Clouds (Photo credit: Wikipedia)

PubMed Central continues its efforts to impose a uniform repository experience on published content deposited there.

One of the most obvious attempts to put the repository version ahead of the publisher version has been PMC’s search results interface, which provides a link to the PMC version of content within the list, but suppresses the publisher’s version into the actual abstract view. This has resulted in more usage of the PMC version and less usage of publisher’s versions, resulting in a repository experience for more users rather than a multi-publisher experience. PMC is trying to build the PMC brand to obscure the brands of publishers depositing content into PMC.

In late December, the team at the NCBI, which runs PMC, rolled out a “PubReader” view of deposited content, a view that while technically respectable, effectively strips the publisher’s brand from the content, leaving it only in the most token way possible — as a one-time, vanilla statement of fact, which has no visual resonance or brand power.

Because PMC wants to push its brand, the NCBI even created a snazzy page touting PubReader.

While their marketing of PubReader uses all sorts of glitz, their presentation of journal content goes the other direction, removing branding elements from published content and replacing these with a homogenous PMC experience. Here are four examples from completely different journals:

In each case, the name of the journal is presented on a field of stylized clouds — a visual metaphor for coming from the cloud, of ascending to heaven, or of coming from heaven like manna? No color, font, design, or brand elements from the source material exists. The flare of JCI is gone. The red of JBJS is gone. The green of Pediatrics is gone. The blue of PLoS Pathogens is gone. It’s all transmuted into the PMC cloud and branding.

Readers depend on branding so that they can know the source and provenance of content. Sources are not equivalent. They are not substitutable. Brand promises vary. Readers have fairly sophisticated knowledge of the brands in their fields. But readers also include journalists and the public in the case of PMC, people who often do not have a sophisticated awareness of journal branding. It is tripping journalists up. It is also confusing across the board, conferring equivalency where there is none.

I was not aware PMC was rolling out a new content presentation format for my journal’s deposits. How many publishers were? I’m guessing awareness of PubReader is low. PMC seems to have no sense of responsibility to the publishers they work with, treating our content as a commodity they can use however they see fit, even to our detriment.

PubReader is just another example of how PMC is putting its brand ahead of the brands of the publishers who deposit content there. The competition with publishers continues.

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.


29 Thoughts on "PubReader — Obscuring Journal Branding for the Sake of Repository Branding"

“… resulting in a repository experience for more users rather than a multi-publisher experience.”

Could this be due to wanting to give readers the free-to-access version?

Or are the publisher versions free to access in all cases?

What does obscuring or diminishing branding have to do with the price of content?

I’m asking you whether it’s the case that the journals’ own (branded) copy of the content is paywalled while PMC’s (unbranded) copy is open access. I don’t know whether this is generally the case. It seems to be so for the JBJS article you linked — when I follow the DOI link to I hit a paywall — but I might be missing something.

The question here isn’t what goes on at publishers’ sites, but what goes on at PMC. Why strip away journal branding? It is an important signal to readers, especially in a repository setting.

I understand your article asks broader questions about why PubMed Central’s own view of articles does not include journal branding. My question was in response to a specific and more narrow part of your concern — this part:

“One of the most obvious attempts to put the repository version ahead of the publisher version has been PMC’s search results interface, which provides a link to the PMC version of content within the list, but suppresses the publisher’s version into the actual abstract view. This has resulted in more usage of the PMC version and less usage of publisher’s versions, resulting in a repository experience for more users rather than a multi-publisher experience.”

So taking it as a given that the formatting of the PMC-internal version of a paper is the way it is, I am proposing that the reason for PubMed Central’s emphasising their own version ahead of the publishers is that it wants to point users to the version that they will be able to read rather than the version that is behind a paywall. That seems eminently reasonable to me.

(I will respond to your broader question in a separate comment.)

No the phenomena occurs even if the journal version of the article is free/open access.

Note that the search-result display issue is potentially more of a problem for OA journals, which might depend more heavily on advertising revenue that would be lost by views occurring at PMC rather than the journal site.

On the main question … you write that PMC’s “presentation of journal content goes the other direction, removing branding elements from published content”. I doubt that they are removing journal branding, merely not adding it. The on-site versions of papers are surely generated from the NLM-formatted XML, which has no information about cosmetics at all — only semantics. In other words, it’s not necessary to assume malice, or indeed any deliberate motive, for the way things are presented. It’s more likely a simple consequence of what data they have available.

“Not adding” a vital element of identification and differentiation is essentially “removing.” Imagine if they did “not add” the abstract or the author affiliations. Would you consider that “not adding” or “removing”? Just because you don’t like or respect brands doesn’t mean they aren’t important.

It seems like PubMed is attempting to provide (though I will not comment on whether it is succeeding in achieving) a uniform, high-quality user experience for readers of article content. I can think of no benefit to users of having to access articles in a variety of formats and styles on different publisher platforms which vary in quality of interface design. As Mike Taylor suggests above, this seems more like an attempt at providing a good user experience using the information (XML files) at hand, rather than a concerted attempt by PubMed to obscure the branding of individual journals.

This isn’t about article format, but about article source branding. Why obscure the source by not bringing it across? Brands are more important than ever to help readers differentiate content at a glance. People skim, follow information trails. Why make those information trails harder to follow?

Technically the source of the article is the authors, not the journal. And the information trail is in the references, not which journal the paper appeared in.

I concur.

I think this also points to whom PMC considers their primary customer: the reader. The undercurrent I get from all of Kent’s posts on various PMC-related topics is that I think he would prefer for the journals to be the primary customer.

(I don’t have the self discipline not to comment on the PubReader format, though. I think emulating a two-page spread is the most base form of skeuomorphism. And why emulate octavo format, which isn’t particularly common in scientific publishing? It make it look like a humanities journal [which might actually be part of Kent’s point]. And why doesn’t my mouse scroll wheel work to go to the next page?)

If you believe the reader is the primary customer, which I also do, then you have to acknowledge that branding is a very important way that readers initially assess published material’s quality and relevance. I am actually arguing that PMC is doing a disservice to readers by diminishing brand signals to the point of making them a non-factor.

I don’t think it’s impossible that PMC considers evaluating work by what journal it’s in is a bad habit (as so I) and that they are deliberately trying to lead people away from the misleading short-cut. (But I still think it’s more likely that they merely haven’t made any particular effort to reproduce branding.)

I don’t think it’s impossible, but it would also be severely misguided. Brands are useful shortcuts for busy scientists.

I can’t agree that “brands are useful shortcuts”. I agree that they are seductive. But any scientist who is too busy to evaluate a paper on its content instead of its brand is too busy to do science. From the perspective of a research, losing this branding information is probably a net win, or perhaps neutral. I can certainly see that from the publisher’s perspective it’s a loss. And I guess I agree with Joel that this whole issue comes down to who PMC sees as their customer. (As a researcher, I am glad they seem to consider it to be researchers!)

Whether you agree or not is moot. The reality is that they are, and they are becoming more important. In fact, I could have argued that PMC is shooting itself in the foot as a repository by obscuring brands because that makes it LESS useful to readers.

I personally don’t think PMC has much insight into reader behavior. I think this was a quick technical demonstration, without much craft on the presentation side. It reveals a tin ear for branding, for readers’ needs, and for usability, ultimately.

Because I have experience inquiring about this with hundreds of researchers and practitioners in many fields and disciplines over many years — both as a publisher and as an invited strategic advisor — in addition to having worked with some of the finest minds in branding and design over my career — both in B2C and B2B. I’ve helped to manage major scientific brands. I’ve reviewed dozens of reports on the topic, all based on rigorous field work. I’ve seen its power firsthand, and used it effectively dozens of times. If you think your personal experience can measure up to that, I’d be interested to know why.

You’re asserting something that is demonstrably false. So, I was being nice. Your opinion isn’t just moot, it’s wrong.

You give cogent reasons why you can speak for publishers. That’s not what I am trying to do. I am a researcher, and my interests are different from those of publishers. The evidence is that PubMed Central wants to serve researchers.

You really are missing the point. I have tons of experience seeing brands through the eyes of researchers and practitioners in a number of scientific disciplines. I’m not speaking for publishers. I’m trying to share my experience. You, apparently, aren’t open to being educated about these matters. Enjoy your bliss!

So just to be completely clear: your point is that you (with your extensive publishing experience) understand what I (a researcher) need, better than I do?

I understand what the majority of scientists and researchers want and expect in branding better than you do, yes.

It is not a new concept at all for third parties devoted to analyzing behavior or phenomena to end up understanding the principles and drivers affecting those behaviors or phenomena better than those involved directly. That’s the difference between observations and opinions, after all.

I don’t know that I can speak for Virginia or Mike, but I suspect that the point we’re all trying to make is that we feel that for most users, the benefits “uniform repository experience” trumps the issue of individual journal branding that you bring up.

Maybe I’m more clever than the typical user, but the inclusion of the article’s citation at the top of each page in the PubReader view, including the name of the journal, makes the provenance of each article rather clear to me.

Maybe I just don’t think branding is that important:

As you mention, specialists know which journals in their fields are top tier and which are not, and I don’t think a lack of branding is going to confuse them. The general public (including, perhaps, journalists) don’t know the difference between different journals in the first place, so no amount of branding is going to be meaningful.

I think the key question here is whether PMC is better served in accomplishing its mission with a cooperative relationship with journal publishers or an antagonistic one. It would seem that fostering good relations would be useful–journals that automatically deposit articles on behalf of authors and go beyond policy requirements, allowing the final, published version of the article to be deposited rather than the author’s final draft manuscript, serve as positive forces on PMC’s behalf.

If PMC’s goal here is to present a uniform interface, is that goal harmed by including some journal branding? I’m sure every journal on earth would be happy to upload a quick file providing a logo and a link (though PMC likely already has this material for most).

What is gained by not allowing the banner at the article top to have journal branding information? For most readers, the journal brand provides a useful filtering mechanism. I know that an article published in a journal with a strong reputation for good judgement and high quality is likely a better use of my time than an article in a journal with a poor reputation. Isn’t the reader better off with more information, rather than less?

I believe they actually do what you suggest for PMC. It’s used for the button to link to the published version of the article. I think some publishers submit up their publisher logo rather than the journal’s but I guess that is their choice.

I suppose there is no reason NCBI couldn’t add that at the top in pubreader format but my guess it is more of an oversight rather than a conscious decision to obscure the brand. Joel above is probably right, as publishers you are very sensitive to this, NCBI and readers probably not so much.

Pubreader is a brand new format and all the ramifications haven’t been fully thought through. I bet if publishers lobbied NCBI about this, they would add it to the format.

I think that’s fair. And I really think that the bigger problem for publishers is the preferential linking given the PMC version on PubMed search results.

NCBI’s appears to be attempting to better support smart phones and tablets.

” Designed particularly for enhancing readability on tablet and other small screen devices, PubReader can also be used on desktops and laptops and from multiple web browsers.”


I think developing a format that works well on these devices makes a lot of sense. It may have the effect of obscuring branding but I don’t think it is fair to automatically assume that is their intent.

Comments are closed.