Costs and Benefits of OA (part 3) - Berlin 6

Fred Friend offered some insight into the costs for Institutional Repositories, as well as some closing remarks.

He pointed to Alma Swan's study of these costs (found in the DRIVER guide to European Repositories) - "The Business of Digital Repositories". The study shows that there is a huge variation in start-up costs according to whether free software is used, the number of staff days required to set up the system, and which functions are included in these costs. Another source of information, Friend says, is the UK LIFE project - which speaks of costs as illustrative, rather than absolute. This provides good methodology, Friend says, that we need to create a methodology for analyzing costs.

It's very difficult to find a norm, since there are so many factors that affect cost. A wealth of data has been aggregated on this, but outside of very general assumptions, it's difficult to come up with a definite conclusion.

"If there's little known about IR cost ... there's even less known about IR benefit"

Benefits are currently described in very general terms. Are these general statements sufficient to justify to a university finance office an expenditure of tens of thousands of dollars / euros / pounds per year? Can these benefits be quantified in financial terms? Should they be? We need to be more specific in describing the benefits.

"It can be done," Friend says. There are benefits. We all know that IRs serve an important role, but we do need more data and a way of messaging this if we are going to make a case for the long term support of a repository.

Cost and Benefits of OA (part 2) - Berlin 6

On to the other panelists ...

2. Matt Cockerill (Publisher, BioMedCentral)

BMC is one of the most prominent OA publishers and a profitable one - ever since its launch in 2000.

Cockerill starts by teasing out exactly what is fundamentally different / what distinguishes an OA publishing business model versus a traditional model. Traditional models, as we all know, involve a transfer of rights, and publishers cover their costs largely by selling access to the content. Under an OA publishing model, there are no barriers to access. The publisher does not acquire any exclusive rights, he says.

In terms of revenue streams for OA - in many ways they are similar to traditional publishing:
- via publication fees
- subscription content (for OA, sometimes special reviews issues that synthesize and add value to OA content)
- services (for BMC - they provide an open repository service that is a hosted digital repository platform)
- advertising (though it covers a very small percentage of cost, usually)
- implicit / explicit subsidy. All forms of scientific publishing are based on these kinds of support (take editors as volunteers for an example). Some costs are completely covered by the organization behind them. Some are not.
- some operate mixed models

He refers to IBM as an example of a change in business model - where in the early days the main revenue stream focus was on hardware and software, where now the majority of business these days are through services (more and more given away as open source)

OA fees do cover largely the same costs as traditional publishers (editorial, technical, production, marketing etc.) But BMC's price per article are much lower than some other journals in the space.

Cockerill tosses out the following numbers, price per article (BMC maintains this list, FYI). It's interesting to see it in context ...

BMC- $780-2100
PLoS- $1300-2850
Company of Biologists- $3100
OUP - $3000
Royal Society- ~$3000
Springer- ~3000

He closes with stating that the health and sustainability of OA as a model is really demonstrated by the number of options or avenues currently available. In 2000, BMC was the only choice - now, there are a multitude of options.

3. Salvatore Mele (CERN, SCOAP3)

"Open Access solutions in High Energy Physics must be a single, global solution, because the discipline is a global, scientific endeavor."

Mele gave a presenation on CERN's SCOAP3 effort (Sponsoring Consortium for Open Access Publishing in Particle Physics). They have been working to set up a consortium whose main goal is to support OA to High Energy Physics (HEP) by "re-directing" subscription money.

HEP is a field that is years ahead in terms of their ingrained practices for sharing data and literature. This approach pools the money normally budgeted out for subscription journals - the top in the field - and uses it to pay OA fees for authors and to make the information available.

More information about this approach can be found here on their Web site.

Philippe Aigrain on OA - Berlin 6

"Instead of "open science", we should refer to it as commons-based science. What science isn't open, anyway?"

Commons-based science is developing quickly in terms of data, protocols, software, research instruments including materials.

There are some significant challenges for defining commons regimes. In terms of data, Aigrain posits it's a paradoxical challenge -- for software or text, we have defined appropriate regimes (ie., GPL, CC licenses or in broader view, copyright). When you look at data (some exceptions, data protections ie., sui generis provisions in Europe). There is no title of proprietary tied to data (again, there are some caveats, but speaking in the broad sense). When you want to define some sort of commons based on data the only power you have is either saying I want to use a commons or through limited power of disclosure (imposing terms of use). Aigrain believes regulated commons are needed.

For biological information and materials, there is a great debate as to whether the process should be copyleft or not, and also how to manage the materials. Biological information is normally worth nothing without the physical materials.

Methods and protocols are important as well (ie., Biobricks standards), dictating best practices and existing standards on the use of certain materials and what not.

Aigrain refers to Science Commons but with a caveat regarding our technical work. He emphasizes that the Semantic Web is only one way to do open science, and to not mistake for others. Aigrain speaks solely about our NeuroCommons work - creating an open source knowledge management system, initially implemented in neuroscience. What Aigrain fails to mention (and later comments about the need for) is that our SW work is only *one* component of our work - a very powerful and compelling one, but still just a piece of the puzzle.

Prior to the Science Commons shout out, he references indirectly our Biological Materials Transfer Work, as well as our work in data and copyright, though speaking about it broadly. But for some reason, may believe we are just working towards Open Science via the Semantic Web :) (I cleared this up during the Q&A)

Regardless, thanks to Philippe for the shoutout, both in directly pointing to us, and also pointing to our work indirectly ... We're agreeing loudly here in having a multi-faceted approach towards Open Science, since there are a number of very particular areas that call for a variety of solutions and approaches. Data, biological materials (genes, plasmids, lab mice, cell lines), and content are all different beasts - but incredibly important to scientific research and to the discussion. Thanks to Philippe for this engaging keynote.

OA and the Role of Societies - Berlin 6

We're on to the last panel of Day 2 of Berlin 6. I have a few blog posts that still need some cleaning up, but will get those up tonight. Apologies for the mix up in order. Some sessions are easier to synthesize on the fly than others :)

This session is on the role of scholarly societies in Open Access. The panelists include Ingrid Gogolin from EERA (European Educational Research Association, Gene Sprouse from the American Physical Society, and Stephen Anderson from LSA.

1. Ingrid Gogolin, EERA

background about EERA ( is a scholarly society founded in the 1990's and is made up of 24 national associations and some regional associations as members.
need for an international perspective on European educational research. Traditionally it is very national in nature, with a scope limited by geographic boundries. In order to foster ideas between European researchers and to better promote reserach quality in collaborative projects, saw EERA as a necessary complement to national research associations. They focus on increasing communication amongst research institutions and better disseminating the findings.
.
They have an electronic journal (current issue only "OA" to members, FYI).

Gogolin spoke specifically about a new EERA project - EERQI, funded under FP7 (I'll have to google the acronym). The project sets out to develop new indicators and methodologies to determine quality of educational research publications, as well as to produce a search and query engine for resource harvesting and text analysis.

Why? European educational research, social sciences and humanities is not adequately covered under traditional methods of measuring research quality. Instruments do not cover these due to a number of reasons, including, among others, multi-linguality in Europe. The project doesn't rely on citation frequencies but also incorporates qualitative criteria, which I'd be interested to hear more about as the project develops (very much so still in its infancy).

2. Gene D. Sprouse, American Physical Society

"We're not-for-profit, but also not-for-loss."

A new manuscript arrives approximately ever 3 minutes to the APS' door. APS, for a bit of background, is the largest publisher of physics articles. They have 10 OA publications.

Though they have a set of OA journals, they do maintain some journals that are subscription based, which helps support their OA efforts. Someone's gotta pay the bills.

In terms of their copyright policies, APS allows authors to post APS' version of their article on the author's or institution's Web site / repository immediately on publication. The author can post the final manuscript with corrections made on peer-review in pre-print archives, and all authors are now allowed to reuse up to 50% of their works in derivative works, which they then own the copyright for. This is also retroactive, so any APS article now has this capability.

3. Stephen Anderson, LSA / eLanguage Agora project

Anderson talked about open access and the eLanguage Agora project of the Linguistic Society of America. My notes on this are skimp, since he largely walked through the history of the project, which you can find on their Web site. IMO (and also voiced by a few other participants over coffee), this project - out of the 3 discussed - is the only real OA example. APS has a split model - with some of its journals being subscription based, and EERA is only available to members (which, thankfully, David Prosser of SPARC Europe pointed out -- it's not OA and should not be classified as such).

Interesting panel, covering an array of scholarly societies and their work in increasing access to their materials, whether truly OA or not.

OA and Open Standards - Berlin 6

Moderated by Christoph Bruch, one of our fellow OA contacts from the Max Planck Digital Library, Day 2 of Berlin 6 was kicked off with a dialogue on open standards.

Now, I find standards, norms and what not fascination (don't judge :) ) but chose to focus a bit more on the standarization theory detailed by Dr. Stefan Weisgerber in this conversation. I would have loved to have heard more about the standarization of metadata, standard forms of describing information in the digital space (RDF, ontologies, standardized languages) -- largely since I view describing the info on the Web in a machine readable way as the base on which all of this in built, but so it goes.

So with that, here are some notes on Weisgerber's presentation about standardization theory. I need to do some additional reading to really grok this and be able to build upon it, but here goes ... My apologies for this being so informal.

- intern'l standards are for the global market
- provide reference framework and universal language for suppliers and customers, facilitate trade and tech transfer, ensure tech compat, provide scientific basis for legislation, ensure that trade agreements can be successfully implemeneted - technical know how

World Trade Organization - standards codex
- no preferential status for domestic products
- no trade barriers as a result of national standards
- Adooption of relevant international standards
- participation of nat'l delegations
- DUplicate work should be avoided
- National consensus-building followed by voting
- Consistency in standards
- Publication of work programmes - should be easy to see what's ongoing in standard work
(www.iso.org)
- Public review of procedures published (at least 2 months)
- Appropriate consideration of comments submitted, to the extent possible

A word from the European Commission - Berlin 6

"Results of publicly funded research should be accessible." - Horst Forster

Over the last year especially, with the passing of various Open Access mandates, this phrase has been used more and more - today by Horst Forster of the European Commission.

His perspective (and many of his colleagues in various areas of the EC, from my experience) on the accessibility of publicly funded research is in line with the leading OA advocates. Though still, as good as access is today in comparison, there is still an access problem.

Forster briefly touched on the landscape of OA currently, citing statistics from ROARMap of there being 3,700 OA journals to date, a 15 percent increase since last fall; and current figures of there being 57 self-archiving mandates, a 42 percent increase since last year.

But questions still remain -- how do you fund the dissemination of scientific knowledge? Why is there a limited uptake of Open Access?

Mandated depositories are not as rosy of a strategy as it may seem, he said. Deposit compliance rates in other areas are lower than the figures cited (NIH - 60+% current compliance and Wellcome Trust - 37% of their work accessible).

His view, researchers and funding institutions are not pursuing this rigorously enough -- and points the finger at himself as well (Note: Forster is responsible for a 1 billion euro funding program.)

The EC currently offers fair remuneration for those who try to make the OA system work. They've also recently enacted a large scale experiment in OA to peer reviewed articles, as part of their Framework Programme 7. In this pilot, articles resulting from certain areas of funded research under FP7 will be made available within 6-12 months. This OA pilot covers close to 37% of the budget of the research group until 2013 - a figure that sits at approximately 20 billion euro. Yowza.

Costs and Benefits of OA, part 1 - Berlin 6

Time to tackle synthesizing Tuesday's "Cost and Benefits of OA" plenary session at Berlin 6. Apologies for not getting this out sooner.

This session seemed to be of particular interest to the crowd, since business models and sustainability for OA publishing is an ever growing concern.

John Houghton (Center for Strategic Economic Studies at Victoria University in Australia) kicks off the session, with some of the results from his latest joint-study with JISC. Following Houghton were familiar faces Matt Cockerill (Publisher of BioMedCentral, a profitable OA journal), Salvatore Mele (CERN, speaking on the SCOAP3 project), with Frederick Friend (JISC) serving as the moderator and offering his own comments.

Here are my notes:

1. John Houghton (Center for Strategic Economic Studies at Victoria University in Australia)

Houghton approaches the costs and benefits of OA from an economic perspective, where the aim is to have the most cost-effective system, not necessarily the cheapest. The key aim in his latest Australian study looks at the benefits, as well - which he says are often overlooked in discussions about cost.

"It's not only about cost, it's about added value" - Fred Friend, speaking about Houghton's presentation

For the purposes of this talk, he speaks about a JISC funded study on the Economic Implications of Alternative Scholarly Publishing Models. The aim of the study is to explore the institutional, budgetary and wider economic implications of the major emerging models for scholarly publishing. Phase 1 of this study (which has not been made public yet - still a work in progress) is to describe the publishing models and identify all associated costs and benefits. Phase 2 addresses quantification - comparing costs and models.

The study is extensive, take my word on this. Instead of going in to this in fine detail, here are a few main points.

"There is a lag between R&D expenditure and the realization of the return to research."

There is a vast amount of information on returns to R&D - which while varied, largely show that social returns to publicly funded research is high. The standard neo-classical model assumes that all R&D generates useful knowledge and that all knowledge is equally accessible - which is unrealistic.

Houghton states that it's not possible to compare toll and OA publishing directly at the national level, since they perform completely different functions. He describes this difference in function as toll access seeking to provide UK subscribers (the JISC study studies on UK researchers, FYI) access to worldwide research, whereas OA seeks to provide worldwide access to UK research.

His study approaches it from both sides and will explore the following ideas, among others:
- the diversion of research funding to author pays fees
- the impact of delayed OA
- the impact of speeding up the research and discovery process

Houghton is currently working to finalize the report, which is expected to be released in early 2009.

There's a podcast up on the JISC site on this that I haven't had a chance to listen to just yet. But, to access, click here.

Open Access in the Developing World - Berlin 6

(Note: My notes for some panels may not be as extensive / as much of a data dump as others. This one will focus on a few main themes and concepts that resonated particularly.)

"We are never completely contemporaneous with our present. History advances in disguise; it appears on stage wearing the mask of the preceding scene and we lose the meaning of the play." - Régis Debray, "Revolution in the Revolution?"

Arun (Subbiah Arunchiam, Center of Internet and Society, Bangalore, India) led his individual presentation with a snippet of this quote from Debray. Most developing countries have not yet made the transition from the past to becoming contemporaneous with the present -- neither have they seen the same levels of transformative impact of science and technology as other more advanced countries.

A couple of main themes from this session include the local importance that knowledge from developing countries serves -- based on local evidence, often having a direct impact on local policy making. Clearly, access and resources are main issues for those in the developing world, whether they are characterized as "middle economy" players, "developing countries" or what have you. Access and scarcity of resources plague a significant percent of the world's population, though this panel provided examples more focused on different publishing models (SciELO, MedKnow) that better helps get the scholarly canon into the hands of their countries' researchers - as well as integrate into the broader global scientific knowledge exchange.

Solange Santos gave a wonderful presentation on SciELO (Scientific Electronic Library Online) - an electronic library based in Brazil covering a wide array of the countries' scientific journals. SciELO has been up and running since 1997, serving as a "meta-publisher" somewhat of over 500 journals. We have been working with our contacts in Brazil and at SciELO to integrate the Creative Commons licenses - conversations that continue to this day.

I'm glad Santos was able to make the haul over to Dusseldorf to join us. SciELO is a very interesting example of a publishing model in a developed country.

New Modes of Scholarly Communication: Blogs, Wikis and Web 2.0 - Berlin 6

Time for the first plenary session at Berlin 6 -- New Modes of Scholarly Communication: Blogs, Wikis and Web 2.0 in Academia. This session explores new forms of scholarly communication beyond the paper metaphor (an idea that jtw and I speak on often), with the focus of this being specifically on hypertext publishing.

My notes on this session will be brief, since a lot of this is common knowledge, but important nevertheless.

OK, on to the panelists:
1. Lamber Heller, TIB / UB Hannover

"We're experiencing a radical shift in how we establish authority, significance, and even scholarly validity" - Michael Jensen, New Metrics of Scholarly Authority

Heller's presentation was on the challenging gap between community feedback and academic recognition, specifically in terms of blogs and blog aggregators. He views blog aggregators (public ones, rather than Google Reader which I don't believe is publicly searchable?) as a new communication pattern in and of itself. Many in the research community, Heller says, are encouraging this new form of discourse on top of the existing infrastructure for scholarly communication. Blogging and aggregators will never replace traditional media - digital or paper, but are meant to act as a complement or added layer to pre-existing knowledge.

2. Lilia Efimova, Telematica Instituut, currently finishing her PhD and blogging it as an exercise

"Writing itself is thinking ... "

Efimova began with a brief introduction (she's actually blogging her PhD research currently, hoping that she's granted the title in the coming weeks :) We'll see if blogging her PhD really affected the committee's decision!), and then detailed in general terms the process of growing ideas. First, you sense there is something interesting, start to articulate insights (very fuzzy and difficult to grasp, usually pre-funding, etc.) -> start to connect the dots, pull together disparate knowledge sources, start to conduct experiments, getting the formal setting to do research -> lastly, you work on the publication.

Efimova believes that blogging the research provides a place for ideas that "do not fit" necessarily, allowing one to park the research in a trusted repository as the research is being conducted, analyzed, written up.

This form of research documentation allows for fragmented bits of information to be linked, information readily reused, creating yet another area for networked research. The process is transparent - observable and traceable, can at times increase one's visibility (important for junior researchers), as well as providing baked in mechanisms for attribution and ownership.


3. Lorrie LeJeune, from the OpenWetWare (OWW) project / MIT

"It's about open collaboration, as well as Open Access. Our mission is to be a premier and lasting environment in which biologists and biological engineers can work together in real time to produce, share and improve knowledge."

LeJeune is the managing director of OWW - a web-based knowledge sharing source, currently utilizing wikis to document lab reports, experiments, protocols and other information in the biomedical sciences.

"We want to have an impact on the dissemination of biomedical information."

There's a real incentive to sharing in some disciplines, because you want the visibility. Many scientists fear their work being scooped, but say, if you're on the fringe of a scientific discipline (think "synthetic biology" 3-5 years ago), that visibility may lead to your next article, your next collaboration or even next grant.

This is a project that SC is collaborates with (John's on the Board, as well as the rest of us offering our support as members of the same community.) In many ways, as LeJeune stated, OWW is the embodiment of what this panel is about - and one I continue to marvel at, as a user base of OA and open collaboration in the purest sense.

Tony Hey (MSR) on e-science and OA - Berlin 6

Tony Hey of Microsoft Research, formerly of the UK e-Science initiative, former Dean of Engineering at Southhampton, kicks off the discussion with the first keynote.

Researchers are facing moving from data to information to knowledge, at a very rapid pace. I think it's safe to say that the digital deluge in research is a reality uncontested largely, and there is an ever important need for ways to collect, analyze, process, visualize and archive this mass of digital information.

Hey speaks about the fourth paradigm, one of data centric science. He lists the other paradigms / trends of scientific research as the following:

- thousands of years ago - experimental science
- Last few hundred years - theoretical (ie., Newton's Laws)
- Last few decades - computational (ie., simulation of complex phenomena)
- Today - data-centric science (or as described at the SEED/WEF brainstorming session - a bioinformatics revolution, correlation based science)

We're experiencing a new kind of publishing - where data is often published before the actual paper. This is a different paradigm for publishing.

Hey tosses out the example of the Sloan Digital Sky Survey, the first major astronomical survey project, with 5 color images of 1/4 of the sky and pictures of 300 million celestial objects. The SkyServer, Hey says, is the posterchild in 21st century data publishing, an example of what you can do if you make the data available, and also an example of citizen science (Galaxy Zoo). Galaxy Zoo is a project where they ask the community to participate and help classify galaxies.

In speaking about his time as the Dean of Engineering at SouthHampton, Hey says that traditional means for scholarly publishing is a model in crisis. Journal subscriptions are rising faster than library budgets, web technology and digital media now make dissemination of knowledge "easy" and "Free" without traditional paper journals.

His e-science work continues at MSR, colored also by his role on the NSF advisory committee on cyberinfrastructure (ACCI).

"In order to help catalyze and facilitate the growth of advanced CI, a critical component is the adoption of Open Access policy for data, public [literature] and software" - NSF ACCI

"Make sure we walk the talk." - Hey

Cloud Computing is, to Hey, where we're going in the future.

We're moving towards a world where all data is linked ... and where data is stored/processed/analyzed in the cloud.-- Data is interconnected through machine interpretable information - social networks as a special case of these data networks, or "data meshes".

Cloud computing will affect all sorts of research, and OA - copies of articles can be trusted to the cloud, perhaps IRs still keeping local copies as well. Hey believes the OA movement will succeed, presenting a new rapprochement with publishers.

Open Access and Franz Kafka - Berlin 6

'Allo from the Northern Rhineland. I'm currently sitting at the Berlin 6 Open Access Conference - Changing Scholarly Communication in the Knowledge Society in Dusseldorf, Germany.

In the organizers' opening remarks, one gentleman from Heinrich-Heine referred to a children's fable from Franz Kafka (specifically from "A Little Fable", not trying to sound redundant). The story is a familiar one:

'Alas,' said the mouse, 'the whole world is growing smaller every day. At the beginning it was so big that I was afraid, I kept running and running, and I was glad when I saw walls far away to the right and left, but these long walls have narrowed so quickly that I am in the last chamber already, and there in the corner stands the trap that I must run into.'

'You only need to change your direction,' said the cat, and ate it up."

That's open access, the organizer said, trap or cat.

The internet allows for global and interactive representation of knowledge, preserving cultural heritage and allowing for worldwide access. Openness being the new base or new foundation for global scientific exchange.

And with that, I'm going to do my best to post notes here. Bear with me, this live blogging thing is relatively new to me :)