Journal of Number Theory - Minutes of Editorial Board Meeting

Present: 		David Goss; Kenneth Ribet; Ambrus Pál; Gebhard Boeckle;
By phone/skype:	 Avner Ash; Peter Schneider; J. Cogdell; Urs Hartl; K. Soundararajan; Jean-Louis Colloit-Thélène; Matthias Beck; Mark Kisin; David Burns; M. Bennett; Florian Breuer; John Hsia; Dinesh Thakur
From Elsevier: 		David Clark & Laura Hassink
Date: 			Chicago, 24 March 2012
------------------------------------------------------------------------------------------------------------------------------------------
Note:  Some of the attendants by phone reported that after about a half hour of the meeting (which lasted in total approximately three hours and 45 minutes) the sound quality became so bad 
that it was very difficult/ not possible to participate fully in the rest of the meeting. Some other participants were able to participate by phone until the end.
Main issues mentioned upfront by JNT board:
It is felt that the unwritten contract that exists between publishers and editors/researchers is broken
There is a lot of anger and distrust
It is not just about JNT but about the whole community
Mathematics is distinct from most other communities and Elsevier needs to realize this ; that said there are also many similarities as in all disciplines scientists work as authors, editors and referees and the commercial publishers earn an extreme amount of money from their work. Maybe mathematicians are the most angry about this.

Statement by Laura/ David:
We want to discuss all concerns raised by the JNT board and by the Cost of Knowledge petition
We want to explain on all topics raised what current situation is, what steps we are taking and the plan forward,
We will be open and transparent and if we don’t know the answer we will try to find out.
At the end of the meeting we hope to have an agreed way forward.
Steps we have take so far to address some of the concerns in the community:
Our target is for all of our core mathematics titles to be priced at or below US$11 per article (equivalent to 50-60 cents per normal typeset page). Where journals are more expensive than this, we will lower our prices, as we already have in recent years
To make clear that we are committed to wider access, we have made the archives of 14 core mathematics journals open, from four years after publication, back to 1995, the year when we started publishing digitally.
We are in the process of creating a scientific council for mathematics, to ensure that we are working in tandem with the mathematics community to address feedback and to give greater control and transparency to the community
We dropped our support for RWA as our community was concerned that the RWA seemed inconsistent with Elsevier’s longstanding support for widening access

Access:
The 30 days free access to ScienceDirect and Scopus that Elsevier is providing to reviewers is not suitable for the maths community as reviewing takes much longer.
David Goss:  nobody that he knows uses Scopus given how great Math reviews is.
Laura/David propose to give loyal reviewers (to be determined by the editors) one year free access to the mathematics subject collection. This feedback was received by various Elsevier maths editors.

David Goss: this proposal is impractical for a journal like JNT which has so many submissions and that anything useful would need to be automated.

Jean-Louis Colloit-Thélène mentioned that mathematicians do not need any reward for reviewing as it is part of their commitment to the community.

Response from Laura/David: it is not mandatory nor is this intended as some form of payment, but as something of use to reviewers, particularly those, such as retirees, without access to the literature

Author posting policies: & copyright
Elsevier policies were not clear but they are on par with other e.g. Springer;  we have recently updated our website to make them more clear. Authors can deposit their accepted author manuscripts to the ArXiv.

Ken Ribet: there are mathematicians who refuse to transfer copyright; how do you deal with that?

Answer: There are merits in publishers having the copyright (e.g. in legal cases,or to make sure we can transform manuscripts into formats suitable for the future etc.) but the most important thing is that authors retain the rights that enable them to do what they need to do.

Gebhard Boeckle: Can you return parts of the copyright back to the authors after a certain number of years as in the book industry?

Answer: With books that normally happens when a publisher decides to no longer make the book available or put it out of print whereas in journals the intention would be to have articles permanently have them available.

Pricing & Subscription models “ bundles”
David /Laura: explained about the historic trends of pricing; in the 1990s institutes typically had between 200-300 print subscriptions and this went down year on year. When Elsevier and other publishers started to digitize journals this offered opportunities to broaden access to many more titles. This was very much welcomed at the time as it ensured campus wide access to a very large number of titles for modest additional cost.  Such agreements were based on a core of print subscriptions, with additional electronic access to many more titles.  This applied for Elsevier but other publishers too.

Ken Ribet: you are giving us additional journals but many seem to be useless and of bad quality

David Clark: There will be a wide range of titles in different subject areas.  Their usefulness will depend on the University and what they hold, so in some Universities these ‘unsubscribed’ titles may only account for 20% of usage, while in others it will be 70%.  Across all sites, these unsubscribed content account for 40% of the usage.  Usage is, of course, just one measure but it does give some indication and librarians follow it.

David Goss:  would Elsevier work with the community to lower costs as, for instance, there
may be many services the community does not need

David Clark: yes, we absolutely do this. For example if an institute does not want print, this can be cancelled.

Urs Hartl: The pricing and bundling is a huge concern: there is complete lack of transparency. Urs addressed the situation at the University of Muenster where the price of their Freedom collection was increased from 115k EUR to 197k EUR to 226k EUR. These are price increases of 15-20%. Is this normal practice?

AP: LH to find out more details about this deal and explain to the board.

Urs Hartl:  German librarians gathered in Bonn to discuss their relationship with Elsevier. Although the libraries had signed non-disclosure agreements, the librarians exchanged the prices they pay. The reason was Elsevier's determination for transparency and that libraries found transparency in prices of utmost importance. Nevertheless, the Elsevier sales person Thomas Edelmann heared about the disclosure and very angrily expressed his disapproval towards the librarian of Muenster.

AP: LH to find out more details about this deal and explain to the board

David Goss: Agreements are extremely complicated; Even the Ohio Chief Librarian (who is on the Elsevier Library Advisory Board) is highly frustrated: prices are set and there is no room for negotiation. His librarian said that she felt that no matter what happened, the cost of Elsevier products stayed constant!

AP: Ken Ribet inquired about the situation at the Univ of California. To how many titles do they subscribe?

David Clark: the University of California has an agreement to, I believe, more than 1400 titles, with the holdings of different campuses shared across the whole system. [Update: we checked this and for last two years, the University of California has moved to a system accessing all of the titles Elsevier publishes].

Avner Ash: The librarians at Boston College tell me that they are very upset at the lack of transparency on the part of Elsevier and the difficulty of what they view as fair negotiations with respect to bundling.  
It is important for Elsevier to improve both its actions and how these actions are perceived.
AP:  LH to find out situation at Boston College.
Update from Laura Hassink after the meeting: BC is in the 4th year of a 5 year agreement, organized under the umbrella of the Boston Libraries Consortium. For 2009 they had originally opted for a 3yr agreement but this was extended in 2011 for  2012-2013. BC  has a SciencDirect-Complete type of agreement:  subscribing to approximately 370 journals (back to 1995), with online access to about 850 additional titles.

Gebhard Boeckle: Although not all librarians might be  against “ bundles”  in itself, there needs to be more transparency and enough flexibility.

David/Laura explained that there is a range of subscription options (Freedom collection, subject collection, unique title collection) and explained how the Freedom collection works: (slide 13) customers have a certain set of print subscriptions; the list prices of these journals are added up and (depending on the size of the institute) need to reach a certain minimum percentage (10%, 20% or 30%)of the total Freedom collection value.  The remaining journals can then be accessed at a highly discounted price (>90% discount).
The subject collections work according to a similar principle but then different discount percentages.

Gebhard Boeckle:  Do I understand correctly that within the Freedom collection that subscribed journals are sold at subscripton prices or are there discounts.
David Clark: The value of subscribed journals forms the basis for the Freedom collection so yes in principle they are sold against the list price and access to the “non  subscribed”  content is then added at a nominal fee.

Florian Breuer pointed out that the most urgent transparency  is to allow disclosure of prices.

It was acknowledged by David/Laura that the subject collections are still very large. Is there a need for a smaller maths subject collection, similar to what Springer recently offered? Laura/David are open to that and mentioned this is typically a topic they would like to discuss with the Scientific Council. Also to determine which titles would then be part of such a smaller collection.

Definition of small, medium and large customers as mentioned on slide 13:  For academics the number of staff and students is taken as the population and for government and corporate customers, the number of researchers. In practice this can be a proxy of research intensity and thus usage within a region.

David/Laura explained that customers only have archival rights for their subscribed journals.
Consensus that therefore more Open Archives, in addition to the 14 already created in mathematics, are important.

We discussed that there is sometimes a discrepancy between the purchasing flexibility individual departments/ faculties want to have and the way the university has chosen to organize themselves when purchasing subscriptions.

More and more frequently, the agreement is reached with Elsevier and other publishers for a whole university, even a group of universities, for a complete state or country. It is acknowledged that in these cases individual libraries do not always have the flexibility they would like to have. The JNT board members stated they should have more influence on what is being bought and it should be easy for them to fix this within their universities.

David /Laura explained that we are evolving our current subscription models and that we are piloting with various alternatives to offer more flexibility and transparency.

David mentioned that University libraries, in choosing which titles to subscribe or to add to incremental collections, might not always prioritise mathematics which has lower usage than other fields.

Ken Ribet: Why is the mathematics subject field different in terms of usage?
Answer: in maths we see lower usage because articles are used in a different way. Mathematicians typically read fewer articles but often in much greater detail, whereas in e.g. materials science or chemistry an article can be used for just a specific part in that article or as part of a very extensive literature search ranging across many journals. Some of the differences in mathematics are not appreciated within Universities, as well as with publishers

Urs Hartl: What are the subscription models and prices for developing countries.

Answer:  2000+ Elsevier journals (and 6500+ electronic books and Scopus) are available to developing countries at no cost through the Research 4Life program (slide22)
More information: HYPERLINK "http://www.elsevier.com/wps/find/intro.cws_home/research4life"http://www.elsevier.com/wps/find/intro.cws_home/research4life
List of countries eligible: HYPERLINK "http://www.research4life.org/institutions.html"http://www.research4life.org/institutions.html

Urs Hartl: Due to the practices of selling entire bundles, a price reduction of JNT will have almost no effect for subscribers and non subscribers of JNT
Answer: this is only partly true. The price reduction will have an effect for the customers who buy individual subscriptions. But also the list price is the foundation for all subscription models (Freedom collection and subject collections) so if the prices of the mathematics journals are lowered this will have an effect on e.g. the mathematics subject collection price. Some universities who purchase the Freedom collection have a charge back mechanism to the departments which is based on the list prices. If the list prices for the mathematics journals are lowered this will mean a reduced cost for the maths departments.  The existence of discounted packages, subscribed and unsubscribed options and consortia agreements, all make this complex, but addressing the list prices remains a necessary first step.

David Goss: Will the price reduction affect existing subscriptions or just new ones

David Clark: All subscriptions

Agreements with libraries are the outcomes of detailed negotiations and Elsevier is keen to ensure the widest possible access for our titles, so that they can easily be accessed.  There are some reasonable and well argued comments made by the proponents of the Cost of Knowledge but we objected particularly to the idea that we willfully terminated access as it is something we would only do with great reluctance.  

It is should be noted that regardless of the current economic downturn, there is a group of customers who is receiving a significant relief because of the current economic situation in these countries so that they retain access.

Florian Breuer:  Do countries outside China receive a greater discount?

David Clark: There are a group of countries, along with China, who get different terms as they did not have individual subscriptions in the past. We don’t know the detailed terms of such agreements and it is in all likelihood not something that we could publicly discuss.

Slide 16: The graph shows that for 2010/ Elsevier’s price increases are in the lower quartile of the industry.  This has been the case for the last 10 years. Price increases include the growing number (on average between 3-5% per year) of accepted articles.

AP: The board would like to know where Springer is in this picture; LH to let them know (done; see annex3)

Urs Hartl: asked whether we would increase the prices by 50% if the mathematics community would produce 50% more articles?
Answer: No, although for the prices of individual journals we do take into account volume growth or decrease, but, as with our lower average prices, we are aware that Universities are not able to cope with dramatic changes.

Florian Breuer:  with better technology, one would expect that the costs go down; e.g. hosting fees?

Answer: some costs indeed go down ,but we also see some costs increasing; e.g. IT and competitive landscape requires a continuous investment in new technology such as CrossRef, Applications, Orcid (an author ID that works across publishers); features and functionalities that help the researcher to find the right content)

Gerhard Boeckle: you have prices in different currencies and I noticed that with the current rates the prices differ. Why is that?

Answer: One criticism of publishers in the past was that they passed on the impact of currency fluctuations to customers, making it hard to manage their budgets.  To avoid shocks, we have not adapted prices to reflect changes in currencies but over the last ten years there have been major changes (so, for instance, the Euro was worth US$0.81 in February 2002 and had risen to US$1.59 in June 2008 and is now at US$1.31, and is likely to change in the future. Had we adjusted price to the exchange rate it would have caused lots of problems for libraries.

David Goss: how many subscriptions did you lose over the last 10 years?

Answer: we lost few subscriptions, this had been quite stable.

Slide 17, list price of JNT:  The list prices are paid by customers who pay individual titles and are the foundation of all agreements. The list price divided by the number of articles published is used to compare pricing levels of various publishers. Elsevier wants to be around US$11 for the core maths journals, which is similar to other publishers. We will address the outliers; JNT being one of them.
Slide 18 shows the price per page compared to other publishers

Mark Kisin:  Why do you use cost per article? Some journals (like Crelle’s journal ) publish much longer articles so cost per article is then obviously higher. Cost per page is more transparent (although it differs from publisher to publisher how much text is published on a page.)

David Clark: since typesetting models vary very much, we don’t feel that price per page always works well, but we’re open to talking about other models and to reporting against them.
AP: JNT  board would like to see price per page and price per character.  LH to arrange for that
AP: include some other journals e.g. a journal from AMS and LMS and International Press of Boston.

Question: Is there room for true innovation in the way the articles shown.

Answer: yes there is and we are working on a few of these innovations ourselves. For example  MathJax and the STIX font (slide 30 and 33). We also would like to hear from you and the scientific council what other innovations you would like to see.

David Clark mentioned, since this had come up in pre-meeting correspondence, that Elsevier does not work with page budgets; we simply publish when a paper is accepted and do not build up backlogs, as we think it is not in the author’s interest if his/her accepted article needs to wait on the shelves because there is no page budget left.

Image
Members of the board felt that the lack of transparency is a big issue.  Also the inflexibility of some of the agreements with big publishers, including Elsevier.

K. Soundararajan:  After speaking to my librarian, in comparison to Elsevier, Springer offers better value for money and better quality.

David Clark: at a University – wide level, that is normally said of Elsevier over Springer.

David Goss: But we are focused on mathematics here.

David Clark agreed that this was the key issue.

Open Archives
Elsevier sees the open archives of the 14 maths journals as a starting point to determine whether we can move forward with this both amongst Elsevier but also other publishers. The archives are available after four years of publication and back to 1995 and will be available permanently.

This approach gives authors and referees the confidence that the final published version of their article will be in the public domain after a reasonable time period and that four years is a reasonable time period, we believe.

David Goss: why not shorter?

David Clark: Since mathematicians use articles over a much longer time period than other fields, it may be hard to maintain a subscription base with a short subscription window which is why other publishers have been reluctant to do it. There is some risk of that but we think that the wider access is worth it.

David/Laura explained that Elsevier has opened up the archives back to 1995, when we started to produce digital editions of journals: the older (pre 1995) files are sold as a one off, separate collection because we invested very heavily to digitize this older content.  40% of the articles in this mathematics backfile collection are in fact not owned by Elsevier but are articles that Elsevier published on behalf of third parties. In principle, we would like to do more here and will explore options as we recognize the need for wider access to earlier articles, but we need to consult with customers and with society partners.

Open Access
The various options of Open Access were discussed. (slide21). There is not much appetite for author pays journals in the mathematics area.

Independent digital archives
Laura Hassink explained that to take away any concerns that Elsevier books and journal content will no longer be available if Elsevier were ever to cease publishing operations or  could not for some reason  maintain the archive, that Elsevier deposits a copy of everything we publish with three independent third parties around the world. They then have the right (obligation) to make the archival files available.
Elsevier deposits everything in PDF and XML (eXtensible Markup Language, is a markup language designed to structure, store, and transport data; this ensures that it is also applicable with future format standards regardless of what happens with Acrobe, LaTeX or other formats).
ArXiv
Question: can Elsevier not sponsor the ArXiv?
David Goss: Can Elsevier make a point to have the final tex version deposited on the arXiv as well as the data on the published paper so that the theorems may be used and cited in their final form.

Answer: we and other publishers have been exploring working with the ArXiv and have discussed sponsoring with them (as well as other publishers) but that has not progressed so far. Elsevier supplies metadata directly to the ArXiv and has for many years.


Key activities of the publisher
Laura Hassink explained some of the key activities of the publisher.

These include

developing and launching new journals. Every year we receive many proposals, which we investigate; only a few materialize into a journal in the end. For mathematics not many new journals are launched.  The pay-back period for new journals is normally between 8-10 years. In the future we would involve the Scientific council in decisions whether or not to launch a new journal in mathematical areas.

Outreach for new authors including pre-checking on many journals and workshop programmes for authors in developing countries;

Financial support for editorial offices and overheads;

Maintaining on-line submissions systems including handling queries and issues for authors submitting manuscripts, including the many who will be ultimately be rejected;

Plagiarism, legal and ethical issues support;

Edit and prepare accepted articles including conversion of LaTeX into XML formats, and working with articles to address issues and make corrections – our experience is that the quality of such manuscripts is not always as high as authors think.

Link and prepare metadata, including sharing with major A&I (abstracting and indexing) services, google and the arXiv, as well as Mathematical Reviews and others.

Print and distribution

Maintenance and development of electronic platforms

It was asked whether publishers would continue to maintain print copies?

David Clark: as long as customers want them, then we will do so and there remains a demand especially in mathematics.

Slide 26: Investments of publishers (2000-2009). There was a long discussion about this slide. Laura/David explained that the costs of 2 billion GBP is the cumulated cost of all commercial publishers in the last 12 years. (correction by Laura/David: figures in the slide are over 2000-2009 period). 
Mark Kisin said that this comes down to 200 million EURO per year of which Elsevier's share is maybe less than 50 million EURO. So the figures which look big are actually very low costs.
Urs Hartl said that the profit of Reed Elsevier solely in 2011 was 760 million GBP "largely driven by strong performances from Elsevier's electronic subscription" as one can read on NASDAQ.
Mark Kisin pointed out that the profit is not the right figure to compare the costs with, but that revenue is and that the revenue is even much higher.
Urs Hartl concluded that the proposed price of US$11 per article is still too high by magnitudes.
David Clark objected against the word "magnitudes".
Urs Hartl relativised by saying that the comparison between actual costs and revenue demonstrates that the proposed price of US$11 per article is still too high by a factor 10.
Other board members pointed out that the labor of authors, editors and reviewers is provided to Elsevier free of charge and in response Elsevier creates a huge revenue out of it.
Urs Hartl said that this is a core point of criticism and Elsevier has not responded to it. He directly asked what the response is.
David Clark answered that Elsevier is targeting a price for the math collection of US$11 per article and that’s competitive to what other publishers offer.

EES
Overall consensus amongst the editors that EES is not well liked in the maths community as an online submission system. It is not flexible and there are no short cuts.

David/Laura responded that we are working on better system and also interested to hear which systems are perceived as good submission systems for the maths area. Also we are open to investing in more support (managing editors) to help editors dealing with the system.

Answer:  the system used by Maths Sci Press, Documenta Mathematica and International Journal of Number Theory, which is similar to Elsevier’s system but involves more administrative support.

Giving back to the community
Laura/David presented some of the initiatives already done by Elsevier (e.g. Elsevier Foundation. JMAA Ames fund, more on slide 37), but open for further suggestions from the Board on where we can give back to the community, as we have in other areas.

Suggestions made:  fellowship for individual students, travel grants for young researchers, sponsoring of the National Academy of Sciences award in mathematics, establish a fund for the US National Committee for Mathematics or sponsor the Bowen Lecture series. (more info about each of them was received from Ken Ribet after the meeting by email).

RWA
Laura/David explained that we withdrew from the RWA in response to the negative criticism from editors, reviewers and authors.

We think that voluntary partnerships are better than government mandates, but supporting this legislation was a step in the wrong direction. Dropping our support for RWA is a message to the community that we are open to your feedback, will admit when we are wrong and act on it.

It is worth noting that for many years Elsevier have been supplying the NIH with manuscript versions and we began doing this as a voluntary service before the author mandate was introduced.

Scientific Council (slide 36):
We are in the process of launching a Scientific Council.

This will be a very important group of people for us to make sure we work in tandem with the maths community, hear their feedback, and enable us to be more transparent and to do the right things. The intention is that the majority of the council will consist of people not associated with Elsevier and we would be open to discussion and feedback on all issues.

Transferring ownership of the journal
Urs Hartl asked whether Elsevier would transfer ownership of the journal to Ohio University.

David Clark: no, transferring ownership is not something we have done in the past or that we, or any other publisher, be they society or commercial, would normally do. Ownership of a journal is not only an asset but also a liability.

We do realize that the question of ownership is also a moral one for the community rather than a financial or business one, with a wish for greater control over the literature to maintain standards and ensure the heritage of mathematics.  That’s one of the reasons why we set up our Scientific Council to give greater control and transparency to the Community and that’s something which we’d like to discuss further with the IMU among others.

 Ethics (slide34)
Laura/David explained that Elsevier have recently invested a lot in providing tools and guidelines to help editors dealing with the increasing number of plagiarism cases.

One example is Crosscheck a cross publisher initiative to detect instances of plagiarism in scientific articles. These reports are being checked by Editors and internal staff for actual plagiarism. For journals with many substandard submissions a technical check is done before the papers are sent to the editors to filter out the papers that are technically not correct (e.g. not written according to the author guidelines).

David Clark explained about what happened with Chaos, Solitons and Fractals and which measures have been taken from the lessons that we have learned.    We believe that the Journal is being turned around thanks to the efforts of the new Editors, supporting by an in house team at Elsevier.  We will support these Editors in their work. Whatever happens to the journal in the future, our primary discussion will be with the Editors of that journal.

David Clark added that although there had been much discussion about retracted papers, last year Elsevier retracted about 110 papers. (We publish 270,000 papers per year and have published 11 million papers in total ; some retractions go to articles older than one year). To a certain extent, each retraction is a failure but it is also important that we do this and that we do this in a proper way. This can often by complex and legalistic to manage and we are providing increasing internal support to deal with these matters.

Ethics would also be an important topic to discuss with the Scientific Council.

Next steps:
The board would like to see the detailed minutes and also David/Laura to respond in writing to all questions raised in the letters from Urs Hartl and Jean-Louis Colloit-Thélène. (Note from David/ Laura: we have tried to answer all questions in this document as there was some overlap with the questions raised during the meeting and the ones in the letters).
Laura/David will reach out to the individual board members to ask for their support for another 6 months; after 6 months they will go back to them again to ask again whether they want to continue on the editorial board  of JNT
Avner Ash mentioned that in principle he thinks it is a good idea to agree to Laura/David’s proposal to commit to stay for another 6 months but only after he has seen the minutes and responses to the letters in writing.

Annex 1 – responses to letter by Jean-Louis Colloit-Thélène on questions that were not answered during the meeting

The Elsevier Editorial system is most impractical for members of the editorial board, also for referees and authors. I want to see a system that works (like the one from MSP)

Answer: We are working on better system but it will take some time before it goes live. As also discussed during the meeting we are interested to hear which systems are perceived as good submission systems for the maths area and why and learn from that. This will also be one of the topics we hope to be able to discuss with the Scientific Council. One intermediate option is to have greater administrative support between the system and the board members so that Editors and reviewers can use their time most efficiently.

Complaints on the format of JNT papers, whether printed or online.

Answer: We agree with you that the format of the mathematical papers can be further approved. We expect that some of the initiatives we are currently working on (like MathJax and the STIX font (both discussed during the meeting) will address this. We will also look into whether the other format changes you propose are possible.

Sending the proofs to the author
Answer: In general we think it is important to publish articles as quickly as possible as this is in the interest of the author, but since the reviewing times in mathematics are much longer than in most other areas the time authors are given to correct a proof is something to look at. We will work with the Editor to identify what alternative time periods might work better.
Issue with the price
Answer: Please see the minutes (page 2/3) and also our response to the letter sent by Urs Hartl (annex 2)

Open Access
Answer: please see the minutes (page 6) and also our response to the letter sent by Urs Hartl (annex 2)

On 5.2, other than Topology, these are society journals so we would discuss this with the societies involved.  We can certainly see such journals being included in further extensions of the Open Archives policy.

Keep the paper copy of the journal
Answer: We will continue to print paper copies of the journal as long as customers want them, and there remains a demand especially in mathematics (see also page 9 of the minutes).

David Goss: are not some printed copies kept for archival purposes?
Laura Hassink: Since we now use printing on demand we can always print more copies if we want so therefore no printed archival copies are kept.

Annex 2  - responses to letter by Urs Hartl on questions that were not answered during the meeting

Elsevier makes the archives of all its mathematical journals open access, from the first issue published, after at most a four-year window and releases all content under an appropriate license (CC-BY-SA 3.0) that guarantees a broad right to distribute.

Answer: We have opened the archives of 14 of our core mathematics journals. Elsevier sees the open archives of the 14 maths journals as a starting point to determine whether we can move forward with this both amongst Elsevier but also other publishers. The archives are available after four years of publication and back to 1995 and will be available permanently. Elsevier has opened up the archives back to 1995, when we started to produce digital editions of journals: the older (pre 1995) files are sold as a one off, separate collection because we invested very heavily to digitize the older content.  40% of the articles in this mathematics collection are in fact not owned by Elsevier but are articles that Elsevier published on behalf of third parties. In principle, we would like to do more here and will explore options as we recognize the need for wider access to earlier articles, but we need to consult with customers and with society partners.
Wrt to your request to release all content under an appropriate license: our licensing policy for our open archive content is still under development and we are open to pilot more flexible licensing models  for the maths titles.  We have discussed this with our colleagues responsible for copyright and licensing models  and we are now working with them on a proposal.

Elsevier must cut the prices for all its mathematical journals by 30%,For the next 5 years the annual price increase is limited to 2%. There must be significant discounts for institutions and scientists in developing countries.

Answer: What we are planning for mathematics is to rebase our core mathematics titles at approximately US$11 an article which we think is reasonable and competitive. This will mean that for some journals the prices will indeed be cut by 30% (even more), but not for journals that are already at that price point. Access for developing countries: please see page 4 of the minutes.

Bundling practices:
The 30% price reduction of JNT must be fully taken over into reducing the price for all bundles containing JNT. The same is applied to all other mathematical journals of Elsevier.
For the next 5 years the annual increase in all bundle prices negotiated worldwide is limited to 2%.
 All institutions must be allowed to publish the price they are paying for their current bundle and all further bundles.
Elsevier must offer sub bundles for each scientific discipline at no extra cost
It must be possible to cancel journals from the bundles and receive the corresponding price reduction.

Answer: We negotiate in detail with libraries and enter into complex and highly detailed agreements. In general licensing agreements are between two parties that have agreed to keep the terms confidential. Published information can be used by competitors and abused, and there may be competition issues.

The terms of negotiations will always be with libraries, or consortia representing libraries, and they will contain different conditions and terms as appropriate. We cannot easily open up such agreements and change the terms, and would not do so. The nature of the agreement is such that they are often multiyear and that at time of renewal librarians will go over the actual needs of the institution. For instance the closure of a department might lead to actual cancellations and reduction of invoice.

We are targeting a price for our core mathematics journals equivalent to US$11 per article, or 50-60 cents a page. This will not result in a price reduction of all journals by 30%, as many of our titles are already at or below that level.  Also we will (in cooperation with the Scientific Council) investigate the need for a smaller mathematics subject collection.

Academic oversight of Elsevier journals
The scandal around the Elsevier journal Chaos, Solitons & Fractals" is extremely bad for Elsevier's reputation and thus also that of its journals. As a reaction to the scandal the journal Chaos, Solitons & Fractals" has to be ceased.

Answer: please see our response in the minutes (page 9).

5. Transfer of ownership
We require that on January 1, 2013 ownership of JNT's name is transferred to the mathematical department of Ohio State University (OSU), because of its historical role in founding the journal. OSU will exercise academic control over the journal but will subcontract the production of the journal to a publisher. The first of these contracts will be negotiated in 2012 and will last three to five years.

Answer:  please see our response in the minutes (page 8/9).


Annex 3 Priced increases per publisher (additional information requested during the Board meeting.