Just read this news story … Scientific Publishers Offer Solution to White House’s Public Access Mandate – ScienceInsider
It reports on an effort by various scientific publishers to create something they call “CHORUS” which stands for “Clearinghouse for the Open Research of the United States.” They claim this will be used to meet the guidelines issued by the White House OSTP for making papers for which the work was supported by federal grants available for free within 12 months of being published.
This appears to be an attempt to kill databases like Pubmed Central which is where such freely available publications now are archived. I am very skeptical of the claims made by publishers that papers that are supposed to be freely available will in fact be made freely available on their own websites. Why you may ask am I skeptical of this? I suggest you read my prior posts on how Nature Publishing Group continuously failed to fulfill their promises to make genome papers freely available on their website.
See for example:
- Calling on Nature Publishing Group to return all money received for genome papers and article corrections
- A Solution to Nature Publishing Group’s Inability to Keep Free Papers Free: Deposit them in Pubmed Central
- Please help keep the pressure on Nature Publishing Group to restore free access to genome papers #opengate
- Today is a day to be annoyed with Nature (Publishing Group that is) #NatureFail
- The Tree of Life: Nature’s publishing machine really wants you to pay for stuff even if it is supposed to be free.
Saw this Tweet
We just published the story yesterday about the 700.000 year old horse that we sequenced. Check it out ! http://t.co/jAym3HLAC0
— Bent Petersen (@bentpetersen) June 27, 2013
//platform.twitter.com/widgets.js Seemed potentially really interesting. Read the story and got pointed to a new Nature paper on the ancient horse genome. I guess not so surprisingly, despite the fact that they report a new genome sequence, it is not openly available. We really cannot trust Nature on this can we? They could say “Well, this is a draft genome, and we did not mean to apply our policy to draft genomes.” Well, that would be weird since, well, they have applied this to draft genomes before. And then I decided to search for other examples … and in about ten minutes I found a few. See
And this paper too? http://t.co/rCC1A4AKj0 – promises from Nature Publishing Group http://t.co/vlDAd7KFOR not being met #CHORUS
— Jonathan Eisen (@phylogenomics) June 28, 2013
Correction: Here is another non open genome paper http://t.co/X6TgWBKR9u promises from Nature http://t.co/vlDAd7KFOR not being met #CHORUS
— Jonathan Eisen (@phylogenomics) June 28, 2013
20 thoughts on “I am highly skeptical of the CHORUS system proposed by scientific publishers as an end run around PubMed Central”
Right. As I noted in a comment on the Scholarly Kichen's post (we'll see if it makes it past moderation):
This does look potentially positive, though I think there is a big trust gap to be bridged before researchers, librarians and indeed the government will be happy entrusting all this to the very publishers who up till now have made themselves roadblocks in the path of all such initiatives.
Not so sure what is potentially positive about this … but I will ponder it again
Really? At face value, publishers seem to be offering to host the equivalent of Green OA themselves. If I believed wholeheartedly that they would do it, and do it well, then I would consider that a good thing for sure. But the “but” is a big one.
I don't see anything in this that seems particularly useful. The publishers are going to make stuff freely available on their websites right? And make some sort of searchable system? Is there anything else? Seems pretty minimal. And if they use this to justify killing PMC (as they are clearly arguing to do), that would be a bad thing. Plus who knows what kind of ads, non-uniformity of format, download delays, and other crap will be implemented. And the lack of centralization will cripple broad, new, integrative uses of the literature. Seems like a lame idea to me. Hey – I am all for the publishers that innovate. For profit. Non profit. Whatever. This seems like a major step backwards not forwards.
I don't see anything in this that seems particularly useful. The publishers are going to make stuff freely available on their websites right? And make some sort of searchable system? Is there anything else? Seems pretty minimal.
That's my understanding. But that would be a good thing. At least, a better thing than the current scheme, which I need hardly remind you entails the publishers not making stuff freely available on their websites.
And if they use this to justify killing PMC (as they are clearly arguing to do), that would be a bad thing. Plus who knows what kind of ads, non-uniformity of format, download delays, and other crap will be implemented.
Well, exactly — that's the big “but”.
And the lack of centralization will cripple broad, new, integrative uses of the literature.
The ideas is that the open APIs will enable this kind of thing. Unfortunately the problem with that is that it requires us to trust that publishers will both (A) act with good faith, and (B) be competent at providing these open APIs. And their track record on both counts is extremely poor.
That's the reason I won't be supporting CHORUS: not because it it's a good proposal (I think it is a good proposal) but because I don't believe the publishers either either willing or able to follow through on it. It'll all be adware, crudware, different incompatible incompletely implemented subsets of the APIs, new kinds of barriers we've not thought of yet, and so forth.
They have every incentive NOT to act in good faith. They clearly don't like public access, which is why they've been fighting it tooth and nail. So now they're going to implement it? Their interests clearly are with making it as minimally functional and difficult to use as possible.
We'll pay a little money and pretend to be open in order to maintain the control we currently have. Who could be skeptical of that?
This whole thing reminds me of Microsoft's attempt about ten years ago to subvert Open Source by releasing code with restrictions that they called “Shared Source”
Do these groups really think that just by using some smily wording everyone will believe them?
Note – check out the original Nature promise to make genome papers free … here
Absolutely. What's more, the feigned motives of the CHORUS advocates are glaring. One is quoted in the Science Insider article saying “We're anxious for agencies not to spend their precious research funds on the OSTP directive.”. Oh, please. Whatever motivates publishers to propose CHORUS, spending their own money to save government funds ain't it.
Jonathan, I'm not going to comment on CHORUS. I would, however, like to address your reminder of NPG's issues with making genome papers freely accessible on nature.com. I see the point you are trying to make. I also acknowledge that this was a problem you've had to flag several times. Yes, we've had some technical problems, which we regret and we've apologized for and explained here and elsewhere. But we've never intentionally broken commitments. Given that our policies and procedures make 1000's of articles and accepted manuscripts publicly accessible in PMC, perhaps some 'carrot' to encourage that behaviour would be more productive than continuing to beat NPG with this particular stick?
NPG deposit open access articles into PMC, and we also archive accepted manuscripts on the author's behalf into PMC, free of charge. These articles and accepted manuscripts are also included in the open access subset on PMC, and the accepted manucripts licensed so that they are available for data- and text-mining.
Our self-archiving policy means that any author of a research article can self-archive the accepted manuscript in PMC or elsewhere 6 months after publication. And for CC licensed articles the author can of course also deposit in PMC. This includes the genome sequence articles published in Nature and Nature research journal, where no APC has been paid, but the CC license has been applied under our policy on genome sequence papers.
Hi, Grace. I think it's great that you're chipping on this.
Jonathan will have his own view, but from my perspective the key point here is that even though Nature Publishing Group is not a Bad Actor, and even though it would be widely considered one of the most competent of the established publishers, it's required (at least) five reminders from Jonathan alone to get it to do what it promised.
What this tells me is not that there's any bad intent on NPG's part, but that it just isn't capable of getting the job done. I'm as surprised as anyone by that, but evidently open access is not sufficiently high on NPG's agenda for it to be paying attention to its own OA promise.
And the question that raises is: why would we expect that to be any different for other established publishers? Especially those like Elsevier and Cambridge University presses with a history of abusive behaviour?
The core problem with CHORUS is that it only works if all the publishers involved are both well-intentioned and competent. And I'm afraid the evidence is that few if any of them fit both those criteria.
Heather Morrison stated the issue most pithily: public access needs public stewardship.
Mike, I don't accept your evalution of our capabilities or our commitment.
I've explained the problems we experienced in my responses to Jonathan's earlier blog posts, and apologized for the errors, I don't intend to go through that in detail here but will recap briefly.
For OA articles made OA under CC license on publication, we have not experienced this problem. We published over 2000 articles last year under CC licenses, and there are no issues there.
The issue is specific to these genome papers. It took us a while to get our metadata/XML tagging right for these genome sequence articles (many of which were made OA retrospectively some years after publication, when we introduced the policy). This exception to our processes made this difficult to implement correctly initially, and we have resolved it some time ago.
“Mike, I don't accept your evalution of our capabilities or our commitment.”
Well, I am a bit perplexed by that assertion. It's a matter of historical fact than NPG did, repeatedly, fail to make genome papers open access. If it wasn't because of either a failure to care or a failure to execute, what was it? I honestly don't see what other possibility there is.
Thanks for all the comments Grace. When I wrote this post I worried a bit about what you would think and probably should have commented about how NPG really seemed both committed to getting this correct and also that this was a technical glitch of sorts. I note – there is one thing I suggested when this was going on that I still think would be good – depositing the genome papers in PMC would be a way to guarantee that whatever NPG does, the papers were still available.
Thanks, Jonathan, I appreciate you thinking of that and saying so. We're not infallible or perfect, and we could have done better in this particular instance. I'll check in on whether we can do a bulk deposit to PMC and let you know. In the meantime, if authors of those CC licensed genome sequence papers wish to put the final version in PMC, they of course can and should feel free to do so (they are CC licensed, after all!)
Not only do they have a very strong incentive to act in bad faith, but the CHORUS proposal would eviscerate institutions that could punish them for bad faith, or in the case of PubMed Central, make their bad faith actions largely irrelevant.
I'm surprised that more people haven't pointed out that this sort of “self-regulation” idea is what every industry group has floated as an alternative to public regulation since 1880. It might kinda work for a little while, but the last 130 years is a smoking junkyard of counterexamples.
Jonathan, responding to your 27 June 2013 update to this blog post.
NPG's genome sequence policy applies “to those articles in Nature journals that are publishing the primary sequence of an organism's genome for the first time”. This has been the case since we introduced the policy in 2007, and is available on our website here: http://www.nature.com/authors/policies/license.html
With respect to the two Nature Genetics papers you specify, I've checked with the editors who say “NG applies CC licenses only to first reference genomes and community standards papers in accordance with Nature guidelines.”
These are not first reference genome papers, that is why the policy was not applied.
I'm checking with colleagues on the ancient horse genome and will let you know.
Revealing Dialogue on “CHORUS” with David Wojick, OSTI Consultant