The journal PLos ONE turned one year old yesterday (although the first papers were published in December 2006). PLos One is an open access scientific journal that takes a different take on interestingness. Most journals will reject scientific papers even before peer review based on a perceived interestingness, that is the paper is not of general interest, doesn’t fit the scope of the journal or is not the first paper on the subject. PLos ONE is different because it judges a paper solely on the quality of the science:

All papers that make a valuable contribution to the scientific literature, that are replicable, that are clearly written, and whose conclusions are supported by the data deserve publication.

The interestingness or perceived impact of a paper is not judged by the editor handling the submission, but will be judged by open peer review after publication of the paper. The journal uses tools provided by the open source Topaz Project to let readers rate and comment on the published paper. This approach is of course commonplace for other online content, e.g. Digg, but had not been tried before for scientific publications. Nature Precedings has since taken the same approach of open peer review.

PLoS ONE is radically different from other scientific journals in that it doesn’t filter that much before publication. The acceptance rate of submitted papers will therefore be much higher, but the impact factor (i.e. average number of citations) much lower. As long as the perceived reputation of a journal – which closely correlates with the rejection rate and the impact factor – is the driving force in the decision of the author to submit to a particular journal, PLoS ONE will have a hard time attracting very good papers. But if an author is mainly judged by the number of citations for his papers, and some positive ratings and comments thrown in, PLoS ONE all of the sudden looks very attractive.

Two important scientific journals recently stated that they cannot accept manuscripts written with Microsoft Word 2007, the .docx format: Nature and Science. The .doc format of previous versions of Microsoft Word hasn’t changed for close to 10 years and has become the standard for manuscripts submitted to STM (Science, Technology and Medicine) journals.

The reasons behind this move by Science, Nature and other journals are explained in a post on Nature’s Nascent Blog. Basically, journals have over the years developed workflows around the .doc format. Documents written with Microsoft Word 2007 are not compatible with these workflows in two important areas: equations and special characters.

What does this mean for scientific publishing? There are probably two answers. (1) We are at the beginning of a transition period to the new Microsoft Word format. It will take some time (probably years) until not only the publishing workflows of the journals but also the majority of scientists have moved to the .docx format. (2) This situation creates a window of opportunity for new document formats that might either coexist with .docx or even become a new standard. What are the contenders?

• Publicon Launched by Wolfram Research two years ago, it hasn’t seen a product update since (it’s still at version 1.0.1). A good idea, but most likely a dead product.

• OpenOffice and the OpenDocument format The other Office software. ODF has been adopted by the US state Massachusetts as standard document format. It is important to remember that ODF is supported not only by OpenOffice but also other software such as KOffice.

• Web-based tools such as Google Docs or Zoho Writer Web 2.0 technology now allows fairly sophisticated word processors within your web browser. Not quite as comfortable as your stand-alone word processor, but with one killer feature: collaboration. Almost all scientific publications are written by many authors, sometimes thousands of miles apart. One of the most important aspects of writing a scientific publication is the integration of references, which again are almost all found online. And the publishing workflow (peer review, printing, online publishing) would be much easier with web-based documents.

The most likely scenario is the gradual transition to Microsoft Word 2007. But the best scenario for STM publishers is clearly the last one, simply because it is the cheapest. Price is important for everyone, but expecially so for Open Access publishers that typically charge the author and not the reader. Let’s hope that someone builds the infrastructure for this. Google would be a likely contender as they already have both Google Docs and Google Scholar already in place. Or a publisher such as Public Library of Science picks up the ball.

The Nature Publishing Group started Nature Preceedings, a new preprint and document-sharing service last week. The new service was announced by Timo Hannay in Nature’s Nascent blog, Tim O’Reilly was one of the first to blog this.

Nature Preceedings is a searchable and citable document repository of preprints and presentations. It is very similar to the ArXiv.org service that covers Physics, Mathematics, Computer Science and Quantitative Biology. The main features of Nature Preceedings are:

• The service is free for readers and authors
• All content is released under the Creative Commons Attribution License and will be mirrored on other sites
• There is no peer-review, but readers can comment and vote on the released documents
• Papers submitted to a Nature journal can be prepublished on the service

This is a bold move and we wish them good luck. Time will tell whether enough authors are willing to submit their conference presentations, preliminary findings or supplementary material.

A technoraty search this morning for asco 2007 oncology lists 425 blog posts. Most blog entries cover drug and biotech companies, as the clinical trial results presented at ASCO have a substantial impact on the future of these companies. Finding blog entries with information for cancer researchers or clinical oncologists is more difficult.

I would like to see more blogs like Dr. Len’s Cancer Blog. Dr. Len Lichtenberg, Deputy Chief Medical Officer for the national office of the American Cancer Society, covers the ASCO meeting from a professional, but still very personal perspective.

OncTalk is a blog by Dr. Howard West that focuses on lung cancer. His expert opinion allows him to not only report the news, but also filter out what is important and comment on the findings.

The Cheerful Oncologist reports about one of the biggest news of this year’s meeting, the tyrosine kinase inhibitor sorafenib extending survival in patients with hepatocellular carcinoma.

Let’s hope that there will be more blogs in ASCO 2008. Until then, those of us can’t go to ASCO themselves have to find other ways to figure out what is new in clinical oncology.

The latest issue of the journal Cell (May 4, paid access) has an article by Laura Bonetta entitled Scientists enter the Blogosphere. In her article she describes some of the most popular science blogs and the people behind them.

Blogs are unfamiliar to most scientists

The article is interesting for several reasons. Its main purpose is certainly to introduce the typical reader of the journal Cell (one of the most prestigious journals in the Life Sciences) to the concept of a weblog. Scientists traditionally have been well connected to the internet (remember the days of gopher and usenet?) and have used it for at least 20 years to exchange their ideas with colleagues in different parts of the world. But blogging has been slow to catch on – in sharp contrast to the huge popularity it has quickly gained elsewhere. A few reasons for this obscurity among scientists are obvious, e.g. lack of time, age (most bloggers tend to be younger) and reluctance to talk about unpublished results in public. But blogs are still a fairly new medium and I expect science blogs to become much more widespread over time.

Scientific journals are the traditional source of information

The fact that this article was published in a reputable scientific journal is quite interesting as well. Journals have traditionally been one of the most important sources of information in science (besides scientific meetings and personal communications) and all scientific journals have managed the transition to electronic publishing and online distribution during the past 10-15 years. But the format and business model remained essentially the same. The almost universal access to the internet allows for very different ways of scientific exchange and the traditional journal paper (or talk at a scientific meeting) might not always be the best way of communicating a new scientific finding.

Why peer review is important

Open Access Publishing is one important driving force behind newer journals that take advantage of the new technical possibilities offered by the internet. PLoS ONE takes the idea of Open Access one step further and adds new features: annotations and discussions. These features were obviously stolen from the blogosphere where online comments are an important part of almost all weblogs. But what is stopping scientist of going the whole way and publish their research findings in their weblog? The answer is peer review: without the critical appraisal of your colleagues (or competitors), your scientific work will not be accepted in the community. Since production and distribution of scientific articles have (almost) become commodities in the age of the internet, peer review became the most important asset of science journals – and nobody knows this better than the editors of Cell. Peer review guaranties that scientific articles are interesting. Interestingness translates into revenue for the journal and scientific reputation (and consequently grant money) for the author. Interestingness of a journal is measured by it’s impact factor, which roughly measures how often it’s articles are cited by other journal articles.

Can blogs compete with scientific journals?

The blogosphere works differently. The interestingness of a blog entry is measured by the number of readers, comments and other blog entries that link to it. Because the cost of setting up a weblog is so low, there is no need for peer review. The blogosphere has shown again and again that for news, politics, culture, information technology and other topics their blog entries can compete with peer-reviewed articles in traditional media. Only time will tell whether this also holds true for science.

Papers is the new application from the folks that brought us 4Peaks and EnzymeX and other fine Macintosh programs for molecular biologists. Papers again is Macintosh-only (both PowerPC and Intel) and the first of their programs that is not free ($39 or €29 for a single user license).

Handling PDFs before Papers

Before the release of Papers, most scientists had devised their own system of storing the scientific publications most relevant to them. This could be folders with printouts sorted by topic or author or, as practically all papers are now distributed as PDF files, electronic copies stored on a computer in some filing system or another. PDF files could also be stored in reference managers such as Endnote, programs that are needed when you write your own scientific paper. But PDF support in these programs has always been something of an afterthought. For Macintosh users – and many scientists use the Mac – Sente is a wonderful reference manager with good PDF support, e.g. Sente automatically renames imported PDFs to something meaningful, e.g. by first author and publication year.

How Papers handles PDF files

Papers has taken PDF support to a new level, and the comparison to iTunes is not too far off. It is by far the best application to import all the PDF files of the scientific papers scattered around your hard drive and than match the publication record in the PubMed database. If Papers is like iTunes, than PubMed is like Gracenote, a central database of all scientific papers, maintained by the US National Library of Medicine. The nice integration with PubMed also shows when you want to import the PDF file of a publication you found in PubMed – it just couldn’t be easier.

Once you have your papers in Papers, it is a breeze to find and print out a paper. This is what you will end up doing most of the time, and it is faster and less space-consuming than reaching for that folder with the printouts in your bookshelf. What makes Papers such a pleasure to use is not just the basic idea of handling the PDFs of scientific papers in the best possible way, but do it with the look and feel and attention to detail that we love in a good Macintosh application. One of the features I like in particular is the PubMed search interface. It makes it very easy to use the more advances search features of PubMed.

Still at version 1.0

Papers is at version 1.0.1, so expect a few version 1.0 quircks – from sometimes sluggish behaviour to little interface bugs. What I would like most to see in version 1.1 is a better way of handling the way most scientists look out for new literature. One common approach is to scan new issues of the most important journals in your area of interest for articles relevant to your work. Most journals send out emails with the table of contents (TOC) once a new issue is released. More and more journals are also offering RSS feeds of the same information. In addition to scanning journal TOCs, most people will also regularly search PubMed for particular keywords, either manually or again by email notification or RSS. Papers is trying to help you with both strategies (e.g. by listing the most recent publications of a journal and linking to the table of contents), but the execution is far less elegant than the PDF handling.

Another obvious feature for version 1.1 is library sharing (again think iTunes) so that you can easily exchange references (with or without PDF files) with your coworkers. This of course touches on an aspect which is as bitterly fought in scientific publishing as is the distribution of mp3s in the music world, namely what are you allowed to do with scientific papers as almost all of them are copyrighted material. A growing number of Open Access journals allow you to freely distribute their publications, but most scientific journals don’t, not even when you are the author of the paper.

Papers is not a replacement for Endnote or other reference managers. When it comes to writing your own scientific paper, you still need them to integrate the references into your manuscript. For now, Papers can export to Endnote and other bibliography formats, but we don’t know whether Mek & Tosj have any plans in that direction for Papers 2.0.

We spent most of March traveling in New Zealand. Milford Sound is one of the must-see stops, and we were very lucky with the weather. More pictures can be found on my Flickr pages.