Cameron Neylon

*The third installment of the paper ([first
part](http://cameronneylon.net/blog/a-personal-view-of-open-science-part-i/),
[second
part](http://cameronneylon.net/blog/a-personal-view-of-open-science-part-ii-tools/))
where I discuss social issues around practicing more Open Science.*

Scientists are inherently rather conservative in their adoption of new
approaches and tools. A conservative approach has served the community
well in the process of sifting ideas and claims; this approach is well
summarised by the aphorism 'extraordinary claims require extraordinary
evidence'. New methodologies and tools often struggle to be accepted
until the evidence of their superiority is overwhelming. It is therefore
unreasonable to expect the rapid adoption of new web based tools and
even more unreasonable to expect scientsits to change their overall
approach to their research *en masse*. The experience of adoption of new
Open Access journals is a good example of this.

Recent studies have shown that scientists are, in principle, in favour
of publishing in Open Access journals yet show marked reluctance to
publish in such journals in practice
\[[1](http://www.bio-diglib.com/content/4/1/1)\]. The most obvious
reason for this is the perceived cost. Because most operating Open
Access publishers charge a publication fee, and until recently such
charges were not allowable costs for many research funders, it can be
challenging for researchers to obtain the necessary funds. Although most
OA publishers will waive these charges there is anecdotally a marked
reluctance to ask for such a waiver. Other reasons for not submitting
papers to OA journals include the perception that most OA journals are
low impact and a lack of OA journals in specific fields. Finally, simple
inertia can be a factor where the traditional publication outlets for a
specific field are well defined and publishing outside the set of
'standard' journals runs the risk of the work simply not being seen by
peers. As there is no perception of a reward for publishing in open
access journals, and a perception of significant risk, uptake remains
relatively small.

Making data available faces similar challenges but here they are more
profound. At least when publishing in an open access journal it can be
counted as a paper. Because there is no culture of citing primary data,
but rather of citing the papers they are reported in, there is no reward
for making data available. If careers are measured in papers published
then making data available does not contribute to career development.
Data availability to date has generally been driven by strong community
norms, usually backed up by journal submission requirements. Again this
links data publication to paper publication without necessarily
encouraging the release of data that is not explicitly linked to a peer
reviewed paper. The large scale DNA sequencing and astronomy facilities
stand out as cases where data is automatically made available as it is
taken. In both cases this policy is driven largely by the funders, or
facility providers, who are in position to make release a condition of
funding the data collection. This is not, however a policy that has been
adopted by other facilities such as synchrotrons, neutron sources, or
high power photon sources.

In other fields where data is more heterogeneous and particular where
competition to publish is fierce, the idea of data availability raises
many fears. The primary one is of being 'scooped' or data theft where
others publish a paper before the data collector has had the ability to
fully analyse the data. This again is partly answered by robust data
citation standards but this does not prevent another group publishing an
analysis quicker, potentially damaging the career or graduation
prospects of the data collector. A principle of 'first right to publish'
is often suggested. Other approaches include timed embargoes for re-use
or release. All of these have advantages and disadvantages which depend
to a large extent on how well behaved members of a specific field are.
Another significant concern is that the release of substandard, non
peer-reviewed, or simply innaccurate data into the public domain will
lead to further problems of media hype and public misunderstanding. This
must be balanced against the potential public good of having relevant
research data available.

The community, or more accurately communities, in general, are waiting
for evidence of benefits before adopting either open access publication
or open data policies. This actually provides the opportunity for
individuals and groups to take first mover advantages. While remaining
controversial
\[[3](http://www.bmj.com/cgi/content/full/337/jul31_1/a568),
[4](http://www.bmj.com/cgi/eletters/337/jul31_1/a568)\] there is some
evidence that publication in open access journals leads to higher
citation counts for papers
\[[5](http://dx.doi.org/10.1371/journal.pbio.0040157),
[6](http://eprints.ecs.soton.ac.uk/12906/)\] and that papers for which
the supporting data is available receive more citations
\[[7](http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0000308)\].
This advantage is likely to be at its greatest early in the adoption
curve and will clearly disappear if these approaches become widespread.
There are therefore clear *advantages* to be had in rapidly adopting
more open approaches to research which can be balanced against the risks
described above.

Measuring success in the application of open approaches and particularly
quantifying success relative to traditional approaches is a challenge,
as is demonstrated by the continuing controversy over the citation
advantage of open access articles. However pointing to examples of
success is relatively straightforward. In fact Open Science has a clear
public relations advantage as the examples are out in the open for
anyone to see. This exposure can be both good and bad but it makes
publicising best practice easy. In many ways the biggest successes of
open practice are the ones that we miss because they are right in front
of us, the global databases of freely accessible data in biological
databases such as the [Protein Data Bank](http://www.pdb.org),
[NCBI](http://www.ncbi.nlm.nih.gov), and many others that have driven
the massive advances in biological sciences over the past 20 years. The
ability to analyse and consider the implications of genome scale DNA
sequence data, as it is being generated, is now

In the physical sciences, the [arXiv](http://www.arxiv.org) has long
stood as an example to other disciplines of how the research literature
can be made available in an effective and rapid manner, and the
availability of astronomical data from efforts such as the [Sloan
Digital Sky Survey](http://www.sdss.org/) make efforts combining public
outreach and the crowdsourcing of data analysis such as [Galaxy
Zoo](http://www.galaxyzoo.org/) possible. There is likely to be a
massive expansion in the availability of environmental and ecological
data globally as the potential to combine millions of data gatherers
holding mobile phones, and sophisticated data aggregation and
manipulation tools is realised.

Closer to the bleeding edge of radical sharing there have been less high
profile successes, a reflection both of the limited amount of time these
approaches have been pursued and the limited financial and personnel
resources that have been available. Nonetheless there are examples.
Garret Lisi's high profile preprint on the ArXiv, *An exceptionally
simple theory of everything*, \[[8](http://arxiv.org/abs/0711.0770)\] is
supported by a comprehensive online notebook at
<http://deferentialgeometry.org> that contains all the arguments as well
as the background detail and definitions that support the paper. The
announcement by Jean-Claude Bradley of the successful identification of
several compounds with activity against malaria
\[[9](http://usefulchem.blogspot.com/2008/01/we-have-anti-malarial-activity.html)\]
is an example where the whole research process was carried out in the
open, from the decision on what the research target should be, through
the design and in silico testing of a library of chemicals, to the
synthesis and testing of those compounds. For every step of this process
the data is available online and several of the collaborators that made
the study possible made contact due to finding that material online. The
potential for a coordinated global synthesis and screening effort is
currently being investigated.

There are both benefits and risks associated with open practice in
research. Often the discussion with researchers is focussed on the
disadvantages and risks. In an inherently conservative pursuit it is
perfectly valid to ask whether changes of the type and magnitude offer
any benefits given the potential risks they pose. These are not concerns
that should be dismissed or ridiculed, but ones that should be taken
seriously, and considered. Radical change never comes without
casualties, and while some concerns may be misplaced, or overblowm,
there are many that have real potential consequences. In a competitive
field people will necessarily make diverse decisions on the best way
forward for them. What is important is providing as good information to
them as is possible to help them balance the risks and benefits of any
approach they choose to take.

*The fourth and final part of this paper [can be found
here](http://cameronneylon.net/blog/a-personal-view-of-open-science-part-iv-policies-and-standards/).*

1.  Warlick S E, Vaughan K T. [Factors influencing publication choice:
    why faculty choose open
    access](http://www.bio-diglib.com/content/4/1/1). *Biomedical
    Digital Libraries*. 2007;4:1-12.
2.  Bentley D R. Genomic Sequence Information Should Be Released
    Immediately and Freely. *Science*. 1996;274(October):533-534.
3.  Piwowar H A, Day R S, Fridsma D B. S[haring Detailed Research Data
    Is Associated with Increased Citation
    Rate](http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0000308).
    *PLoS ONE*. 2007;1(3):e308.
4.  Davis P M, Lewenstein B V, Simon D H, Booth J G, Connolly M J. [Open
    access publishing, article downloads, and citations: randomised
    controlled
    trial](http://www.bmj.com/cgi/content/full/337/jul31_1/a568). *BMJ*.
    2008;337(October):a568.
5.  Rapid responses to David et al.,
    <http://www.bmj.com/cgi/eletters/337/jul31_1/a568>
6.  Eysenbach G. [Citation Advantage of Open Access
    Articles](http://dx.doi.org/10.1371/journal.pbio.0040157). *PLoS
    Biology*. 2006;4(5):e157.
7.  Hajjem, C., Harnad, S. and Gingras, Y. (2005) Ten-Year
    Cross-Disciplinary Comparison of the Growth of Open Access and How
    it Increases Research Citation Impact. *IEEE Data Engineering
    Bulletin*, 28 (4). pp. 39-47. http://eprints.ecs.soton.ac.uk/12906/
8.  Lisi G, An exceptionally simple theory of everything,
    [arXiv:0711.0770v1](http://arxiv.org/abs/0711.0770v1) \[hep-th\],
    November 2007.
9.  Bradley J C, [We have antimalarial
    activity!](http://usefulchem.blogspot.com/2008/01/we-have-anti-malarial-activity.html),
    UsefulChem Blog,
    <http://usefulchem.blogspot.com/2008/01/we-have-anti-malarial-activity.html>,
    January 25 2008.


<em>
 The third installment of the paper (first part, second part) where I discuss social issues around practicing more Open Science.
</em>
Scientists are inherently rather conservative in their adoption of new approaches and tools. A conservative approach has served the community well in the process of sifting ideas and claims;

Messages de Rogue Scholar

A personal view of open science – Part III – Social issues