OER World Map Project Interims Report

Provided to the William and Flora Hewlett Foundation

Reporting Period: 01.12.2014 – 31.07.2015

1 Introduction

This report is an interim’s report for The William and Flora Hewlett Foundation. It summarizes the overall progress about the project achieved in the reporting period. It focuses on four areas:

• platform development,
• data model & collection strategy
• community development
• legal inquiry and business model development

The report reviews the major achievements, focusing on central design decisions and connected lessons learned. Additionally, an updated overview planning for the upcoming months is presented. Technical details are described within the “Platform progress report” and the “Platform progress report v0.2” blog posts (both written by Felix Ostrowski).

2 Platform development

Within this section achievements regarding the development of the technical platform will be described. Within the reporting period two major milestones have been achieved. On the 11^th of March we published version 0.1 of the system (“landing page”). On the 10^th of July followed the publication of version 0.2 with significantly enlarged functionality.

2.1 Launch of landing page

The launch of the landing page was a consequent implementation of the agile approach in the sense that from the very beginning of the project functionality was delivered, which could be used for interaction with the OER community. The landing page presented a world map and allowed users to register. Registered user were counted and visualized for each country by using different shades of green. Even if the resulting first version of a heat map was not a result of a representative statistic procedure, the result delivered indication, which countries are most active in the field of OER. At the time of writing, 325 people have registered themselves on the page. The countries with the most registrations are:

Nr.	Country	Number
1.	USA	57
2.	Germany	54
3.	UK	37
4.	Italy	32
5.	Spain	28
6.	Canada	19
7.	Russia	17
8.	Australia	15
9.	Brazil	12
10.	South Africa	12
11.	India	10
12.	Portugal	9
13.	Rumania	9
14.	France	8
15.	Greece	8
16.	Belgium	7

Table 1: Numbers of registrations per counry

Additionally a call for “country champions” was launched with the landing page. For more information about the country champions see 3.3.1 and 4.1.

2.2 Launch of version 0.2

Version 0.2 of the platform included significantly more functionality than the landing page.

2.2.1 Stories

Deviating from the original planning, stories of OER activities were included in this version. The stories concept was developed during a workshop at the Hewlett Grantees Meeting and can be seen as an example of agile development, which picked up a suggestion from the OER community. From a functional point of view the stories can be seen as another front end module which allows to describe parts of the OER ecosystem in a narrative way. Until we will provide input templates (planned for version 0.3) the stories will be the primary way to input new data into the system. Using this approach will lead to a gradual and iterative development of the data base.

According to our experiences the story approach offers several advantages:

1. Stories are easily written and understood, and can be used to stimulate interest and participation of the community.
2. Stories provide real world examples of OER activity and can be used to develop the data model. We found especially this aspect very rewarding, since it gave us a possibility to check our data model with reality in a relatively early stage of the project. Later on the stories could be used for training purpose, by demonstrating examples of data modelling with country champions, e.g. in blog posts and webcasts.
3. Stories can be used as a means of assuring the quality of the data, assuming that projects, services and institutions mentioned within a story are more likely to be relevant to the community.

Working with the stories approach also triggered two lessons learned which are important to the strategic direction of the project:

1. Quality over quantity: Instead of “flooding the map with pins”, it seems to make sense to concentrate on high quality data, which provide real value to the community.
2. Activities over Actors: While during phase I of the project our emphasis was on institutions and persons (=actors), we learned from reflecting on the activity focused story approach, that projects and services (=activities) probably will be more important for the community. Services provide solutions to concrete problems (e.g. finding resources), while projects provide lessons learned, which can be shared with other OER projects. Compared to this the value of data about institutions is relatively small. The most important aspect of collecting institutional data is, that it allows to locate projects and services on the map. One important conclusion of this discussion was that the project will concentrate on the collection of services and projects during phase II.

Nevertheless it has to be stated that editing stories requires considerable resources, since not all provided stories are of high (editorial) quality. Editorial operation for stories was not planned within the project budget planning. This has to be kept in account for the continuation of the project. In the long run one solution might be to shift this effort to the country champions.

2.2.2 Enhanced backend development

As a side effect of the stories approach backend functionalities of the platform are quite advanced, since extracting the data included in the stories required that the backend already can handle all major data types (organizations, persons, projects, services).

2.2.3 Country views

One major deliverable of version 0.2 were the country views.

Figure 1: Screenshot of Country View as implemented in Version 0.2

The country views combine the geographic map with basic statistical information of the regional OER activity and will provide a fast overview of the status of the OER movement within individual countries. They also provide information about the country champions in a very prominent way.

3 Data modelling & collection strategy

One of the most crucial goals of the project is to provide answer to the questions, which data has to be collected, how it should be modelled and how it should be collected.

3.1 Which data to collect?

3.1.1 Ill-defined concept

Every library approach to collecting data needs a collections scope or strategy, which determines which data to collect and which to ignore. One of the fundamental lessons learned of the project is that this collection strategy is quite difficult to define precisely. Despite the good OER definitions which have been provided by UNESCO and The Hewlett Foundation the concept remains ill-defined in many regards:

1. Openness: Though it is widely agreed that licenses are central to the question of openness, it is still not agreed, which concrete licenses are open and which are not. For the OER World Map Project we will assume that while “Non Commercial” (NC) licenses will still meet the requirements of real openness, “No Derivatives” (ND) licenses will not. Another challenge will be that the Creative Commons framework is not the only provider of open licenses.
2. Educational: There seems to be no definition for what makes a resource an educational one. One answer to this question could be that all resources which can be used for learning are educational resources. Though this answer makes quite sense on the one hand, it also extends the collection scope into the indefinite. In fact one could argue that in this case the project should not be named “OER World Map”, but “Open Content World Map”. Another answer could be that OER require a certain didactical coinage or intention. Nevertheless this would exclude important content collections. Also it will be difficult to check this didactical intention in practice. Stories of OER use can help to provide information about the use of resources and so provide educational context while a clearer picture emerges.
3. Resources: Compared to the other two questions, the concept of resources seems relatively clear. Nevertheless it is e.g. discussed, if MOOCs are OER themselves, or if they are courses, which include OER, if being open licensed. We also found that it might be unwise to concentrate on resources only, since there seems to be a shift from resources to practices. Therefore the platform will be prepared to incorporate Open Educational Practices and other aspects of Open Education in the future.

In conclusion it has to be stated that this question remains unanswered for the moment, though it should be noted that we have flexibility within the data model to allow for an agile approach. We would argue, naturally, that the OER World Map provides excellent opportunity to refine the existing concept of OER.

3.1.2 Openness indicator

Bearing in mind the need for an iterative approach to understanding openness we decided to extend the scope of collection. Originally we aimed at restricting the collection scope to services which provide resources which are either CC BY, CC BY-SA or CC BY-NC licensed. We now believe it makes sense to include services which provide (arguably) less open content, on the basis that the “gratis” services of today may become the “open” services of tomorrow. Also we learned that there are many services, which do not make use of the Creative Commons licenses toolkit, which will make it even more difficult to decide if a certain service can be collected or not. Instead of restricting data collection we decided to assure appropriate focus on openness by the integration of an “openness indicator” in combination with a standard configuration of the display restricted to open licensed material. By doing so we hope that providers of “gratis” services will be motivated to move into the direction of real openness.

3.2 How to model it?

One of the major deliverables provided within the reporting period is an initial version of a detailed data model. Generally our decision to use schema.org as a basis metadata scheme was validated, there being little need to add fields to our data model, which where not already included in schema.org.

Collecting data always carries the structural risk of including too many fields, which in theory increases the informative value of the dataset, but in practice might prevent its development due to the increased collection and curation effort. To make sure that only necessary data is collected we include only fields which are connected to a concrete user story (see column “user story”).

Through close cooperation with country champions it will be assured that the data model does not get too complex to handle. Nevertheless, we already learned that the distinction of actors (=organizations and persons) and actions (=projects and services), though being logically sound, is time consuming and sometimes intellectually demanding. Since the OER movement is characterized by many self-organizing grass root initiatives, we also learned that sometimes it even can be difficult to distinguish between actors and activities.

Though we still aim very much at intuitive usability, it therefore might be necessary to provide basic training opportunities for country champions and data collection services to increase data quality. The complexity of the data model will also significantly influence the degree up to which non trained community members can participate in the data collection process.

3.3 How to collect the data?

There are generally two ways to collect the data needed for the OER World Map. On the one hand it can be collected by trained experts (e.g. librarians), on the other hand it can be done by volunteers (crowdsourcing). While the former contains top down elements, the latter can be classified as a bottom up approach. Our original strategy, to combine both approaches was overall confirmed but refined in several ways.

Our goal is now to provide a skeleton of high quality data with the help from OER data curation services (see 3.3.1) and country champions (see 3.3.2). We believe that this will be a good basis for crowdsourcing the further enrichment and updating of the data. Since providing the basic data can be both time intensive as well as intellectually demanding (see 3.2), we believe that this approach is more promising than expecting the community to provide complete data sets.

3.3.1 From data curation projects to data curation services

From analyzing “data curation projects”, we learned that many of them (e.g. POERUP, EMUNDUS) did not succeed to provide regularly updated data, while others (e.g. ROER4D) rather aimed at indeep analysis of single projects than at extensive data collection. At the same time it appeared that other kind of data providers (“OER data curation services”) might be able to provide qualified data on a sustainable basis:

• OER conferences & festivals (e.g. OEC-Global, Open Education Week, Open Ed, OER16, OERde16, etc).
• OER Funders (e.g. The William and Flora Hewlett Foundation)
• OER Consortia (e.g. Open Education Consortium)

The OEC membership data already have been prepared and can be imported any time. Additionally, we started to develop a concept to include data from Open Education Week 2016 as well as from OEC-Global 2016.

3.3.2 Country Champions

Our call for country champions (for more information on the country champion network see 4.1) received a quite strong response. One of the major responsibilities of the country champions will be to take care of collecting data as well as assuring its quality. One of the advantages of this development is, that in contrast to the “data curation projects”, the OER World Map country champions will use the World Map platform from the onset, so that it will not be required to develop potentially expensive interfaces for them.

3.3.3 Crowdsourcing

Crowdsourcing data collection was of minor importance within the reporting period, since the therefore needed input templates will not be available before version 0.3 of the project.

4 Community development

Beneath the development of the technical platform community development is the second component, which will be essential for the success of the project. Core activities in this field, which were performed in the reporting period, were the development of a multiplier network, the presentation of the project at major OER conferences and ongoing information by the use of social media.

4.1 Multiplier network

At the heart of our efforts was the development of a multiplier network, which consists of two levels.

On the first level are the country champions, which should ideally not only collect and check data, but also raise awareness of the project, e.g. by the use of social media. With regard to the question, who can become a country champion, we took a pragmatic approach, which allowed us to make even single persons a country champion, especially for smaller countries less active in the field of OER. Nevertheless, we prefer institutional partners or evolving “country champion networks” for bigger or more active countries. Examples of “ideal” country champions we recruited are the OER Special Interest Group in the Netherlands, OER Africa in South Africa, the Project Cyfrowe Project in Poland, the OER Research Hub in the UK and the Opening Up Slovenia Project. The main motive for participating in the project seems to be the wish to get an overview of the evolving OER movement in their own country. We also learned that the planned “gadget exporter”, which will allow to include a filtered map into another website or blog seems to be of high interesting for the country champions. Our goal for phase II is to find for each major OER country a reliable country champion. Until now we found 17 country champions, a complete list can be found in the appendix.

On the second level we are looking for one or two regional points of contact for each continent, which oversee the development on a multinational level and who can help finding regional country champions. Ideally the regional points of contacts will actively use the world map as well. In doing so we hope that experience and commitment will be gained that can be used to support strategic management decisions.

4.2 Strategic Partner Network

Besides the country champions based multiplier network a strategic partnering network will be implemented. In order to do so promising conversations with many leading OER actors and institutions were held at OEC Global in Banff. The strategic partnering network is planned to be launched together with Version 0.3 of the platform.

4.3 Conferences

During the reporting period the project was presented at following conferences:

• Open Education Conference 14 in Washington, USA
• Hewlett Grantees Meeting in Sausalito, USA
• OER15 in Wales, UK
• Open Education Global Conference in Banff, CA
• European Library Automation Group (ELAG) Conference 15, Stockholm, SE

4.4 Social Media

To support communication with the community following social media accounts are being used:

• Blog
• Twitter Account
• Facebook Account

To support the technical development we are using a github account, which allows to see and comments all included issues. Important information was also disseminated on the major OER E-Mail lists.

4.5 Project Budget

One major lessons learned is that especially within the field of community development the project is underfinanced. To provide dynamic and intensive communication with the community one additional full time employee would have been desirable. As a countermeasure a trainee will be hired to support community building within the upcoming weeks. In the future it will also be desirable to provide funding for micro-grants to support country champions and regional points of contact.

5 Legal inquiry & business model development

5.1 Legal inquiry

In order to comply to German and international data privacy laws a specialized lawyer with extensive experience in open approaches was consulted. Together with him we are taking a “privacy by design” approach which assures that the technical development considers legal requirements right from the beginning of the project.

5.2 Business model development

During the Open Education Global conference in Banff an initial version of a business model canvas was developed during a workshop with Paul Stacey. We are planning to refine this version and participate in the Open Business Model project of Creative Commons.

One important goal will be to achieve additional funding for the project. In this context a positive development is that the German Government provided a first statement on its direction concerning OER, which proposes the development of a platform which provides reference to all OER sources in Germany.

6 Updated Milestone Planning

Name	Version 0.1	Version 0.2	Version 0.3	Version 1.0
Date	11.03.2015	10.07.2015	15.09.2015	01.12.2015
Goal	An initial web presence is installed.	World Map Module and general search module are available.	Data input & statistics	Data export & help system
Features	– A landing page is available (done) – Users can register (done) – Necessary prearrangements and initial analysis have been undertaken  (done)	– A high level view on the OER world (heatmap) (done) – A detail view for individual countries (exposing institutions, services, projects and people as pins on a country map) (done) – A search interface (done) – An initial set of real-world data (done) – publication of OER stories	– User registration (second iteration) – Registered users can enter projects, institutions and services – Registered users can create their user profile – External data can be ingested via the API – Statistics on grade level, country and language are available – Calendar – (Timeline) – launch of OER World Map strategic partnership programm	– Data formats for export – Help system / User documentation – A German OER editorial team has been installed

Table 2: Updated Milestonplanning. Blue colour indicates changes from the orignal planning

Picture: “Concord House – Holloway Head, Birmingham – Work in Progress” by Elliott Brown, CC BY-SA 2.0