From Bjoern Hassler's website
Jump to: navigation, search

Mixed Methods UJ
RFF Zambia
OER Guidance for Schools
OER4Schools (since 2009)
ORBIT (2011-2012)
ANTSIT (2010-2011)
UNESCO Access2OER (2009)
Samfya (2008-2011)
Steeple (2008-2010)
CamTV (2008-2010)
ICTP WS (2007)
OpenLearning (2007-now)
VideoUnit (2006-2013)
Video Hosting (2005/2006)
ScienceLive (2005-now)
MediaPlayer (2003-2010)
BlueSci (2002-2005)

1 The Access2OER UNESCO OER Community Discussion

One of the Access2OER cartoons developed during the discussion.

The Access2OEROER Community discussion sought to bring awareness to digital divide issues that effectively stop access to otherwise open educational resources. Open licensing and public distribution of OERs is necessary, but not sufficient, for end-users to make good use of such OERs. An important consideration is the connectivity available to the end user.

Often, the issue of (lack of) connectivity is seen purely to do with the available technical (national) infrastructure, while capacity building and human resources aspects are neglected. However, roviding OERs in appropriate ways, as well as managing local networks carefully to allow access are paramount to ensuring access to OERs. Without such skills and best practice guidelines, even in scenarios where technical infrastructure is plenty, access to OERs is difficult.

Two concrete suggestions:

Unfortunately it seems that many of these guidelines haven't yet penetrated into the OER community, hence the suggestion to set up the present 'Access2OER' project.

To raise awareness, UNESCO Open Educational Resources Community launched a new three-week discussion on the subject of access issues. launched a new discussion on Open Educational Resources (OER): Access to Open Educational Resources, 9-27 February 2009. During the first week the community focussed on identifying - and attempting to classify - the main problems in accessing OER. The theme of the second and third weeks was "tried and tested solutions", as participants were invited to share their experiences in working around such problems. The discussion was facilitated by Bjoern Hassler of Cambridge University’s Centre for Applied Research in Educational Technologies.

About the community: UNESCO’s international Community on Open Educational Resources has been active since 2005. It connects over 700 individuals in 105 countries to share information and discuss issues surrounding the production and use of Open Educational Resources – web-based materials offered freely and openly for use and reuse in teaching, learning and research. UNESCO’s work on Open Educational Resources is generously supported by the William and Flora Hewlett Foundation.

2 Discussion report

The report is available as pdf here: Access2OER final report.pdf (info), and you can also read it online on this wiki (Access2OER:Home). The report was originally online at UNESCO OER Community wiki Subscription to mailing list here:

3 Proposals that emerged from the discussion

The proposals emerging from the dicussion are reported in the proposals section of the report.

Having outlined the principle issues that affect access to OER, and surveyed a broad range of potential solutions in the previous two weeks, the third week of the discussion was dedicated to developing proposals in areas where further work is needed. It was beyond the scope of the discussion to develop fully finished and detailed proposals; instead the aim was to explore which of the ideas from the previous weeks could be framed as viable proposals. In this sense what follows are really "proto-proposals". They are collections of ideas in proposal form, to help stakeholders think creatively about how to move forward on the question of access to OER.

The suggestions for proposals can be grouped into three overlapping themes:

  • training;
  • Open Educational Resource centres;
  • OER delivery.

Readers may also find it helpful to refer to the original discussion and accompanying notes on the OER community wiki:

3.1 OER training proposals to address lack of awareness and skills

The need for training is evident. Many participants testified to the fact that awareness of OER remains low among educators (see, for example, the case studies in Chapter 2. Teacher skills and behaviours required for identifying, using, and sharing digital teaching and learning content, such as technological and information literacy, and cultural shifts, need to be supported and incentivised.

Training is needed particuarly in the areas of:

  • "technology literacy" for users that are not familiar or comfortable using computers or the internet and computers, and at a higher level for IT officers to enable them to evaluate and choose appropriate educational platforms;
  • "information literacy" for users that do not know how to find and select information online;
  • "design of learning environments", for example for face-to-face instruction with the support of technology (blended learning), and for educational strategies like "active learning";
  • "evaluation and assessment of educational programmes", to evaluate the success of in-course OER use;
  • "OER policy issues" for decision makers, covering production, dissemination, adoption, infrastructure, etc., as well as copyright issues.

In terms of the classification of access issues, training proposals would address:

  • Social, awareness, policy, attitude, cultural:
    • Access in terms of awareness;
    • Access in terms of local policy/attitude;
  • Technical: receiving OER:
    • Access in terms of internet connectivity/bandwidth;
    • Access in terms of discovery;
    • Access in terms of ability and skills.

The key element in the design and execution of any training activity would be the adoption of participative principles: participative development of course proposals; participative teaching of courses; and a train-the-trainer approach to maximise scalability and the opportunity of independent replication. Workshops would need to be structured so that participants become trainers immediately, able to hold similar workshops themselves. Course could be delivered online initially; trainers could later expand this to face-to-face training in their respective countries.

Finally, any new training initiatives would need to integrate and build on existing initiatives, for example those of OER Commons.

3.2 A proposal to build documentation to support the setting up of OER centres

The following proposal is related to the OER training proposal outlined above, but it takes a more holistic approach to OER access and use, including addressing infrastructural needs. While the proposal goes beyond just OER, OER plays a key role in it.

Scalable OER deployment depends on ICTs, so it can be argued that successful deployment of OER needs to take into account improving existing structures and rural connectivity. There are various strategies to address those questions. It was proposed that we should focus on what a comprehensive and coherent set of Open Educational Resources that address those issues would look like. Here are some of the questions that would need to be addressed:

  • How do you connect a rural school? How do you get affordable connectivity? How do you share a connection?
  • How do you manage the connection to make optimal use of the bandwidth?
  • How do you design robust, maintainable, low-power ICT equipment? (What are the cost implications of solar power, deep cycle batteries or a generator?)
  • How do you obtain OER materials for local use?
  • What training do you provide on ICT and OER?
  • How do you overcome "brain drain", whereby trained people leave for better jobs?

There is of course much documentation available, but it focuses generally either on the ICT/hardware/connectivity aspects, or on the OER aspects - rarely both. For instance, there are plenty of instructions available for installing Ubuntu and plenty for installing Moodle. However, at least to the knowledge of the present author, there is no comprehensive set of tutorials to take a novice user from a blank computer, through installing Ubuntu/LTSP, installing Moodle and adding OER packages into Moodle, to conducting meaningful training/learning with the installation and content packages. Individually, there are guides and instructions on the Internet for each of these elements, but they come from and are written for different communities. It would be quite rare to find a single individual that could bridge the whole spectrum. Moreover, in a rural ICT situation, the process described above is only one element. More skills would be needed. Overall, there is a need for training (and associated OER materials) that give the whole picture, taking the user from an empty building, to an OER-equipped training centre.

In terms of the classification of access issues this proposal addresses:

  • Technical: receiving OER:
    • Access in terms of infrastructure;
    • Access in terms of internet connectivity/bandwidth;
    • Access in terms of ability and skills.

There was also a significant amount of discussion around the One Laptop Per Child initiative. It may thus be justified to frame a proposal that focuses more on infrastructural issues.

Relation to existing institutions

Any proposal would need to build on the structures that already exist in different countries. To take the example of Nigeria, the country has the National Educational Technology Centre in Kaduna, Education Resources Centres in regional capitals, audio-visual centres in universities and local school management boards. Any new initiative should consider networking them in a useful way.

Selection of existing materials

Kim Tucker suggested that it would be useful to list existing resources which could be adapted and updated each implementation. Links to the following resources were shared with the community:

Resources could be classified according to their suitability for different learning situations. A tool could then be designed to select and present the most suitable set for a given situation. Compiling a full list of resources, however, might be a significant challenge. The task of designing a solution and selecting relevant learning and other resources could be carried out when needed. The project could focus on methodology, research, collaboration, preparation of the next custom guide and mobilising multidisciplinary teams for implementation.

There is a seperate wiki page for notes on an add-on proposal.

3.3 An OER exchange infrastructure

The third area of proposals focussed on improving infrastucture for OER. This was motivated by a number of "What if?" questions:

  • What if any computer you bought (be it a netbook, desktop, server, harddrive, memory stick) could come preloaded with a free content collection?
  • What if when you placed an order for a device, you could choose from a large catalogue of preloaded OER/OCW/free content with no further connectivity needed? However, when and where you had connectivity, your chosen content could be updated and extended automatically. (And content collections freely installed, transfered and shared from the internet.)
  • What if any computer could also be pre-ordered with a free content production suite? Not just OpenOffice, but a complete set of key OER applications, including Audacity for podcasting, Moodle and EduCommons to enable course development, and a set of training materials?
  • What if the content you created could then be contributed back to the global community, even where there was little or no connectivity?

In principle, many of these ideas could be realised now, but for the moment there is not enough coordination within the worldwide OER community. Participants proposed the following steps to move the community towards those goals:

  • hybrid information delivery strategies (North-South, South-North, South-South, North-North; seamless online/offline content delivery; caching);
  • suitable strategies for content packaging so that content could be delivered in this way;
  • content transformation/transcoding methods (would also include wiki content transformation options for Wikipedia, WikiEducator, etc.), including bandwidth management and bandwidth managed resource delivery.

In terms of the classification of access issues this proposal addresses:

  • Social, awareness, policy, attitude, cultural:
    • Access in terms of local policy/attitude;
  • Technical: provision of OER:
    • Access in terms of file formats;
  • Technical: receiving OER:
    • Access in terms of internet connectivity/bandwidth;
    • Access in terms of discovery;
    • Access in terms of ability and skills.

3.3.1 Scenarios

The worldwide OER community has yet to realise the goal of downloading OER materials easily, without bandwidth problems. However, there are a number of strategies that could be implemented quite quickly and relatively cheaply, as they just depend on software.

The possible options can be illustrated with a number of scenarios. Let us suppose that we are based in a rural school and have only slow connectivity available. We are using Miro - a free, open source application that can download podcasts. However, we are using an enhanced hypothetical version of Miro, called "SuperMiro", that is geared for use with low bandwidth connections. SuperMiro is also be able to understand specialised very low bandwidth formats, such as AMR narrowband.

Suppose we use the hypothetical SuperMiro application to download a podcast. Normally, the connection would be made straight to the podcast server and would put immediate strain on the network, preventing others from browsing the web, or sending and receive email. However, with the new and improved SuperMiro application, the subscription does not go straight to the podcast server. Instead it goes first to a local server at the school, then via a national school gateway run by the national research and educational network operator (NREN) to provide an internet exchange point for schools and universities). Only then does it go to the podcast server.

Diagram 1. The downlink

Diagram 1 illustrates this for generic OER/OCW content. (Diagrams illustrating more scenarios can be found on the wiki on the OER exchange cartoon page.) Content providers have content on their website that is also mirrored into a global OER/OCW mirror. From the OER/OCW mirror the content is pushed to a national mirror (for instance with the NREN), and then to a local mirror in a school or university. To the user of the content this might be "transparent": the user thinks that they are accessing (e.g. MIT OCW) content through their browser directly, just like any other web content, except for some reason the OCW content is faster than the rest of the web.

Coming back to the SuperMiro podcast scenario, it is important that the "network" talks back to the SuperMiro application, so that SuperMiro does not take up all the available bandwidth. Instead, SuperMiro should be able to find out the total bandwidth available and restrict itself accordingly. The user is informed of the total download time and has the option to get a low resolution preview, while waiting for the high resoluation file. This preview need not be generated by the podcast server itself: it could be generated "on the fly" on the global mirror server. The user chooses audio/image preview and has the file in an hour. When the user has listened to and/or watched this, SuperMiro says: "A higher resolution version is available - do you wish to download it?" If the user proceeds they will receive an email in a day or so to notify them that the high resolution file is now available on their school server. When the download is complete a copy of the audio/video file is kept by the full chain of servers: the school server, the national gateway server, and perhaps another regional internet exchange point. Others requesting the same file will not need to go back to the podcast server (or the global cache) to get the file. However, every time a file is requested from any of the servers, the originating podcast server gets a "ping" so that they have good statistics about how their media are being used.

The same mechanism would work for open courseware and other content packages, as well as audio/video files. For content packages (provided as zip files), SuperMiro would be able to look inside the package and - just like the audio/video file - the user would have the option of downloading a lower bandwidth version of the materials first, before downloading the whole content package. (That is to say, the content package could be downloaded in pieces, to be reassembled by SuperMiro on the user side.)

Finally, the system would not just work online, but could also be "primed" with content packages downloaded elsewhere. A Zambian school server, for example, would not need to be on the internet. Teachers would be able to request content packages from the national Zambian school server that would be put on a DVD/memory stick/hard drive and mailed to them. Those content packages would be installed on the school server and be available to teachers "as if" the server were on the internet.

3.3.2 Uploading content

Diagram 2. The uplink

What happens when the user in the Zambian school becomes a remixer and wants to share content that they have created or adapted? They would normally just put it onto their school server, perhaps for other teachers to use. In the scenario outlined above, there would now be a channel back from the school, to the national server and the global mirror. This means that other users, whether elsewhere in Zambia, Africa or the rest of the world, would now have access to the content.

In this way the system is bi-directional: content produced by the Zambian school user is uploaded to their school server, but automatically mirrored to the national Zambian server and perhaps to the server near the African internet exchange point. When somebody from the North wants a learning resource from the Zambian school, they need not put any strain on the school network; the content comes instead from a server near the African internet exchange point.

3.3.3 Offline content

Diagram 3. No connectivity

In the above situation there is at least a little connectivity. Let us now assume there is no connectivity and/or national server. (The server is greyed out in Diagram 3.)

In this scenario, the content is taken off the global server in one location where there is connectivity. This might be in the nearest city, or while a teacher is abroad on an OER course. A content bundle could be transferred to a memory stick or perhaps fetched automatically using the hypothetical "SuperMiro" application. The teacher then returns to their school, where the content can be put onto the local server.

One could imagine a scenario where this happens seamlessly using SuperMiro. The teacher clicks a few buttons while they have connectivity and SuperMiro downloads the required content packages. Once back at the school, SuperMiro talks to the school server and automatically transfers the content. The content could even have been requested by other teachers and preselected with SuperMiro before the teacher left. As before, to the user this might be "transparent": they think that they are accessing (e.g. MIT OCW) content through their browser, just like any other web content, but for some reason the OCW content works, unlike other web content.

To complete the scenario, the teacher could also pick up local content from the school server, to share with the global server once the teacher is in a place with good connectivity.

3.3.4 Peer-to-peer sharing of offline content

Diagram 4. Peer-to-peer content distribution

Centralised infrastructure may be missing in many locations, but it is also possible to envisage full peer-to-peer sharing of content, without central infrastructure at the local school or nationally.

Even over very low bandwidth, a content directory could be shared in SuperMiro, which would list all globally available OER content. Each resource would have peer-to-peer sharing information next to it, to inform the user as to whether it:

  • had been downloaded already and was available on his or her SuperMiro;
  • was available on the local network and could be downloaded (estimated time, etc.); or
  • was not currently available on his or her network.

3.3.5 Mobile access

Finally, a more organised presentation of OER content would make it possible to provide better mobile access. For instance, there could be a SuperMiro application for mobile phones, including for the most basic phones supporting Java. There may also be a need to provide access for even more basic phones, via voice and/or SMS. This may be particularly relevant for OER that has been produced as bespoke resources for a certain community. (For more information see the wiki talk page for the OER exchange proposal.)

3.3.6 The proposal

How could such an infrastructure be reliased?

Initially, a small consortium of key stakeholders (such as content providers, NRENs and content users) would need to come together and carry out some action research to find out whether the system outlined above would be usable and acceptable by schools and other educational institutions in developing countries. Following this, the consortium would need to develop guidelines for content providers (e.g. to explain how to make their resources automatically downloadable by SuperMiro), as well as to raise awareness about the system and obtain support from widely used content respositories (including Wikipedia, WikiEducator, Wikiversity, Connexions, installations of EduCommons, Le Mill, Kewl/Chisimba, OER Commons, and many more).

A strong feature of this proposal is that it would place very little strain on the actual OER user. All they would need to do is to download SuperMiro. The rest would be taken care of by the application and content providers.

It should also be noted that this proposal has not been plucked out of thin air. There is related work being carried out at the moment, including by the eGranary and the Global Grid for Learning, as well as various ideas in the OpenCourseWare Consortium and OpenCast community. Nevertheless, there is still a significant need to bring these efforts together in a way that is usable for developing countries.

3.3.7 Certification

Such a proposal, as well as the general clarification of access-related technical issues for OER, might be helped by promoting OER "accessibility" standards, through which content could be marked as accessible. There are already criteria in some areas, such as licensing and disability access, from which a comprehensive accessibility rating for OER could be derived.

In the classification of access issues in Week One, the community primarily discussed issues qualitatively, rather than looking at how different elements of accessibility might be measured or assessed in a formal way. A nice application of our OER classification was provided by Stephen Downes, who used it to look at various aspects of the CCK08 open course. Another example of the use of a classification to evaluate a project can be found in a blog post by Jared Stein . The post develops criteria for estimating reuse and remix value, and applies this to a number of existing OER projects, including the Open University's OpenLearn, Carnegie Mellon Open Learning Initiative, MIT OCW, webcast.berkeley and Connexions. The following criteria were used to evaluate resources:

  • technical openness of media (e.g. Java applet vs. Javascript);
  • quality of source;
  • variety of media sources;
  • semantic/standard structure (e.g. HTML tables vs. semantically-correct XHTML, IMS);
  • Creative Commons license compatibility;
  • hosted tools and support for remix.

The ratings process used in that example was to some extent subjective, but the criteria could be used to devise a more formal measure of accessibility that covers more than just legal openness. Bandwidth is another area that would lend itself well to objective accessibility criteria.

Finally, certification could be used in a top-level domain structure (e.g. All content within this domain would have to meet the criteria that had been established (such as Creative Commons licensing, a minimum amount of metadata, resource discovery, etc).

4 Further links

This paper picked up some ideas from the discussion:

To some extent, the

On this wiki: