1. Building anopen sourcebased, open standards, infrastructure for the large scale provisioning ofreusable opencontent . , . ,. , . , .. , . , . email: pstath; [email_address] / 2011, /... National Documentation Center /N.H.R.F.

2. EKT: The National Documentation Centre 3. & . Internet Building anopensource based ,open standards , infrastructure for the large scale provisioning ofreusable open conten t

reusable and open content:

• functionality ?

4. standards ?

5. software ? 6. emerging requirements ?

An open sourcesoftware stack for delivery of

reusable, open content

7. in large scale, using open standards 8. and addressing emerging requirements 9. Content types and characteristics (I)

Different content types but similar functional requirements,examples:

Research output , e.g.papers ,reports, patents, documentation, conferences, books, data:didaktorika.gr ,helios-eie.ekt.gr , etc.

10. Cultural - Humanities repositories : from research papers to commented cultural artificats:pandektis.ekt.gr 11. Educational reference material :studies, books, multimedia courses, project outcomes, e.g.repository.edulll.gr 12. Digitised and/or born digital e-books :ebooks.serrelib.gr( pre-release ) 13. Library automation systems and OPACs: linking the digital with the physical realm,abekt.gr 14. Digital objects in general 15. Content types and characteristics (II)

Common characteristics:

• Online databases providing access todigital to objects(e.g., books, articles, images, mutlimedia) accompanied withrich metadata

16. Collection, dissemination, preservation of material that hasarchivalvalue for future reference

Different requirements and emphasis in features implemented in comparison toWeb Content Management System s 17. Fast evolving landscape. Full range of features required:

copyright information management,metadata-bibliographic data management ,metadata organisation and interoperability mechanisms ,highly automated backoffice digitisation processes support,digital content interoperability and reuse frameworks,persisted identifiers ,end user interfaces and intuitive content interaction ,highly efficient storage and computing resourcesmanagement

18.

Standards ( so far)

Openandreusable :

• Digital content, file formats, encoders, etc.

19. Metadata description:

UNIMARC , MARC21toDCand MODS

Interoperability protocols and interfaces:

• • OAI-PMH ,OAI-ORE , Z39.50, SRU/SRW,Europeana ESE , etc.

• Unique and persistent identifiers

• Independent from repository software,

20. HANDLE.NET (RFC3652), openURL

21. An open content open source software stack No easy task to tackle the full rangeof theseissues, in a scalableand viable manner

Custom built - problem specific solutions pay ahigh toll for standards compliance, locked in a specific solution/architecture

Solution :best of breed open sourcesystems for the assembly of a fullyopen sourceopen content delivery software stack

Not trivial task -or just an integration/deployment issue

22. High quality FLOSStools are available: pluggable and modular stack, do one thing but do it at world class FLOSS/Open Standards is the enabler forscalable, viable, large scale open content infrastructures . Just like the Internet no long term alternative to open standards/FLOSS seems to exist! 23. viewers and enhanced user experiences 24. novel reading devices 25. mashups and linked data 26. The open software stack components best of breed: open source, setting the standards pace, world class features, scalable and robust communities 27. Case studies

Examples of what can open source do for you

28. 3 case studies:

The Education and Life Long Learning Repository

Persistent Ids, rich user experience/custom tag bilingual tag clouds based on documentation metadata

The e-OPAC alpha version of openABEKT

Connecting the digital with the physical realms

Public Library e-books repository pilot: the Serres Public Library pilot

Backwards interoperability, repositories, enhanced viewers and image servers infrastructure

29. repository.edulll.gr 30.

http://ebooks.serrelib.gr

31. http://hdl.handle.net/10812/1929 32. ABEKT e-OPAC 33. Enhanced image viewer/server infrastructure Data webnode(s) pooldatanode(s) poolShared Storage: 1 Master (RW) + N slaves (RO) Currently: 300K pages, 3M imminentfrom didaktorika.gr, 10 resolutions/page EKT's virtualised infrastructure Shared Storage Backend Processing Frontend Load Balancer Backend Processing Internet Requests 34. Characteristics Caching Load Balancing Error compensation through VCL Backend Health Check Varnish Lightweight & Simple Stable Right for the job NginX On the fly Image Processing No need for several Images withdifferent resolutions Open and Extensible API Djatoka Stateless Infinite Scalability Highly Available End to End FLOSS Layer Security Apache Tomcat/DSpace, NGINX and Varnish cache engine in high availability configuration, PostgreSQL DB, on CentOS VMs 35. (open)ABEKT e-OPAC

ABEKT new e-OPAC

Different need: manage the library (digital and physical) using bibliographic tools and a librarian approach

36. The interoperability layer between the physical and the digital library 37. Richer metadata and information ROR over OSS components 38. Digital repositories software

Dspace 2.0: with Fedora inside

EKT participates at the design team

Dspace + Fedora = Duraspace 39. Advanced content provisioning and preservation functionalities 40. The core element for an interoperable repositories infrastructure 41. Preparing for operation as SaaS and over IaaS Cloud

Delivers as SaaS a complex service

42. Future work and informational links

Current and forthcoming work

Viewer with full text search: OCR XML

43. Back port features to existing repositories 44. Integrate to OJS the page by page functionality 45. openABEKT 46. Automate already existing economies of scale:offer SaaS capabilitiesfor the whole range of tools Links

http://www.duraspace.org/

47. http://openlibrary.org/dev/docs/bookreader 48. http://www.indexdata.com/zebra 49.

Thank you for your attention!