Sunday, January 13, 2008

the arxiv and publishing

Journals seem to have a love/hate relationship with the arxiv. For years Nature in particular used to campaign actively against electronic preprint servers, though they have given in and no longer consider posting preprints to compromise a journal submission. Science still forbids web posting of preprints before publication, presumably because at the end of the day they think that people won't buy the journal if they can get versions of the paper for free. This comes up most often when someone has a hot result and they want to get it out there to the community quickly (perhaps to establish priority), rather than wait weeks or months for the editorial process at a journal. Other publishers have attitudes that run the gamut from embracing preprint servers to ignoring their existence. Does anyone out there know if there is any actual evidence to support or refute the idea that preprint posting harms journal circulation?

8 comments:

Schlupp said...

I do not know any evidence, so I'll engage in some aimless speculation.

I think there are two different questions:

1) Can Archive harm a specific publisher? E.g., one specializing in overpriced journals of not-so-great quality, at least for physics? You know, the ones where most people only ever publish conference proceedings? Here, the answer might be yes, and I would not mind.

2) Will Archive bring down the whole concept of peer reviewed publications done by publishing houses (i.e. with editors and so) making at least some money from the journals? In don't think so. At most, it might lead some journals to switch to the free-to-read / pay-to-publish model. Authors might still pay, despite archive's being free, because they want the refereeing process as a sort of quality certificate for their work. And readers would probably still prefer to read journals, because of the - hopefully - better signal to noise ratio.

NONE said...

I don't have any evidence one way or the other. But I think that threat presented by arxiv is way overrated.

For example - imagine every paper in Science or Nature can also be downloaded through Arxiv. Does it mean you will download them through Arxiv, rather than directly through Nature or Science? I don't think so. It is much more convenient (and natural - pun intended) to get the original through Science/Nature. I would still want my university to maintain a subscription, even if I could get a poorly formatted arxiv equivalent of the same article (with low-res images), after a few minutes of googling around.

The same situation exists in respect to PRL/PRB - a lot of papers can be found on arxiv, often well before publication, but it doesn't mean people don't read PRB or PRL anymore.

There are other reasons to subscribe to Nature or Science, but even if we are limiting ourselves to research physics articles, my suspicions is that arxiv is not much of a threat to Science or Nature.

Douglas Natelson said...

Schlupp - I don't think anything can hurt Elsevier. They're like the borg.

The one that surprises me a bit is that the ACS and APS have diametrically opposing views on the arxiv. APS (and AIP) don't care, and if anything they encourage the arxiv by giving you the opportunity (though I've never tried this) to do a web-based submission of a preprint based on arxiv number. The ACS, in contrast, strongly implies (and if asked directly, will confirm) that posting a result on the arxiv can be considered compromising the originality of your submission. I really don't understand this attitude. If anyone was going to be hurt by electronic preprints, it'd be the APS, and yet the ACS seems to be much less progressive.

Perhaps this shouldn't be surprising, though. The other evidence that the ACS is fiercely protective of copyrighted material is their lack of inclusion in the AIP's virtual journals. Science, Nature, PNAS, New Journal of Physics, etc. all are willing to allow the AIP to reproduce their abstracts with web links to the original source papers. ACS is not, even though this may lead to more click-through traffic on their journal pages. Presumably they have some logical reason behind this.

Anonymous said...

Elsevier is like the Borg? Wow, that is about the coolest put-down I've ever heard directed at a publishing house. I wonder if their corporate offices are housed in a cube-shaped building...

Douglas Natelson said...

Dan - I thought about comparing them to the Terminator (They don't feel pity, or remorse, or fear, and the absolutely will not stop, ever, until you are dead!) but that might be over the top.

Anonymous said...

A point in favour of Elsevier though was that they did publish Galileo when nobody else would because of fear of the church...

Bracelets for women said...

But perhaps the most malignant rumours are those spread by scientists themselves — in all good faith but erroneously. One such is about prepublication.

Anonymous said...

Incoherent Ponderer said:
"It is much more convenient to get the original through Science/Nature."

Au contraire mon ami, navigating the simple HTML page of arxiv is much more quicker. Even if sometimes they have bandwidth problems and it takes 30 seconds to get a PDF, that is much better than navigating thorugh all the corporate BS that fills the editing companies websites. Amongst them elesevier is possibly the worst, where you cannot get a paper just simply by page/volume number and you need to navigate though all the Archive/Volume/Issues, etc.