extremist seem to have control over media and power.
Sure, this topic is "severely" misrepresented by the alt-right, but it's also "severely" misrepresented by the alt-left. The alt-right wants to pretend it is pervasive and all-important. The alt-left wants to pretend it doesn't matter and doesn't tell us anything of use. Well here is a shocker. Both of you guys are wrong. The alt-right and alt-left is wrong.
The only difference is that the alt-left are trying to stifle research, debate and discussion because the alt-left extremists have more power. Hopefully that doesn't last much longer as sane people wrest control from the alt-left extremists.
I swear, the feeble minded always use "alt-right" and "white supremacist" as a crutch because they can't accept facts, data and truth.
Can you please not use HN for ideological battle? We ban accounts that do that, because it destroys thoughtful discussion and curiosity. Those are what we want here.
Starting ideological flamewars like this will get you banned here. In fact, I just banned your account. But then I saw that you've posted interesting comments on other topics, so unbanned you. Please don't use HN for political or ideological battle, though.
The Gates Foundation incidentally is one of the strongest voices in the push for Open Access. As a large and influential scientific funder, they were one of the first to demand research they funded to be OA, and recently one of the first non-founding parties to sign up for the strongest push for OA so far: Plan S.
Unfortunately, it's not a problem you can solve just by throwing money at it. It's primarily an incentive problem, and the only influence the Gates Foundation has on that is through its funding of research. In that regard, it's doing very well.
Github could be crucial in helping to win this war. The problem is not only that journals are behind paywalls (luckily sci-hub is helping here) but also the format we use for publishing and review is from last century. Github could help to make it easier to share articles with the code and data used to obtain it, and to have better process of peer review.
I heard arguments where US Government employees were working on creating open access policies for the research they funded. People would lament, "Nobody will work with us if we force them to make their research open access." I would reply, "Who are they going to work with? The US Government is the only place with these big piles of money. Yes, they will take a principled stand until their next mortgage payment is due."
This issue could be solved over night if the USFG (NSF/NIH/DoD) stepped in and said "all publications supported by our grants must be published open access and we'll pay no more than $N/page in publishing fees."
You probably need to explicitly ban publishing in publications with publishing fees, otherwise money from other sources will be used to pad out the difference.
No, that's letting the perfect be the enemy of good enough.
Outright bans make open access harder; editing+publishing with reasonable quality and archival levels of access guarantees can be cheap but it's never free.
Just limit it to a very reasonable $/page. Even upper bounding it at some obscene amount like $10/page would be a vast improvement and a completely trivial expense (you don't want to know what plane tickets to IJCAI cost this year...)
To clarify, I'm saying any research funded by the NSF should have this requirement imposed on all publications regardless of funding source. I.e., DON'T say "NSF $ can't be used for more than $X in publication fees", say "NSF $ can't be used AT ALL if you ever pay more than $X in publication fees".
>But there’s a big thing getting in the way of a revolution: prestige-obsessed scientists who continue to publish in closed-access journals.
That's blaming the victim. It's not the academics who are prestige-obsessed, it's the universities that assess and rank their staff by publication counts within such journals. Most academics I know just want to keep their job.
34 comments
[ 2.4 ms ] story [ 62.2 ms ] thread>Put another way: Publishers are still going to get paid. Open access just means the paychecks come at the front end.
Firstly, the fees imposed by journals are thousands of dollars, which is far too much for many researchers to pay. It would seemingly largely prevent the publication of independent research within such journals.
This was mentioned in the article:
>In fact, many academics still don’t publish in open access journals. One big reason: Some feel they’re less prestigious and lower quality, and that they push the publishing costs on the scientists.
However, the article seemingly (and contradictorily) earlier implies that Gold OA is a solution to pushing the cost onto the researchers:
>Academics are not paid for their article contributions to journals. They often have to pay fees to submit articles to journals and to publish.
However, under Gold OA this is only exacerbated, with large fees being everywhere on the publication-end. The readers don't have to pay, but now the authors do.
Additionally, this may create another pro-industry publication bias, as industry-funded studies may be more likely to have the money to publish in pay-to-publish journals, and this apparently has now been dubbed "e-publication bias" (bottom of https://www.bmj.com/content/bmj/340/7753/Letters.full.pdf, also see http://doi.org/10.1136/bmj.l1544).
Lastly, the article mentions predatory publishing, however fails to note that this phenomena is caused by Gold OA in the first place. In fact, it is sometimes specifically called "predatory open-access publishing". The idea behind predatory publishers is that Gold OA incentivizes publication (as they now get paid per-paper), leading them to seek out and accept as many papers as possible regardless of quality.
While open science certainly is in-line with my views, I'm not convinced that Gold OA is a good solution here.
I'm not going to say that they don't add any value, but the existence of overlay journals like Discrete Analysis, and other high-quality "pure play" open-access journals like JMLR, and JAIR, etc., suggest to me that it's possible to create a journal system where very little money is required on either end. I think our aspiration should be to see most scientific publishing move to such a model.
Still, the cheapest in my field is AIAA journal at $2000 per article. That's still outrageous. Given my understanding of what happens behind the scenes I find it hard to believe any journal actually needs that much money to stay profitable.
I'd be happy to pay an order of magnitude less, say $200, if that could ensure the article stays online for a long period of time. I don't know anything about starting new journals, but I hope competition helps this situation. I might start a journal someday...
Thus, when OA advocates use the term Gold OA, that gets interpreted the way you do above - whereas they usually intend for the fees to be low or non-existent, for authors. Some have started to use Diamond or Platinum OA for that, but it's hard to get that to stick now.
The point is: there are definitely Open Access models possible where publishing does not entail thousands of dollars of publication costs. This has been proven by many quality journals already.
Definitely citation needed on the “(illegal)”.
FYI, this site is very useful and keeps track of the ever-changing sci hub links: https://sci-hub.now.sh/
Also, the actual DNS system doesn't have a way of redirecting you to another site when the record has been censored, so it's not very resilient in the first place.
Relevant similar site: https://dark.fail
[1] https://news.ycombinator.com/item?id=16591962
[1]: https://www.vox.com/policy-and-politics/2018/3/27/15695060/s...
[2]: https://www.vox.com/2018/4/10/17182692/bell-curve-charles-mu...
What happens when Vox aims their influence at scientific publications publishing research with uncomfortable implications? For example, the genetic factor in IQ, violence, and development.
That's all I'm saying. I don't think their proclivity for censorship will merely stop at social media platforms.
Besides, these distributions are such that they can only ever matter in the large, as in "explain other large-scale social phenomena"; they don't tell you ANYTHING about any individual person that you couldn't garner way more easily and more reliably within a few seconds of interaction or even observation.
Most people (as in 99.999%) who bring up "race-IQ" connections as if they were some sort of forbidden truth are either white supremacists or clueless memebots, spreading a memetic infection of white-supremacist origin without even realizing that this is what they're doing. It's pure race baiting and white supremacy, not a serious attempt at social debate.
I'm happy to address your comment though, because I'd like to relieve you of the mental gymnastics you're engaged in:
> there's no consensus whatsoever that all or even most of that is due to any sort of "genetic factor", and the plausible contribution of something like that is well within the "too small to matter" range!
Where did you get this information? A quick glance at the "Heritability of IQ" Wikipedia page [1] clearly states that "the heritability of IQ for adults is between 57% and 73% with some more-recent estimates as high as 80% and 86%."
There are probably hundreds of twin/adoption/family studies on the heritability of IQ within and between races. This is simply one of the most researched topic in this field, and claiming that it's "too small too mater" and "there's no consensus whatsoever" seems disingenuous to me.
For example, some scientists have claimed that the heritability of blacks is significantly lower than that of whites [2], and this is due to poverty, the history of slavery, etc. This was addressed by Osborne [3] with his Georgia Twin Study comparing 123 black and 304 white pairs of twins (12-18 years old). Osborne used three different aptitude tests and found an overall heritability of 50% for both blacks and whites. Another study done by Rowe [4] studied the correlations of background variables and outcome measurements using cross-sectional correlational matrices with both independent variables (homes, peers) and outcomes (achievements, criminality). The black and white matrices were almost identical.
The scientific consensus is that IQ (specifically g) is both environmental and genetic.
> Besides, these distributions are such that they can only ever matter in the large, as in "explain other large-scale social phenomena"; they don't tell you ANYTHING about any individual person that you couldn't garner way more easily and more reliably within a few seconds of interaction or even observation.
I totally agree with you here. You can't judge an individual by group averages.
I should also add, I'm not saying people are subject to a kind of race/IQ determinism. But in large-scale, yes these things matter.
> Most people (as in 99.999%) who bring up "race-IQ" connections as if they were some sort of forbidden truth are either white supremacists or clueless memebots, spreading a memetic infection of white-supremacist origin without even realizing that this is what they're doing. It's pure race baiting and white supremacy, not a serious attempt at social debate.
This appears to me as an attempt to discredit those with uncomfortable arguments. If the truth is uncomfortable, we still have to address it and pursue it. There are people who dedicate their entire academic career to studying these questions, with the goal of people discussing and arguing over their work.
The science shows that there are racial differences in IQ. In fact, it also shows that east Asians and Ashkenazi Jews on average have much higher IQ scores than whites. Am I a Jewish-supremacist for bringing that up?
[1]: https://en.wikipedia.org/wiki/Heritability_of_IQ
[2]: https://brocku.ca/MeadProject/Cooley/Cooley_1897.html
[3]: https://www.gwern.net/docs/genetics/heritable/1980-osborne-t...
[4]:
People like you are why we can't have a serious discussion about anything nowadays. You instantly try to shut down any discussion by "alt-right" or "white supremacist" attacks.
Unfortunately for you, I'm not white or a white supremacist, so your ad hominems don't work on me so you'd better find another tactic.
For all the nonsense about "alt-right", it's the "alt-left" extremists that I fear as the alt-left extremist seem to have control over media and power.
Sure, this topic is "severely" misrepresented by the alt-right, but it's also "severely" misrepresented by the alt-left. The alt-right wants to pretend it is pervasive and all-important. The alt-left wants to pretend it doesn't matter and doesn't tell us anything of use. Well here is a shocker. Both of you guys are wrong. The alt-right and alt-left is wrong.
The only difference is that the alt-left are trying to stifle research, debate and discussion because the alt-left extremists have more power. Hopefully that doesn't last much longer as sane people wrest control from the alt-left extremists.
I swear, the feeble minded always use "alt-right" and "white supremacist" as a crutch because they can't accept facts, data and truth.
https://news.ycombinator.com/newsguidelines.html
https://news.ycombinator.com/newsguidelines.html
Calling Bill Gates and Warren Buffet and all others : please make science Free, Forever, For Everybody.
There's no bigger creator of human development opportunities than free science (imho).
More than a decade? ago, Bill Gates freed the Feynman lectures, only to lock it in Microsoft's Adobe Flash-competing technology, silverlight.
Now do it properly : make it free as in beer and free as in freedom. Yes Bill, you pay everything, we nothing.
Unfortunately, it's not a problem you can solve just by throwing money at it. It's primarily an incentive problem, and the only influence the Gates Foundation has on that is through its funding of research. In that regard, it's doing very well.
This issue could be solved over night if the USFG (NSF/NIH/DoD) stepped in and said "all publications supported by our grants must be published open access and we'll pay no more than $N/page in publishing fees."
Outright bans make open access harder; editing+publishing with reasonable quality and archival levels of access guarantees can be cheap but it's never free.
Just limit it to a very reasonable $/page. Even upper bounding it at some obscene amount like $10/page would be a vast improvement and a completely trivial expense (you don't want to know what plane tickets to IJCAI cost this year...)
To clarify, I'm saying any research funded by the NSF should have this requirement imposed on all publications regardless of funding source. I.e., DON'T say "NSF $ can't be used for more than $X in publication fees", say "NSF $ can't be used AT ALL if you ever pay more than $X in publication fees".
That's blaming the victim. It's not the academics who are prestige-obsessed, it's the universities that assess and rank their staff by publication counts within such journals. Most academics I know just want to keep their job.