Are chemists really grinches?

With well known and respected open science projects coming out of chemistry as well as cool tools like pubchem and emolecules... it seems a bit unfair of me to ask if chemists are grinches. But there has been and there continues to be a lot of study of data/information/knowledge sharing in chemistry - or, really, the lack thereof.  In general, pre-prints are not passed around or self-archived, there is very little data sharing (there are counter examples in crystallography), and details are withheld from conference presentations or the conference slides are not made available (Milo used to have a post on this, but I guess he's dropped from teh netz). These things might be quite striking to someone who is familiar with scholarly communication in high energy physics or bioinformatics.

Most recently, there is a commentary article out by Theresa Velden and Carl Lagoze [1] that summarizes things that are different about chemistry that probably impact the adoption and use of some of the newer and more open data/information/knowledge sharing - communication tools:

• long tail science - elsewhere this is described as resource concentration, among other names. In other words, there are lots and lots and lots of small and medium sized labs funded from all sorts of different places.  If you're more familiar with physics, this is like comparing HEP (with a very few, very big experiments) with some of the bench-top optical research areas (like what Chad studies, I think). The idea is that if there are lots of little and highly competitive labs around there are disincentives to share. In the case of big science - some areas of astro, HEP - you *have* to share, it's built into the funding and access to equipment/data.
• longevity of data - chemistry information doesn't go bad so quickly. It's sort of like math that way. With some areas of biomed, it's pretty easy to give away content for free when it's a year old. Immediacy is important there. Also, because there's a lot of data around, and it's all locked up in closed proprietary systems, there's a huge amount of inertia in trying to change anything.
• diversity of research cultures in chemistry
• proprietary information - big money closed databases. I have to say that I think ACS and CAS evolved the way they did because that's what the community thought it wanted. ACS is a member organization so... (oh, and AIAA and SAE have both said that it's the pub board made up of society members that are requiring their grinch-y behavior).
• proximity to industry - or as Velden and Lagoze call it - the industry-academia balance. There are two pieces to this - the amount of work that's being done in industry and how sell-able the fruits of scientific labor are. Even in academic chemistry there's industrial funding and the push to patent and license stuff in chemistry. Industrial chemists might not publish as much and keep more information secret.
• ACS and CAS's global dominance and iron fist control. (but they say that's not true - see pp60-1 in [2], "points of dissent"

But this discussion really builds on work that has been done over a number of years. Jeremy Birnholtz' dissertation [3] and JASIST article [4] studied collaboration propensity.  Resource concentration, agreement on quality, and the need for/availability of help were significant in predicting collaboration propensity.  Agreement on quality is also mentioned decades before by Zuckerman and Merton [5]. The acceptance rate of HEP journals is much, much higher than in fields like history because submitters know what is expected and the article gets reviewed within the lab before being submitted.  The vast majority of the pre-prints on ArXiv do end up published in peer-reviewed journals or conferences so it's pretty safe to use them. This might not be the case in other fields.

A whole bunch of articles on the adoption of e-mail, bulletin boards, listservs, etc., came out in the mid-1990s.  To sum these up, it's not just a matter of time [6], the practices and social features of the research area matter.

So, are chemists grinches?  It does appear that in many areas of chemistry, there is not a tradition of sharing.  In these areas, it's not a matter of the availability of appropriate technology, it's more related to how much individuals or labs need to share to produce scientific knowledge and how much they are concerned about being scooped. New researchers are brought up and trained to control their data - so it's not just a matter of time.

If chemists are grinches, but are happy that way, is there a need to change that?  Is there a need to tackle the Sisyphean task of culture change?  Well, chemistry information is not just for chemists. Physicists and biologists who have to deal with chemical information get pretty annoyed at the hassles. Should there be parallel systems of chemical information used by folks who like to share (maybe that's some of what is going on).

So I hope I've made some people angry enough to comment :) 

References

[1] Velden, T. & Lagoze, C. (2009) Communicating Chemistry. Nature Chemistry 1, 673 - 678. DOI:10.1038/nchem.448

[2] Velden, T. & Lagoze, C.(2009) The Value of New Scientific Communication Models for Chemistry. Retrieved November 28, 2009 from http://hdl.handle.net/1813/14150.

[3] Birnholtz, J. P. (2005). When Do Researchers Collaborate? Toward a Model of Collaboration Propensity in Science and Engineering Research. Unpublished Doctor of Philosophy (Information), The University of Michigan. 3186579.

[4] Birnholtz, J. P. (2007). When do researchers collaborate? Toward a model of collaboration propensity. Journal of the American Society for Information Science and Technology, 58(14), 2226-2239. doi:10.1002/asi.20684

[5] Zuckerman, H., & Merton, R. K. (1971). Patterns of Evaluation in Science: Institutionalization, Structure and Functions of the Referee System. Minerva, 9, 66-100.

[6] Kling, R., & McKim, G. (2000). Not just a matter of time: Field differences and the shaping of electronic media in supporting scientific communication. Journal of the American Society for Information Science, 51, 1306-1320. doi:10.1002/1097-4571(2000)9999:9999<::AID-ASI1047>3.0.CO;2-T