Friday, 25 July 2014 00:00

For Scientific Publishing, the Future is Now

Written by 
Rate this item
(2 votes)

publishingWill traditional publishing go the way of the printed journal?
Photo by Robert Cudmore (CC BY-SA 2.0)
A collaborator and I are about to submit a manuscript, a process that is deeply satisfying. However, it also leads me to reflect on the inefficiencies of the current publishing system. For example:

● Traditional academic publishing is extremely expensive, and much of its cost goes to filling corporate coffers rather than paying for services necessary for publishing. For example, Elsevier alone pulled in $1.1 billion in profit in 2010, an astounding 36% of its total revenue. In an age of tightening budgets, this is probably not the best use of public funds.
● The rigid format of a standard journal article suppresses all but the most "important" findings, excluding negative results, or results with unclear impact. Because only a certain type of work ever sees the light of day, this forces scientists to place undue emphasis on perceived importance, perhaps at the expense of accuracy.
● Multiple rounds of review can delay manuscripts for a year or more, leaving useful knowledge to languish and slowing the pace of innovation.

[Note that ASCB's journal, Molecular Biology of the Cell (MBoC) is only $116 per page for members to publish (no extra charge for color or supplemental data), it boasts "civil and constructive peer review" and getting a peer-review decision takes an average of 27 days].

One can fantasize about totally different models for publishing and evaluating science. For example, we could reduce the minimum publishable unit to a single figure, or enable version control so that works could be revised over time. Perhaps altmetrics could be used to rank papers according to their popularity or rating, much like reddit. In the last few years, however, we've finally started to see the promised jetpacks of the information age: new services that disrupt traditional publishing models. Here's an overview of some of them.

Preprint servers
A preprint server is a place to deposit a manuscript, making it publicly available before peer review. You can deposit your work before, during, or after it is under consideration at a traditional journal, thus freely sharing your work with the scientific community months before it would otherwise become available after a traditional peer review and editorial process. Some journals will not consider papers if they have previously been submitted to a preprint server, so be sure to check with your publisher, or see this list of journals' preprint policies (MBoC also allows preprints!). Because these servers are not publishers, work posted to them does not get indexed by PubMed, but some do get picked up by Google Scholar. A fantastic and more detailed write-up on all the options from Ethan Perlstein can be found here.

ArXiv
The classic preprint server is well-known and respected in physics, math, and computer science communities; most manuscripts from those disciplines are first are circulated here before submission to journals. There is a section for quantitative biology, but this category excludes much of the biology community, and it makes up a tiny fraction of the ArXiv's content.
Total preprints: Currently about 900,000, with about 8,000 in quantitative biology
Cost: Free
PubMed Indexing: No
Google Scholar Indexing: Yes

BioRxiv
Sponsored by Cold Spring Harbor, this service was created specifically for the biology community.
Current preprints: >500
Cost: Free
PubMed Indexing: No
Google Scholar Indexing: Yes

PeerJ Preprints
This is the Preprint service associated with PeerJ, a journal that is described in more detail below. One benefit is that scientists may be incentivized to offer feedback on PeerJ Preprints to maintain their PeerJ publishing plans (see below).
Current preprints: >428
Cost: Free
PubMed Indexing: No
Google Scholar Indexing: Yes

Figshare
Not a publisher, but a public and private repository of data. You can upload an image or movie and get a DOI, which will enable that data to be easily and permanently cited. Currently, all PLoS supplemental data is already hosted there, and they have a rich collection of posters.
Cost: Free for the standard plan
Total figures published: >960,000
PubMed Indexing: No
Google Scholar Indexing: No

Post-publication peer review journals
In post-publication peer review, your article is made publicly available shortly after submission. After reviewers make comments and suggestions, you can upload a new version of the paper. While third-party services like PubPeer, Open Review, and of course PubMed Commons aim to stimulate discussion through their independent services, some new journals are integrating post-publication peer review into their own processes.

The Winnower
The Winnower (which, in the interest of full disclosure, was founded by COMPASS member Josh Nicholson) offers a simple article submission process and open post-publication peer review. It has a convenient interface for paragraph-level comments, allowing discussion to happen in the context of the manuscript. It's also offering tools to easily publish blog posts, useful if you want an DOI that can be cited.
Cost: Currently free, eventually <$100 per paper
Total articles published: >12 (beta testing has just opened)
PubMed indexing: No
Google Scholar Indexing: Yes

F1000Research
F1000Research is an open-access, online-only journal that's part of the F1000 brand (also including F1000Prime, which contains recommendations of published papers, and F1000Posters, for....well, posters). Articles are published rapidly after submission, then peer-reviewed later.
Cost: $1000 for normal papers, $500 for single-result papers
Total articles published: >514
PubMed indexing: When articles have two "Approved" reviews, or one "Approved" and two "Approved with reservations." Currently, >286 F1000Research articles are on PubMed.
Google Scholar Indexing: Yes

ScienceOpen
This service is a combination of a preprint server, aggregator of other open-access articles, and post-publication peer review journal.
Cost: $800 per article
Total articles published: >8
PubMed indexing: No
Google Scholar Indexing: Yes
Nontraditional journals
Neither a preprint server nor a post-publication peer review platform, these organizations nevertheless support a different model of sharing science.

PeerJ
The peer-review process is traditional, but this journal is innovative in its approach to cost-reduction. Each author pays a relatively low lifetime membership fee, and is expected to review or comment on at least one paper per year to maintain it. Publishing itself is free, but the number of papers that an author can be listed on annually is a function of their plan level.
In another innovative measure, PeerJ has a platform that allows scientists to get credit for nontraditional, but useful scholarly discourse, like commenting on papers or preprints.
Cost: Starting at $100 per author (for lifetime membership)
Total articles published: >469
PubMed indexing: Yes. Currently, >429 PeerJ articles are on PubMed.
Google Scholar Indexing: Yes

So, what will my collaborator and I do with our latest manuscript? Balancing idealistic desires with practicality, we will probably find a middle ground between these new models and the immediate respect of an established brand—most likely by using both a preprint server and a more traditional journal. Surely, the growing popularity of alternative publishers will make these decisions easier and easier in the future.

Have I missed any nontraditional publishers? If so, please leave a link and short description in the comments!

Jessica Polka

Jessica Polka is interested in the spatial organization of the bacterial cell. Having studied a plasmid-segregating actin homolog during her PhD with Dyche Mullins at UCSF, she is currently working on a natural and engineered bacterial compartments during a postdoc in Pam Silver's lab at the Harvard Medical School.

Email This email address is being protected from spambots. You need JavaScript enabled to view it.

submissions

COMPASS Blog