Temptation (1946 film)
Temptation (1946 film) (Photo credit: Wikipedia)

In March 2012, executives within the US National Institutes of Health (NIH) and National Library of Medicine (NLM) began discussing with their counterparts at Wellcome Trust how PMC could secretly help eLife launch early.

On June 19, 2012, the PMC National Advisory Committee (NAC) met, where Mark Patterson of eLife, at David Lipman’s invitation, gave a presentation about eLife.

It was an unprecedented presentation — the first in the NAC’s history to focus solely on an unlaunched journal. At no time during that meeting was the NAC told that eLife and PMC were in the midst of abetting the launch of eLife. Patterson initiated the idea of PMC helping eLife, you might recall.

I wrote about much of this yesterday, sharing information gleaned from reams of printed emails, memos, and presentations generated by a Freedom of Information Act (FOIA) request I filed in November 2012.

Today, I’m going to explain finding something in those 500+ pages that I didn’t expect — evidence that Lipman and others, as well as executives at eLife, were also discussing PeerJ, with Lipman at one point considering bending the rules again for PeerJ, despite PeerJ not expressing any desire for special treatment. In fact, PeerJ asked pointedly for an explanation of eLife’s sweetheart deal once the secret deal’s results were on the market.

The email chain begins on June 12, 2012, at 2:45 p.m., about a week before the PMC NAC meeting. Lipman emails Binfield with the subject “congrats!!!”

Peter —

congrats on PeerJ. This is very exciting and I’d love to learn more about it. Let me know when would be a good time to connect by phone.

Best regards,

David

The two then arrange through some brief email back and forth to speak at 2 p.m. EDT on June 19th, and Lipman provides Binfield with his phone number. After that call, Binfield provides Lipman with a list of Advisory Board members PeerJ is signing. He also asks Lipman for more:

We discussed creating an Advisory Board (aiming for the biggest names), and I would appreciate it if you could give me your opinion on the attached invitation text. I also append the more generic invitation text that I have been using to invite ‘regular’ Ed Board members – again, any feedback you have on that doc would be welcomed.

So, here we have the head of the NCBI providing advice to a new publisher — this time, it’s about how to attract editorial board advisors — after reaching out spontaneously to that same publisher a week prior. This is reminiscent of Lipman giving eLife business advice at the June 2012 PMC National Advisory Committee (NAC) meeting. In both cases, his actions seem to go far beyond his remit.

Lipman’s spontaneous advice for PeerJ continues, through emails on July 3 (“This is really good Pete – this will keep the buzz going.”) and September 4:

Congratulations Pete – this is excellent. I’ve taken a look at the site and you’ve done a great job getting a strong and comprehensive group of scientists for the Editorial Board. I’ve cc’ed Chris Kelly who can help you on the steps needed to participate in PMC and PubMed.

Chris – Pete was the Editor at PLoS One and has moved to help start up peerj, a new journal that builds on some of the ideas behind plos one. Let me know if I need to be involved in all this.

Pete – we might want to touch base very briefly by phone or skype sometime soon as there are some related issues worth discussing.

It’s important to register Lipman’s spontaneous and ongoing enthusiasm for PeerJ, as it seems to color his judgment. Rather than being an unbiased manager of an index and repository of the scientific literature, Lipman is a cheerleader of certain initiatives, and shows tendencies to throw his weight behind splashy new approaches.

Chris Kelly, a production manager at PMC, then takes over communications for a time, cc’ing Lipman.

Binfield notes to Lipman in an email the same day:

David – happy to Skype anytime if you have related items you want to discuss.

I emailed Lipman and Binfield to ask about what “related items” they discussed. Binfield didn’t recall the discussion, and had nothing on his calendar showing any call occurred. Lipman’s reply is below (received February 4, 2013):

In my follow up with Binfield back in September, I spoke with him to learn a little more about their plans.  Related to that, I’d heard that they were developing their own software system for manuscript submission, review management, & online presentation and wanted to know more about that.

In this same September 4, 2012, email, Binfield notes to Kelly that PeerJ hasn’t launched yet. Kelly then begins to explain to Binfield the requirements for inclusion, in an email dated September 5, 2012:

Peter,

Thanks for clarifying. Approximately 30 articles published initially *should* be adequate for the NLM reviewers to assess the journal. Keep me posted as you get closer to the launch date.

This returns us to the confusing and inconsistent standards being promulgated by the NLM — they can’t seem to get this right. In one location, they state:

The journal must have a reasonable number of published articles in order for NLM to make a decision about its scientific quality.

In emails with publishers, they’ve stated:

A journal needs to be included in the NLM Collection to meet PMC’s scientific quality standard. This review can take place once [the title] has published 15 articles.

Now, we have another standard being promulgated — 30 articles.

Kelly plays by the rules he understands, even if the rules aren’t entirely consistent throughout the NLM or NCBI. He and Binfield talk on September 5, 2012.

On October 15, 2012, eLife content goes live on PMC. On October 17, 2012, Binfield sends Kelly this email (Lipman is not cc’d):

Chris

I have a question about the PMC approval process. When we discussed, it seemed that the process couldn’t start until a journal launched, and even then it would take approx. 2-3 months to complete (and only once a critical mass of articles has been published).

However, I noticed that eLife now has articles up on PMC (without having launched yet, and with only 8 articles published). I’m not trying to be difficult, but is there a similar fast track process we can be put through to get indexed sooner than 2-3 months post launch?

Thanks

Peter

Binfield told me via email that his inquiry was an honest effort to see why one journal had gotten in so fast:

When eLife appeared on PMC (and before your posts came out) I was surprised, as I had been told that a journal had to be publishing, had to have published a certain critical mass, and had to go through a content and technical evaluation which could take several months to complete before being accepted into PMC. Therefore, I emailed them to find out how eLife had seemingly been evaluated and accepted into PMC when clearly they hadn’t been up and running for that ‘several months’ time period (to be quite honest, the aspect of PMC acting as the publisher of the articles, which you have been drilling into, had not occurred to me – it was simply the apparent speed with which they had been evaluated and accepted, when they weren’t yet formally publishing). Therefore, I was asking if our evaluation could be similarly prompt, so that when (and only when) we started publishing, then our content would start appearing in PMC sooner than the normal several months delay.

Binfield was as perplexed as anyone, apparently.

On the same day as Binfield’s email inquiry (October 17), Kelly emails Lipman (note, this was prior to my post; it appears PMC had already received complaints about its actions with eLife from others):

More eLife blowback.

I spoke with Pete on the phone at the beginning of September, gave him some pointers on what the NLM selectors want to see for new journals — details are included in my note to him below. . .

Chris

Kelly is referring to that day’s email inquiry from Binfield.

That same night, at 9:58 p.m. on October 17, 2012, Lipman replies:

thanks Chris.

It’s important to separate the issues here. I don’t understand the “launch” issue. If a journal is getting articles from authors , reviewing them, copyediting, tagging, etc, then they are publishing. Whether they show them on their website, I don’t care.

So the only issue is LO’s criteria. Do they really want 30 articles? Is that what we’re telling publishers?

I will email Pete after I hear from you.

“LO” is Library Operations.

There are two aspects to this email:

  1. Lipman doesn’t know the criteria the NLM is using, yet claims in other communications that the usual process was followed. Is there a usual process?
  2. Lipman considers non-public (i.e., editorial) review and approval of articles to be “publishing.”

Interestingly, this inexplicable idea — that completing the editorial process without making the information public equates with “publishing” — occurs later in another NLM communication. In an email sent to a librarian on December 7, 2012, in response to an inquiry citing my post on PMC and eLife, Joyce Backus of the NLM echoes this mysterious state of private publication:

Our policy on inclusion of a journal in PMC is governed by two factors: (1) A determination of the scientific quality of the journal and (2) conformance with technical requirements of submitted files such as the XML format specification. . . . As soon as any journal meets these standards, PMC provides access to the journal’s articles in accordance with a timeline agreed upon with the publisher. It does not matter to NLM or PMC how publishers disseminate their published articles. It’s irrelevant to us whether their content is available via subscription or OA, on no website or many websites, or in print. Our primary consideration is the quality of the scientific content and the editorial process that results in the published articles.

This statement, which I have never seen outside of this email acquired via my FOIA request, is a contradiction of stated PMC policies and is not reflected in PMC practices in general, based on reports from the field. It also seems to be a revisionist history provoked by “eLife blowback” — that is, PMC suddenly didn’t care if the journal was actually publishing, and may include the content in PMC even if it’s not, as long as the organization has a good reputation and a decent XML DTD.

This obfuscation creates a paradox — published articles have to be evaluated before they can be added to PMC per published policy and documented communications, but the articles don’t need to be published to be evaluated. It’s a particularly convenient paradox — PMC asserts it is not a publisher; by defining “publication” as “editorial review and approval,” then using US government infrastructure to make information public for the first time, even to the extent of providing the primary output for eLife content for two months, PMC is, via this clever redefinition of reality, not publishing.

So we have PMC stating it’s not a publisher, and moving the line of what a publisher is so that it only involves the editorial review process, not the act of making information public. This is pure evasion.

The exchange between Lipman and colleagues discussing PeerJ continues on October 18, 2012, with Kelly confirming:

David,

Yes, 30 published articles is the standard line for most new journals; but there are always exceptions.

When I mentioned this as a typical number suitable for review, Pete didn’t seem concerned — sounds like Peer J will have much more than 30 once the journal goes live.

Chris

Lipman weighs in, with a voice that is very much like a businessman wanting to make an important acquisition:

Chris & Ed –

I have no problem with getting articles from peerj prior to them having live articles on their site. In this case I think Binfield has a track record — but it’s reasonable to get more than 8.

Peerj could end up being a very important journal for us if they can catch some of the PLoS One types of authors.

djl

This email is worth parsing. Here is the Director of the NCBI inventing rules again for a new, high-profile publishing initiative. But the more troubling and revealing part comes when Lipman states, “Peerj could end up being a very important journal for us . . . .” Who could the “us” possibly be in this sentence? Why would it matter to PMC which journal an article is published in? It’s a repository, after all. I can’t find a rational explanation for this wording.

Fundamentally, Lipman is making a competitive statement, a proprietor’s statement, and has a whiff of “publishing mogul” to it. He may think of himself in this way, to some extent. I saw glimpses of this as far back as 1998, when I visited the NCBI. As we’ve written about before, PMC is competing with publishers, by design. This has negative financial impacts on publishers — suppressing usage data and stealing advertising inventory. Now we have even more evidence that the Director of the NCBI thinks competitively, thinks of PMC as a publisher competing for content, and is willing to bend rules to gain what he considers to be an advantage.

This surprising set of emails reveals that just days after publishing eLife content on PMC as part of the launch of that journal — and making US government-funded infrastructure available to a new multinational journal publisher — Lipman and the NLM were ready to bend the rules again for PeerJ. Remember, journals are supposed to be vetted by a clearly stated and (actually) published process. The head of NCBI seems to have become comfortable short-circuiting this process if something struck his fancy.

For its part, PeerJ wanted nothing to do with special treatment from PMC, as Binfield wrote to me:

Even if PMC had offered to “eLife” us and host our articles early (something which was never discussed), we would *not* have allowed that to happen (i.e. we would not have allowed them to host our articles before we had published them, as that is simply not a good idea for all sorts of reasons).

eLife Enters the PeerJ Discussion
The emails also show that there is more of a cabal here than I’d imagined, as demonstrated by this email from Mark Patterson of eLife to Ed Sequeira on October 18, 2012 — contemporaneous with many of the emails above. Here, we have Patterson and Sequeira working on language again, in an effort to give PMC’s publication of eLife articles a benign appearance:

Hi Ed

Just a quick heads up – an ex-colleague (now at PeerJ) just got in touch to ask how we managed to get eLife content into PMC so fast. I was planning to reply as follows, but want to make sure that this way of describing things is OK with you. I’m just planning to keep it brief.

Please let me know today if you can.

Cheers

Mark

“We contacted PMC a few months ago when it was becoming clear that we’d have some pretty stellar content to share before our website would be ready and launching the journal on PMC looked like a good option. We also considered making the content available via arXiv, and authors can deposit there too of course..”

Sequeira responded at the end of the day:

Pete Binfield wrote to us too. We’ve also had complaints from a couple of others already. It would be easier to chat than address this through email.

The two then collaborate on what Patterson should write back to Binfield. The last draft in the FOIA documents reads:

Thanks again, Ed

Here’s a slight redraft of my message to Pete

We contacted PMC a few months ago when it was becoming clear that we’d have some pretty stellar content ready to publish (XML, PDF, etc) before our website would be ready. So we ended up launching the journal by listing the articles on our wordpress site and making the full text available via PMC. One the HighWire site is ready in a few weeks, we’ll repost the same content on our site, along with some new content. We also considered making the content available via arXiv, and authors can deposit there too of course. We felt it was important to get these articles out there as soon as possible.

Sequeira reponds:

Mark

I’d just add that, given that you had editorially ready content, you’ve been working with PMC for a couple of months getting your XML up to par.

Ed

I don’t believe Binfield knew that Sequeira had a hand in Patterson’s response to him. Here is obvious collusion between PMC and eLife on how to present their deal to the public and to other publishers.

Conclusion
The inquiries from PeerJ should have been answered normally by staff, and shouldn’t even have been necessary, frankly — the eLife situation should never have occurred. Instead, the same activism at the top of the NCBI that led to the publication support for eLife nearly took over communications with PeerJ. In addition, the inappropriate relationship between eLife and the NLM is clear once again — these two unrelated organizations should not have been coordinating how they would respond to a third organization. Finally, we have evidence of how the Director of NCBI (and the Executive Secretary of the PMC NAC) thinks — as a competitive publisher, not as a US government employee managing a non-competitive index or repository.

When the organization running the systems to objectively and dispassionately catalog the literature has been polluted by inappropriate activism and conflicts of interest, where do we turn for objective and fair treatment?

(Note: Tomorrow, I’ll explore the conflicts of interest on the PMC NAC, and new evidence showing how mismanaged these have been.)

Enhanced by Zemanta
Kent Anderson

Kent Anderson

Kent Anderson is the CEO of RedLink and RedLink Network, a past-President of SSP, and the founder of the Scholarly Kitchen. He has worked as Publisher at AAAS/Science, CEO/Publisher of JBJS, Inc., a publishing executive at the Massachusetts Medical Society, Publishing Director of the New England Journal of Medicine, and Director of Medical Journals at the American Academy of Pediatrics. Opinions on social media or blogs are his own.

Discussion

19 Thoughts on "Why Were PubMed Central and eLife Discussing PeerJ?"

Kent, who is responsible for making the rules here? It certainly seems like Lipman thinks he can “change” policy as he wants but is that really the case? It seems that it would be harder to officially change the policies of a government agency than this case has shown.

I think PMC can set its own rules. However, when there are different rules for different people, or inconsistent rules, or imagined rules, or fudged rules — well, then I’m at a loss. I think the feeling of unfairness and bias emanates from the capriciousness of PMC’s actions.

I think if PMC had to recast its own rules to match the practices outlined here, there would be a hue and cry — a lot of objections. I think that’s why they thought they had to keep some things secret.

This is just what I meant when I suggested the problem was vague PMC procedures on another PMC post comment. They may call it flexibility and claim that the capricious variations are based on informed judgements about the publications in question. That eLife is sponsored by major funders gives this argument some leverage. Absent legal requirements federal agencies have broad discretion, a principle the courts have repeatedly maintained. So while what they did may be grossly unfair it may not be unethical.

If the “choosing” were the only issue, that would be one thing. But to knowingly be providing one publisher with competitive advantages while shutting others out (“Seriously, we’d like to play down the idea that there’s anything special about what you’re doing in PMC. Chris Kelly, who manages PMC production, is constantly fending off publishers — generally new, small journals — who push to get into PMC as early as possible because it raises their credibility and, often, their chances of survival. . . . And it’s not just the little guys. We’d had people like Peter Ashman at BMJ complain about inequitable treatment in a slightly different context.”) as a government agency crosses ethical, professional, and possibly some legal lines.

Keeping this secret from their own advisory committee, even when discussing how eLife would be hosted, is another ethical lapse.

Imagine this were a big new journal jointly launched by Elsevier, Wolters-Kluwer, and Springer, and it had gotten this level of “cooperation” from US government employees. The issues are the same, I believe. And they’re significant.

Again, if the record were limited to PMC making an objective, uncompromised choice to index eLife early, your argument might hold water. But to provide primary publishing support? To operationalize conflicts of interest? To try to conceal the collusion instead of being transparent? It’s not just “latitude” we’re dealing with here.

Federal agencies make these sorts of choices all the time. DOE gives one solar manufacturer a half billion dollar loan guarantee while turning others down. Nor do they have to explain why except to say it is their considered judgement. You would have a better case if you made this a policy issue rather than personal attacks. Demand to know what the policy is.

That’s precisely the point — they have a published process and a public position in the market, but they deviated significantly from both of these for eLife. If another agency were to award a contract to a firm on their advisory committee, while not telling the advisory committee it had been awarded during a presentation from that company’s representatives about their future contracts, keep it secret within the industry, and deviate from published and stated policies and approaches the entire time (not even requiring an application, it seems), that would go far beyond “choices.” I mean, really, this is so clearly beyond the policy realm it’s not even funny.

You left out a crucial sentence in your email quote. It is “I’m comfortable justifying what we’re doing with eLife, ….”

This suggests that PMC thinks their actions are legitimate. The question is why? That is what I am trying to figure out.

Kent, You are doing a marvelous job on this. Now, what can be done? Is the STM Assoc looking into this?

I’m starting to wonder whether the publishers who did go through the prescribed steps have standing for a class-action lawsuit.

Dear publishers,

If you want to win back the hearts and minds of alienated academics, then suing PMC for making biomedical research freely available is not the way to do it. If you initiated such a suit, and if you somehow contrived to win, it would be a PR own-goal of the most spectacular kind.

Has the RWA debacle taught publishers nothing?

Remember, Mike, this isn’t about OA. Many OA publishers and journals were mistreated by PMC’s apparent favoritism toward Wellcome and eLife. A class action suit could easily include many OA publishers and journals.

I take your point that this isn’t fundamentally about OA. But it will inevitably be perceived that way, not only because of your own well-known position, but because what PMC is alleged to have actually done in this case is further the cause of an OA flagship. So I think my assessment of the PR implications is nevertheless right.

What PMC is alleged to have done is to provide one organization/publisher with repeated and inappropriate assistance in an unfair manner, with abundant conflicts of interest, and a high degree of secrecy, even from its own advisory committee. Do you really not see that these issue are about fairness, the appropriate management of a government agency, cronyism, abuse of authority?

As you say, this is fundamentally not about OA.

But, if you’re a hammer, I guess you view everything as a nail.

I’m not saying anything about the issue itself. I am saying that the perception will be that you are going after this because of your well-known anti-OA agenda.

Well, now that you know the real issues, I hope that if things do heat up, you’ll lend your voice in an effort to restore fairness and objectivity.

The “gain” would be to recover damages to both reputation and business because eLife gained a competitive advantage and other journals’ endorsements by PubMed listing were unfairly delayed. I’ll leave the “against whom” to the lawyers — possibly both the entities and individuals named in these posts.

Comments are closed.