From: <[log in to unmask]>
Date: Thu, 18 Dec 2014 18:26:42 +0000

Ann,

Thanks for your reply and kind comment. Very happy to be probed a bit.

I think the number of journals perspective is, if you'll allow me, a
collections-librarian perspective - by which I mean the interest/duty
a librarian has in building a collection for the long-run for their
institution. In the analogue world, this was measured by the number of
titles - even if one title might contain 100 articles per year and
another 10000 articles per year - because it was easy to measure and
collect by title and very hard to measure and impossible to collect by
article.

I absolutely understand the importance of collection building because
ensuring long-term access to a professionally curated collection of
content has tremendous value for an institution. (It is something
that, I think, will become more and more important because users will
need a lot of help sorting the wheat from the endless and increasing
volumes of chaff that appear on the internet - but that's another
story).

In a digital world, the question then becomes, build a collection of
what? Do you need to build a collection of journals/bundles of what
could be a mixed bag of articles some of which are of marginal value?
Or should we be moving toward an 'article economy' where an
institution collects the units of knowledge that matter to them? Two
things here - can one collect at the article level? How can one know
what matters?

On the first point - can one collect at article level? - Well, I'm
sure it's technically possible, but the issue may be moot because the
cost of digital storage is low - so collecting 'everything' might be
more efficient in the long run.

On the second point - knowing what matters - this is where usage kick
in, usage metrics can help a librarian shape their decisions on what
they collect. But how does that fit with the previous point? Since the
cost of storage is so low, then wouldn't it make more sense to collect
everything? This comes back to the original problem - the cost.
Clearly, if articles are acquired by the bundle or indeed one-by-one,
it would be prohibitively expensive to collect everything. If,
however, the cost of building the collection is linked to usage, then
everything changes. Fundamentally, it's a switch to a just-in-time
model from a just-in-case model - which does mean thinking in a
different manner.

This conversation has got me thinking about another angle - content
that is open access. How does the concept of 'collection' work with
content that is open access? Will librarians want some sort of service
level agreement with OA publishers to ensure long-term access and
service support to OA content (to guard against the potential risk
that the content could disappear at some point in the future following
financial or technical failure or against a deterioration in the
service offered)?

Sorry, more questions and not many answers!

Best regards
Toby

-----Original Message-----

From: Ann Shumelda Okerson <[log in to unmask]>
Date: Wed, 17 Dec 2014 20:38:42 -0500

Toby, appreciate as always your thoughtful reply, but let me probe
your answer a bit more.

Are you suggesting that so long as overall usage stats of a given
journals package of, say, 1000 or 2000 titles stay much the same
overall or increase, and if the customer service is acceptable, then a
library would not need further to worry about what is in or out of the
package from one year to another?  To me, that feels like a publisher
perspective (and it's what we hear from some), but in terms of
institutional "fit" with useful content, it doesn't seem like a
satisfying or institutionally savvy approach.

If you're right, I'm wondering why we librarians put such clauses into
our contracts if they really don't mean so much after all and it would
be old-fashioned to analyze and/or take action?

Ann



On Wed, Dec 17, 2014 at 8:24 PM, LIBLICENSE <[log in to unmask]> wrote:
> From: <[log in to unmask]>
> Date: Wed, 17 Dec 2014 07:06:39 +0000
>
> Ann,
>
> I wonder if the question is more a legacy from the past when
> everything was measured in terms of what was delivered - a hangover
> from the print world and understandable in that print was the only
> reference point when digital first arrived. In the print world, when
> usage was devilish to measure accurately and much of the cost was
> linked to manufacture and delivery, the value was understandably
> linked to the volume of content supplied.
>
> In today's digital world, I think we can do better. Surely the value
> isn't in the volume of content delivered but in the volume of usage
> (which is more valuable, 100 journals that are barely used over the
> course of a year or one journal used by hundreds daily?).
>
> I also think value lies in the combination of the volume of usage and
> the non-content elements of the package offered (e.g. MARC records,
> customer support etc). Now, if usage falls or the service element
> deteriorates, then the value of the product is clearly lower and
> there's an argument for a rebate. Framing agreements in this way would
> 'catch' the volume element too because if a publisher fails to deliver
> enough articles, for whatever reason, presumably usage will fall and
> thereby trigger a rebate clause.
>
> But as you say, I don't think there's a 'right' answer for everyone's situation.
>
> Toby Green
> Head of Publishing
> OECD
>
>
>> On 17 Dec 2014, at 02:23, LIBLICENSE <[log in to unmask]> wrote:
>>
>> From: Ann Shumelda Okerson <[log in to unmask]>
>> Date: Tue, 16 Dec 2014 20:15:48 -0500
>>
>> Dear liblicense-l readers.  Your listowner/moderator (me) has a
>> question for you.  I would very much welcome the views of anyone on
>> this list, whether publisher or librarian or someone in the scholarly
>> communications chain.  There's no right answer; in fact, I'm not sure
>> there is even an answer, but I was in a group that started discussing
>> this matter and we felt caught short.  And we felt we should have a
>> reasoned opinion, when we did not.  Please read on.
>>
>> Most many big deal journal packages contain language [such as that
>> below] re. modification to "portions of the Licensed Materials."  The
>> contracts say that if any of the changes make the materials less
>> useful, the institutions may seek to terminate this agreement for
>> breach.  And, there will likely be language of this sort: "If any
>> such withdrawal renders the Licensed Materials less useful to
>> Licensee or its Authorised Users, Licensor shall reimburse XX for the
>> withdrawal in an amount proportional to the total Fees owed."
>>
>> My question is this:  if my library has a "big [or medium] deal,"
>> let's pretend it's 300 or 500 or 1000 or 2000 titles, what is a
>> reasonable expectation for the numbers or percentage of content that
>> will leave the package before the library or consortium would either
>> seek reimbursement (more likely) or total termination (less likely)?
>>
>> Do libraries (or consortia) review the big-deal lists each year to
>> look for changes?  Every 3 years? If there were a loss of previous
>> titles in the amount of 5%, would it be a concern? How about 10%?
>>
>> Of if not a percentage "bright line," then what would cause a review
>> of the list and a concerned conversation with the big deal publisher?
>> Would it be the loss of a couple of absolutely key titles?  the loss
>> of a particular smaller publisher's journals list?  a disciplinary
>> impact? a dollar impact?  If "it depends," what does it depend on?
>>
>> Do libraries care very much about what's actually in these large
>> packages, or are we too busy to pay attention to their changes? What
>> would it take to get libraries' attention?
>>
>> Thank you, Ann Okerson
>>
>>
>> *******
>>
>> "Notification of Modifications of Licensed Materials. From time to
>> time, Licensor may add, change, or modify portions of the Licensed
>> Materials, or migrate the Licensed Materials to other formats. When
>> such changes, modifications, or migrations occur, the Licensor shall
>> give notice of any such changes to Authorized Institutions as soon as
>> is practicable, but in no event less than thirty (30) days in advance
>> of modification. If any of the changes, modifications, or migrations
>> renders the Licensed Materials substantially less useful to the
>> Authorized Institutions or its Authorized Users, the Authorized
>> Institutions may seek to terminate this Agreement for breach pursuant
>> to the termination provisions of this Agreement.