@Joao I agree, but I am not sure what form post-publication review must take.
The problem is that we need some form of criticism. I want people to tell me I am mistaken. Yet, they will not openly do so for fear that I will get angry.
Well. There are plenty of people who come to my blog to tell me how wrong I am, so there is hope.
Stevan Harnadsays:
Salut Daniel!
You don’t have to pay a publisher for Open Access (OA), nor do you have to “self-publish” to provide to your papers. Publish with your preferred journal and self-archive your final peer-reviewed online free for all (as you already do). Self-archiving is “Green” OA — as distinct from publishing in a “Gold” OA journal: http://www.nature.com/nature/focus/accessdebate/21.html
The best place to self-archive is neither your home page nor Arxiv, but your own institutional repository (IR) — in the case of UQAM: http://www.archipel.uqam.ca/
I’ve often thought that a system of open (if non-binding) reviews could easily replace current peer-review in databases as arXiv. An arXiv article is not reviewed prior to publication, and for that reason, as you put, it lacks “reputation, quality control”, etc…
However, if post-publication reviews from peers is allowed, articles can, in time, gain the reputation and credibility and exposure required for science communication.
In fact, such a system would prove to be much more reliable than having anonymous reviewers dictate the acceptance of an article. As an example, research that departs from the mainstream would not be suppressed, although it might be initially unpopular. And if it is proven to be valid, it stands the chance to be reevaluated and to get general acceptance.
What are your thoughts on this issue?
Francois Rivestsays:
Daniel, what do you think of the PLOS One approach? Paper are peer-reviewed for soundness, not importance, and then people can comment them as for blogs.
Although researchers do not yet tend to leave comments, the next generation might well do. I think good comments about papers are as usefull as the paper themselves, especially for future readers.
The best place to self-archive is neither your home page nor Arxiv, but your own institutional repository (IR) — in the case of UQAM: http://www.archipel.uqam.ca/
At some point, a few months ago, I was the fourth most prolific author on UQAM’s archive. So you can’t blame me for not using archipel. 😉 Also, I aggressively promoted our repository with colleagues, I really did try my best but people see it as a burden. Thus, institutional repositories, at least UQAM’s, are not proving very disruptive.
The fact that you must try to compel professors so that the institutional repository works is a good sign that it is not convenient enough. Arxiv, on the other hand, has over 500,000 papers, and it covers few fields. Arxiv has had a much deeper effect. In Mathematics, Computer Science and Physics, it is changing the game. There is no chance that an institutional repository will do the same.
That is not to say that institutional repositories cannot be useful. I love the way archipel is now indexing student thesis. Great idea. But they are not changing the game. Not the way arxiv does in many fields.
@Rivest PLOS One has a somewhat disruptive effect, I agree. But I have not studied it enough to understand what makes it tick.
Stevan Harnadsays:
DL: “The fact that you must try to compel professors so that the institutional repository works is a good sign that it is not convenient enough. Arxiv, on the other hand, has over 500,000 papers, and it covers few fields. Arxiv has had a much deeper effect. In Mathematics, Computer Science and Physics, it is changing the game. There is no chance that an institutional repository will do the same.”
(1) Computer scientists and physicists have been self-archiving for two decades, spontaneously – computer scientists on their local websites (c. 750K papers, as harvested by Citeseerx), physicists centrally in Arxiv (c. 500K papers).
(2) Other disciplines have not been self-archiving spontaneously in this way (although their own institutional repositories’ deposit interface is infinitely more convenient and user-friendly than Arxiv’s — as you must know, Daniel, being a veteran depositor in both ways!)
(3) The 768 institutional repositories registered in ROAR contain 4,752K deposits, which totals far more than Citeseerx + Arxiv, but falls woefully short, considering that at least 2.5 million papers are published every year in the planet’s 25,000 peer-reviewed journals. http://roar.eprints.org/index.php?action=browse
(4) That is why mandates are needed (UQAM has one for theses, but not yet for articles; however, Harvard, Stanford and MIT and over 80 other institutions and funders do): http://www.eprints.org/openaccess/policysignup/
(5) International surveys, across all disciplines, have shown that 95% of researchers report they would comply if deposit were mandated by their institutions and/or funders, over 80% of them willingly: http://eprints.ecs.soton.ac.uk/11006/
(5) The principal purpose of mandates themselves is to reinforce researchers’ already-existing inclination to maximise access and usage for their give-away articles, not to force researchers to do something they don’t already want to do. http://opcit.eprints.org/oacitation-biblio.html
(6) Researchers need to be reassured that their departments or institutions or funders are indeed fully behind self-archiving, and indeed expect it of them; otherwise researchers remain in a state of Zeno’s Paralysis” about self-archiving year upon year, because of countless groundless worries, such as copyright, journal choice, and even how much time self-archiving takes.)” http://www.eprints.org/openaccess/self-faq/#32-worries
Yay Arxiv! It works definitely better than my institution’s preprint service.
When (if?) I get to ten papers on it, I will be able to “endorse” my advisor. 🙂
I’m already boycotting non-open-source journals and conferences by not reviewing for them and not submitting to them (not that I’ve submitted a journal paper in ages).
Our small company can’t afford proprietary journals, so why should I contribute my time for free?
I’m also boycotting low acceptance rate conference reviewing, because I think it leads to too much redundant work for reviewers.
I have the same issue with books. I wrote two books for mainstream publishers (Cambridge and MIT), but the next book I write will probably be self-published. Cambridge is still an option, since they seem to let folks continue to distribute PDFs for free (like the new Manning et al. IR book and Marti Hearst’s new search book).
@Joao I agree, but I am not sure what form post-publication review must take.
The problem is that we need some form of criticism. I want people to tell me I am mistaken. Yet, they will not openly do so for fear that I will get angry.
Well. There are plenty of people who come to my blog to tell me how wrong I am, so there is hope.
Salut Daniel!
You don’t have to pay a publisher for Open Access (OA), nor do you have to “self-publish” to provide to your papers. Publish with your preferred journal and self-archive your final peer-reviewed online free for all (as you already do). Self-archiving is “Green” OA — as distinct from publishing in a “Gold” OA journal:
http://www.nature.com/nature/focus/accessdebate/21.html
The best place to self-archive is neither your home page nor Arxiv, but your own institutional repository (IR) — in the case of UQAM: http://www.archipel.uqam.ca/
Most institutions now have IRs: http://roar.eprints.org
More and more institutions (but alas not yet UQAM) are adopting mandates to self-archive all published articles: http://www.eprints.org/openaccess/policysignup/
A+ É
Daniel,
I’ve often thought that a system of open (if non-binding) reviews could easily replace current peer-review in databases as arXiv. An arXiv article is not reviewed prior to publication, and for that reason, as you put, it lacks “reputation, quality control”, etc…
However, if post-publication reviews from peers is allowed, articles can, in time, gain the reputation and credibility and exposure required for science communication.
In fact, such a system would prove to be much more reliable than having anonymous reviewers dictate the acceptance of an article. As an example, research that departs from the mainstream would not be suppressed, although it might be initially unpopular. And if it is proven to be valid, it stands the chance to be reevaluated and to get general acceptance.
What are your thoughts on this issue?
Daniel, what do you think of the PLOS One approach? Paper are peer-reviewed for soundness, not importance, and then people can comment them as for blogs.
Although researchers do not yet tend to leave comments, the next generation might well do. I think good comments about papers are as usefull as the paper themselves, especially for future readers.
The best place to self-archive is neither your home page nor Arxiv, but your own institutional repository (IR) — in the case of UQAM: http://www.archipel.uqam.ca/
At some point, a few months ago, I was the fourth most prolific author on UQAM’s archive. So you can’t blame me for not using archipel. 😉 Also, I aggressively promoted our repository with colleagues, I really did try my best but people see it as a burden. Thus, institutional repositories, at least UQAM’s, are not proving very disruptive.
The fact that you must try to compel professors so that the institutional repository works is a good sign that it is not convenient enough. Arxiv, on the other hand, has over 500,000 papers, and it covers few fields. Arxiv has had a much deeper effect. In Mathematics, Computer Science and Physics, it is changing the game. There is no chance that an institutional repository will do the same.
That is not to say that institutional repositories cannot be useful. I love the way archipel is now indexing student thesis. Great idea. But they are not changing the game. Not the way arxiv does in many fields.
@Rivest PLOS One has a somewhat disruptive effect, I agree. But I have not studied it enough to understand what makes it tick.
DL: “The fact that you must try to compel professors so that the institutional repository works is a good sign that it is not convenient enough. Arxiv, on the other hand, has over 500,000 papers, and it covers few fields. Arxiv has had a much deeper effect. In Mathematics, Computer Science and Physics, it is changing the game. There is no chance that an institutional repository will do the same.”
(1) Computer scientists and physicists have been self-archiving for two decades, spontaneously – computer scientists on their local websites (c. 750K papers, as harvested by Citeseerx), physicists centrally in Arxiv (c. 500K papers).
(2) Other disciplines have not been self-archiving spontaneously in this way (although their own institutional repositories’ deposit interface is infinitely more convenient and user-friendly than Arxiv’s — as you must know, Daniel, being a veteran depositor in both ways!)
(3) The 768 institutional repositories registered in ROAR contain 4,752K deposits, which totals far more than Citeseerx + Arxiv, but falls woefully short, considering that at least 2.5 million papers are published every year in the planet’s 25,000 peer-reviewed journals.
http://roar.eprints.org/index.php?action=browse
(4) That is why mandates are needed (UQAM has one for theses, but not yet for articles; however, Harvard, Stanford and MIT and over 80 other institutions and funders do): http://www.eprints.org/openaccess/policysignup/
(5) International surveys, across all disciplines, have shown that 95% of researchers report they would comply if deposit were mandated by their institutions and/or funders, over 80% of them willingly: http://eprints.ecs.soton.ac.uk/11006/
(6) And outcome studies confirm that they do:
http://fcms.its.utas.edu.au/scieng/comp/project.asp?lProjectId=1830
(5) The principal purpose of mandates themselves is to reinforce researchers’ already-existing inclination to maximise access and usage for their give-away articles, not to force researchers to do something they don’t already want to do.
http://opcit.eprints.org/oacitation-biblio.html
(6) Researchers need to be reassured that their departments or institutions or funders are indeed fully behind self-archiving, and indeed expect it of them; otherwise researchers remain in a state of Zeno’s Paralysis” about self-archiving year upon year, because of countless groundless worries, such as copyright, journal choice, and even how much time self-archiving takes.)”
http://www.eprints.org/openaccess/self-faq/#32-worries
(7) What will generate the desired “disruptive” effect will be the globalization of these growing Green Open Access Self-Archiving Mandates:
http://www.eprints.org/openaccess/policysignup/sign.php
Yay Arxiv! It works definitely better than my institution’s preprint service.
When (if?) I get to ten papers on it, I will be able to “endorse” my advisor. 🙂
I’m already boycotting non-open-source journals and conferences by not reviewing for them and not submitting to them (not that I’ve submitted a journal paper in ages).
Our small company can’t afford proprietary journals, so why should I contribute my time for free?
I’m also boycotting low acceptance rate conference reviewing, because I think it leads to too much redundant work for reviewers.
I have the same issue with books. I wrote two books for mainstream publishers (Cambridge and MIT), but the next book I write will probably be self-published. Cambridge is still an option, since they seem to let folks continue to distribute PDFs for free (like the new Manning et al. IR book and Marti Hearst’s new search book).
@Carpenter Ideally, books should be published under a Creative Commons licence.