IEEE Robotics & Automation Magazine - September 2017 - 4

FROM THE EDITOR'S DESK

Publication Impact Factors
and Submission-to-Decision Times
By Bram Vanderborght

he 2017 IEEE International
Conference on Robotics and
Automation (ICRA) was impressively well organized;
congratulations to the entire organizing team! During ICRA, Eugenio
Guglielmelli (vice-president of publication activities), Frank Park [editorin-chief (EiC) of IEEE Transactions on
Robotics], and I were invited to the
Young Professionals luncheon to discuss topics related to publications in
our Society. One of the questions
raised was, "Why are the impact factors of robotics journals relatively low
compared to
other fields?" In
the 2016 inforIn the 2016
mation that was
information that
recently announced, the impact
was recently
factors of roannounced, the
botics journals
approximately
impact factors of
doubled (for
robotics journals
IEEE Robotics
and Automaapproximately
tion Magazine,
doubled.
e.g., 3,276 versus 1,822), so
there appears
to have been a correction of sorts.
We are still trying to figure out what
changed in the course of a year and
whether this is an anomaly or a
trend that we can expect to see for
some time.

Digital Object Identifier 10.1109/MRA.2017.2719418
Date of publication: 13 September 2017

IEEE ROBOTICS & AUTOMATION MAGAZINE

Nevertheless, the question remains relevant and
especially important for
those of us who must compete with researchers from
other fields for grants, promotions, and awards. The papers in
our field do not cite any fewer references, but, in many cases, conference
papers are cited instead of their extended journal versions. Most other
fields do not have our tradition of publishing high-quality, fully reviewed sixpage papers in ICRA, the IEEE/Robotics Society of Japan International
Conference on Intelligent Robots and
Systems (IROS), and other conference
proceedings. For example, there is no
equivalent of the ICRA or IROS proceedings in chemistry. Rather, only
extended abstracts are submitted and
archived for the conference; the full versions of the work are published in
archived journals. Citations, therefore,
naturally go to the journal papers, leading to the much higher impact factors
of their journals.
This state of affairs can be traced in
large part to the conference-to-journal
evolutionary publishing model that has
been popular within our community for
a long time. Initial results are typically
submitted to a workshop, then extended
to a full conference paper, and eventually to a journal article. When researchers in our community only cite
the conference paper rather than the
extended journal version, it is sometimes because they are not aware of the
existence of the extended journal version, other times it is simply because of

SEPTEMBER 2017

negligence. There may be
good reasons for citing the
conference paper-e.g.,
accurately tracing the temporal history of the past
work-but there is little reason
not to cite the extended journal version as well. This practice of citing only
the conference paper clearly impacts our
journal metrics in a negative way.
Another raised question was, "What
are the obstacles that delay the submission-to-decision times?" Throughout
our IEEE Robotics and Automation
Society (RAS) publications, we are uniformly observing that, more and more,
reviewers are declining to review papers
that match their areas of expertise and,
in some cases, never even bothering to
reply to a review request. One reason
for this phenomenon is that we are
inundated with more and more review
requests, from conferences as well as
journals. Another factor is that writing
both archival-quality conference and
journal versions of our work puts even
more demands on our time. Yet, for
many of us, it is important not only to
publish our work in journals but also to
promote our work and obtain peer
feedback at conferences. Often, our
researchers struggle with the dilemma
of how much to put into the conference
paper. Not enough and it may be
rejected, too much and there may not
be enough left for an extended journal submission.
The enormously successful IEEE
Robotics and Automation Letters was
launched in part to mitigate these
dilemmas, and it has been extremely

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2017