Archives

Categories

Ad Lagendijk Ad Lagendijk 12 August 2009

Elsevier is going the wrong way

Tags: , , ,
Posted in Getting published, Web 2.0

Summary
Reed-Elsevier’s daughter Elsevier has introduced as an experiment a new way of publishing science. The “paper” is now basically a website, in which the idea of a linear text is abandoned. The web interface implements access to text fragments, graphs, supplementary material, interview with an author, through hyperlinked tabs and mundane hyperlinks. In my opinion this development is a step backward and scientist should avoid publishing their material this way.

Elsevier’s solution to a non-existing problem
Desktop publishing revolution made the beautiful IBM type balls obsolete Scientist agree that way too many papers are being published. In addition commercial publishers keep on launching new journals in an already overcrowded market. The desktop-publishing  innovation has radically improved the productivity of scientists. There are many factors that hamper the progress of science, but the alleged inadequacy of present-day science publishing is not one of them.

Elsevier and the Internet
For about ten years, until 2002, I  have been an editor of the Elsevier journal Physics Letters. I have good memories of that time. Capable journal publishers with full scientific training and a journal with a beautiful CYCLING-FRA-TDF-2009-ANDORRA-CANCELLARA typography. In those days Elsevier was lagging behind as far as using the Internet for its communication with authors, referees and editors was concerned. I remember vividly that I warned them regularly to take the world wide web seriously. I am sure many of my fellow Elsevier editors uttered the same anxiety.  I couldn’t understand why Elsevier did not react. This lack of perception on my side shows my ignorance about how multinational companies cope with major technological advances: they do not reinvent the wheel, no they just buy it. I do not know how many web-developing companies Elsevier Reed acquired, but it must have been quite a number. All the back-issues of their journals were scanned and put online. As a result Elsevier has a reasonable web performance. My respect for the management of Reed Elsevier.

I say reasonable, because it is still not great. The interface of the Elsevier portal ScienceDirect is clumsy and ugly. And if you really want to see a distasteful page, it is here: the official Elsevier web page for the press.

Elsevier and Dutch reporters
Every Dutch journalist gets a bitter taste in his mouth when the discussion comes to Elsevier. One of Reed-Elsevier’s daughters, Dagbladunie, once owned a few high-quality Dutch newspapers with a return on investment of about 13%. This return on investment was considered to be too low by the management of Reed Elsevier in 1995. The publishing company was used to higher returns with science publishing, So the company sold the Dutch newspapers. From then on these Dutch newspapers did not fare well as far as their finances were concerned. Amongst other mishaps they suffered from a raid by private equity firm APAX.

Problem 1 for science publishers: open access
The open access movement is gaining ground. The general public is getting interested in the issue. Why should public scientific libraries pay a fortune to get access to papers in which results are is reported of research financed by the tax payer?  It is useful to make a distinction between journals, like Nature, which are published by pure commercial publishing houses and journals, like Science, that are published by learned societies or other not-for-profit publishers. I must add that some learned societies, for instance the Optical Society of America, are in their behavior not any longer discernable from pure commercial enterprises.

The non-commercial companies have much fewer problems with open access as the commercial ones. However all publishers realize that full open access will become a fact of life.

Problem 2 for science publishers: open standards
Standardization is always a hot issue in a market economy. The market leader likes to impose his standards on the market. From the moment on No single printer company succeeded in getting its cartridges universally accepted standards that the company has succeeded iTravelers realize every day the problems with lack of standardst will start to continuously change its standard, making it very cumbersome for their competitors to get a market share.These lagging companies will complain and ask governments to interfere and are continuously trying to to influence public opinion. But as soon as the situation is reversed, and one of the plaintiffs gets dominance over the market he would start to behave exactly in the same way in protecting his own standard.

Adobe is an admirable company. It has been in computer graphics and typography right from the start. It has produced postscript, a computer language to drive laser printers. Postscript was a revolution in desktop publishing. As postscript was not open-source, companies making laser printers had to pay Adobe a license fee to be able to use postscript in the firmware of the laser printer. These licensing fees made the company Adobe big.

But Adobe also introduced – in 1993 –  the portable document format (pdf). An absolute blessing for science. This format can easily deal with graphs, figures, mathematical formulas, chemical formula’s etc. Its linear and inter-page independency makes for very fast web viewing. Although Adobe still holds patents for the pdf standard it is now officially an open standard.  A pdf file is by the way the only file format that can be reasonably protected with encryption and passwords. In this respect it is superior to all Microsoft Office products.

All scientific papers are these days available as pdf files. A “reprint”  of a scientific paper is identical to a pdf file copy. A publishing company that starts a new journal will have to supply its articles in pdf format. Submitting authors regularly have to submit their paper, including figures, list of references, supplementary material as a pdf file. Referee reports are sent as pdf files. This free exchange of scientific information through pdf files is an ideal situation for scientists. But it is a nightmare for companies like Elsevier. Monopoly seekers would like to control the scientific market and they would like to impose their own standard and get rid of the pdf standard. In the present experiment Elsevier still supplies the pdf version of the paper. But for how long? This new effort of Elsevier, if successful – which god forbade –  would mean you cannot send reprints around any longer. You will have to supply your colleague with a web address of a commercial company, with – in the future very likely – paid  access to be able to access the “web-paper”.

Context of discovery
Scientists are human beings. Some get heir inspiration in church, and others while watching a ball game, or by going to a conference. One can write novels about the life of a scientist. Richard Feynman wrote amusing books about his life. By reading such literature one can learn a lot about fermatthe psyche of a scientist and the sociology of the scientific community. These activities are part what is called the context of discovery. The proof of Fermat’s last theorem made John Wiles famous and a cult figure, featuring in many tv programs. All these accounts will never make it into a physics, chemistry or biology journal. And happily so.  But Elsevier’s experiment is an attempt to compromise the hard core of science. Any scientist can give an interview. The next thing is a scientist talking about his religious feelings as an explanatory introduction to his paper.

Context of justification
The body of accepted knowledge, that is the content of scientific papers when reproduced, when survived many challenges, and when finally widely accepted as true, belongs to the context of justification. Progress in science concerns increase of this knowledge. Brilliant scientific discoveries are part of this context. An interview with a scientist is not.

Didactics
If a proof of a theory is known and accepted it often can be simplified. Explaining science to outsiders is not part of context of justification Initial mathematical derivations can take tens of pages and after a couple of years simplified proofs can be produced that take only a page or two. This simplification can be part of the context of justification. But didactics, defined as expressing the same thing in a simpler form without adding any new science, is not part of the context of justification. If the content of a paper could indeed have been better presented and explained as being done in the paper itself the authors wrote the wrong paper. Explaining the content of a scientific paper for lay people is not part of the context of justification and should be kept separate.

Mathematics has no video
I recently bought a new wireless router. I like the brand Linksys, but this company has been acquired by Cisco, so now the brand is Linksys-Cisco. The contribution of Cisco is certainly that the “improved” and more “timely” manual has become of terrible quality. The manual is extremely modern, so it is not just a simple pdf file that I could read and use to install my router. No the manual has become modernized: it is a video. I had to run this video maybe twenty times. The reason was that something in the vido was unclear, at least to me. So I had to go through the whole video over and over again. Oh I would loved to have had a linear text. With a linear text I would just have gone to the specific location in the text and read it, may be several times. That would be done in seconds in stead of the rerunning the video costing me half an hour.

The ultimate dull linear text is a pure mathematical treatise. This is a sequence of lemma’s and proofs. No interview with the author. No video. 

What is wrong with linear text?
A virtue of a linear text is its extreme inflexibility. The first sentence is supposed to be the first sentence and the whole text is a serial line of arguments and presentations. Inflexibility can be a great virtue. On toll ways in France there are almost no exits. This inflexibility makes transport along these toll ways very effective. In my country, the Netherlands, every village requires its highway exit and gets supported by Parliament. As a result maximized flexibility but speed zero.

Technical problems
Present day (x)html rendering is still poor. This is easily seen in the experimental text of Elsevier. I will give just one example. The web text of the Cell paper uses tens of times the chemical formula “Ca2+”, whereas the pdf version tells us that it should be “Ca2+”. As you can see in this post the superscript is possible in html but then ugly varying line spacings are introduced.

The text lines in the Elsevier’s web texts are much to long. This makes reading tiresome. Narrowing the window does not help as the Elsevier developers have prevented the text from wrapping. In addition the text is not fully justified but ragged right. It is well-known that fully justified text can be read quicker.

Elsevier is afraid of open discussion
I am not at all saying that the context of discovery is not important for science. In this respect I like the forum discussions and comments in which scientists participate. Elsevier started its new experiment, “the article of the future” and says it welcomes feedback. Reactions can be given in two ways: through a web form or via email. But these are old-fashioned one-way communication channels. Why not open up  a forum and allow people to discuss openly? The company that claims to have invented the article of the future communicates with the community in a previous-century way.

Related Posts:
- - - - - -
If you like this post why don't you email subscribe to our new posts. Or subscribe to our RSS feed.
  1. Unregistered

    12 Aug 2009 22:58, Bram van Ginneken

    Nice post. There is little doubt that the Article Of The Future will quickly become yet another Thing Of The Past. The money-making companies and societies keep hassling us with their supposedly great new inventions. Journals always ask me endlessly if I want to submit additional material (data, videos, whatever) to their web site when a paper has been accepted. They have added some clumsy crummy new feature that nobody wants but of course they’ve raised their subscription rates by ten percent for it. They claim that they are much more capable of keeping material online indefinitively than we, the ignorant scientists. We’ll see about that. I’d rather put my own stuff on my own website. Many scientists keep pre-prints of their papers on their homepage because most journals mess up about everything that can be messed up when they prepare your camera ready submission for publication. Last year I was asked four times during the idiotic web-based upload process of a conference paper if I wanted to submit additional material like mp3s. I decided to upload a punk rock song from my band. Six months later I received a nice email from a girl who told me they all laughed a lot in the office while listening to Poodle Destroyer, but they thought it would not be a good idea to publish this alongside with my paper on the detection of pleural nodules.

    In the August 1 magazine of NRC Handelsblad there was an article about Open Science. What various people in that article (university managers, librarians, and, yes, a manager from Elsevier) said about scientists was downright insulting. We scientists do not understand anything about publishing, we do not understand what open access means, and we have no idea how expensive it is to run a journal. Maybe, my idea is that it cannot possibly be very expensive at all. All the relevant work is done for free by scientists. So I think the amount open acces journals like PLOS charge could be much lower (not that they are too unreasonable now). I have to stop now. I still have twenty conference papers to review.

  2. Mirjam

    13 Aug 2009 15:19, Mirjam

    Frankly, I indeed have no idea how much it costs to run a journal, but just printing costs must be significant (maybe these days we can do without print?), as well as the typesetting, layout and administrative overhead; the delivery of content by the scientists probably isn’t the main expense (for comparison: inquire at your institute how much they spend on overhead to pay your salary – you will be shocked).
    With respect to it being better to keep additional material on your own website: I don’t think I agree on that. My website is hosted by the university and will change everytime I change my affiliation (with my new move this month I will have been at 5 different institutions in the last 9 years). And who is going to maintain your site after you are dead? A centralized server with the journal seems better to me… (that way I also don’t need to search in the crap on people’s unstructured homepages for the right material). Also, sometimes supplementary material can convey information that is difficult to put in words (especially for experiments), so it is not always bullshit. What I don’t like though is when people use it to keep their written text within the length criteria of the journal, that is bad practice.
    With respect to the ‘web page article of the future’ (as per the description given above, I haven’t looked at it): I hate navigating between multiple tabs and much prefer a straightforward structure as in an old-fashioned paper. Moreover, most of my reading I do when I am traveling (on airplanes etc), for which I bring old-fashioned printouts (much easier on the eyes too). I agree with Ad that ‘interviews’ in scientific publications are clearly out of place.
    With respect to open access: obviously journals need some source of financing. If the argument is that research results should be open to the taxpayer, then isn’t the logical conclusion that the journals should be financed from tax money as well to provide this open framework?
    Finally, if Elsevier doesn’t listen to negative feedback and proceeds with the web-based articles it of course is very easy to make ourselves heard very clearly: don’t publish in or cite that journal (I wonder how many scientists have a strong enough backbone for that).

  3. Unregistered

    13 Aug 2009 15:48, Bram van Ginneken

    @Mirjam. Good points. 1) Yes, we can do without print. But I also print out a lot of articles myself because reading and sorting papers on the floor is easier that way. And I read a lot of papers in my bathtub. 2) I know exactly how expensive researchers are in the Netherlands. This is a major difficulty in collaborations with US companies. A US PhD student costs 20% of what a Dutch PhD student costs. Dutch PhD students are, on average, much more productive but probably not five times. This makes it harder to survive in science in the Netherlands if you do applied research. 3) Websites moving is no problem. Give your supplemental material a name and make sure it can be found with the major search engines, and that it can be found if people search for your name. You can also give it a doi. And make sure your homepage is not unstructured crap (yeah, that is hard). If you do not have the technical skills to do this, then you may indeed be better off leaving it to a publisher. And also if it is indeed stable data in a simple format and the publisher’s site is open access. But in our case it is for example software that may change in the future, or data for which descriptions and annotations may require updating. If you leave it at the journal’s site, do not believe it will still be around when you are dead (unless you are very old maybe). 4) Open access is not that important. I’m in favor of it, but to survive in science I need to publish in high impact journals and I take the fact that these are not open (in my field) for granted. For those who have no access: Everybody can download the papers from our institute’s web site, and this is legally allowed the way we have set it up (at least we think so). We monitor the number of downloads: it is a service that is well appreciated.

  4. Unregistered

    14 Aug 2009 8:52, Klaas Wynne

    I understand that the American Chemical Society will now stop printing its journals except two of their flagship journals. As for the cost of publishing, of course, I don’t know the details either, but my experience with the online journal Optics Express is that the authors do all the typesetting and conversion to PDF and the editors (scientists themselves) decide on publication. That means that the journal simply provides secretarial service in the form of sending PDFs out for review and posting the PDFs on their website. Wouldn’t that imply that the cost per paper is on the order of $100-1000 per paper? (More likely on the lower end of that range.) And that is more or less the price you pay for publication in open access journals.

    I too prefer to publish in the highest profile journals irrespective of whether they are open access or not. Therefore, it is the duty of big organisations or countries to force open access publication. Didn’t some big US universities close a deal to provide open access to all their papers? Anyway, clearly it has to be paid for somehow.

    By the way, I also hate the proliferation of ‘additional material’. Why can’t they just all stick it all in one big convenient PDF that I can store in Papers on my computer? Oh yeah, and print out for reading on the comfy sofa in my office…

  5. Unregistered

    16 Aug 2009 1:55, David Stern

    These new features are similar to what Nature and Science have been doing all along with their articles discussing the papers that appear in the journal. So I don’t see any problem with that. The concern is that Elsevier wants to drop the pdf for revenue reasons as you suggest. Already university libraries, which in the past were accessible to anyone they would let in the door to browse the journals on the shelves are only accessible to members of the university who have the requisite passwords to the journal databases.

  6. Unregistered

    19 Aug 2009 18:05, Headline Commentary Aug 14-Aug17 | Health Content Advisors

    […] » Elsevier is going the wrong way | Survival Blog for Scientists […]

  7. Unregistered

    2 Sep 2009 23:08, Bibliotheken, google en zoeken augustus 2009 overzicht « Dee'tjes

    […] Elsevier gaat de verkeerde kant op met dat New way of publishing science (scientists should avoid publishing this way zegt Ad Langendijk. […]

  8. Unregistered

    12 Aug 2010 2:35, Wolfgang

    I like the idea of an article written not in a linear style but rather like a wikipedia entry, because it allows people much more easier to read exactly that part of the story, that is most interesting to them. How many pdf articles are out there which are filled with technical details not interesting anyone? How many articles have to repeat descriptions because it is expected that readers of this article do not necessarilly know the preceding one? In fact I would throw out scientific publishers alltogether. Give every scientist his own homepage, skip peer-review as it never proved to do the things acclaimed for and teach people to decide for themselves which information is good, and which is nonsense.