The JISC ebooks universe

My colleague Ben Showers has recently been looking across the work taking place around digital books in all their forms: open textbooks, digital monographs, epub, web-based books. For educational institutions the need to keep up with the content needs of learners and researchers is paramount but so much is happening at the moment, with hardware, content formats, the emergence of new authoring tools and rising user expectations, so where do you start?

We have pulled together some key information for decision-makers, with a distinct JISC flavour. Particular thanks to Caren Milloy from JISC Collections and Zak Mensah from JISC Digital Media for their help.

 

Legal (Licensing, IPR, DRM)

Business Models

 

Technology and Standards

 

User behaviour/Requirements

What have we missed?

Please let us know what resources you find most useful, from JISC and elsewhere, in meeting the challenge of ebooks in your institution.

 

Ben Showers and Amber Thomas, JISC Digital Infrastructure Team

May 2012

(last updated 28th May 2012)

Beyond Grid vs Cloud – EGI Community Forum 2012

‘The grid? Shouldn’t they all be doing cloud computing now?’ As a JISC programme manager working with the National Grid Service (NGS) project people ask me this question more and more often. ‘Absolutely’ and ‘not at all’ is the seemingly contradictory answer I usually give, for instance the other week when I mentioned I would attend the European Grid Infrastructure Community Forum 2012 in Munich.

the delights of cloud and grid computing: cake

conference coffee break revelations: EGI delivers -- cloud, grid and cake!

I give this answer because the question originates from a double misunderstanding. The first is about the nature of cloud computing that, despite some marketing claims, is not the answer to everything and in some ways more a new business model than a new technology as such. The cloud is neither the solution for all computing needs, nor is it always the cheapest option – as a recent study commissioned by EPSRC and JISC (PDF) shows. The second misunderstanding relates to branding and the history of research computing. When projects like the National Grid Service were envisaged, grid computing was the dominant paradigm and their names reflect that. These days however, they are looking at a broad range of technologies and business models for providing compute resources, and mostly refer to themselves by their acronyms: NGS and EGI in this case. So at least for the initiated it was no surprise that the EGI conference was as much about the cloud as it was about the grid.

The conference, hosted by the LRZ supercomputing centre and the Technical University of Munich, was a five day event to bring together members of the research computing community from across and beyond Europe. With several parallel strands and many session to attend I won’t summarise the whole conference but instead pick out a few themes and projects I personally found interesting.

First of all I noticed there was a lot of interest in better understanding the use of e-infrastructures by researchers and, related to that, the impact generated by this. In some ways this is a straightforward task insofar as easy to capture and understand numbers can be collected. The EGI for instance now has over 20,000 registered users. You can count the number of cores that can be accessed, monitor the number of compute jobs users run and measure the utilisation of the infrastructure. However, this becomes more difficult when you think of a truly distributed, international infrastructure such as the EGI. Will national funders accept that – while the resources they fund may used by many researchers – much of that usage originates from abroad? If we want to support excellent research with the best tools available we have to make it as easy as possible for researchers to get access to resources no matter which country they are physically based in. Thinking in terms of large, distributed groups of researchers using resources from all over the world, often concurrently, the task of understanding what impact the research infrastructure has and where that impact occurs (leading to who may lay ‘claim’ it in terms of justifying the funding) can make your mind boggle. We need business models for funding these infrastructures that don’t erect new national barriers and address these problems from the angle of how to best support researchers.

Business models, not surprisingly, was another theme I was very interested in. Complex enough already, it is made even more difficult by commercial vendors now offering cloud nodes that for certain, smaller scale scenarios can compete with high performance computing – how do you fairly compare different infrastructures with different strengths and very different payment models? Will we see a broadly accepted funding model where researchers become customers who buy compute from their own institution or whoever offers the best value? Will we see truly regional or national research clouds compete against the likes of Amazon? What the conference has shown is that there are emerging partnerships between large academic institutions and vendors that explore new ways for joint infrastructure development. One example is a new project called ‘Helix Nebula – the Science Cloud’, a partnership that involves CERN, the European Space Agency and companies like T-Systems and Orange. Such partnerships may have a lot of potential, but finding legal structures that allow projects based in academia to work in a more commercial environment is not always easy. A presentation from the German National Grid Initiative explored some of these problems and also the question of developing sustainable funding models.

In order to develop good models for funding e-infrastructure we also need to understand the costs better. As far as institutional costs are concerned these are mostly hidden from the researchers, whereas the costs of commercial providers are very visible – but not always easy to understand in terms of what exactly it is you get for a certain price per hour. As our cloud cost study shows this is an area where more work needs to be done, and so I was happy to find a European project that aims to address this. e-FISCAL works on an analysis of the costs and cost structures of HTC and HPC infrastructures and a comparison with similar commercially offerings. It already lists a useful range of relevant studies and I hope we will see more solid data emerge over time.

In the commercial/public context I found it interesting to see that some academic cloud projects aim to take on commercial players. BeeHub, for instance, was presented as an academic DropBox. Now, to be fair to the project it aims to be an academic service for file sharing in groups and to address some of the concerns one might have regarding DropBox, but I wonder how successful they will be against such a popular offering.

I was also very interested to learn more about initiatives that address the critical question of training. Usually these are researcher training or more technically focussed programmes, but the EU-funded RAMIRI project offers training and knowledge exchange for people (hoping to be) involved in planning and managing research infrastructures.  Because of the increasing complexity of this task in terms of legal, cultural, technical and other issues better support for those running what often are multi-million projects is highly desirable.

As I cannot end this post without referencing the more technical aspects of research infrastructure let me point you to a project that shows that grid and cloud can indeed live together in harmony. StratusLab is developing cloud technologies with the aim of simplifying and optimizing the use and operation of distributed computing infrastructures and it offers a production level cloud distribution that promises to marry ease of use of the cloud with grid technologies for distributed computing.

To sum up, it is not a question of grid versus cloud. It is about selecting the technologies that are best suited to facilitate great research – and then do deal with the non-technical issues from training to sustainability and cultural change that will decide how well we will be able to make use of the potential the technology offers.

The OER Turn

At ALT- C I participated in several discussions around open content, and then this week we ran the closing programme meeting for Phase Two of the HEA/JISC OER programme.  I feel that a new perspective on academic-created open content is emerging.

I think it’s sometimes useful to think in terms of foreground and background: most of the elements are there and have been there all along, but some take centre stage. It’s a question of weight and attention given to the different activities, a question of where the discussions happen.

2009 Foreground 2011 Foreground
Focus on provision Focus on use
Focus on educator as provider and user Focus on non-educators as users
Open courseware, the course online Rich media, beyond text
Embedding in courses or free searching online Tutor or peer or social recommendation
CC BY NC ND  for editing by educators CC BY for remixing by anyone
Focus on licensing is key Focus on licensing might be distracting
Institutional workflows Institutional support for open practice
Storage and presentation services Brokerage services
OERs Open content

Just to stress, all of these views are evident in part in 2009, 2010 and 2011, and even before that. My interest is just in what is shifting in the foreground. This is just my first take and I hope it will stimulate discussion. It would be really interesting for others to write their own lists.

I think that enough of the focus has shifted that the space has changed.  I christen this The OER Turn. The OER Turn is characterised by the working through of the following topics:

Visible and Invisible Use

The Value of Reuse Report by Marion Manton and Dave White of Oxford, produced for the OER programme, uses an iceberg analogy of visible and invisible use. It helpfully guided our discussions at the end of phase two programme meeting. Previous to this, David Wiley’s “toothbrush” post was the example cited of this conundrum. The fact that it was so divisive amongst OER people shows that it hit a nerve (excuse the pun). I think the iceberg diagram draws on this and illustrates the problem of visibility.

The general consensus amongst people involved in UK OER Programme is that reuse of web-based resources does happen, all the time, but it is usually private, mostly invisible to the providers and often not strictly legal. So there is above waterline use and below the waterline use.

The visible use of open content tends to be institutional-level use that is itself more visible because it is openly shared again: it is being relicensed out again. Where institutions want to reuse content in aggregate open resources, it may influence the types of content they reuse, and they way in which they do it.

Messages:

On reflection, perhaps the top of the iceberg fits most closely with 2009 foregrounded conception of OER, the majority of the iceberg is what is being given more attention in the 2011 conception.

The importance of licensing?

Naomi Korn’s video introduction to licensing and IPR issues illustrates the concept of rights in > rights out. The more rights you want to give out to users, the more you have to restrict yourself to content you have obtained broad rights in for. As the risk management calculator illustrates, if you use content licensed as CC BY NC ND, you cannot licence it out as CC BY NC, because that doesn’t not include the ND clause. And CC BY SA can only be licensed out as CC BY SA, so cannot be remixed with anything less than CC BY or more CC BY SA. Share-Alike is quite restrictive. This is counter-intuitive but true. Play with the calculator and find out for yourself.

It may be only when reuse is more visible, such as formal institutional adoption of third party resources (above the waterline) that the risk of reusing un-cleared content is high enough to make the open license a key aspect of OER.Institutions as providers of content may wish to choose different licences to what institutions as  users of content want. They may want to publish content as (c) all rights reserved. If it is a choice between that or nothing, what should they choose?  Note that they could publish something as (c) all rights reserved unless otherwise stated, and have clearly marked CC elements. Polarising a simple open or not open isn’t helpful. As Naomi Korn put it, “what is the opposite of open”? Paid-for? Authenticated? Not editable? Open content needs to be viewed as part of the wider ecosystem of content. What it “affords” will be specific to the use case. Interestingly this is a good parallel with accessibility: “accessible to who to do what?”

Reflecting with Naomi Korn we shifted our recommendation from “whichever CC licence suits” (1) in phase one to “ideally CC BY” in phase two. John Robertson has summarised licensing choices made by the projects.  This is an example of how the programme has pushed forward our understanding of this area, including that of the legal experts. If we knew the answers at the start, it wouldn’t be an innovation programme!

Thinking above to the points about visibility: if the open content is not shared out again under an open licence then it might be being used but not so visibly. It might show up as web stats, but even then, once the content has been copied to another place, as any CC licence lets you do, then the content proliferates, as does the usage. Usage becomes even harder to see.

Another implication of described use as above below the waterline is that the risk is significantly less. The feeling was that this is an appropriate risk evaluation.  So, open licensing does matter for formal institutional use, less so for practice that is “under the waterline”.

Messages:

We are genuinely learning together on this: as open licensing spreads and models of use develop, we will need to be alert to the mechanics of remixing content. And also open to the possibility that the end point of the supply chain need not be open: not all content will be licensed out again.

A great example of the OER Turn is how  OER Commons includes resources described as “read the fine print” licenses. Or Therese Bird’s work on iTunesU.

However … I still want to hear more about the rights that Creative Commons licenses grant to translate and shift formats, for above the waterline activities. Examples please! This could be key.

The role of the university

Alongside the previously dominant narrative of OERs as exchanges between educators and each other, and educators and their learners, there is also the potential for academic open content to connect academics with the public. I explored that a little in a previous post, Tony Hirst further in his post on OERs and public service education, University of Nottingham, Oxford University and other institutions also see their open content as part of their public engagement.

In some scenarios opening up universities through open practice is about OER, in the top of the iceberg sense. But much of the benefits of opening up knowledge to the public can be achieved without open content licensing (unless the brokering services need it). Initiatives like Warwick Knowledge also draw on the notion of the university as a pubic knowledge organisation.

So there is a strong argument for focussing more on the public and global use of open content.

Sidenote: Critics of the toothbrush analogy might say that this is what they meant all along. I’m not sure that is true. If it is, it wasn’t very well articulated. Because we still need to understand the drivers behind provision and how the benefits of public engagement can be articulated.  Academic staff are a university’s most precious resource. The needs of institutions, who pay academic’s wages, need to be factored in to how open practice can be supported.

The pursuit of global knowledge is not owned by universities. Wikipedia, Slideshare, YouTube, Twitter, Delicious have all seen a blossoming of thoughtful quality input from a huge range of sources. The role that academics play in this open content space is as one type of contribution. Learners contribute too. But so do millions who are outside of formal education.

OER is dead. Long live academically-created content appropriately licensed and formatted to support intended users

Not quite as catchy, is it? However I am increasingly hearing suggestions that OER is not a useful term any more, aside from a supply-side term relating to the visible tip of the iceberg. I have recommended for some time that we drop the term and focus instead on open content and open practice.

Having asked a year ago what the “O” in OER means, now I find myself asking what the “Open” in Open Content means. Well, it definitely means free (not paid). And it means easily linkable, which means not authenticated (not closed). However what about near-ubiquitous controlled access through gmail or facebook? Sometimes the format matters, sometimes the licensing matters. Maybe this matters a lot for content to cross language boundaries, maybe it matters a lot for accessibility. In which case do we need to articulate the costs and benefits of open content for those use cases? We don’t want to kill open practice dead by focusing too strictly on definitions of openness any more than we want to kill open content by diluting the promise to users seeking editable re-licensable content. What questions should be asking about open content?

What do you think?

Amber Thomas

Creative Commons Licence
OER Digital Infrastructure Update by Amber Thomas is licensed under a Creative Commons Attribution 3.0 Unported License.
Fromt infteam.jiscinvolve.org.
Permissions beyond the scope of this license may be available at http://www.jisc.ac.uk/contactus

Footnote (1) The wording was  “Examples of suitable licences include those with the “Attribution” and “Share-Alike” clauses. We would encourage projects not to use the non-free Creative Commons variants (such as “Non Commercial” and “No Derivatives”), as these negatively affect the reusability of resources. However, we do recognise that in some specific circumstances this may not be possible.” (Thanks to David Kernohan for extracting)

Halfway through the US Blue Ribbon Meeting

Greetings from Blue Ribbon Meeting again. http://brtf.sdsc.edu/index.html

Some interesting angles emerging from a variety of participants. Heard from Thomas Kalil this morning who works as a policy staffer at the White House in the Office of Science and Technology. (Is he perhaps what they call a ‘policy wonk’ … I don’t know. I should have paid closer attention to the West Wing). Anyway – he talked about how the preservation community might be able to get this whole area of digital sustainability onto the Presidential radar. What we don’t do is to present this as a problem that needs to be tackled in ten different ways, at different levels, by diverse stakeholders. We need to realise that the White House is in the business of saying ‘no’ most of the time, so we need to make it as easy as possible for them to say ‘yes’. We need to be realistic about what we’re asking for and we need to be very directed about who in the Presidential office we approach to champion the cause.

What we probably do is get a message to the President that person x at trusted organisation y, has a low cost/high benefit measure that the President really needs to hear, which fits with his broad agenda, and has the backing of many thought leaders across the expert community.

I’m assuming we can transpose that to Downing Street.

 Right … we’re back in session. More later.

#brtf

The Economics of Sustaining Digital Information

I’m in Washington to attend the US Symposium of the Blue Ribbon Task Force on Sustainable Digital Preservation and Access.

Symposium programme – http://brtf.sdsc.edu/symposium.html

This is the end of a two year process of enquiry and analysis where  14 experts have had a series of meetings to consider what the economic implications are, and what economic frameworks are required, to ensure that all of the expensively acquired knowledge that we commit to digital formats is available for as long as we need it … or in some cases, even longer! – given that there is a great deal of information around that subsequent generations might find useful even though we have no clear idea what we should do with it!

The panel of experts is mostly drawn from the US although the UK has been represented by Paul Ayris (UCL) and Chris Rusbridge (DCC). The final report is now available at: http://brtf.sdsc.edu/index.html

The report focuses on four types of information: scholarly discourse; research data; commercially owned cultural content; and collectively produced web content, and uses these categories to frame some recommendations for a range of different stakeholders. These are presented as a number of bullet-pointed lists and tables which can (and no doubt will) be extracted from the report by way of summarising some of the detail contained in the main body of the text.

For those of us not au fait with the language of economics, hearing digital materials referred to as ‘non-rival depreciable durable assets’ makes you stop and think for a moment … but as the concepts are explained and the principles become clear, this well written report starts to give you a slightly new take on the long-term management of digital resources.

#brtf