Re: Separating Quality-Control Service-Providing from Document-Providing

From: Donald King <dwking_at_umich.edu>
Date: Sun, 5 Dec 1999 18:37:10 +0000

dk> Evidence suggests that the total cost of
dk> authors' time, publisher services and library (and other intermediary)
dk> services are roughly equal, but each is less than one-tenth of the cost
dk> of readers' time. Since institutions are willing to compensate authors
dk> (and referees) and pay for libraries (and publisher services through
dk> libraries), there is no particular reason that they should not directly
dk> purchase the article QC/C services; as you suggest.

sh> However, I don't suggest paying for them ON TOP OF paying
sh> Subscription/Site-License/Pay-Per-View (S/L/P) but INSTEAD!

Absolutely!! Nor do I!!!

dk> The point is that we should all support whatever distribution/access
dk> means are best for scientists and the scientific community and this may
dk> not necessarily be only one version or another. If the
dk> author/institution payment for article QC/C services is adapted, this
dk> can result in little risk, but with huge benefits.

sh> Yes, but if it is adopted ON TOP of S/L/P and without public
sh> self-archiving, it is only adding insult to injury!

But that's not my point NOR would I suggest S/L/P on top.

It seems to me that we only disagree on one major point, which is how
scientists will access/obtain articles in the near and distant future.
We totally agree that the article processing costs are what should be
the focus of attention and I like your approach to it.

Over several decades of surveying scientists by personal interviews,
focus group interviews and tens of thousands of self-administered
questionnaires (13,500 of which are readership) I have come to the
conclusion that scientists are very conscious of the "price" that they
pay for information services and information. However, the price they
pay consists of three components: (1) the purchase price, (2) the time
they spend accessing/obtaining information, and (3) the time they spend
in reading/using the information. The purchase price tends to be by far
the least amount paid and reading/using by far the greatest -- when
dollars are assigned to resources used by them.

You mention that "I cannot imagine users will prefer the for-fee
version over the for-free version, once public self-archiving makes the
latter available". However, information is never "free" because of the
time necessary to identify, locate and access/obtain the information
from whatever source. I have thoroughly examined answers concerning
thousands of readings and the multitude of combinations of means used
to identify articles (browsing, automated searches, citations, referral
by colleagues, etc.), to locate them (here, the journal title most
often leads one to sources such as one's own subscription, internal
library and departmental collections, external library collections, and
other scientists) and to access/obtain articles that have been read. In
many of the surveys I asked scientists to indicate how much time they
(or someone on their behalf) spent on each of the activities of
identifying, locating, accessing/obtaining, and photocopying articles
that are read.

The point is that, while the average amount of time for these
activities is only about 6 minutes (total) per reading, many of the
readings involve a great deal more time (depending, for example, on the
purpose of reading and the age of the articles read). It may well be
that public self-archiving will reduce scientists' time to identify,
locate and access/obtain needed articles, but I kind of doubt it.
Hopefully, I'm wrong.

I can think of many examples in which scientists exhibit a valid
economic rationale in optimizing among purchases, their time, and use
of alternative services and sources. For example, when personal
subscription prices increased they shifted their reading to library
access -- yet the average number of personal subscriptions and
proportion of readings from them go up as distance to the library
increases. They also began spending more time at the library per visit
and more time the further away they are from the library. Outside of
universities, scientists often choose to use librarians/information
specialists to conduct their automatic searches, contrary to the belief
of many at the time searching went online and, even though they were
trained to do their own searching. The principal reason given by them
is that "it saves them time", followed by "my searcher can do it faster
and better". There is a tendency for them to do "easy" searches, but
delegate the more difficult ones.

It is for these reasons in our book we say "Our studies show that both
scientists and libraries are well aware of the advantages of choosing
from among sources and they generally select them in an economically
rational manner. This bodes well for the future of electronic
publishing and its value-added capabilities because scientists and
libraries will ultimately recognize and utilize the beneficial aspects
of electronic journals and text databases".

Again, if by some great magic (and I, if possible, would gladly be the
wand waver) the authors-institutions would reimburse "suppliers" for
the QC/C costs, I believe many of the current problems of article
communication would be resolved. On the other hand, I think that we
should keep open the availability of as many distribution means as
possible in order to minimize scientists' time. I can't think of any
reason that a variety of potential distribution means would be
detrimental to the up-front payment for QC/C or the benefits derived
from this scheme. Rather, I think that up-front payment will open the
way to optimizing the overall system through inexpensive (and sometimes
"free") distribution means, where scientists can choose from among them
in such a way as to minimize their time and expenses.

Best regards,

Don.

Donald W King
<dwking_at_umich.edu>
Received on Wed Feb 10 1999 - 19:17:43 GMT