IEEE Solid-State Circuits Magazine - Spring 2015 - 6

a ssoc iate editor's vie w

Marcel Pelgrom

How to Stop a Tsunami

He gets up before 6 a.m. After his
jogging round, he eats a super-food
breakfast alone without waiting for
the family, as he has to be at the university before 7:30 a.m.
There, he opens his mail, deletes
four requests for reviews, and sees
with satisfaction that another paper by
his group was accepted by the "Third
Annual Conference" on "Circuits, Integration, and Systems" (choose any permutation of these terms).
This conference is held in a place
where the only link to silicon is on its
tropical beaches. But that's not a consideration for our quantity-focused
professor. With an average production of 100 papers per year, he and his
students need to start, write, review,
and submit a paper every two days. He
runs them at a schedule of 12 minutes
quality time per student per week,
and if he speeds up, a few minutes
remain to sip from his pu-erh tea. Having delegated much of his teaching to
his more experienced students, he can
attend conferences to stay in touch
with the latest hype, as that generates input for even more publications.
He goes to bed far after midnight,
dreaming of the day where his h-index
passes the 50 level.
The adage "publish or perish" is
arguably the most widely shared principle of the academic world. Until recently, publishing in a journal required
peer reviews, legal and patent offices,
and sometimes substantive or smallscale revisions. A conference submission required as many as 45 copies
sent by express mail; acceptance could
Digital Object Identifier 10.1109/MSSC.2015.2418152
Date of publication: 25 June 2015

s p r i n g 2 0 15

result in an expensive intercontinental
trip. A published journal paper could
lead to a stream of postcards requesting reprints. Publishing was a decision,
not a right. Reviewing, on the other
hand, was desirable: the reviewer belonged to a limited group of peers
who would get insight into the newest
developments.
In the decades after the Second
World War, fresh scientific ideas, concepts, and inventions sprung up in
waves. In this century, however, science
and technology have come into calmer
waters that promised the pruning of
scientific conferences and journals.
To the contrary! A tsunami of publications started when technology
removed the intrinsic costs of publication at the same time that influential
business schools began promoting
management by quantifying performance in all aspects of society. So rankings, citation lists, impact factors, and
indexes of various sorts sprung up.
And some university management
departments and funding agencies
came to believe that such counts alone
should determine prioritization.
The professors who have joined
this control loop act accordingly: if
the acceptance rate of funding proposals drops by 5x, they will submit 5x more proposals. And if more
papers are required to stay on budget, they will produce more papers.
Result: exponential growth and scientific progress drowned in a sea of
debris. Conferences run seven parallel tracks, at times with fewer audience members than speakers. Editors
can only accept 10% of submissions,
and reviewers refuse to spend time
on "insignificant" papers.

IEEE SOLID-STATE CIRCUITS MAGAZINE

A wise man in our field once said,
"No exponent lasts forever"-which
holds true for publication as well: this
exponential increase must stop. In the
long term, the reduced information
value of journals compares unfavorably with the increased cost of producing them. Compounded by open
access to publicly funded research,
this lopsided ratio has spurred large
publishing companies to introduce a
different business model: the reader
no longer pays for reading, but the
author for being able to publish. And
there is little hope that open access
will shift the quantity-quality balance in the right direction, as every
submitted paper will generate revenue for the publisher.
In a more sensible system, conference and journal publishing would
be balanced by services rendered to
the (academic) community. If scholars had to earn publication rights by
reviewing, editing, and managing submitted manuscripts-activities that
do not rank high on our professor's
agenda-the system could be rebalanced. A way to implement such a
tit-for-tat arrangement is virtual "pubcoins." A reviewer would gain a pubcoin, while a publication submission
would require four as payment. For
volunteer-driven organizations like
the IEEE, such a mechanism would
reward membership loyalty, quality,
and effort. The ridiculous number
chasing would stop, or at least abate,
and the value of papers would go up.
Our publication culture would become
healthier. And our storied professor
could have a long night's sleep.

Table of Contents for the Digital Edition of IEEE Solid-State Circuits Magazine - Spring 2015