Re: Response to lib-license email

[Robert Michaelson <rmichael@nwu.edu>]

At 02:54 PM 8/18/99 -0400, Gay Dannelly wrote:

>Colleagues,
>   Following is a message from Tom Sanville that I am forwarding, with his
>permission, in response to the email message from Bob Michaelson regarding
>cancellation of Elsevier titles and Bob's comments about OhioLINK's lack of
>cost benefit analysis.  As a member of OhioLINK and, probably, the longest
>standing member of the consortium's collections committee I wanted to be
>sure that a response was made to this list (and Tom has also replied
>directly to Bob).  In my view, we are providing extensive access to a broad
>array of titles that no single institution could afford.  In point of fact,
>our last year's usage figures indicate that user behavior, when provided
>with materials not previously available, changes dramatically.  Our data at
>present is only 15 months long and it remains to be seen how usage will
>change over a longer period of time.

When you give something for free, people will take it -- that is not a
valid measure of its value.

>   And we are all frustrated with the price increases we have seen from the
>"usual suspects" over the past several years.  As a result, we find our
>OhioLINK solution to be one way of dealing with the situation that also
>allows greatly expanded access for the smaller institutions in OhioLINK.
>OSU is clearly saving money both by avoiding paper subscriptions that we
>would have had to place, seeing a decreased inflation factor and decreasing
>reliance on ILL (Elsevier has consistently been our highest borrowing ILL
>publisher for several years).  

Your solution is to push increased costs directly onto the State of Ohio,
and since you don't see the bills for this you no longer think they
matter?

>With the use data now available we expect to
>begin a review of our paper subscriptions for potential cancellation.   I
>think it's safe to say that OSU, in particular, and OhioLINK are very
>pleased with the Electronic Journal Center as one way of coping with
>commercial scholarly publishing.  Like Tom we are encouraged by and
>interested in a variety of new publishing mechanisms of which SPARC is only
>one.

This isn't coping, it is caving in -- but if the State is willing to pay
the bills perhaps you don't care?
  
>    Now, if any wishes further discussion on this, I just want to notify
>you that I will be away from the office and email for the next several
>days, but perfectly happy to continue a discussion beginning next Tuesday.
>

[following is the message forwarded from Tom Sanville]:

>>Here is an amalgamated version of 2 emails I sent to directly to Bob
>>Michaelson in response to his email  forwarded to me from lib-license.
>>Please share this as you think necessary. The questions and issues are
>>complex and my response through email can only dig so deep, but I hope it
>>addresses questions raised adequately.
>>
>>I think assumptions about what the OhioLINK community has or hasn't done in
>>terms of cost-benefit analysis should be made.  I am confident that you will
>>be hard pressed to find any other group or library that has analyzed this as
>>extensively as we have and to which analyses our library directors and
>>collection development committee are privy.  Before even moving into this
>>and other licenses an extensive amount of analysis and discussion was held
>>before the community agreed to proceed.

This remains to be seen -- without open cost-benefit and
cost-effectiveness studies, there is no way for the library community to
critically evaluate your claims.

>>We opened up the electronic Journal Center in April 1998 with Elsevier and
>>Academic Press. We added MUSE and are now adding Springer, Kluwer, Wiley,
>>Am. Physical Society.  and more in the future.

We all know about Bradford's Law and bibliometrics:  the most important
titles will account for significantly more usage than will the second or
third rate (or lower) titles. Logically, then, OhioLINK should have
started by supplying electronic access to the most important titles:  
American Physical Society, American Institute of Physics, Science, Nature,
American Chemical Society, Royal Society of Chemistry, Journal of
Biological Chemistry, etc. etc. Instead, OhioLINK started with Elsevier
and Academic (which publish a few important titles but mostly second-tier
or worse) and is only now starting to add _one_ of the major journal
producers. Thus OhioLINK is in effect starting with the dregs -- this is
cost-effectiveness?  This is cost-benefit?

>> We have shared the usage of
>>the system with our libraries and their directors know what our licenses
>>call for economically. So we as a management community are able to look very
>>clearly at the cost-benefit.

It would be interesting to learn how you are measuring benefit.  Number of
journal articles supplied is not, by itself, a good measure of benefit,
since it treats them all the same.  A serious benefit study would sample
the dollar benefit to the end-users of the delivered articles. You would
also want to look at cost-effectiveness: e.g. what would be the cost of
delivering the same information by alternative means (ILL or document
supply).  You would certainly want to determine the relative
cost-effectiveness and cost-benefit of delivering Elsevier and Academic
vs. delivering (for example) American Chemical Society titles -- in short,
is the benefit obtained as great as the benefit in spending the same
amount in some other (dare I say more logical) way?

>> There is no rush to publish in that we only
>>have 15 month's data in the bag. It continues to grow so rapidly and will be
>>affected by additional publishers and links in our systems that we have been
>>satisfied with using this internally up to this point.  There is now a draft
>>paper reporting on our last year's use that may see light of day in a
>>journal. Also I made a preliminary presentation at ALA on these results. Go
>>to http://ala8.ala.org/ascla/cancel_program.html for a copy of my PowerPoint
>>slides used.
>>It remains to be seen what level of financial detail we provide in our
>>public statements. Being on the front edge of this a certain amount of
>>discretion must be applied.  But I will add the following for you since
>>there is the  implication that we have not done our financial homework:

No doubt Elsevier and Academic gave OhioLINK significant reductions in
pricing because they needed to get some highly visible users.  But there
is no such thing as a free lunch -- they aren't being accomodating
publishers, this is just a part of their cynical marketing strategy.  You
are playing along with them; one result is that you can't give us real
dollar figures -- there is no way to give us an analysis that we could
critique. Moreover, the figures that you have aren't realistic, since they
are based on the Elsevier/Academic "loss leader".

>>In short so far:
>>
>>1. We are delivering  a dramatically greater volume and breadth of journal
>>information than ever before across all our institutions- and with an
>>increasing number of journals publishers - both profit and non-profit. The
>>patrons at each of our universities use 2 to 5 times more titles
>>electronically than they have ever had in print. And because it's
>>patron-driven desktop delivery they do a  lot of it, over 430,000 downloads
>>so far. Over half of the articles downloaded are from titles not available
>>on the patron's campus.  Our smaller schools see totally new access at
>>significant levels compared to the paltry print collections held in most.
>>

But we know that they'd be better off with desktop delivery of more
important titles (see above) -- quite apart from the fact that the
selected, more important titles would cost the State less.

>>2. Our annual statewide cost increases to electronically license all a
>>publisher's titles to all of our schools are likely lower than most schools'
>>annual cost increase to maintain its specific selected subset of the
>>publisher's titles. Basic math says greatly increased access and use divided
>>by no
>>change in expected cost is better cost-benefit than you have now. Even  a
>>small increase in expected cost is overwhelmed by the expanded breadth and
>>volume of use.

So if you go into a store that has a huge 20%-off sale, and randomly spend
$1 million, you've saved $250,000!! I'm being frivolous of course, but so
are you.  A minor cost increase on a huge base-price, for material that is
mostly of mediocre quality, is no bargain.  A larger cost increase on
carefully-selected material, at a modest base price, can be a bargain.
(BTW, the American Chemical Society is significantly _lowering_ its
electronic journal surcharge for 2000. Those of us who buy their Option B
package will actually be paying less money. This is a bargain on top of a
bargain!)

>>3. We are less and less information rationers and more information enablers,
>>letting users select and use information of their choice with much greater
>>ease. We are collecting comprehensive data on use by title and school which
>>will help us long term to understand which of the publishers material we
>>really need long term.

Is your idea of responsible collection management to spend huge amounts of
money to sweep in all the garbage that is published? That would be the
ultimate "information enabler".  I would say that any library that almost
never needs to get articles from outside the system is a library that has
more money than sense.

>>I found that the email fails to recognize that there may be multiple ways to
>>approach the problem and that the solutions are evolutionary. We are not
>>wedded to any publisher forever.  But a problem that has been caused by
>>years of bad buyer behavior which has allowed Elsevier  and other
>>publishers, both profit and non-profit, to act as they have does not have
>>simple solutions.

To be sure many libraries have been bad consumers. More significantly
(IMO) many publishers, Elsevier prominent among them, have abused the
academic market.  Abrogating our role as selectors is not a way to become
better consumers; expecting the Elseviers to mend their ways because you
abrogate that role is, in my opinion, foolish.

>>There is no simple way out of this situation. The actions of OhioLINK and
>>other groups and schools like U Toronto improve our ability to meet patron
>>needs now, create an environment that increases information use, provide a
>>better means to measure information use, and provide tools to better
>>negotiate and change publisher behavior  in the future. With the status quo
>>you have no idea what your patrons will really use in an electronic
>>environment - we are
>>beginning to find out. You must find a way out of the box.  You must think
>>evolutionary. We ask to license all the titles now so we can find out what
>>we really need. This information will help us much more in the future than
>>having no idea at all.
>>

Spending huge amounts of money without selection is a pretty poor way of
thinking "out of the box", IMO. Wouldn't it be better to wait for
Elsevier, Academic, etc. to offer title-by-title selectivity, and then
create an electronic package of the titles that are most important for
your audience (as well as most cost-effective)? What is the rush?  
Wouldn't it be more cost-effective to offer a first-rate package (Science,
AIP, ACS journals, even selected Elsevier titles) than to go with a
decidedly mixed bag from the world's publishing cartels?

>>At the same time I applaud the efforts of SPARC and others with alternative
>>publishing models.  These too will bring pressure on existing publishers and
>>help evolve to a better equilibrium of information access and economic
>>reality.
>>
>>One may not agree with what we are doing- but I think you are in error to
>>conclude it is without merit. I'd talk to our librarians and users around
>>the state and you may find out most think otherwise. Our solutions are not
>>risk free and not without question marks. But IMO we will advance. learn,
>>and redirect efforts as needed.

Users and librarians around the state see what they are getting, but they
don't as readily see what they are not getting (lots of higher-quality
journals), nor do they see what they, as taxpayers, are paying. Most of
all they do not see the deleterious effect on the world of scholarly
publishing -- until it is too late.

Bob Michaelson
Northwestern University Library
rmichael@nwu.edu