Is the “request copy” button good for OA?
An oft-quoted statistic is that 63% of journals endorse immediate open access (OA) by allowing the Accepted Manuscrips of peer-reviewed papers to be deposited in institutional repositories. This of course leaves 37% for which an embargo is required, ranging from a few months to, well, infinity. To help with access to this embargoed material, many repositories make use of an “eprint request” button, allowing visitors to contact the main author from that institution and ask to be sent a copy by email. Essentially, this cuts out the time the user would otherwise need to invest in browsing the institution’s website, looking for the email address of the author, and generating a request themselves. It also makes use of the fact that, in most copyright assignment agreements, authors are allowed to distribute personal copies, for non-commercial reasons, to their peers and colleagues. There are some that argue the technicalities of this, but I won’t go into that here.
Instead, my reason for posting on this topic is to raise the possibility that, far from being a useful partner to OA, could the eprint request button, in fact, end up being an unforseen hindrance? What! (I hear you all shout). How can something that provides more efficient access to otherwise closed-access research be a problem? Well, the devil is in the detail… quite literally…
You see, with the eprint request button comes information. At the very least, that information will be a name, an email address (and thus an indication of the requestor’s institution), and a reason for wanting a copy of the paper in question. All of a sudden, rather than just being told his or her paper has been downloaded 50 times in the last month, the academic is finding out who wants to read their work and why. This is really valuable stuff, especially as we (in the UK at least) move increasingly towards having to justify the impact of our research. Which is going to go down better when it comes to the REF… that this or that paper, or this or that research theme, seems to have attracted a lot of traffic in our repository; or that we know for a fact that our research in such and such a field was requested by someone from local government to help develop a new policy. You see where I’m going with this…
Basically, I wonder whether there might come a point where academics prefer the closed-access/eprint-request option because it ends up telling them what they want to know, and, perhaps more importantly, what their institution needs to collect for research assessment purposes. I’m not saying this is something that will definitely happen, nor perhaps should I be planting the idea in the minds of OU academics who read this blog (!)… but, I do think it is an important issue to raise for discussion. So, feel free, go ahead and comment!
February 3rd, 2010 at 4:33 pm
We could do user sampling, like commercial websites. We can ask 1% of downloaders to identify themselves by occasionally adding a form to the splash page.
Mind you, I think this is a temporary itch. Once you have seen a dozen requests for your stuff, then you “get it”.
February 3rd, 2010 at 6:02 pm
A good point to raise, Colin. My perspective is that when requesters have to identify themselves, this raises privacy issues. There are also many potential fairness issues when researchers choose who can view the work. The sooner we move to full, immediate, and preferably libre OA, the better, in my view.
Facilitating voluntary and welcome author / reader communication is another matter, such as through the auithor contact options and commenting features in OJS. Another means of connecting authors with users of their information would be track-forward to citations and trackbacks.
February 3rd, 2010 at 6:10 pm
Our experience is that authors get “tired” of replying to copy requests, especially when requests are very frequent.
The consequence is that some start not replying at all, and others ask to change to Open Access articles/papers/theses there were in closed/embargoed access. We had more than 20 of those requests just on the last year…
February 3rd, 2010 at 7:01 pm
I am grateful that I am not the only one fighting against the request button or discussing its disadvantages. You can read my contributions at
http://archiv.twoday.net/search?q=button
(i) The button is making scholars to beggars.
(ii) There is evidence that asking directly the author is more successful. Having used the button without success a following direct mail would diminish the chances.
(iii) My own experiences with the button (for test purposes) have shown that the success rate is low.
(iv) DON’T LET THE AUTHOR DECIDE IF DARK OR OPEN. If there are no cogent copyright reasons DON’T ACCEPT dark deposits.
(v) DON’T ACCEPT DARK DEPOSITS WITHOUT FIXED EMBARGO TERM.
February 3rd, 2010 at 7:29 pm
Ad Dr. Graf (who is certainly not discussing the point Colin raised, but rather rising his habitual hobby-horse (that the button is “evil”!):
(i) Requesting an eprint by button makes scholars beggars?
(ii) Requesting an eprint by email is less beggarly?
(iii) Stay tuned for a forthcoming paper with objective data on the success rate of the button.
(iv) The beggar (without a mandate) is the repository, and beggars can’t be choosers.
(v) Dr. Graf seems to be systematically incapable of understanding that the purpose of the button is to maximize deposits and deposit mandates, not to minimize them.
February 3rd, 2010 at 7:33 pm
At Université du Québec à Montréal’s Archipel repository, we decided not to ask requesters to give the reasons of their request (there is no field to that effect). They are simply informed that by clicking the Request Copy button, they certify that the copy is to be used for research, study, criticism or news reporting (purposes falling under fair dealing provisions of the copyright Law, not only in Canada but in other Commonwealth jurisdictions). The same information appears in the request e-mail received by the author, i.e. that the request is for a one of these purposes.
The goal was to dissipate any ambiguity (or fear) about the legal character of the requests, especially when requesters wouldn’t t state any reason. We wanted also to decrease the risk of authors ignoring requests because they are not really sure about their legal status.
I thought at first that we would lose something, not as much as it can be satisfactory to read “I find your work interesting” or “Your work is useful for this or that…” but because a request could be a way to start a discussion, which could ultimately lead to collaboration.
I personally enjoyed such discussions twice recently, when I emailed to authors to request copies of their papers (I did it manually, as the articles were not available on a web site or in a repository, and it was not difficult to find their email addresses). But I’m not sure I would have got much attention if these authors have had only to click on a hyperlink to send me the copy.
As to the “research assessment” issue, one has to ponder, on the one hand, the advantage of knowing who uses your work and for what purpose, and on the other hand, the possible decrease in the number of downloads (a metric that can also be used for assessment), as it is may well be that a fraction of the potential users (or readers) will turn towards more readily available papers.
February 3rd, 2010 at 9:44 pm
The point of Colin and the privacy issue is perfectly clear and I support these doubts.
It’s Prof. Harnads Hobbyhorse to propagate one holy way and denouce all others. I am not convinced of the button and my personal preferences and psychology are harnad’s obsession. I would like a more topic-centered argumentation.
Against the often repeated Harnadian Orthodoxy I would like to underline that
(i) dark deposits without embargo term are worthless
(ii) scholars prefer to deposit (and quote) publisher’s PDFs with the result that these are not free until the end of time (or 70 years pma, which comes first)
(iii) archives should carefully choose the conditions of deposited materials – nothing other is true in the IR (= electronic archives) case. Archivists don’t like deposits without clear embargos.
February 3rd, 2010 at 10:30 pm
PRAGMATICS VS BLIND IDEOLOGY
“(i) dark deposits without embargo term are worthless”
Tell that to the would-be users who (without “dark deposit” and the button) would not be able to access the paper at all. And to the authors whose paper swould lose that usage and impact.
“(ii) scholars prefer to deposit (and quote) publisher’s PDFs with the result that these are not free until the end of time (or 70 years pma, which comes first)”
Tell those preferences to the above users and authors, when it is a choice between the author’s refereed final draft, today, or no access at all (or, if you are tilting against the button rather than the author’s refereed final draft: when it is a choice between the button, today, or no access at all).
“(iii) archives should carefully choose the conditions of deposited materials – nothing other is true in the IR (= electronic archives) case. Archivists don’t like deposits without clear embargos.”
We are not talking about archives’ “choices” here, nor about archivists’ likes and dislikes. We are talking about filling near-empty institutional repositories (archives), with as much of OA’s target content (refereed research articles) as is possible, today, by mandating deposit (because most authors are not depositing spontaneously, unmandated, for various reasons, including worries about copyright and embargoes).
So the rationale for Closed Access (“dark”) deposit and the button is the following: 63% of journals already endorse immediate Open Access self-archiving. That 63% of articles can already be mandated for immediate OA self-archiving, today. But what about the other 37%? Wait till the end of the embargo (or 70 years, whichever comes first)? Or mandate immediate deposit anyway, for them too, and allow Closed Access (“dark”) deposit, providing the button to tide over usage needs until deposit becomes universal, OA prevails, and mandates die their natural and well-deserved death.
Klaus Graf is so angry with embargoes that he simply cannot follow the reasoning behind this practical strategy. He sees only that “dark deposit” is evil. He cannot see the light this provides for would-be users who need access today, and authors who want their findings to be used today, and not only after the final battle against the Dark Side has been won.
February 4th, 2010 at 9:33 am
Just to clarify, I am not against use of the request button. If I was, I would not have it as a feature in our repository here at the OU. My intention in writing this post was simply to raise a question: do people think that academics, or their institutions, might prefer to receive regular information on who is accessing their work and why? If so, my worry (as an OA advocate) is that we may see more depositors choosing closed-access just to have the request button added.
Clearly, one opinion is that academics get fed up with dealing with requests, and of course that may well be the case. Indeed, here at the OU we too have had our fair share of academics adding the full text retrospectively for this very reason, which is great for OA. However, anecdotally, I have also had conversations where it is clear that certain individuals really do like the information provided by the request copy button. This, combined with the fact that one only has to make a single click of the mouse to release the file from the repsitory to the requestor (i.e. it is not that burdensome), made me wonder if we are setting ourselves up for a fall.
February 4th, 2010 at 10:02 am
I think you are missing the point of the “button” here. The correct way to phrase a mandate is that the “button” is solely a back-up to ONLY be used when there is a publisher embargo. I think the OU is correct in keeping the information requested to a minimum – a statement that the requester is requesting it for purposes within the legal allowances for authors to distribute it, and a contact address. anyone rally concerned with privacy can use short-term single-use email addresses to make their requests. personally, I find that I want other schlars to know when I am reading their work because it opens up the possibility of interaction, the next stage beyond just reading people’s work is to discuss it in more depth with them, but this is a sideline compared to the mass access that is the purpose of OA. The solution to the original question is for mandates to make clear what the purpose of the button is, and to not allow depositors to place a barrier to access where one is not necessary.
February 4th, 2010 at 10:32 am
Andrew: Thanks for your comments. However, we do not have a mandate underpinning our repository here at the OU. Therefore, staff are free to either include their email address (and thus the button) or not. I must admit, our policy when reviewing/verifying items before they go live is currently to add an email address (for giveaway works – journal papers, conference papers etc.) if one has not been supplied… perhaps we will need to review that.
But even so, without a mandate, depositors can choose between open and closed access. If it occurs to them that the closed-access option gives them what they want, then perhaps they will actively choose that option.
Essentially, what you are saying is to not provide a choice (i.e. implement a mandate telling people exactly what they must and must not do when depositing their work in the repository). I confess I’m a fence-sitter when it comes to mandates, so I’m not going to agree or disagree with them here… but for the majority of institutions that don’t have one, I think the issue I’ve raised here could come to the fore. I’m sure Stevan will argue this is all the more reason to press on and introduce a mandate!
February 4th, 2010 at 11:08 am
I agree with the (almost certainly correct) opinion youascribe to Stevan here: mandates are necessary. This supposition (and at this point, without evidence, it is merely a supposition) if true would re-inforce the need for a clear maximally efficient mandate.
February 4th, 2010 at 7:39 pm
(i) I am not sure whether mandates are necessary. In Germany we cannot have mandates according legal discussions. Japan has success with IR without one mandate in the whole country. We should try several ways and not follow Harnadian Orthodxy.
(ii) It is absolutely WRONG to give scholars the choice dark or open deposit if the open deposit is possible because of copyright reasons. If scholars want deposit dark if open is possible DON’T ACCEPT THIS. Without a mandate you can set your rules.
(ii) IR managers who are thinking that the button is’nt evil should make their statistics of the use of the buttons transparent. Anecdotical evidence is worthless.
(iii) Nobody has discussed my experience that authors like to deposit publisher’s PDFs as dark deposits instead of (mostly free) final drafts.
(iv) There is evidence that the button is a damage for open access because scholars would make their deposits more open if it doesn’t exist.