About

MIRA: Mapping of Open Resource Initiatives

The MIRA project is the result of a Hewlett Foundation grant aimed at the creation of a prototype for a global map for Open Educational Resources (OER). This call was prompted by a virtual discussion held at the end of 2012 to help define the necessity and model for such an undertaking. In this initial phase, three groups were selected to create a functional prototype between the months of February and April 2014. MIRA is one of these projects. MIRA had as its objective not only software development, but also content gathering. We aimed to identify and map OER-related initiatives in an area that is not well known, because of the languages (Portuguese and Spanish), geographical region (Latin America), and the level of education chosen for this survey (K-12, or basic education). For this project, because of the time constraints, we focused on 24 countries: Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Suriname, Uruguay, Venezuela.

Based on this list, we developed three parallel lines of action: identifying and filtering trustworthy data; second, the definition of a structure for metadata/vocabulary to be collected; and third, constructing a functional prototype of the mapping system. Below, we delineate the three steps.

Data gathering

We compiled a listo of contacts, people with knowledge of open access, open educational resources and basic education in Latin America from whom we could ask help in identifying relevant initiatives in each of the countries. We ask each person, via email, to fill out an online form specifying the (5) most important OER-related initiatives in their country and in other Latin American countries, and a suggestion of another person we could get in touch with. From our list of 70 contacts, 23 replied, from nine countries (Venezuela, Peru, Chile, Colombia, Brazil, Argentina, México, Ecuador, Uruguay). In parallel, we identified initiatives through web searches, repository listings and published documents, including:

      1. OER Brazil (http://rea.net.br/site/rea-no-brasil-e-no-mundo/)
      2. Educação Aberta Working Group (http://educacaoaberta.org/wiki/index.php?title=Lista)
      3. Commonwealth of Learning (http://www.col.org/SiteCollectionDocuments/Forms/AllItems.aspx?RootFolder=%2fSiteCollectionDocuments%2fcountry-information&FolderCTID=&View=%7b54D20EF6-7EA0-4B00-9087-C58D284E6E2F%7d)
      4. WSIS (http://www.wsis-community.org/pg/directory/view/672996?offset=0)
      5. Governmental and Ministry of Education sites in each country
      6. TEMOA (http://www.temoa.info/)
      7. Dictalia (http://didactalia.net)
      8. RELPE (http://www.relpe.org/)
      9. InfoDev Report – Survey of ICT and Education in the Caribbean (http://www.infodev.org/infodev-files/resource/InfodevDocuments_441.pdf)
      10. CKLN Report – Use and development of OER (http://www.ckln.org/home/sites/default/files/OER%20Workshop_Resources_Links_0.pdf)
      11. Instituto Salamanca – Recursos educativos digitales abiertos (http://institutosalamanca.com/recursos-educativos-digitales-abiertos-libro-gratis/)
      12. Recursos educacionais abertos no Brasil: O estado da arte (http://www.cetic.br/publicacoes/2012/rea-andreia-inamorato.pdf)
      13. Uma análise dos termos de uso de repositórios de recursos educacionais digitais no Brasil (http://editorarevistas.mackenzie.br/index.php/TDig/article/view/5892)
      14. UNESCO Iniciativa de Acesso Aberto – (http://www.unesco.org/new/en/communication-and-information/portals-and-platforms/goap/access-by-region/latin-america-and-the-caribbean)
      15. BID, portal institucional (http://www.iadb.org/en/indes/open-education-resources,7016.html)
      16. BID, blog/comunidade (http://blogs.iadb.org/education)
      17. OEI, (http://www.educoas.org/default2.aspx)
      18. OEA (http://www.educoas.org/default2.aspx)
      19. Comunidad Educativa de Centroamérica y República Dominicana (http://www.ceducar.info/).

A first collection resulted in 80 initiatives. We filtered those initiatives not aligned with the Project scope, leaving behind projects focused on higher education, thesis and dissertation repositories, those which were merely informative, among others. Our exclusionary criteria was based the defition of OER expoused by UNESCO/COL (2011)

OER are teaching, learning and research materials in any medium that reside in the public domain and have been released under an open licence that permits access, use, repurposing, reuse and redistribution by others with no or limited restrictions (Atkins, Brown & Hammond, 2007).  The use of open technical standards improves access and reuse potential. OER can include full courses/programmes, course materials, modules, student guides, teaching notes, textbooks, research articles, videos, assessment tools and instruments, interactive materials such as simulations and role plays, databases, software, apps (including mobile apps) and any other educationally useful materials. The term ‘OER’ is not synonymous with online learning, eLearning or mobile learning. Many OER — while shareable in a digital format — are also printable.

In light of earlier studies conducted in Brazil (Amiel & Santos, 2013; Rossini, 2010), we knew that even the best repositories and initiatives present difficulty in showing, with clarity, which licenses are being used, and aligning the licenses presented for the site with those of the resources. We ignored the use of open formats and standard protocols as criteria, since this is an area of reduced priority and awareness, not only in Latin America, but also around the world. Because of the considerations, we had a porous filter: we aimed to include rather than exlcude. We added to our list any site that had an open license or contained resources with open licences, and had a connection to K-12 education. An important exception: we decided to keep initiatives which were clearly publicly-funded, even if they had no license or a clear copyright, so that we could open up a discussion on publicly funded initiatives and the opening of resources.

An analysis of the projects demonstrated heterogeneous implementations, starting the software system (static site in HTML, CMS, or structured repositories), the use of licenses (disparity between site licenses and resource licenses, lack of clarity in terms of use, in general, among others), as well as differences in how data was categorizes/metadata used to organize the resources. We found few initiatives making use of structured repositories (such as DSpace). The project currently holds approximately 50-60 initiatives with detailed data, and new data are and will be added regularly.

Categorization

We analyzed five existing OER projects, which were part of the virtual discussion in 2012 and the Hewlett call, in order to understand the systems and functionalities in these initiatives, but with a particular interest in their categorization schemes and which type of data they collect. The results of this detailed analysis is available openly. The analysis of these projects helped us identify varied methods for initiative categorization. Beginning with these data, we created categories and a vocabulary in order to start a collaborative project to define the metadata scheme for our project. We acted in partnership with the eMundus (http://wikieducator.org/Emundus) and POERUP (http://poerup.org) projects, which are also aiming to map their initiatives. The final scheme is made up of 11 core, and 14 extended categories and a controlled vocabulary. We aligned each of the field with Dublin Core in order to guarantee further interoperability. We used standards such as ISCED in order to categorize schooling levels, as well as IMS LODE categories in thinking about the expansion of the project.

Mapping system

We built a detailed analysis of existing software systems and frameworks which we considered for this initiative, and already thinking about the second phase of the project. We based this detailed analysis based on customization of the JISC/RSP project for the analysis of repository software. The complete table and a critical analysis of these systems is available openly (only Portuguese, translation to English is ongoing) and can help others in defining platforms for mapping initiatives based on dozens of criteria.

Considering the development timeframe and the desired functionalities we decided to make use of JEO, a WordPress-based template, which is openly licensed, in order to create a prototype of our system. Among the existing functionalities, we highlight:

A multilingual interface (Portuguese, Spanish, English);
Detailed text searches and cross-category selections on all of the metadata vocabulary and categories;
Customizable pins (unlimited) based on initiative characteristics;
User contribution and moderated comments, as well as a demo of the possibility of “agregatting” points by the user (OER collections);
Use of Open Street Maps;
Federation of data and extraction in standard formats (JSON);
Ready to use in mobiles;
Demontration of how multimedia content can be integrated to each iniciative (see Banco Internacional/Brazil example);
Advanced editing interface for administrators to customize every aspect of the system and its data, with native multilingual support.

We also used one example (Banco Internacional de Objetos Educacionais) to demonstrate functionalities which we aim to deploy in phase 2, listing detailed information on the resources themselves. These data could be extracted automatically through APIs, OAI-PMH, RSS or other connection/extraction format or protocol. For now it is a static page demonstrating this potential.

From the start, we decided the mapping system should be open and distributed, so any initiative could consume the data produced by MIRA or replicate the software in their installations. All instances of the software should be able to share and synchronize their contents with any other installation. This configuration creates a reliable network of information repositories, without a central node or single point of failure. All mapping information is stored in a CouchDb database, which provides out-of-the-box the necessary capabilities, enabling peer-to-peer synchronization and native support of JSON formats.

Diagrama do sistema de software

In this model, we currently focus on WordPress as the content management system. It holds the multilingual content in the prescribed metadata scheme. The structure of the database and the data model can be expanded easily to include further metadata gather about the resources themselves. In the future, moving away from traditional CMSs will allow use greater flexibility in making use of linked data model, through “centralizing” the content management function on CouchDB databse. We don’t present a visual wire frame of the site as we consider the current implementation to satisfy the current requirements for users (and the RFP), but for further development (as seen in the demonstration example of Banco Internacional/Brazil). We envision making changes to the design and new prototyping based on fruther user-testing/considerations and as prototypes of the resource-based collection and mapping expantion take place.

The software (customized JEO) is available for download at (https://github.com/fksr86/rea/).
A download of the MIRA data in JSON format can temporarily be view here (http://www.mapasdigitais.org/rea/?geojson=1).

Second phase of the project

The prototype and an initial analysis of the data point to a higher than expected number of initiative that have open resources for basic education in Latin America. We also identified the many possibilities for improvement in these initiatives. The data collected during this first phase serve as a diagnostics, giving us an opportunity to further investigate this scenario, and as pathway for further development. For phase two, we aim to:

      1. Expand the scope of the initiatives in Latin America, as well as a global level;
      2. Improve on the existing layout through user testing;
      3. Expand on the decentralization scheme, facilitating the exchange and synchronization of data amongst simple and complex repositories, investigating the use of possibilities including ResourceSync and SWORD.
      4. Make it easier to include data and simplify the moderating procedures;
      5. Develop an application and software interface that will allow us to go beyond initiatives (mezzo level) to the resources themselves (micro), as long as these initiatives have some sort of data output mechanism (API, XML, JSON, etc.)
      6. Improve on the use of linked data, using URIs (schema.org, FOAF), and others;
      7. Disconnect the system from its WordPress-dependency, used here for the prototype. We will aim to develop a similar interface, but without the dependency on an SQL database, creating a lighter CMS-independent system (Worpdress/Joomla/Drupal).
      8. Find ways to dynamically organize and present content and allow interaction with users, including information about OER, and other possibilities which might extend know-how in regards to OER.

The project therefore, achieves the aims which were listed in the call (section III)

      1. Develop a funcional prototype, using open source software;
      2. A scheme that allows for further developments;
      3. The possibility of data extraction that, in the future, might lead to visualizing the data in other ways and formats (using, for example, D3.js);
      4. Produce new information about OER (beyond existing content), with a focus on Latin America, Portuguese and Spanish;
      5. Determined categories and a vocabulary which were appropriate for current and future data, using a CC-BY license for the content;
      6. Evaluated existing databases and publications for contente;
      7. Create a report of our activities and detailed future development.

Coordination
Tel Amiel (coordinator) – University of Campinas – Brazil
Everton Alvarenga, Open Knowledge – Brazil
Priscila Gonsales, Instituto Educadigital – Brazil
Xavier Ochoa, Escuela Superior Politécnica del Litoral – Ecuador

Team
Andre Deak | Felipe Lavignati – Customization of the JEO system
Carlos Villavicencio – Implementation of the decentralization prototype
Gabriel Fedel – Analysis of the mapping systems
Tiago Soares – Survey and data systematization

We thank those who collaborated with the project:

Argentina
Dominique Babini (Universidad de Buenos Aires)
Paola Bongiovani (Universidad Nacional de Rosario)

Brazil
Andreia Inamorato dos Santos (Researcher and Educational Consultant)
Carolina Rossini (Public Knowledge)
Débora Sebriam (Educadigital/REA-br)
Hélio Kuramoto (IBICT)
Itana Gimenes (UFMT)
Rosa Maria Vicari (UFRGS)

Chile
Luis Alberto Alvarez González (UACh)
María Francisca Gatica Cádiz (Mineduc)
Werner Westermann Juarez (Educalibre)
Walter Antonio Santander Wanhoff (Universidad Catolica Norte)

Colombia
Andrés Chiappe (Universidad de la Sabana)
Blessed Ballesteros (Universidad del Norte)
Juan Carlos Bernal (Ministerio de Educación Nacional)
Néstor Darío Duque Méndez (Universidad Nacional de Colombia Sede Manizales)

Ecuador
Andrés Ycaza Mantilla (IEPI)
Jorge Maldonado Mahauad (Proyecto MED)

Mexico
Marcela Morales (OCWC)

Peru
Fernando Ardito (Universidad Peruana Cayetano Heredia)
Libio Huaroto (UNMSM)
Mariella Cantoni (Universidad Inca Garcilaso de la Vega)

Uruguay
Graciela Rabajoli (Plan Ceibal)
Luis Alonzo Fulchi (Creative Commons)

Venezuela
Antonio Silva Sprock (UCV)

Contact – MIRA project (mira@okfn.org.br)

References
AMIEL, T.; SANTOS, K. Uma análise dos termos de uso de repositórios de recursos educacionais digitais no Brasil.
ROSSINI, C. Green-Paper: The state and challenges of OER in Brazil: From readers to writers?
UNESCO/COL. Guidelines for Open Educational Resources (OER) in Higher Education.