Trends in Research and Publication: Science 2.0 and Open Access

Abstract

This paper considers current trends in academic research and publication, in particular as seen from the control community. The introduction of Web 2.0 applications for scientists and engineers is currently changing the way research is being conducted. In the near future, participants in the research community will be able to share ideas, data and results like never before. They will also be able to manage the rapidly increasing amount of scientific information much more effectively than today through collaborative efforts enabled by the new Internet tools. However, an important premise for such a development is the availability of research material. Many research results are currently shielded behind expensive subscription schemes that impede the sharing of information. At the same time, an increasing amount of research is being published through open access channels with unrestricted availability. Interestingly, recent studies show that such policies contribute to an increased number of citations compared to the pay-based alternatives. In sum, the parallel development of new tools for research collaboration and an increased access to research material may help spur a revolution in the way research is going to be conducted in the future.

Full text

Page 1

Modeling, Identification and Control, Vol. 30, No. 3, 2009, pp. 181–190, ISSN 1890–1328
Trends in Research and Publication:
Science 2.0 and Open Access
Morten Breivik1Geir Hovland2P˚ al Johan From3
1Centre for Ships and Ocean Structures, Norwegian University of Science and Technology, NO-7491 Trondheim,
Norway. E-mail: morten.breivik@ieee.org
2Department of Engineering, Mechatronics Group, University of Agder, NO-4898 Grimstad, Norway. E-mail:
geir.hovland@uia.no
3Department of Engineering Cybernetics, Norwegian University of Science and Technology, NO-7491 Trondheim,
Norway. E-mail: from@itk.ntnu.no
Abstract
This paper considers current trends in academic research and publication, in particular as seen from the
control community. The introduction of Web 2.0 applications for scientists and engineers is currently
changing the way research is being conducted. In the near future, participants in the research community
will be able to share ideas, data and results like never before. They will also be able to manage the rapidly
increasing amount of scientific information much more effectively than today through collaborative efforts
enabled by the new Internet tools. However, an important premise for such a development is the availability
of research material. Many research results are currently shielded behind expensive subscription schemes
that impede the sharing of information. At the same time, an increasing amount of research is being
published through open access channels with unrestricted availability. Interestingly, recent studies show
that such policies contribute to an increased number of citations compared to the pay-based alternatives.
In sum, the parallel development of new tools for research collaboration and an increased access to research
material may fundamentally transform the way research is going to be conducted in the future.
Keywords: Web 2.0, Science 2.0, Open Access, Journal Ranking, Print On Demand, Publication Methods
1 Introduction
Many important activities related to science and re-
search are currently located in the offline world of yes-
terday and have not kept up with an increasingly on-
line reality. The possibilities offered by the Internet
and its World Wide Web have not been fully taken
advantage of by the scientific communities around the
world, including those involved with the fields of mod-
eling, identification and control. Although on a digi-
tal format, most new research results are still reported
through traditional outlets as if paper was the only
publication medium. Despite having access to e-mail
and instant messaging, most researchers are still collab-
orating within relatively small and confined research
groups. Situated at the beginning of the 21st century,
it seems strange that key scientific activities such as
those related to publication and collaboration haven’t
changed much since the middle of the 20th century.
At the same time, a few large publishing conglomer-
ates have spent the last decades buying up small but
prestigious journals in order to create monopoly-like
situations regarding research material. This situation
has enabled them to significantly increase the subscrip-
tion fees for providing access to such material to univer-
sities and research institutions. Through a remarkable
publication cycle, work primarily funded by the public
thus ends up not being publicly available.
doi:10.4173/mic.2009.3.8
c ? 2009 Norwegian Society of Automatic Control

Page 2

Modeling, Identification and Control
Enter Science 2.0 and Open Access. Promising to
address the above-mentioned issues by unleashing the
full potential of the Internet, these concepts are chosen
as the topics of this paper, whose goal is to introduce
the control community to exciting new developments
which can fundamentally change our work day. Key-
words are access, collaboration and content.
2 Science 2.0
Wikipedia defines the term Web 2.0 as “...commonly
associated with web applications which facilitate in-
teractive information sharing, interoperability, user-
centered design and collaboration on the World Wide
Web.” (Wikipedia, 2009a).
such applications include blogs, social-networking sites,
video-sharing sites and wikis.
users to change their content and interact with other
users, as opposed to just being passive receivers of pro-
vided information. In fact, the user-generated encyclo-
pedia Wikipedia itself represents one of the most used
Web 2.0 applications around.
Relevant examples of
These websites allow
2.1 Web 2.0
The most popular social-networking site in the world
today is Facebook, which currently has over 300 mil-
lion active users. These users can create their own,
personal profiles where they can publish information
about themselves. However, the most popular features
relate to how they can interact with each other by ex-
changing messages, sharing pictures and videos, play-
ing games, etc.
LinkedIn represents the business-equivalent of Face-
book and is used for professional networking. It pro-
vides its current user base of over 50 million people
the possibility of creating personal profiles and sharing
work-related information such as job and business op-
portunities. Companies and professional organizations
also have a presence on LinkedIn.
Among the video-sharing sites, YouTube tops the
list. It lets users upload, view and share videos, and
has revolutionized the way in which such content has
become available on the web. In 2008, YouTube was
awarded a Peabody Award as “...an ever-expanding
archive-cum-bulletin board that both embodies and pro-
motes democracy.” (Wikipedia, 2009b).
Other popular and interactive websites include the
Internet Movie Database (IMDb), which provides on-
line information about movies, televisions shows and
actors. Registered users are able to rate movies and
write reviews, which other users find useful when eval-
uating whether to watch a specific movie or not.
What all these websites have in common is the
framework they provide for ordinary people to easily
generate and share content with each other within spe-
cific areas of interest.This feature is also the key
to their enormous success.
vard Law School professor and Creative Commons co-
founder Lawrence Lessig, this development enables the
return of a read-write culture, as opposed to the read-
only culture that dominated the 20th century (Lessig,
2009). People no longer only consume but also become
producers of creative content.
How does this web-based development relate to the
world of academic research and publication, where sci-
entists have always been both consumers and produc-
ers? It turns out that it represents a unique opportu-
nity which is just beginning to take shape.
As pointed out by Har-
2.2 Web 2.0 for Scientists
In 1945, having coordinated the activities of 6000 lead-
ing American scientists during the Second World War,
the great science administrator Vannevar Bush wrote
(Bush, July 1945): “There is a growing mountain of
research. But there is increased evidence that we are
being bogged down today as specialization extends. The
investigator is staggered by the findings and conclusions
of thousands of other workers – conclusions which he
cannot find time to grasp, much less to remember, as
they appear.” Continuing: “Professionally our methods
of transmitting and reviewing the results of research are
generations old and by now are totally inadequate for
their purpose. If the aggregate time spent in writing
scholarly works and in reading them could be evaluated,
the ratio between these amounts of time might well be
startling.” Then concluding that: “The difficulty seems
to be ... that publication has been extended far beyond
our present ability to make real use of the record.”
To help solve such problems, Bush suggested the de-
velopment of a device he called a “memex”, coined as
a combination of the words “memory” and “index”,
which he defined as “...a device in which an individ-
ual stores all his books, records, and communications,
and which is mechanized so that it may be consulted
with exceeding speed and flexibility.”
Today, scientists have much easier access to research
material than in 1945, in particular due to the Internet
and its World Wide Web, which provides memex-like
services to all. Conferences and journals make their
content available online for convenient and easy down-
load for subscribers, and search engines such as Google
make it easier than ever to find new material.
However, the amount of material is also growing as
never before. With an emphasis on quantity over qual-
ity, current financing schemes encourage scientists to
182

Page 3

Breivik, Hovland and From, “Trends in Research and Publication: Science 2.0 and Open Access”
publish more often and with fewer increments in re-
sults between successive publications.
With such opposite trends working against each
other, the challenge for any scientist or researcher be-
comes the same as in Bush’s days: To locate relevant
literature as fast as possible.
In solving this problem, inspiration can be sought
from the dominating Web 2.0 applications of today.
Imagine a mix of Facebook, LinkedIn, YouTube and
IMDb, only specifically tailored for scientists. A frame-
work for user-generated content, supplied by the scien-
tists themselves. A way for scientists to conveniently
share and organize information about their work. A
tool for them to collaborate in classifying, reviewing
and managing the large and rapidly increasing body of
research material in their respective fields of work.
Launched in 2005 by graduate students collaborat-
ing on synthetic biology, the MIT-based OpenWetWare
project shows how Web 2.0 infrastructure can be used
advantageously for research purposes (Waldrop, 2008).
A collaborative wiki, OpenWetWare has rapidly grown
into a social network serving a broad range of biologists
and biological engineers. Laboratories on five conti-
nents currently use the website, it encompasses dozens
of courses and interest groups, and hundreds of discus-
sions about lab protocols and techniques are ongoing.
According to lab manager Maureen Hoatlin: “I came
to love the interaction, the fact that people in other labs
could comment on what we do and vice versa. When
I see how fast that is, and its power to move science
forward–there is nothing like it.”
Adressing concerns about possible vandalism, Open-
WetWare allows users to make changes only after they
have registered and established their affiliation with
a legitimate research institution. Regarding the dan-
ger of getting scooped by someone stealing ideas from
a wiki, chemist Jean-Claude Bradley believes that:
“The Web provides better protection than the tradi-
tional journal system.Every change on a wiki gets
a timestamp, so if someone actually did try to scoop
you, it would be very easy to prove your priority–and
to embarrass them. I think that’s really what is going
to drive open science: the fear factor. If you wait for
the journals, your work won’t appear for another six
to nine months. But with open science, your claim to
priority is out there right away.”
More recent initiatives include the scientific network-
ing site ResearchGATE, which was launched in 2008
and whose underlying motivation is stated as: “The
vision of Science 2.0 is promising: Communication be-
tween scientists will accelerate the distribution of new
knowledge. ... Science is collaboration, so scientific
social networks will facilitate and improve the way sci-
entists collaborate.” Another website in the same vein
is Mendeley, which started out as a tool for manag-
ing and sharing research papers, but now also has
evolved into providing online collaboration and social-
networking features for researchers. Additional facil-
itators for web-based communication and collabora-
tion include Google Wave, which is designed as a mix
between e-mail and instant messaging. According to
Google, a wave “...can be both a conversation and a
document where people can discuss and work together
using richly formatted text, photos, videos, maps, and
more.”
Combined, these tools can help scientists both man-
age their research literature better as well as connect
and interact efficiently with colleagues all around the
world. The sharing of ideas, data and results represents
one of the key cornerstones of science, and the rapid
development of Web 2.0 applications for scientific pur-
poses enables more efficient sharing than ever before.
Ultimately, the development of Science 2.0 applications
recognizes the fact that individual researchers are col-
lectively better equipped to organize information and
connect the important dots than a few selected pro-
fessionals who work full time with such endeavors but
who have obvious restrictions on their level of overview,
personal preferences and work capacity.
Science 2.0, also frequently referred to as open sci-
ence or open research, is nevertheless still only in its
infancy. However, according to online community man-
ager Bora Zivkovic at the open access resource PLoS
ONE, experiments in Science 2.0 are currently prolifer-
ating so fast that it’s impossible to keep track of them
(Waldrop, 2008): “It’s a Darwinian process.
99 percent of these ideas are going to die. But some
will emerge and spread.” Cancer researcher Bill Hooker
adds: “I wouldn’t like to predict where all this is going
to go. But I’d be happy to bet that we’re going to like
it when we get there.”
About
3 Open Access
“Writing researchers get paid by the public, their ed-
itors get paid by the public. Afterwards, the publish-
ers are left with the copyright to the research articles.”
This citation is taken from a news article in the Nor-
wegian newspaper Aftenposten focusing on the current
scientific publication regime (Bjørkeng, 2009). Accord-
ing to the article, it is not unusual that the largest sci-
entific publishing houses achieve a 30% profit of their
turnover.In comparison, the rest of the publishing
business must typically settle with 3%.
During recent years, the largest academic publish-
ers have significantly turned up the price of their ser-
vices, forcing university libraries to spend up to 70-80%
of their budgets on electronic resources. According to
183

Page 4

Modeling, Identification and Control
(Gripsrud, 2006), the University of Bergen now spends
as much as 99% of its purchasing budget in medicine
and natural science on electronic journal subscriptions.
Librarians Kari Garnes and Ole Evensen comment on
the development: “We are concerned ... The journals
are eating up our economy while also putting severe
limitations on our rights.” In the period 1990-2005, the
price of the major journals increased by 200% while the
price index only rose by 57% (Gripsrud, 2006).
According to librarian Leif Longva at the University
of Tromsø (Bjørkeng, 2009): “The problem is that jour-
nal publishing in academia is still stuck in a model that
goes back to the time of paper-based journals. Then
it was quite reasonable with subscriptions. However,
when publishing on the web, there is no reason why any-
body who wants access to scientific articles shouldn’t get
it. It’s primarily publicly funded research we are talk-
ing about here, and then it is optimal, seen from the
viewpoint of society, that the cost of reading is zero.”
What Longva is calling for is referred to as open
access (OA) publishing.In general, OA means free
access to any kind of electronic material available on
the Internet. In a scientific context, OA publications
typically refer to freely available scholarly work as rep-
resented by scientific articles and books. Such publi-
cations can thus be downloaded by anyone, anywhere,
anytime, which matches perfectly with the collabora-
tive vision of Science 2.0.
3.1 Open Access and Science 2.0
A necessary condition for the arrival and impact of Sci-
ence 2.0 is that scientists are allowed to publish and ex-
change their research material freely. As already men-
tioned, most of their scientific output is currently pub-
lished through channels with restricted access, requir-
ing increasingly expensive subscriptions to gain admis-
sion. This regime means that researchers working at
institutions which cannot afford the subscription fees
are effectively left out in the cold. In particular, this
situation affects researchers in developing countries.
However, although OA is necessary for Science 2.0,
Science 2.0 will also support OA. When scientists catch
on and increasingly start using the new web-based ap-
plications as part of their daily routine, they will see
the need for and demand an increase in freely available
material. Science 2.0 and OA will thus help push each
other forward to the benefit of both.
3.2 Open Access Publications
At the moment, open access publications can be di-
vided into two main categories:
1. Electronic archives established by individual uni-
versities or research institutions which offer open
access to the publications of their employees.
2. Peer-reviewed electronic journals which provide
freely available articles for download.
Regarding institutional archives, two main factors
currently limit their use. First, the institutions need
exemption from the copyright of the publishers in
order to provide free access to the articles.
ever, today over 90% of journals permit such archiv-
ing (Hugubakken, 2006). Second, encouragement to
deposit articles in institutional archives has so far not
been effective. International experience shows that re-
quirements will have a much larger impact (Norges
Forskningsr˚ ad, 2009).
Regarding newly established OA journals, a limiting
factor is their acceptance as valid publication channels.
The Norwegian DBH database (Database for statistikk
om høgre utdanning) lists all accepted academic publi-
cation channels in Norway. As of June 2009, only about
150 of a total of 23 000 publication channels were regis-
tered as open access. On the other hand, the Directory
of Open Access Journals (DOAJ) currently lists more
than 4300 OA journals worldwide. While it is expected
that the number of OA journals will increase sharply
in the future, the increase of such journals in the DBH
database is predicted to be smaller. It takes time for
a new journal to build its reputation, especially since
most researchers are required by their institutions to
publish in journals that are already accepted as valid
publication channels. The growth of OA journals in
the DBH database is thus more likely to come from
established journals converting to open access.
How-
3.3 Funding of Open Access
In the coming years, national funding agencies are ex-
pected to play a more important role regarding open
access policies by demanding that publicly funded re-
search should be made freely available. As an example,
Det Frie Forskningsr˚ ad | Kultur og Kommunikation
(FKK) in Denmark already requires journals receiv-
ing funding to become OA within one year. Also, the
Swedish research council Vetenskapsr˚ adet recently an-
nounced that all researchers who receive funding must
guarantee that their publications are available as OA
within maximum six months of publication. Similarly,
on January 29th 2009 the Norwegian Research Coun-
cil (Norges forskningsr˚ ad, NFR) announced new OA
guidelines for scientific publications, where all funded
researchers are required to make their results accessi-
ble either through instutional archives or through OA
journals (Norges Forskningsr˚ ad, 2009).
libraries have already supported OA for many years
since its growth will help reduce their subscription fees.
Institutional
184

Page 5

Breivik, Hovland and From, “Trends in Research and Publication: Science 2.0 and Open Access”
On September 14th 2009, the 5 US universities
Cornell, Dartmouth, Harvard, Massachusetts Insti-
tute of Technology (MIT) and the University of Cal-
ifornia, Berkeley (UCB) announced joint support for
OA publications (Richards, 2009).
Thomas Leonard, librarian at UCB: “Publishers and
researchers know that it has never been easier to share
the best work they produce with the world. But they
also know that their traditional business model is cre-
ating new walls around discoveries. Universities can
really help take down these walls and the OA compact
is a highly significant tool for the job.”
The recent economic downturn also underlines the
importance of OA publications.
straining library resources, subscription and licensing
fees for journals have come under increasing scrutiny.
Alternative means for providing access to vital intellec-
tual content are thus explored. According to Harvard
computer science professor Stuart Shieber (Richards,
2009): “Universities and funding agencies ought to pro-
vide equitable support for OA publishing by subsidizing
the processing fees that faculty incur when contribut-
ing to OA publications. Right now, these fees are rel-
atively rare. But if the research community supports
OA publishing and it gains in importance as we believe
that it will, those fees could aggregate substantially over
time. The compact ensures that support is available to
eliminate these processing fees as a disincentive to OA
publishing.”
However, the future funding mechanisms for OA
publications remain an open issue. With a full tran-
sition to OA, publication costs must be shifted from
the subscribers to either the funding agencies, research
institutions, authors or to other sponsors. It will be a
vital challenge in the future to ensure that such costs
do not increase to a level where they prevent certain
scientists and researchers from publishing their work
in high-rank journals in the same way that large sub-
scription fees prevent some of them access to the work
of their colleagues today.
According to
With budget cuts
3.4 Journal Ranking and Citations
An increasing trend in previous years is to quantify
and rank the research impact of both journals and in-
dividual researchers. Two common measures are the
ISI Impact Factor and the h-index.
The ISI Impact Factor for a given year is a mea-
sure of the average number of citations per article for
a journal during the previous two years, and the idea
dates back to Garfield (1955). The h-index is more re-
cent and can be defined as follows (Hirsch, 2005): A
journal has index h if h of all Nppapers have at least
h citations each, and the other (Np− h) papers have
at most h citations each. While the impact factor is
most commonly used for journals, the h-index is used
for both journals and individual researchers.
Table 1 shows some selected journals in the field of
automation and control ranked by the h-index. For ex-
ample, Automatica has a h-index of 124, which means
that 124 articles have been cited 124 times or more.
The h-index is a more stable measure of journal impact
than the ISI Impact Factor, which only considers cita-
tions during the last two years. Also, while the impact
factor can change significantly between two consecu-
tive years, the h-index can never decrease. Note that
MIC’s impact factor has not been calculated in 2007
and 2008, but will reappear in 2009.
Table 2 shows the most cited papers during the last
decade from the top 10 journals listed in Table 1. Not
surprisingly, the highest ranked journals also have the
most cited papers. The research topics dominating the
citations in the area of automation and control during
the last decade include control of constrained systems,
multi-agent systems and time-delay systems.
Since 2005, a significant amount of the research fund-
ing at Norwegian universities has been directly linked
to how much the academic staff publish in certain
accredited journals. As mentioned earlier, the DBH
database lists all the accepted publication channels in
Norway. The channels are ranked in an incentive hier-
archy based on data from the ISI Web of Knowledge,
using measures such as the ISI Impact Factor. Natu-
rally, individual scientists and researchers aim at ad-
vancing their own careers by publishing in the high-
est ranked journals.They cannot afford to publish
in newly established OA journals with low ranking on
purely idealistic grounds. The current incentive sys-
tem thus favors the established journal hierarchy and
prevents the growth of OA publications, thereby con-
tradicting the new OA requirements imposed by the
research council NFR.
3.5 Open Access and Citations
The effect of open access on the number of citations
has received a lot of attention after the OA journals
appeared on the web in the 1990s. Most reports tend
to support the view of an OA citation advantage, which
claims that OA publications cause more citations than
comparable non-OA publications.
Early work supporting this citation advantage is pre-
sented in (Lawrence, 2001), where it is found that con-
ference papers which are freely available for download
have greater impact than those that are not. A signif-
icant 150% citation advantage is claimed. A possible
explanation for this finding might be that freely dis-
tributed papers are more visible. As access becomes
more convenient, papers are more frequently down-
loaded and their usage increases. This effect helps the
185

Page 6

Modeling, Identification and Control
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Journal
IEEE Transactions on Automatic Control∗
Automatica
IEEE Trans. on Systems, Man and Cybernetics
SIAM Journal on Applied Mathematics
International Journal of Control
Chemometrics and Intelligent Laboratory Systems
IEEE Transactions on Industrial Electronics
Systems & Control Letters
SIAM Journal on Control and Optimization
Journal of Optimization Theory and Applications
IEEE Transactions on Control Systems Technology
IEEE Control Systems Magazine
ASME Journal of Dyn. Sys. Measurement and Control
IEEE-ASME Transactions on Mechatronics
IEE Proceedings - Control Theory and Applications
International Journal of Robust and Nonlinear Control
Robotics and Autonomous Systems
Journal of the Franklin Inst. - Eng. and Appl. Math.
Journal of Machine Learning Research
Control Engineering Practice
International Journal of Systems Science
Journal of Process Control
Mathematics of Control Signals and Systems
Engineering Applications of Artificial Intelligence
International Journal of Advanced Manufacturing
IEEE Robotics & Automation Magazine
Mechatronics
IEEE Transactions on Robotics
Int. Journal of Adaptive Control and Signal Proc.
Autonomous Agents and Multi-Agent Systems
Discrete Event Dyn. Systems - Theory and Appl.
Optimal Control Applications & Methods
European Journal of Control
ESAIM: Control Optimisation and Calculus of Var.
Journal of Vibration and Control
SIAM Journal on Applied Dynamical Systems
Control and Cybernetics
Proc. of the Inst. of Mechanical Engineers - Part I
Trans. of the Institute of Measurement and Control
Modeling, Identification and Control
IEEE Trans. on Automation Science and Engineering
Asian Journal of Control
Annual Reviews in Control
Int. Journal of Control Automation and Systems
Simulation Modelling Practice and Theory
Measurement & Control - UK
Assembly Automation
Automation and Remote Control∗
Intelligent Automation and Soft Computing
Journal of Dynamical and Control Systems
Table 1: Some selected journals in the area of automation and control ranked by the h-index as calculated by the
ISI Web of Science. IF08 stands for Impact Factor 2008, TA is the total number of articles and TC is
the total number of citations for each journal. Both TA and TC were recorded on the date 15.11.2009.
DBH signifies the publication level in the Norwegian DBH database.∗Publication data for this journal
only available from the year 1990.
h
134
124
95
91
89
73
67
66
61
60
50
48
48
40
39
39
38
38
37
37
37
36
28
27
27
26
26
24
24
22
21
20
19
18
18
17
16
15
14
14
13
13
12
11
11
11
11
11
9
8
IF08
3.293
3.178
2.361
1.110
1.130
1.940
5.468
2.073
1.517
0.860
2.130
2.827
0.753
1.614
2.106
1.560
1.214
0.616
3.116
1.871
0.634
1.606
1.500
1.397
0.743
3.000
1.434
2.656
1.403
2.125
0.974
0.833
1.013
0.787
0.656
1.211
0.689
0.322
0.500
TA
6208
5960
3507
4556
7768
2351
4008
2311
1995
4393
1406
1105
2797
846
1076
1130
1231
3639
626
2218
4815
990
321
1310
4287
594
1115
630
818
234
263
747
663
385
861
253
589
740
441
440
286
430
127
516
562
806
1230
4003
360
199
TC
121613
111492
72774
64636
77733
31670
43089
28632
24637
33669
16321
11202
19502
8934
9849
9249
10367
13568
7538
14166
17773
9126
3757
5956
13302
3177
5818
3156
4078
1934
1755
2875
1608
1775
2596
1219
1288
1647
1027
869
808
706
550
928
855
540
876
1266
479
366
TC/TA
19.59
18.71
20.75
14.19
10.01
13.47
10.75
12.39
12.35
7.66
11.61
10.14
6.97
10.56
9.15
8.18
8.42
3.73
12.04
6.39
3.69
9.22
11.70
4.55
3.10
5.35
5.22
5.01
4.99
8.26
6.67
3.85
2.43
4.61
3.02
4.82
2.19
2.23
2.33
1.98
2.83
1.64
4.33
1.80
1.52
0.67
0.71
0.32
1.33
1.84
DBH
2
2
2
2
1
1
1
2
2
1
1
1
1
1
1
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1.929
0.562
1.109
0.590
0.586
0.493
0.382
0.236
0.224
0.368
186

Page 7

Breivik, Hovland and From, “Trends in Research and Publication: Science 2.0 and Open Access”
#
1
Citations
1053
Title
Constrained model predictive
control: Stability and optimality
Coordination of groups of mobile autonomous
agents using nearest neighbor rules
Consensus problems in networks of agents
with switching topology and time-delays
Multilevel inverters: A survey of topo-
logies, controls, and applications
The explicit linear quadratic regulator
for constrained systems
A new method for the nonlinear
transformation of means and covariances
in filters and estimators
Time-delay systems: An overview of
some recent advances and open problems
A new discrete-time robust stability
condition
Stability of multiagent systems with
time-dependent communication links
Stabilization of linear systems with
limited information
Authors
Mayne et al.
Journal
AUT, 2000
Vol. 36, No. 6
TAC, 2003
Vol. 48, No. 6
TAC, 2004
Vol. 49, No. 9
TIE, 2002
Vol. 49, No. 4
AUT, 2002
Vol. 38, No. 1
TAC, 2000
Vol. 45, No. 3
2 822Jadbabaie et al.
3 590 Olfati-Saber et al.
4 491Rodriquez et al.
5424Bemporad et al.
6367 Julier et al.
7364Richard AUT, 2003
Vol. 39, No. 10
SCL, 1999
Vol. 37, No. 4
TAC, 2005
Vol. 50, No. 2
TAC, 2001
Vol. 46, No. 9
8318de Oliveira et al.
9312Moreau et al.
10290Elia et al.
Table 2: Ranking of the highest cited papers (by ISI Web of Science on 17.11.2009) in the top 10 journals
from Table 1 during the last decade. AUT=Automatica, TAC=IEEE Transactions on Automatic
Control, SCL=Systems & Control Letters, TIE=IEEE Transactions on Industrial Electronics.
authors obtain more citations and also to obtain them
quicker. However, other studies point out alternative
explanations to the increase in citations such as self-
selection bias since authors typically only distribute
their best work, and the early-view effect since online
publication dates in general are earlier than for print
publications (Frandsen, 2009).
Concerning download statistics, Watson (2009) re-
ports that they provide a useful indicator, two years
in advance, of eventual citations, and that correlation
between downloads and citations is as high as 0.74.
These results are obtained for the open access Journal
of Vision.
In many fields of mathematics and physics almost
all scientific papers are placed in the arXiv reposi-
tory, which is an online archive for electronic preprints.
Started back in 1991, the arXiv was one of the precipi-
tating factors that led to the OA movement. Henneken
et al. (2006) find that papers submitted to the arXiv
are more cited than comparable articles not present in
the archive, and also that, in general, the most impor-
tant papers are submitted to the arXiv before they are
published elsewhere.
In the most extensive study to date, covering around
26 million articles from more than 8000 journals which
went online before 2006, Evans and Reimer (2009) find
that making articles freely available on the web partic-
ularly increase the number of citations in developing
countries. Given that scientists and researchers from
such countries have limited access to the pay-based al-
ternatives, these results are not surprising. Hence, the
proliferation of open access publications is expected to
contribute strongly to reduce the current information
gap which exists between the developed and developing
parts of the world.
3.6 Statistics for 3 Selected OA Journals
Table 3 shows some statistics for 3 selected OA journals
within the fields of modeling, identification and control.
Journal
JMLR
MIC
IJCCC
Est.
2001
1980
2006
TA
626
440
286
TC
7538
869
69
AC
837.56
28.97
17.25
h
37
14
3
Table 3: Statistics for 3 different OA journals recorded
on 17.11.2009. TA is the total number of ar-
ticles, TC the total number of citations and
AC the average number of citations per year.
The International Journal of Computers, Commu-
nications & Control (IJCCC) is a relatively new OA
journal which was started in 2006. With 286 articles
and a h-index of only 3, IJCCC illustrates how diffi-
cult it is for a newly established OA journal to gain
reputation. Having only recently converted to OA, the
citation data and h-index for MIC are included for fu-
187

Page 8

Modeling, Identification and Control
ture reference to evaluate the transition effects.
The Journal of Machine Learning Research (JMLR)
is an OA journal which was founded in 2000 as an
open access alternative to the well-established Machine
Learning Journal (MLJ). In 2001, forty editors of MLJ
resigned to support JMLR, signing a letter stating that
“...our resignation from the editorial board of MLJ re-
flects our belief that journals should principally serve
the needs of the intellectual community, in particular by
providing the immediate and universal access to jour-
nal articles that modern technology supports, and doing
so at a cost that excludes no one. We are excited about
JMLR, which provides this access and does so uncon-
ditionally.”
Hosted at MIT, JMLR is an example of an OA jour-
nal in a field close to MIC which has become a high im-
pact publication in a relatively short time. Figures 1
and 2 illustrate its development. While the number
of published articles since 2004 has been almost con-
stant at around 80 per year, the number of citations
has grown exponentially. JMLR is currently the only
level 2 OA journal in the Norwegian DBH database in
a field close to MIC. Likely explanations for the rapid
and successful development of JMLR are acceptance
and support from the research community in machine
learning in combination with the quality standards im-
plemented by the editorial team.
3.7 Print On Demand
Due to increasing production costs, the financial sta-
bility of journals published by scientific societies is in
jeopardy and the journals need to find ways to survive
(Castillo, 2009): “The problems faced are mainly those
of decreasing membership allocations, lack of growth
of institutional subscriptions, diminishing advertise-
ments, and pressure for open access. Faced with these
situations, a journal can raise subscription rates or
switch to an author/institution pay model. With both
options, the journal risks alienating its audience and
contributors. To reduce cost, some journals opt to be-
come electronic only. Fortunately, a third model exists,
and that is print on demand (POD). By using POD, a
reader or an institution can pay extra to have a print
version of a journal, whereas others may choose on-
line-only subscriptions.”
POD is a technology for which new copies of a docu-
ment are not printed until an order is received. Based
on preliminary figures from US publishers, Bowker
(2009) projected that US booktitle output in 2008
decreased by 3.2% to 275 232 new titles and edi-
tions, down from the 284 370 that were published in
2007.Despite the decline in traditional book pub-
lishing, there was an extraordinary year of growth in
the reported number of POD books published in 2008.
JMLR: Published Articles Per Year
0
20
40
60
80
100
120
140
2001 200220032004 200520062007 20082009
Year
Figure 1: A stabilizing number of published articles for
the Journal of Machine Learning Research.
JMLR: Citations Per Year
0
500
1000
1500
2000
2500
3000
20012002 2003 2004 20052006200720082009
Year
Figure 2: An exponentially increasing number of cita-
tions for the Journal of Machine Learning
Research.
Bowker projected that 285 394 POD books were pub-
lished this year, a 132% increase over the 2007 final
total of 123 276 titles. This was the second consecu-
tive year of triple-digit growth in the POD segment,
which in 2008 was 462% above the 2006 level. The
statistics for 2008 benchmark a historic development
in the US book publishing industry as the point was
crossed where the number of POD books exceeded the
number of traditional books entering the market.
POD technology fits particularly well with the on-
line, electronic nature of OA publications by providing
a flexible option to produce paper copies on demand,
and will probably help further the OA development in
the future. As a recent OA journal, MIC was very
quick to offer POD as an option for its readers.
4 New Publication Methods
Traditionally, the medium of scientific publications has
been restricted to paper. Articles have been published
in conference proceedings or journals, with strict page
188

Page 9

Breivik, Hovland and From, “Trends in Research and Publication: Science 2.0 and Open Access”
limitations and formatting rules. Even though elec-
tronic technologies render possible a much broader
spectrum of publication types, the vast majority of
publications are still made as if paper was the only
available medium. However, there are signs that this
practice is about to change.
SciVee is a video-sharing website for scientists, the
science-equivalent of YouTube so to speak, sporting the
slogan “Make Your Research Known.” Researchers can
upload, view and share science-related video clips and
connect them to various scientific material through a
synchronization feature. For example, synchronization
with an article is called a “pubcast” while synchroniza-
tion with a poster is called a “postercast”. Content
ranges from the elementary-school level and up to the
levels of university and state-of-the-art research.
In (Fink and Bourne, 2007), key contributors from
the SciVee team state: “We believe that the research
community is ripe for a revolution in scientific commu-
nication and that the current generation of scientists
will be the one to push it forward. These scientists ...
have a natural ability to do science in an electronic en-
vironment without the need for printed publications or
static documents and, in fact, can feel quite limited by
the traditional format of a publication. Perhaps most
importantly, they appreciate that the sheer amount of
data and the number of publications is prohibitive to
the traditional methods of keeping current with the lit-
erature.” The authors also claim that: “SciVee makes
it easier and faster to keep up with current literature by
delivering the key points of articles in a portable and
enjoyable medium. A reader can interact with several
articles using this website in the time it would take to
read a single full article in the traditional way.” Fink
and Bourne then conclude with their belief that: “...re-
vitalizing journal articles will have a significant impact
on the scientific community.
format no longer effectively supports the research in
the electronic age.”
Conferences are also starting to let authors upload
videos and animations in addition to their traditional
written contributions. Including MIC, journals now al-
low color pictures in articles as well as clickable DOI
and HTTP links embedded in the PDF files. In the
future, even more interactive articles will probably
emerge, containing not only text and pictures, but also
animations, movies and sound embedded in the docu-
ments. As with SciVee, author presentations can then
be synchronized with such articles to achieve a maxi-
mum dissemination effect. In the electronic world the
possibilities thus seem endless.
naturally be taken to develop standards for such con-
tent and to uphold scientific quality all along the way.
Interestingly, it seems that scientists from some
The traditional article
However, care must
particular research fields are fronting this develop-
ment. Founding members of several of the Science
2.0-supporting websites, including OpenWetWare, Re-
searchGATE, Mendeley and SciVee, have a background
from biology and genetics research, where efficient han-
dling of huge amounts of data is required. These sci-
entists have probably felt a special need to change the
way research is performed and disseminated in the on-
line world of today.
Although still mostly keeping to traditional publica-
tion methods and formats, researchers within the field
of automation and control have also begun to take ad-
vantage of the possibilities brought on by the web. For
example, Karl˚ Astr¨ om and Richard Murray have re-
cently authored the book Feedback Systems: An Intro-
duction for Scientists and Engineers and made it freely
available through a wiki (Murray, 2009). The wiki also
contains lecture materials, source code for relevant soft-
ware, supplemental examples and exercises, as well as
a FAQ database. In addition, the book is available in
print for purchase (˚ Astr¨ om and Murray, 2008).
In the preface of their book, the authors state that:
“This book provides an introduction to the basic prin-
ciples and tools for the design and analysis of feedback
systems. It is intended to serve a diverse audience of
scientists and engineers who are interested in under-
standing and utilizing feedback in physical, biological,
information and social systems.” Taking note of the
fact that research is becoming increasingly multidisci-
plinary,˚ Astr¨ om and Murray have thus written a book
which aims to explain the principles of automatic con-
trol to a broad audience, and they have chosen to re-
lease it through a wiki-based framework which allows
for convenient and unrestricted dissemination. Their
initiative is in spirit with the concepts of open access
and Science 2.0, and hopefully more efforts of a similar
kind will be seen in the near future.
5 Conclusions
This paper has reviewed some recent trends in research
and publication that have the potential to fundamen-
tally transform the way researchers, scientists and en-
gineers carry out their work.
tools for literature management and research collab-
oration are currently being introduced, following the
development seen in the general population concerning
Web 2.0 applications such as blogs, wikis and social-
networking tools.
While most scientists during the last two decades
have been accustomed to perform their work using
computers, their methods of publication and collab-
oration have not adjusted accordingly. However, led
by researchers within biology and genetics who daily
New and innovative
189

Page 10

Modeling, Identification and Control
struggle with large amounts of data and a radically in-
creasing body of research literature, new web-based ap-
plications are being developed to facilitate efficient on-
line collaboration between scientists around the world.
Such a development relies partly on open access pub-
lication policies.
Currently, OA publications represent a break with
the traditional pay-based channels by making research
results freely available on the Internet. These publi-
cations increase the online visibility and seem to give
a rise in citations compared to their pay-based coun-
terparts. The MIT-hosted journal JMLR demonstrates
that new OA journals can rapidly challenge established
journals in terms of citations and other ranking indices.
Having recently converted to OA, MIC also aims to
play a role in this development. Its goal is to construc-
tively contribute within the fields of modeling, identi-
fication and control for the benefit of all.
Recognizing the fantastic possibilities brought about
by the Internet, the movements of Science 2.0 and
Open Access may very well represent the mutually in-
teracting dynamics required to bring science and re-
search into the 21st century.
References
Bjørkeng, P. K. Forskere føler at de blir fl˚ add. Aften-
posten, 2009. URL http://www.aftenposten.no/
kul_und/article2949887.ece.
Bowker. Statistics on U.S. book publishing for 2008,
Press release, May 19.
bowker.com.
2009.URL http://www.
Bush, V. As we may think. Atlantic Monthly, July
1945. pages 101–108.
Castillo, M. Print on demand and scientific publishing.
American Journal of Neuroradiology, 2009. 30(1):1.
doi:10.3174/ajnr.A1298.
Evans, J. A. and Reimer, J.
global participation in science.
323(5917):1025. doi:10.1126/science.1154562.
Open access and
Science, 2009.
Fink, J. L. and Bourne, P. E. Reinventing scholarly
communication for the electronic age.
Quarterly, 2007. 3(3):26–31.
CTWatch
Frandsen, T. F. Scholarly Communication Changing:
The Implications of Open Access. Ph.D. thesis, Royal
School of Library and Inf. Science, Denmark, 2009.
Garfield,
new dimension in documentation through asso-
ciation of ideas.Science, 1955.
doi:10.1126/science.122.3159.108.
E. Citation indexes to science:A
122:108–111.
Gripsrud, S.
2006. URL http://nyheter.uib.no/?modus=vis_
nyhet&id=30523.
Kjenner ikke BORA. P˚ a Høyden,
Henneken, E. A., Kurtz, M. J., Eichhorn, G., Acco-
mazzi, A., Grant, C., Thompson, D., and Murray,
S. S. Effect of e-printing on citation rates in astron-
omy and physics. J. of Electronic Publ., 2006. 9(2).
Hirsch, J. E. An index to quantify an individual’s sci-
entific research output. Proc. National Academy of
Sciences of the USA, 2005. 102(46):16569–16572.
Hugubakken,
med p˚ a opprør.
http://www.universitetsavisa.no/ua_lesmer.
php?kategori=nyheter&dokid=43c2095edae056.
13683998.
T. Bibliotekene vil ha forskerne
Universitetsavisa, 2006.URL
Lawrence, S. Online or invisible? Nature, 2001. 411
(6837): 521.
Lessig,
laws
http://www.ted.com/talks/larry_lessig_
says_the_law_is_strangling_creativity.html.
L.
that
TED
choke
Talks:
creativity.
Larry Lessigon
2009.URL
Murray, R. M. Amwiki. 2009. URL http://www.cds.
caltech.edu/~murray/amwiki.
25-Nov-2009].
[Online; accessed
Norges Forskningsr˚ ad. Forskningsr˚ adets prinsipper for
˚ apen tilgang til vitenskapelig publisering. Vedlegg 3
til KD 29/01/2009, 2009.
Richards, P.
for open-access publication.
tions, 2009.
2009/open-access-091409.html.
Top schools announce joint support
MIT media rela-
URL http://web.mit.edu/press/
Waldrop, M. M. Science 2.0: Great new tool, or great
risk?Scientific American, 2008.
www.scientificamerican.com/article.cfm?id=
science-2-point-0-great-new-tool-or-great-risk.
URL http://
Watson, A. B. Comparing citations and downloads for
individual articles. Journal of Vision, 2009. 9(4):1–
4. doi:doi:10.1167/9.4.i.
Wikipedia. Web 2.0 — Wikipedia, The Free Encyclo-
pedia.2009a.URL http://en.wikipedia.org/
w/index.php?title=Web_2.0&oldid=326313261.
[Online; accessed 17-Nov-2009].
Wikipedia. YouTube — Wikipedia, The Free Ency-
clopedia. 2009b. URL http://en.wikipedia.org/
w/index.php?title=YouTube&oldid=325665878.
[Online; accessed 17-Nov-2009].
˚ Astr¨ om, K. J. and Murray, R. M.
tems: An Introduction for Scientists and Engineers.
Princeton University Press, 2008.
Feedback Sys-
190